XXHighEnd

I know quite a lot of people on the forum are using Gainclones and high efficiency speakers so I thought I’d mention an experience I had setting up Gainclone amps this week end.

Some time ago when I built my amp I decided to build it with 4 Gainclone amp modules but I never go round to using all of these and have been driving my speakers with only two of the modules. The idea of having 4 modules was that I could configure the amp as parallel paired amps, bridged pairs, or run Bi-Amped into individual speaker drivers.

I finally go round to getting some extra speaker cable this week-end so I could Bi-Wire the speakers, then I set the outputs of the amp up to Bi-Amp the speakers using one Gainclone module per speaker driver (In fact the set up is really Tri-Amped because Avantguards have active bass already). Anyway the results were really very eye opening. Lots of "problems" that I had assumed were PC, OS, XX software or DAC related were just fixed.  :xx:

The experience made me rethink what I was doing to tune my system  :scratching: Before this I was assuming my focus should be in the PC and DAC. It turns out they were doing a pretty fantastic job all along and the Amps were the place to be looking. I hadn’t bothered with the amp side of things because general wisdom is that 1 gain clone per channel is great for high efficiency speakers that present an easy load. Now I’m really glad I spent the small amount of extra time and money needed to put the extra gain clone modules in to the amps. I would certainly recommend thinking about Bi-Amping if you are already using Gainclones.

The only downside is this could lead me down an expensive path that I hadn't intended to take. I always said to myself I would not go down the path of an XX software cross-over and a multi channel Phasure NOS1 when (if ?) Peter decides to develop this. With the Tri- Amping now in place and working so well, suddenly software crossovers look like the next logical step. Better start saving now !  :sad:

Nick.

Nick, hi there !

What I wonder in the mean time, is how you set this up really. To make clear what I mean, I better explain how I do it (not really my "invention" btw) :

So, I have (per channel) one GainClone and it's for the mid-high. In parallel output, that GainClone goes into the second amp which is for the Bass driver. However, in between there ("line"), there's the crossover for the Bass. Here too, the bass part can just as well considered to be "active", although physically the both amps are in one chassis, and are both GainClones. But, I see no difference when one of these would have been "actively" in the speakers cabinet, further fed exactly the same. So, bi-amp allright, and no "tri" as seems to be needed in your setup.

Notice that the first amp acts as a preamp for the second one, and will determine the sound largely. Also, think something like 27dB gain into 27dB gain but with a (crossover and) volume in between it. The latter is part of the crossover just the same (and I assume that you know how this works).
Before my crossover was passive, but today it is an active one (also not my idea, and IIRC under 150 euros for two channels). Seems to work as good, but with many more possibilities. 9 dials on it, but 5 are unused.

So, am I using one GainCone too few, or are you using one too much ? :)
Btw, I have lots of headroom left for the bass, and the mid-high is at -15dB most.

Let me know what you think,
Peter


PS: Which all doesn't prevent me of looking forward to software XOvers.

Peter hi,

If I understand your post I think your speakers may have two driver channels a combined High/Mid and separate Bass (?), whereas I have three driver channels High, Mid and Bass.

In my setup, working forward from the each Gainclone module’s output in my amplifier towards the speaker driver, each speaker channel looks like this;

Horn loaded Tweeter

Gain Clone Module 1 > Bi Wire speaker cable 1 > Tweeter Cross Over > Tweeter Driver.

Bi Polar reflex Bass

Gain Clone Module 1 > Bi Wire speaker cable 1 > “High Level” feed to Active Bass amp within the speaker via its high level input, the speaker’s internal base amp has volume control and roll off pots to trim its Bass integration with the other drivers > Bass Drivers.

So yes as you say this Gain Clone Module 1 does two jobs, first it drives Tweeter and second it acts as a "pre amp" for the active base amp within the speaker. The good thing here is that speakers internal bass amp only present a load on the K Ohm range so Gainclone module 1 only really has to “drive” the tweeter.

Horn loaded Mid Range

Gain Clone Module 2 > Bi Wire Speaker cable 2 > Direct connection to mid-rage horn driver without any cross over.


I build the Gainclone amp with two modules per channel as left / right mono blocks electrically going back as far as to the amp’s power cable but everything is in one case. There are two Gainclones modules per channel fed by a separate 300Va R-Core transformer and rectifier. (As an upgrade I guess I might gain an improvement by adding a couple more transformers and rectifiers to get mono blocks on all 4 Gainclones modules back to the power cord ?)

As you say my setup is not quite true Tri-amping because the speakers active bass amp is fed by Gainclone module 1 (above). If I were to go down the active software cross-over route I would tap in to the existing active base amp within the speaker cabinets by passing the internal bass volume control and cross over to drive the amp it at line level from a DAC output.

Ah, ok. I never understood you had three drivers in there. Sorry.

But now I think of it ... what I also have, is the output of the second amp (the bass amp) (besides from going to the bass driver), going to the high (LS) input of the sub woofer - which is active. This itself having a(n active) crossover again.

In the end it all turns out to be pretty much the same I think. So, I don't have separate amps for mid and high, but at the low(er) end I have.
GainClone1 -> LineFilter -> Gainclone2 -> SubWFilter->SubWAmp.

Peter,

This is interesting, I would be tempted to bay pass the SubWFilter, but you have me wondering what the benefit is of the stacked Gainclones ?


Gainclone 2 only sees the bass spectrum passed by the Line Corssover filter which prevents it having to process higher frequencies ?

Gainclone 2 maybe has a PSU feeding it that is optimised for good low frequency response (eg more capacitance or something) ?

The DAC driving the input sees ~20k Ohm load rather than two gainclones in parallel giving say a ~10k Ohm load (but the output impedance of the NOS1 is easily low enought to drive two or more Gainclones).

Some quality of the high gain from having three amp stages ? but I am not sure this in itself would be doing anything. Perhaps you need this to keep up with the 115db response of your Mid High driver ?
Any insights ?

Regards,

Nick.

Hmm. GainClone,

Did some reading, looks nice......

Because i have almost all parts here, 4x psu amplimo 35v, 4x power-delay circuits.
So the only things i need, are some cooling parts and some decent caps, opamps and some other cheap parts.
ok...... I'am in.....

I have to read some more, but if you have some proper links, these are always welcome.

PS: if you want to try using software x-overs, do you need for eg. a second juli@ ? or......?
PPS: if you dont use software x-overs but just a passive x-over and use bi-amping, how would you connect the interlink to the amp
as 1 cable input and internally y wired to the inputs of the 2xamps per ch. ?


Roy

Hi Roy,

My first question would be how efficient are your speakers ? People do use Gainlones with moderately efficient speakers but they give about 35 ish watts which only goes so far I guess. I think Peter St and I have both found you can get a long way using them with efficient speakers and easy load characteristics.

For my amp I bought a couple of Peter Danial’s premium kits giving me 4 Gainclone modules. I replaced Peter D’s supplied main caps with some BlackGate Std 1000uf caps (no BG N Types left anywhere  :sad:). You need + and – 25 ish volt rails to power them, in my amp this is created using couple of 300Va R-Core transformers (people also recommend good quality torroids). Peter D’s kit has all the rectifiers and PCBs so it very easy to assemble. I am very impressed with the kit, its very nicely designed and gives an extremely pleasing sound when built.

http://www.audiosector.com/lm3875.shtml

I have direct and volume controlled inputs. The volume control is via TVC’s wired before the Gain Clones. I agree with Peter St that direct input from the DAC is better using digital volume control. I spent about £1500 (inc £500 for the TVCs) which is way, way over the top because I went for high spec components, but the amp is still incredible value for money. I retired 5Ks worth of parallel 300B mono block amps when I built the new amp and have not looked back once.

Thinking about it using an active cross between the DAC and Gainclones would be interesting but the experience of using the NOS1 DAC direct connected with digital volume makes me feel that if Peter St cracks the software cross over using the NOS1 that would be the way to go.

I think Peter St would have to provide help with the Juli@ config questions, I am not sure what he would have in mind there.

Nick

Roy,

I missed this one...


I think you are describing what I am doing. The DAC feeds via a pair of normal (left and right channel) interconnects from the DAC or preamp into the Amp. Inside the amp (for each channel) the inputs either go via the TVC to the inputs of Gainclone 1 and 2 when I want a volume control or in the case of the direct inputs the input RCAs are connected directly to the inputs of Gainclone 1 and 2.

For the Gainclones I have this gives and input impedance (of both Gainclones 1 and 2) of about 12k ohms so you need a preamp / DAC that can drive this input. Its fine for most solid state output stages but I added the TVC so that I could drive the amps inputs with a Valve stage if I wanted to.

Regards,

Nick.

Correct. But this is not only about prevention from high frequencies or something, it's a cross over. So, in my case the output of GC2 would be lowpassed at 270Hz, but since the SW can output 120Hz while the normal Woofer goes to 27Hz, the SW is lowpassed at 40Hz - and so forms an "xover" with the normal speaker.
Or I didn't understand your questions perhaps ...


Exactly. So GC1 and GC12 are not the same.


Yes, that too. Since the SW is fed with high voltage input (LS output of GC2) - but the SW is made for that - there' more current to flow (and less gain to add by the SW). It just sounds more robust on the SW side.


No ... but for the woofer it does. So, all 'n all (and net) it would come down to the woofer having a sensivity of 115dB. Maybe not completely honest, but I think this is what it comes down to. Anyway, where the mid-high would really have that sensitivity, the woofer would calculate the same, but of course it is really GC2 doing that. But don't try to leave out GC1 or GC2 because one tiny GC really would'nt do it for such a 15" driver.

Peter

It will need another "sound card".

Peter,

Thanks for the explanations, interesting stuff. I had not read about anyone using GainClones in this way. Thinking about it  why not ?

All of this has made me start to think about the possible benefits of different cross-over approaches. No plans to act too quickly but defiantly a topic to think about in the background  :)

Regards,

Nick.

Hi Peter,

I am thinking about bi amping my 2 driver speakers with gainclone amps. So with 2 way speakers I will have 4 gainclone amps one feeding each driver. To do this I will need to connect the output of the NOS1 to two gainclone inputs i.e. parallel. Is NOS1 OK to drive this? Is there a minimum impedance that NOS1 should see?

All the best

PC

Hi Paul,

That will be no problem. Minimum impedance should be around 350 Ohm.

But I wouldn't do it like that, because you'd be having two different reactances to the one source that way. So, maybe it's better to think "serial". Well, see this topic and my part in it ... :)

Regards,
Peter

Hi paul,

Peter's suggestion of serial configuration with input gainclone driving two output gainclone is an interesting one. I have not tried it myself but I can say that the NOS1 driving two gainclone in parallel (per channel) works very well. I have used the NOS1 connected directly (volume control by XX) and via my TVCs with no problems.

I'm guessing you'll get a good result with your AN-Es. Will you build the gainclone if you go ahead ?

Nick.

Hey Nick,

How are you doing?. Yep I have most of the bits and I am waiting for a few more next week before wielding the soldering iron. So the first thing will be to hear the GainClone (mono block config.) versus my home made SET mono blocks. The thought that el cheapo Gainclone's could sound better than the £5k+ worth of SET mono's is intriguing. First things first though - how will the Gainclones sound versus the SET's?   If that works out OK (but I am listening to the NOS1 - SET's now and it is difficult to imagine how anything could sound better) I will start to think about Gclone bi amping. Fist I will try 2 Gclones in parallel on the output of NOS1 but I have to say Peter has a point - NOS1 driving 2 amps in parallel is not ideal. But there again on the other hand a "serial Gclone setup" needs some thinking about. So far it seems that I would be looking at (assuming I can dig out the LF section of the AN crossovers): -

Gclone1 >> HF xover >> HF drive unit

Gclone1 >> Vol Pot >> LF xover >> Gclone 2 >> LF driver unit

These are just very initial thoughts I have to think more about - not sure it will work yet - but somehow I have to "lose" gain with one Gclone in serial with another. The only other way possibly is to reduce the gain of Gclone 2 to 1:1 but that seems a non starter as the LM3875 needs at least 10:1 closed loop gain to be stable. But as I say only just started thinking about it - any ideas would be welcome though.

All the best

Paul

Hi again Paul,

I don't want to intervene with Nick's ideas and responses, but I have a few general remarks which may be of importance for you;

First is that the position of your volpot may have been derived from my own (shown) setup earlier in the topic. In my situation indeed it's at the bass side. But what may have slipped your mind is that in my situation the first GC acts as a preamp for the second one, or in oter words, the gain from the first goes right into the second. Quite obvious that this urges for the volpot at the the second GC, right ?
But I think not so in your parallel setup ...
I even think that you will find yourself in tuning that pot to 0 attenuation in this setup, because both really should be applying the same output to the individual drivers. This means you should leave out the pot, as far as I can reason out (and if not, you'd be tweaking in the wrong direction ?).

You must also realise that it is quite easy for me to explain how the connections are, but which is different from applying a well designed crossover. But let's say you have that in the speaker and can make use of it.
Now draw the cross over slopes (make up something, because it needn't to be right), and now imagine that you'd indeed be having (or *needing* like in my setup) a volume for the bass section. This is needed for two reasons (or better, is applied to two phenomena) :

1. Getting the net bass output equal to the mid/high output;
2. Maintaining the crossover slope as designed.

Ad 2.
This is why you have to draw this line. All you need to do further is put your imagination onto the left part of this line going up and down, at you changing the volume ...

See ? together with changing the volume you would be adjusting your cross over. One good say "not good", but one could also say "hey, right !". Why the latter ? well, because it is your actual means to tune in that theoretical cross over. And, you can bet that when you have it right, the volume in between bass and mid/high also is right. As you can imagine that this won't work the other way around : when you hear the volume is right, the cross over will be right as well by accident.

Of course, the latter should be true, but it requires objective listening instead of "liking a fair amount of bass". And actually it requires your "capability" of hearing that the cross over is tuned in right. But, always START with a theoretically well designed crossover (which takes two electrical sets of stuff of course), or otherwise you'll end up in tears.

In the very end I'm quite sure this all can't go without measuring, so think about this too, before it's too late.
Preserve your old set of coils and resistors and caps and make photos of how they were exactly mounted. Next build new ones as copies from the old. because you know, once you indeed start measuring, you will also start changing them. If now next you can't get it together within on year, you *will* want to go back. This may also happening after months of being satisfied, but in the end not. It is especially this situation that you stopped paying attention to the original situation. Always preserve that ...

If your current speakers are made for bi-amping, it would be my advice to just leave it like that, and bi-amp - period. No volume nothing (preamp allowed of course).
If your speakers are not made for that, well, it could be a nice hobby for indeed a year or so, but my advice would be really to not do anything. Ah, too late eh ? Well, yeah, what ...

One last thing :
If you don't have the crossovers right, this may not show right away at all. I mean, technically things may sound right, but the emotion may be lacking. So, check for emotion. The most tough and indirect thing, because you *will* be focusing on the technical thing first. It goes automatically. And of course, if that ain't right, how to proceed. But when you're at last done with the technical stuff (all sounds balanced etc.), how to tweak for emotion without the experience ?

Peter

Hi Peter,

Thanks for your comments. Well at least I did say I had not thought about it yet :)


The speaker xovers are split and can be bi-amped. So other than the usual bi amp set up (NOS1 into 2 Gclones) the only other way (?) would be NOS1 into Gclone 1 (acting as pre amp) then into Gclone 2 and Gclone 3 one for each driver. But is that solving the problem? won't Gclone 1 see different reactance's of 2 gclones just as NOS1 will in the traditional set up?. Also more components etc in the signal path.

PC

Indeed, NOT the only other way. :)

GC1 outputs in parallel to the mid/high drivers. So, to that and to GC2. Don't forget to sneak in your crossover between GC1 and GC2, so it frees that from dealing with high frequencies (would be a waste of power).
Otoh, your crossover (bass side) won't be made for that, so better stuff it behind GC2's output afterall.

Where's GC3 now eh ? haha

Also, the reactance from GC2 to GC1 is line level against high level output, which is quite different from line level against line level when two GC's would be pondering onto a poor DAC output.
You could say, though, that the combination of high output against line input is the other way around from "good", and although I never gave it some thought really, I think it's all the current which will push it through anyway (I must be very careful what I'm saying here, because it really hasn't been thought over - not by me).

Peter

Hey Peter,

I am sure that I do not understand this properly (for which I apologise and hope you will bear with me)

My speakers actually only have 2 drive units so as I understand what you are saying this means:

GC1 >> HFxover >> HF driver

GC1 >> GC2 >> LF xover >> LF driver

I think I must misunderstand as the gain of GC1 into GC2 would be far too high.

PC

Hey Paul - that depends.

Originally you are right. But what if your mid/high is really high-sensitive, while your bass obviously is less ?
Anyway, that is what the pot is for ...

In the mean time I can blow windows with this, and all the amps are, are 27 Watts or so ... (not at maximum gain)

Just think carefully about yor options. Step away from my setup for a while. Think about Nick's means too.


PS: Also important (at least for high sensitive speakers) : these things can be 100% dead quiet at full gain (which latter just is the case without preamp).

Thanks Peter plenty to ponder. I will try a number of options and post the results.

All the best

PC

Hi Nick,

As a result of reading this thread I bought and have now finished my first Gainclone amp (not sure how I missed all the stuff on the net about GC's!!). It was fired up for the first time last weekend and I am listening to it now (Peter Daniels "Premium Kit" in dual mono with 2 x 300VA transformers). I am really very surprised by how good this amplifier sounds. A very precise sounding presentation with excellent imaging (oodles of sound stage depth on good material) but still with a "liquid" mid range. Even with some of the worst recorded hard sounding digital CD's it still sounds good. A neat trick with such a high level of detail. This little amp seems to work really well with my Audio Note AN/E's and there is loads more volume on tap. The only negative is that it is not quite as relaxed as my old 300B's.

I guess the real surprise is that I am not tempted to return to my mono block 300B SET amps which I did not think I would ever be parted from. Maybe I will give it a bit longer until a final decision but so far the GC's have it. So the next mission is to build another GC and bi-amp. Could be interesting.

Paul

Hi Paul, Thanks for sharing. I suppose you bought them from http://www.audiosector.com What about a nice picture showing your new amps?

Paul hi,

It’s great to hear how things have gone so far. I know what you mean about not noticing gainclones they just passed me by whilst I was spending loads of time working on my AN P4s. They are great fun aren’t they and the costs relative to manufactured alternatives just keep me  :).

A quick tip, in the Peter D’s kit there is an input resistor 200 Ohm I think. It is fine to bypass this (if you have not already) and it’s very well worth doing. Just put a link across then (a 5 min job) and take a listen.
I completely agree with your view of sound, I think of them as super “accurate” sounding  valves in the mids but with low end extension/control and very sweet highs. They don’t sugar coat the sound though, so if the source is not absolutely spot-on you seem to hear everything, both good and bad.

You have me smiling about your much loved 300Bs, I haven’t powered up my much loved P4’s once since I got the gainclones going. It’s not to say that the P4s aren’t really great amps but I just love the gainclones.

When you get time the bi-amping is really worth going for. I was very surprised at the difference it made. In my setup I have two clones per channel running from a 300va R-core per channel, works great and dead easy to setup this way all in one case.

Regarding them not being as relaxed as your ANs this may turn out to be an asset in the end. With the changes I have been trying on the NOS1 they REALLY shine and I have a sound now that is supper detailed whilst being relaxed and not fatiguing. I would describe the gainclones as very “honest”.

The last change to the NOS only happened today but it’s real hum-dinger, and dead easy to apply (it takes about 15 mins). I am waiting for Peter to come back to me but I am quite confident it’s based on sound theory and it is simple to understand what it is doing. It really suits the transparency of gainclones, I will keep you posted.

Kind regards,

Nick.

Per - Yes audiosector is the site. Here are a couple of pictures of the GC so far. Its not quite complete I just need to tidy and final run of the I/O wiring before bolting down the lid.

Nick - I did not install the input resistor just a bypass link after reading comments on the diyaudio site. Gain is high (separate issue to input R) I have just been listening (at quite a reasonable volume) at -30db but with the 300B's I was never much below -6 / -9db. I do not have volume pot on these as prefer to use XX digital volume I may reduce the gain at some stage by changing the feedback resistor(s) ratio but no rush to do that yet it sounds fine. I can get some serious window rattling volume now which I have to say is v welcome (sometimes when the neighbours are all out :-)

I agree exactly with your comments about the sound quality of the GC's what an audio bargain they are!!.

Also I spent a great deal of money upgrading my AN monoblock 300B's I could have saved a lot of time and money had I known then what I learned from this thread. One of the things I love about NOS1 and this forum is that after selling all my old kit I will have more money in the bank (quite a bit more) and a hugely better sounding system. I never imagined my system could possibly sound this good and there are more changes to make  :)

Can't wait to hear the mods to NOS1 - improved sound quality? mouthwatering.

By the way note that the transformers below are secured in the same way as NOS1 - good old blue tack.

I guess one of the reasons the GC's sound so good compared with the 300B's is the GC lack of an output transformer - Mani is coming to see me later in the summer with his output transformer less 300B amplifier - that is going to be a very interesting session and I will report back with our findings.

Regards

Paul

Thanks a lot, Paul for the pics. Nice job  :) Where did you buy the case and the trafos? Best, Per

Hi Per

Trafo's from Airlink    http://www.airlinktransformers.com/    in the UK - good quality units. I found it difficult to find decent looking cases there are quite a few out there but most of them look home made I eventually found these  http://www.modu.it/homeeng.html  which are based in Italy. The cases are v good quality and look nice with lots of sizes to choose from. Nice company to deal with as well. Only problem is that whilst the cases are reasonable price courier charges are quite high but anyway they arrived very quickly so OK. Modu will also engrave and ink the front panels if you really wanted to go that far. If you want any more info please let me know.

All the best

Paul

Thanks a lot Paul. And you are right about cases. I have found a few some time ago - but like with a lot of other stuff I can't find anything when I want to   :( So I am very grateful for the links. I'll give them a nice searchable description and sync them with my gmail account - hoping they won't disappear  ;)

All the best,

Per

Hi Guys,

Can anyone help with ideas about cost effective but high quality speaker cables?. Any ideas would be very welcome. I am now using GainClones in a mono block set up up - one amp per speaker (of course). I would like to bi amp - no problems I have all the bits. BUT speaker cable IS the problem - or rather the cost of it. I have longish runs to my speakers about 6m per speaker. I am currently using CHeLA (http://www.vhaudio.com/speakercables.html) cable and that is very good to these ears anyway. In the hope of finding a decent but cheap cable I did try paralleled satellite coax cable (OK anything is worth a try to reduce costs) but it falls a long long way short of the CHeLA cable currently in use. Bass disappears and all the fine filigree of detail that NOS1 produces also disappears. I could feasibly run bare cable for most of the route with a lot of work and I would be prepared to do that. But before I start it would be great to hear what any of you other Gainclone users have in the way of cost effective speaker cables.

All the best

Paul

I'm not sure if you need a specialty kind of speaker cable or not. If not here are two sites that sell well made inexpensive cables. Monoprice.com (http://Monoprice.com) and http://www.bluejeanscable.com/store/speaker/index.htm (http://www.bluejeanscable.com/store/speaker/index.htm).

-Chris

Hi Scroobius,

I´m not a GainClone user but I would recommend a few of high quality speaker cables at a good price.

Copper:

-  Anti-Cables from Paul Speltz. The cheapest ones:http://www.anticables.com/products.html (http://www.anticables.com/products.html)
Good cables at an extraordinary price. The Anti-Cables may be returned up to 30 days after purchase.

Silver:

- Clear Day cables:  http://www.cleardaycables.com/ (http://www.cleardaycables.com/)
Detailed and smooth, great cables. These are the speaker cables I´m using now in combination with the Grover interconnection cables. You have 30 days to test the cables.

Hybrids

- Grover Huffman cables: made of copper, silver and aluminun http://www.groverhuffman.com/home (http://www.groverhuffman.com/home)
The best interconnection cables I´ve tried. Detailed, natural, big soundstage. Great in my opinion to use with the NOS1. I have to try their speaker cables yet. Gover offers 60 days to test the cables and if you don´t like it you can return it. I bet you won´t.

As you see I´m using a combination of copper and silver. I´ve tried also silver for the bass part of the speakers and copper for the mid/highs with good results. At the end all depends on the rest of the system to balance it well.

I hope it helps, Juan

Juan - Thank you so much for your help these are really useful links I will let you know how I get on.

All the best

Paul

Hi Juan,

This is really interesting I had a look at the AntiCables and as a result tried a temporary home made version bare copper wires of 3sq mm (well 2 x 1.5sq mm in parallel) run well away from each other and away from the carpet etc. So the cables have no insulation which is obviously at this stage just a temporary lash up.

The sound quality is excellent really first class right up there with my existing 48 strand (2 x 24 strand per channel) litz speaker cable. And most interesting is that they sound good straight away no running in needed? (because no insulation ?).

So thank you very much for your help it looks as though there is a (cheap !!) solution to my problem of bi-wiring. More tests to do yet and also the practicality of running the bare cable in an acceptable way in a domestic environment (i.e. acceptable to my wife) but that can be solved I am sure.

Thanks again and it is very interesting as I had always thought that a litz type construction would be needed because of skin & proximity effect. Maybe insulation is a bigger evil - who knows but this solution really sounds great.

Best Regards

Paul

I made a couple of anti-whetever cancelling out coax LS cables a few years back, but I had to lay them aside because the sound was awful. Still, theories were good, and I always thought they were too good for my system.

The other day I dug them up again to try on some perceivedbly misbehaving amplifier, and *that* made it misbehave !

I still think they are the best (theory, + feels like that) and maybe I should try them on my normal system from today. This is not easy though, because or they are too long and I miss half of what's necessary (bi-amp) or I must cut the existing ones thus do some work again.
Btw, the real work is about what to connect to what (one cable consists of two coax cables) and the confusing plus and minus at the ends because of the constructrion. Also they must be consistently bound together (not twisted etc.) so it really needs a few hours to do it well.
This, for the expectation that it will sound disastrous again.

Peter

Hi Peter,

Well all I can say is that I spent a lot of time making the coax speaker cables and I was careful to make them exactly to the recipe (not twisted at all) and with the centre core of one in parallel with the shield of the other etc. I also used very good quality coax (as good as it is possible to get hold of). But the performance was just poor compared with existing 48 core litz (ptfe coated) copper cable. Very disappointing. No amount of running in would have made up that shortfall in performance methinks.

But what I am listening to now - a temporary lash up of bare copper cable festooned around the room to avoid shorting - sounds extraordinarily good. At first I thought they were close in performance to the 48 strand litz ptfe coated cables but after a very short time it became clear that they are much much better. And that is a very big surprise - I was simply looking for something that would be just good enough to drive the base units in the coming bi-amp set up that would be cheap (cheap is important to me at this time!!). But simple bare wires are much better - really special. It does seem that the lack of wire coating (dielectric) (PTFE in this case) makes a huge difference. So how important is skin effect? - single copper wires should suffer from skin effect to some degree (as opposed to Litz wire which should not suffer skin effect) - Either that or the gainclone amp likes the electrical characteristics of the simple bare wire rather the the 48core litz cable but I am not sure about that.

The problem is that my conservatory now has bare copper wire festooned all over the place and my wife cannot get in - she is not happy - not happy at all (even though she likes the sound of the cables  :(   -  she has only just recently stopped mentioning the colour of NOS1 so I have to be careful - very careful.

But I have a cunning plan !! just hope it works - watch this space  :)

Paul

Put them in a large rain pipe ? hahaha

You don't happen to have a picture of this high current danger ?

But that dielectric ... yes ...

Ha Ha Ha  - how did you know that 3 years ago when we made big changes to our house I ran a 6" drain pipe under the floor buried under the concrete just to take speaker cables. Of course then I did not know that it would be for skinny cables with no insulation I thought it would be for big butch insulated cables - I did not know then what I know now  :) (wish I had).

Pictures attached the wires are only 1.5mm2 so hope you can see then - big enough to trip up any unsuspecting intruder though. Not sure if the strategically placed pieces of "insulating" paper act as a dielectric though ha ha.

Paul

LOL

Are you sure there are no shortcuts anywhere ?

Yes, I did the same, though with 1" pipes. All carefully planned. Until I wanted to use them in practice and all of them were too steeply bended in the floor somehow, and nothing will go through them. Well, sh*t happens.

Peter

Hi Paul and Peter, I can´t help technically about the advantages of bear cables over the insolated ones, all I can say is that the Anti-Cables are a bargain considering the quality of sound vs the price which is ridiculously cheap. BUT, although they´re good cables even without considering the price, they are not better cables, at least in my system, than the Clear Day or the Grover Huffman, the other two cables that I recommended. The sound is less refined, detailed and less dimensional compared with these latter cables.

Anyway Paul, using absolutely bare cables can be a good investigation idea but in real life it doesn't seem of much utility, unless you want to see how looks a fried cat. As you know, the Anti-Cables are insolated by a very thin red lacquer coat. I used the Anti-Cables as interconnection and speaker cables and both ways, twisted and spaced, as you can read here: http://www.anticables.com/technical.html. I couldn´t tell any difference in sound.

Another consideration about cables. I think, unlike other opinions, that there is a difference in between different power cords using them with the NOS1. I´m going to confirm it using also some Grover Huffman cables that I´ll order next week, but so far, with some power cords that I used it seems to me that yes, there are differences.

Best always, Juan

Very nice to see some diy-ers around here!
Some 25 years ago I tried the double reverse connected coax with RG213, very tick and strong.
A german magazine advized the RG214 but that was not sold anywhere in Holland AFAIK.
It did not sound convincing so tried a lot after.
Now I use this:

It is cable used to make trafo's and coils, lackered wire.
Twisted to get low capacitance and inductance.
The number of wires and the radius has some influence on the sound so I made a mix that sounded neutral.
I think it is the same idea as your's Scroobius, but electrically and estetically more correct.

BTW. The double coax I use now for the 220V from wallplug to stereo.
Not reversed but both neutral and life inside and the screen connected to the central city heating.
Wires close to it don't pick up the hum and the amps don't get RFI from the mains.
Of course sounds very good as always with DIY. ;)
Could also be a good method for the earth wire as a third coax, should have very low resistance at high frequencies.

I have also made experiments with lackered wire. I use them as connection for the tweeters in my DIY speakers. A twisted pair of lackered wire sounds faster compared to a normal cable. In lower midrange and bass they were a little thin. That is why they are on the tweeters only. Using more of them in parallel like GerardA does, seems to be a good idea to overcome the bass lack.
As XLR or cinch cable they were no winner. No screening seems to be bad in this application.

@Scroobius
Is that an experiment cabling? It looks so funny :)

Hi Gerard and Flecko,

Yes the wire is experimental but also doubles as a trip wire for burglars  :).   My existing wire is 48 strand litz (PTFE insulated) so no resistance problems (4.17mm2 equivalent per conductor). The surprising thing is that with 2 x 1.5mm2 (so 3mm total per conductor) everything has improved. The most surprising is the base which is significantly better controlled.

So a question for Gerard - what are the sizes of the enamelled copper wire shown in your photograph?

Thanks for you feedback guys

All the best

Paul

Hi Paul,
I think the bass is 2 times 4 times  1 mm  and mid/high double times about half size but I'll check this if I can find it again.
At the moment listening to James Blake a couple of times with the new trick , load and reboot.
Bass floating around the room in circles, voice sounds heavenly.
Why listen to the moody blues!?

Hello Nick, Gerard and Flecko,

I have been trying out speaker cables recently and it has been a VERY interesting experience. I owe Nick, Gerard and Flecko a really big thanks for their help because my system is now very much improved because of their help. I owe you all you guys lots of beer - and I would be more than happy to buy it for you and provide a bed for the night if you drop by!!!

So I started by building a GainClone amp (thanks Nick) and what an eye opener. A BIG BIG improvement on my high spec SET amp. So the next step was to go bi amp (see start of this thread) but I could not afford to double up my speaker cable. How crazy my speaker cost far more than my GainClone. Enter Flecko and Gerard. Thanks to them I first tried simple bare copper wire (see above) that cost me nothing (taken out of some old mains flex) and was astounded that it was far better sounding than my existing (49 strand full Litzendraht cable - but coated in PTFE see below). I really could not believe my ears - but confirmed by my wife and hi-fi friend Dave.

So then I moved on to testing the following:-

Stage 1  Single 3sq mm lacquered cable as used in transformers and coils - result was that this sounded to my ears as good as bare wire. So first lesson a thin layer of lacquer seems to have little or no impact on sound quality.

Stage 2  To reduce skin effect I tried 4 x .8sq mm (ie 8 strands for stereo) for each speaker (2 drive units per speaker) twisted as shown in Gerard photo above. I think that this is  better than 1 above but not a massive difference.

Stage 3  4 x .8sq mm (ie 8 strands for stereo) for the bass units and 8 x .5 mm (or 16 strand for stereo) for tweeters. Connected in bi-wire configuration. I think this sounds noticeably better (but I have to be careful here much more controlled testing needed). A lot of care has to be exercised here because bi wiring can have advantages and disadvantages that are nothing to do with cable so too many variables to draw a full conclusion.

It is important to realise here that for my home made cables it was not practical to wind a full Litzendraht winding so I did the second best thing by simply twisting the wire as Gerard shows in his photos above. This is a compromise compared with the full Litzendraht plait (note that twisting cable reduces inductance which is a good thing but increases capacitance which is a bad thing). So if my home made cables sounded better it could not be because of the twisting (inferior in the home made cables) or the number of conductors to reduce skin effect (substantially less in my home made cables) so technically my home made cables should have sounded worse apart from one thing the home made cables did not have any ptfe coating. So it would seem from all of this that ptfe (and probably all plastic coatings) have a VERY VERY BAD effect of sound quality.

I do not understand the reason for this. On the face of it I find it hard to believe that it is directly due to the increase in capacitance due to ptfe. I say this because according to my calculations the increase in capacitance due to ptfe coating is less than that due to winding the cables together (I could be wrong here and probably need to check the calculations) but anyway is there another factor involved?. Comments very welcome. All I can say is that from what I hear ptfe / plastic coating is a VERY BAD THING.

Please Note - if you want to try the above I am sure that you will find it very interesting but be warned that the amount of work involved in twisting long length of speaker cables (6 metre runs in my case) is likely to be much more than you could ever imagine - it is a serious amount of effort.

So now the road is clear for me to try the next step which is where this thread started. I am building my GainClone amp with 4 amps this week so I can use my new 4 way speaker cable (that cost me very little  :)) in bi amp configuration. I will post a note soon with the results. But the important point is that so fas I have had a very unexpected but VERY VERY big improvement in the performance on my system for very little outlay. I started this looking for cheap cables that were as good as my existing cables but ended up with cheap cables that were far better.

Nick, Flecko and Gerard I cannot thank you all enough for your help.

Very Best Regards

Paul

Hey Paul, it looks like it's high time for my promised visit to your place. I'm really interested to hear how your gainclone compares to my Berning 300B. Let's liaise via PM and get the visit organised.

On a slightly different but still related note, does anyone know of a gainclone amp that can output ~100W into 8 ohms, which is happy driving a 2 ohm (and even 1 ohm) capacitive load? I've already tried a class-D amp (Hypex HG700) with my Quads and would now like to try a gainclone, having heard so many good comments about them.

Mani.

Hi Mani,

Do notice that I never sorted out about the load the Gainclones can bear, but getting out 100W of them should be easy. It's similar what I do (also in this topic), which is putting them in series. Also, the other day I measured it all, and saw no degradation (with a pot in between to level out the whole contruction; see earlier in the topic).

Btw, my speakers go well under 2 Ohm, but don't ask me whether that's 1 Ohm or 0.1 Ohm.

Regards,
Peter

Paul hi,

I am so pleased that you are having such a good experience with the Gainclones. I really must do something with my SETs I really don’t think I will be issuing them again, I guess your probably thinking  the same way  ;). My SETs are packed full of goodies like VCaps (loads) tantalium resistors Black Gates everywhere, mesh plate 300B’s and silver. I could have built my gainclones for a fraction of budget for the upgrade parts alone. The strange thing is I am very satisfied with the sound and for the first time in absolutely years I have no immediate plans to change my amp. What a bargain these things are  :)

Your findings with the speaker wires is very interesting. I have always viewed insulation as a necessary evil,  I’m sure the type of dielectric has a massive effect on sound. Your trails with 6 metre runs must really highlight the effect. I also have long runs so you have me thinking now, maybe this could be a project for when the nights draw in.

The next step you have planned to bi-amping was a real real eye opener in my system, a major step, I really hope you have a similar experience. Be sure to let us know how it goes.
 
Mani,

I went down the 2 gainclones per channel route to keep options open to build two balanced input, bridged amps one for each channel. The data sheets show, when bridged Gainclones are able to drive high power outputs and tricky loads.  Looking at the electronics you are considering though (serious stuff  :yes:) I would be surprised  if Gainclones could complete, but then again they do seem to be capable of pulling out a few tricks.....

Best Regards,

Nick.

Talking again about speakers cables I´d like to know your opinion on this kind of Remote Sense cables, used mostly, but not only, with the Spectron class D amps. Have you ever hear about it  and the way this speaker cables are integrated as a system with the amps in order to minimize the distortion and noise?.

http://www.spectronaudio.com/remotesense.htm (http://www.spectronaudio.com/remotesense.htm)

and here for the most technically prepared among you:

http://www.tubecad.com/2010/03/blog0184.htm (http://www.tubecad.com/2010/03/blog0184.htm)

As always I´m more interested in the audio part than in the scientific part of the issue, but just in case I give you all the information I could got about it.
Juan

You mean before they explode, right ?
:)

Are these the same speakers I have? If these go that low in impedance then there must be something wrong with your set-up...

Bert

Well, I didn't want to ask you directly, but since you are responding anyway ...

The speakers you provided (of course :)). So, how low do they go ? I recall 2 Ohms for sure ... or ?

Thanks,
Peter

PS: And for others : it is Bert who provided my Gainclones just the same. And that means of line filtering, and ...
But I'm not so sure Bert likes to receive 25 emails about this ...

And Bert ... knowing you a little ...

To me it occurs that having such a low resistance (at frequencies) must be a negative. Well, I don't know anything about this. Maybe you like to explain a little about this ?

And maybe also how the (your) Gainclones respond to that ?

Thanks !
Peter

Hi Mani,

As mentioned above you could bridge a pair of Gainclones to get 80-100W also bridged pairs can be run in balanced mode if you wanted to. The GainClone chips are just about indestructible so you could test it with your Quad's without worry.

Come to think of it I am looking at my half finished bi amp GainClones why don't I bring them up on my next visit to yours and we could try them in bridged mode.

All the best

Paul

Your speakers should not be lower than 6 ohms over their entire frequency range (impedance curve) giving the amplifier an easy load. Mani's ESL's are basically capacitors giving lower impedance at rising frequency which is a problem for most amplifiers, especially if you like to crank it up a little. Your gainclone is not sensitive to very low impedances in terms of distortion and even capacitance does not bring it off track, they just do not like low impedances limiting their maximum output.

Good thing is that the amps will not gradually start to "complain", they only make a weird (loud!) noise when the limit is reached (as if things are broken!). No problem for the amp or speakers though, it is an internal protection that just makes noise like that. If you do not hear that specific noise then all is well and you have plenty of power...

At your place you will not be able to hear the limiting noise coming from the GC's (and if you try then I do not want to be there without earplugs!!!) unless you connect low sensitive speakers to the amps and turn up the volume.

Bridging the amp gives it more headroom (+6dB), not sure if a bridged combo can be bridged as such again though but if so then a GC with enough power to drive an ESL could be created.

That's all I can tell about your GC's, I do not know about other chips (not even sure which chip mine uses!).

Bert

Hey Bert, thanks.


I'm not sure this is correct. Of course it seems like thinking "twice as much", and although this by itself would be true in Wattage or something, this is not about 6dB. It merely will be about the 21dB or something - which is the gain.
Or ?


Well, this logically follows from my setup of course, but it really is not what I intended to propose. Maybe it works, but it won't be the best I think. So :

Just a normal mono GC for one channel, and behind that another one. Nothing about filtering.
This per channel.

Of course a next step could be about the filtering, but first one must need that.
Similarly it could be about a balanced setup; If you do it all together I'm not sure with how many GC's you will end up with. Haha.


I think I know, but for me it is not important. What is, is that at least I am able to let the output show the same harmonics as the DAC throws out, or IOW it adds nothing to it (except for a small 3rd harmonic - I think I talked about this realier).
Anyway, this looks quite different from any other pictures I see from other amps. I don't care much about "but this is way down the audible level" ... I just don't want to see anything ... (which appears to be possible with the GCs to my own surprise).

I don't know much about the chips as I never dove into them really, but what Paul said is true : they are highly "instructable" (many resistor and cap options). But as far as I know in mine (Bert's) nothing is there on that department. It's really nothing else than the chip and gain resistor. And some filtering caps in the power lines, but that needs to be there.

Peter

Thanks Bert. Much appreciated.

I've sent an email to Peter Daniels of Audio Sector, asking if he can make something suitable up for me. Meanwhile, does anyone know of a gainclone expert in Europe who might be able to help me?

Mani.

If Gain Clone's can be bridged multiple times then I can help (using BD30 amp boards). I stll have enough BG's N types laying around to stick into a few boards...  :pleasantry:

Not sure how much bridging you'll need though... (this totally depends on the size of the room and how loud you want to go).

Bert

This way it is still the last amp in the chain being able to deliver the rated power for one amp where everything in front should be seen as source with more or less added gain...

A bridged amp is connected differently, more in "parallel" or "balanced" to be able to process more current or Voltage at the speaker terminal connectors.

Bert

Mani,

The basic amp is very straight forward only a few design decisions to make like rail voltages and cap types if you start with a good design like Peter Danials kit. Help would be good to taylor the application to Quads but if Paul can configure his setup in bridged mode I would really start with a listen there. My guess is you will know really fast if it is worth going further. Now that would really be something if they do drive Quads well. I loved my old ESL63s but as I mentioned before affording amps that could drive them well was a problem at the time.

Back when I was using the SETs and thinking what next for a while whilst I was moving house I was using a Logitec 5.1 computer audio setup for amplification. The thing was it sounded brilliant driven by XX and a decent SPDIF interface. So 15ks worth of "proper kit" struggling to beat 220 pounds of 5.1 system (WTF !). Anyway I had started to read up about gain clones at about this time and decided to take a peak inside the logitechs. There was a row of 5 gainclones and another chip amp for the sub. I wired in my Duo speakers to the logitech amps and the sound was great.

As I say try to get Paul's amp into your system, or mine if you want to try unbridged. I am sure you'll know in moments if there is potential.

Nick.

Ps this could restart my interested in Quads again there's something I really miss about ESLs.

:) it would be a David and Goliath level performance if by some chance they can compete.

Hi Bert, this would be great and I may well take you up on it. I'll PM you...

Mani.

Nick, I'd love to try either your or Paul's gainclones with my Quads. If I don't end up with a Soulution, DartZeel or bridged-BD30 amp in the near future, I might take you up on it. And of course, it'd give you a chance to take a listen to the two modern-day Quad offerings...

Mani.

HEY ! That is not what I (nor you) said !
manipulator ...

:derisive:

Hey Mani,

I think I mentioned when I came to see you that I used to have a pair of ESL's. What I did not tell you that I drove them with a DNM amplifier. It was a mighty fine pairing the best sound I could find at the time (many years ago). Dennis Morecroft the designer and builder of the DNM used to run a pair of ESL himself so it had to be a good pairing. But the amp was only a 12 watt standard class A/B transistor and when not driving ESL needed efficient speakers.  The Gainclones are in a totally different league to that old DNM so it could well be worth trying the GC on your Quads before you spend any money.
All the best
Paul

Hi Anyone who its interested,

Nearly there, I hope to get my bi amp GainClones up and running in the next week or so but I have to say I totally underestimated the amount of work required (will post as soon as I get them up and running). At the end of the day its not difficult (the electronics are of course very very simple) rather its the hardware layout case work wiring and all that stuff - it takes much much more time than I ever planned for.

However, because my GC is out of action I am again (at least for a short time) listening to my (high spec) SET's. Well they sound SO SO much better with my home made cheap (very cheap) speaker wire - basically the wire that GerardA recommended above. The improvement in sound quality is huge - but wait a minute ...... I was only looking for speaker cable to bi amp my GC's to be at least as good as my previous cable. I was not expecting home made cable to sound so brilliant. So this is a big improvement before I even get to the bi amp GC's. I have also to say this is really important when listening to NOS1 - I can really hear what it can do. Mmmmm but it costs peanuts !!!! but I have to say its a lot of hard work to twist into the final product.

I guess that in this whole exercise it was bi amping my gainclones that was the main point. I was simply looking for speaker cable (at very low cost) to allow me to do it. What I was NOT expecting was that the el cheapo speaker cable (see above posts) would provide such a big improvement. I will never trust a hi-fi magazine, a hi-fi review or a hi-fi dealer again.

Scroob's

Wow ... It really seems that we all can learn something here. I mean, how can it be a coincidence that so many (and a few really is enough !) come to the same conclusion ?!?

Really great ...
Peter

Sorry should have stated the edit. I was trying to sort out my bad grammer.  Complete does read differently to compete which is what I had intended.  :blush1:

How is your french these days Nick ?!

haha

Avez vous une pain s'il vous plait ?
That's all I know, and it still will be wrong.
But all so important in France. Don't forget to get yourself a nice paté !

:)

@ Scroobius:
Glad you enjoy these cables too!
It's a pain in the ass to build them, but it is worth it!
Maybe we have to try this for interconnects and inside speakers and amps too...
Other cheap tricks that work: The lightweight stuff that women use to scrub their feet.
Make three blocks and put it under speakers amps etc..
Specially the record player if you got one. And in that case a record mat of Origin live.
But for XXHE-tricks we have to wait for Peter and now for Mani too....

Paul,

I know what you mean about the time spent fixing fitting and wiring. The modules are together in a couple of hours and then days of case work etc. Should be worth it though :)

Have you gone for one module per driver or are you building bridged ? I have a feeling that for our situation with multiple drivers per channel, one gainclone per driver may work the best.

Looking forward to hearing how the amp sounds.

Regards.

Nick.

My French is not great, but I can get by ordering food and drink so the priorities are covered.

You'v got me thinking about a nice canard pate. I broke 22 years of being vegetarian recently, so there is a world of interesting food to try while were here  :)

Nick

LOL!

Hi All,

Est Arrivee !! (the bi amp GC's that is) and I have to say that is the only French you will get out of me tonight.

Anyway after  a busy soldering day at last the bi amped GC's were up and running this afternoon - it has been a very great deal of effort to build these I have to say but everything worked first time  :). And was it worth it? well read on ...................

So what do they sound like?   Well immediately the difference is obvious - there is a lot more detail - cymbals and brush work on cymbals just sound superb lots of tiny detail that I just did not hear before. The sound is easier more relaxed. Peter has often described how cymbals sound with NOS1 and they always sounded good but now they are really special. But it is everything that is improved sounding more controlled, easier, more stage depth less hash much closer to the live experience..... BUT I have to say that without the new El Cheapo speaker cable I most certainly would not be hearing this degree of improvement.

AND two of the amps are brand new and not run in yet - the 4 PS capacitors will surely improve with running in over the next week or so.

So with a superlative level of performance now from my system that I just was never expecting I have some important thank you's :-

Nick - A MASSIVE THANK YOU for starting this thread because as a result my system is hugely improved. And what have I done? replaced expensive (very expensive) amplifiers and speaker cable with Cheap Cheap home made stuff. WOW I wish I had known all of this before I could have saved so much money.

Gerard - thanks so much for your help with the speaker cables - Wow what a huge improvement they are. What a big lesson for me - for a few pounds and a lot of work something really really special. I spent two full days winding these cables with the help of my wife (he he) and I have to say you need to be well organised and patient to wind 2 x 6 meters of 8 strand and 2 x 6metres of 16 strand then wind the whole lot together - pictures tomorrow will be posted. Mmmm - great fun I had it festooned all around the garden - the neighbours wondered what the heck I was doing.

Juan - it was your recommendation to look at anti cables that started me off with cheap single core speaker cable that blew away my existing expensive cable and then led on with Gerards help to use multistrand varnished cable which was another big improvement. Also I will be reworking my IC's to be like the anticable IC's except in silver. Check out the picture of my GC's when I post them tomorrow for the signal input cables to the GC boards that idea came from anticable IC's thanks again Juan.

BUT MOST OF ALL - A MASSIVE THANK YOU to PETER. The improvements detailed above now allow me to hear what NOS1 can really do and it is truly brilliant. I never thought that this level of sound quality was possible at all let alone actually be able to listen to it in my own house. Wow just superb. Thank you Peter for this Forum because without it I would not have the fantastic NOS1 based system I have now (and here is the best bit) I would not have sold my old HI FI AND MADE A PROFIT. So I am more than a couple of thousand (£) better off and I have a truly brilliant exceptional sounding system.

So the big lesson for me is that the less I spend on hi-fi the better it gets (and oh boy does it get better!!!) - HA HA HA HA - all those hi fi suppliers, manufacturers and reviewers out there what do they really know about high quality music reproduction? - in most cases not very much methinks. How many have applied the intellectual rigour and radical approaches that Peter has applied? - well I don't know of any.

Just a final note on the GC's. They are simply superb besting my expensive SET by a long long way (and my SET's were the best amps I had heard until now). There may be better amp's than GC's out there but oh boy are you going to have to pay a lot of money for them. But then it gets better - for peanuts you can bi amp GC's for another big big improvement. So where does that leave expensive amps and expensive speaker cables? well you would have to spend insane amounts of money to bi amp those. The only sensible comparison in my view is a really expensive amp and speaker cable versus bi amped GC's with home made speaker cable. I wonder how much you would have to spend to best a bi amped GC - eye watering amounts of money I should think. I have not tried the Class D amps recommended elsewhere on this forum but maybe I will sometime.

Also I have just parallel my GC's at this time. I have not yet tried them in the configuration that Peter recommended where one GC acts as a pre amp for a second GC feeding the base unit. It just sounds so good now that I will wait a while before I try that (I will try it though so sure - watch this space).

Anyway - for all those I have offered thanks to above there are some free drink vouchers waiting here and if you would like to collect them PM me.

Mani - looks like all the criteria are filled for you visit to Marlow!!!! just wondering if you should bring your stats with you - well maybe that would be a bit too much but I can still bring my GC's up to yours at some time if you want.

A BIG BIG thanks again to everyone who helped me

Scroob's (aka Paul)

Hi Paul,

You´re very welcome and congratulations for your cables and Gainclone amps. I´m eager to see the final pictures.

Juan

Please let me tell you (to all of you, techi-guys :pleasantry:) that we, the poor of us that don´t have enough technical knowledge nor the ability to build such a great amps would like also to have the opportunity to enjoy it. Wouldn´t be anyone of you that write a "Gainclone for Dummies" to make it possible to build it ourselves or even to pass it to someone who could build it for us?. Same thing about the cables and more of the same about almost everything related to the technical stuff so often present here in the forum. Please make a little effort and spread it in an easy and understanding way.

Best always, Juan

Ditto what Juan said.  I'd love to see a concise summary of how to build Gerard's DIY cables and the gain clone amps.  I'm running an active two way set up and need to run about 8' per side of two pair wires (1 pair for the woofers and 1 pair for the compression driver).  Right now I'm using Home Depot zip cord 12 and 16 gauge!  And you know it doesn't sound too bad, but I know from my past experimentation of using solid core wire in teflon insulators that there is a better configuration than my zip cord.

Thanks for sharing.

A while ago I ran across a web page that explained how to build gainclones and I think another related site even made them up for you. Unfortunately I don't remember the details, but I'm pretty sure if you google a bit you'll find it. There are also several wire construction pages around, again googling should find them.

Having said that, I too would like to see the instructions for dummies version here, if anyone has the inkling and time.

-Chris

What a great story Paul. And many congrats on your achievements.

Since there's talk about the anti cables, I'm not sure anymore whether this actually *is* about anti cables. I mean, like the ones I talked about myself (but sure you made them from different material). So, are they ?
(I apologize if I missed it all)

And yes, less is more, and the GC allows for the least.

Awaitening those photo's too ! ...
Peter

Hi Peter,

I just used plain old "as drawn" copper wire with a lacquer coating as found in transformers and coils. So called "anti cables" are the same but use soft annealed copper (i.e. amorphous crystal structure). But anti cables only come in one diameter whereas mine (thanks to Gerard) comprise many strands which (if it is important - at my age I probably don't hear much above 8-10kHz anyway) should keep impedance down at higher frequencies (due to skin effect) - well who knows but that's the theory. All I really know is  WHO CARES it sounds great ha ha ha.

Paul

Peter Daniels website - http://www.audiosector.com/

PDF building instructions for his Chip Amp Kits

http://www.audiosector.com/nigc_kit-users_guide.pdf

Additional informations in this DiyAudio thread

http://www.diyaudio.com/forums/audio-sector/51381-audiosector-chip-amp-kits-dacs-chassis-4.html?postid=584319#post584319

Regarding bi-amping I remember Peter Daniel mentioning that the sound quality improves considerably using bi-amping - ultimately having one mono gainclone pr speakerdriver / side,

Best Regards,

Per

PS: Really enjoying to read about your experience into gainclone (and cable) territory. Keep it up.

Juan hi,

I think this may be the guide you need (see link below). If you use Peter Daniels kits (this is what both Paul and I have used) the guide breaks the build down to is most basic level with good descriptions and lads of pictures, really nice.

If you go ahead I am sure Paul and I could help with ideas on sourcing cases, transformers and other hardware items you need, once past the soldering stuff covered very well in the guide its mainly DIY to get stuff mounted into the case.

see LM3875 USER GUIDE pdf link top right of page

http://www.audiosector.com/lm3875.shtml

Best regards,

Nick.

Thank you Chris, Paul, Per, Nick and any other of you that are helping us, the ones with less technical abilities, to build the Gainclon amps. In my personal case I have to confess that although I´m going to read carefully the documents I have serious doubts that I´ll be able to build it by myself. To begin with I can´t read electrical diagrams and never soldered any electrical thing. I´ll do my best though, and if I can´t do it I´ll look for someone to do it for me (unless any of you kindly want to make me one that I´ll happily pay for, of course).

The configuration I´d use would be a pair of monoblocks to drive the upper (mids/treble/supertweeter) separated part of my speakers. The impedance of this M/T part is 8 ohm, sensitivity is 94 dB 1w/1m using 2.83 V.

The other part of the speakers, upper bass/bass, would be driven by my Spectron class D monoblocks.

I really don´t know if this configuration could have any inconvenience.

To make this configuration possible will need that the NOS1 have another pair of RCA/XLR outputs with the same signal than the ones already there.

Would it be possible when you upgrade my NOS1, Peter?

Thanks again, Juan

This tiny amp realy looks beautifull
http://www.audiosector.com/chassis_patek2_amp.shtml

Tempting. Thanks Per, Nick.


If you have no experience with electrcity in general, it would be better to look for someone who helps you at least. Soldering is not very difficult but it also needs some practice. If you want a good result, a project with highend goals is maybe not the right point to start. Also it is not undangerous if there is something done wrong. Safety first.

Peter Daniel also sells the ready-made amps, doesn't he?

Hi All

For anyone interested here are some pictures. All this was a lot of hard work and if I had known then what I know now it would be much less work for sure. But the end result is just superb sound quality and I would strongly recommend anyone thinking of building a GC to go for bi amping at least - it wont cost much more and the improvement in sound quality (as Nick pointed out) is substantial. The components and circuit boards I got from AudioSector and links to that site are given above. If anyone does want to build a GC I would be pleased to provide some details of what I did hopefully it could save them some time. Also I could provide a few tips on twisting the speaker wire (oh boy what a pain that job is!!).

Does anyone recognise the plug/socket in the wire harness from the power supply to the amp? it was a 20 way power extension cable out of a PC that I cut down to 12 way - that worked well I could not find a suitable plug / socket anywhere also in line in the cable meant no cut out in the panel required.  :)

And yes you have seen that speaker cable before it is in Gerards post above somewhere. At the end of all of this I am still astounded by just how good this speaker cable sounds a huge difference.

A huge thanks to Nick (for this thread) and Peter (for this forum) these babies would not exist without them. The beers waiting chaps!!!!.

Now where was I? oh yes I have some music to listen to he he.

All the best

Paul

 

By the way I don't suppose anyone has 8 x Blackgate N Type 1500uF 50v kicking around that they don't want  ;)

There is a trick how you can do it nice and easy. I draw a scetch to show it to you.
Sorry for the bad picture quality. I saved it in some horrible jpg.

I have been looking for months. I managed to get a suppliers last stock of 8X1000uf standard black gates but unfortunately the only stocks of BGNs seem to be kept by manufactures now.

I was wondering if an array of faster smaller (say 50uf) silmics might be a good alternative. I have never tried this before but then there was a good supply of black gates to hand.

Nick.

Fleko,

I have used this approach a well, at least with insulated wire. My addition would be to replace the clamp with the chuck of an electric screw driver to do the winding. I wonder if this might work with enameled wire without damaging the insulation ?

Nick

Ps like the diagram a lot  :)

I think the clamp works better, if you use a very big and heavy one. It has to hang free in the air. Due to its weight you just have to give it a nice spin and the cable twists itself! You will have a very homogeneously twisted wire because of the constant gravity force that the weight of the clamp deliveres.

Thanks!

But guys, what are you doing ?!?
This looks like the most ordinary twisted wire. That is, if Flecko is correct with his clamp solution.

Method 1

1. Find a grip somewhere;

2. Knot your end of the strands/wires to it (like Flecko's picture with the stand and horizontal bar);

3. Find a partner;

4. Find a battery driven screw machine (like a drill, but rev controlled). yes, batteries should be charged.

5. Mount the other end of the strands/wires in the head of it. Cause the lenghts to be the same of all wires.

6. Position your partner at "some" distance from you. Remember, you may be 10 meters from that grip the wires are connected to; the partner will be at 2 meters from you or so and guides the wires with arms wide (this suggests two wires, but more will work also (see method 2 below);

7. Start screwing. The partners holds the wires so the tension is and keeps (!) what you want. If the partner nevers moves, the tension gets higher. When the partner walks backwards, the tension can keep the same. This needs "one wire" practicing, and then you feel how it goes. It may need another wire to obtain the overal tension (or IOW density of the twists). Keep notice of the shrinking length during your practice (the gross 10 meters may end up easily as 6 meters net).

8. Do 20 meters in one go if you like. The total distance of the wire is not important; the distance to your partner is, and this never changes.

9. Keep in mind that at your end (the screwer) you may cut off half a meter of waste (there the density ain't right/even.

You will do 10 meters in 2 minutes easily !!
Come on, let's break records (good quality twists permitted only).

Notice that this is the only way to twist the wires evenly. I mean, that not one twists around the other, while the other keeps on being straight (which might be another application ...). When done manually, it will always end up like that (wrongly).

PS: Or am I missing something ?

PPS: We do it all the time here. Look under the hood of your NOS1 (the red/black and blue/black wires). Works with as many strands/wires as will fit in the screw machine's head.

PPPS: It looks too simple to be true eh ? still I wondered how to solve this while taking ages and ages at doing it manuallay ever back (for even one meter). Until I saw this light.

Method 2

Don't use the grip at the other end. Actually this is more more convenient, and you will be able to use up all the wire at that loose end (it won't curl up there). So, the partner is more involved at guiding the wires now; there's also somewhat more control about the tension of each of the wires. This is still about two wires basically, but more will still work.

Method 3

Do it completely on your own. Again use the grip at the other end, but now it is a balance between your other arm/hand doing the guidance, and doing nothing but pull back more and less. This changes per distance left and the length at first;
Once experienced this works as well, and in the end the best for more than 2 wires. This method will incur more for "after curling" than Method 1, because there's nothing there as a buffer at the not yet twisted end. You won't understand this now, but will notice this in practice later automatically.
What it comes down to is that at loosening the wires from grip and screw machine, the whole thing will behave as a curling snake. This is a matter or twisting back the whole lot until the snake is out. This looks more tough than it is, because you can also treat your snake like it's a whip. The curls go out automatically.
With Method 2 there are no curls at all, because the loose ends (behind the partner's hands) will "back twist" on their own automatically.

Have fun, which it really is ! If you see what HUGE amounts of time can be save, you will recognize that you made your own machinery with close to nothing. And the result ? perfect.

Peter

Maybe I did not understand your method right. I tryed to twist the wires in a way you discribed but WITHOUT an automatic screwdriver. I used a doorknob to fasten the cable and twisted them around each other by hand. It was impossible for me to make an even twisted cable. I do not know how it would work, if you will use an automatic screwdriver. For sure it will work much better. But this "horizontal work" can be more difficult than one might expect  :innocent:

The technique I suggested is limited by the height of your stands. Apart from that I had perfect results and it was unbeleaveable easy to do. Just spinning the clamp and watshing the cable twist (It can also be done with more than two wires). I wish I could show you a picture but the cables are hidden in my speakers now.

OK, here is a detailed description I made for Gerard:

Hoi Gerard,

Het is heel simpel, een klosje lakdraad, voor trafo's en spoelen, van de gewenste diameter bijv. 0,35 mm voor hoog en 1 mm voor laag, kopen.
Een aantal keer de lengte die je nodig hebt eraf rollen en naast elkaar leggen, na ze in stukken geknipt te hebben. Dan neem je bijvoorbeeld 4 stukken en die ga je in elkaar twisten, gewoon draaien dus. Dan nog een keer 4 stukken hetzelfde. Dan heb je daarmee een plus en een min-ader. Van de ene geef je met gekleurd plakband aan de uiteinden aan of het de plus of min is. Dan twist je die twee weer in elkaar en heb je één luidsprekerkabel. Voor stereo nog een keer en voor biamping nog twee keer extra.
Nu komt het leuke karwei om de uiteinden schoon te maken.
Met een scherp mes kun je de lak op de laatste 2 cm er af krabben en daarna met een soldeerbout de uiteinden vertinnen waarbij de laatste lakresten verbranden.
Met een multimeter kun je dan kontroleren of alle kontakten echt goed zijn.
Dan alles weer netjes samendraaien en uitproberen.
Check nog wel even dat plus en min geen kortsluiting maken!

Groeten,

Gerard.

OK, OK, you need it in English?

You take a roll of enamelled wire like they use for trafo's and coils. Say .35 mm diameter for  mid/high and 1 mm for bass.
Take the length you need four times, cut them off and lay them together. Twist them together. Now you got one wire for plus or for negative.
Do the same again for negative or plus. These two wires you twist together again. This makes one wire for left or right high/mid or base.
So for stereo repeat this once. For biwiring two times more.
Now you have to clean the enamel from the wire at all ends of all strands for about 2 cm.
You can do this by scr*ping with a sharp knife and after that using a soldering iron to put some solder and this will burn off the last residue of the laquer. Then for convenience put a piece of coloured tape on one of the wires on both ends to identify plus/minus.
After this you check with a multimeter to see if all strands are conducting. Only after that you can solder the seperate strands together for plus/minus seperately. After that check with the multimeter to see there is no shortage between plus and minus.
After that your ready to listen!

I think I got the idea from an english forum in a topic about the anticable where the famous dutch hifi-journalist John van der Sluis gave the tip about enamelled wire. I tried the tips and found that only one pair of single strips sounded very good but also a bit coloured as if there was a specific resonance. Twisting more strands solved this, so I guess the important part of the twisting is the added rigidity that keeps the wire from moving/resonating. And of course the lack of plastic must work the same way as using very expensive capacitors, no pvc but paper in oil and the likes.
You could try oxygen free copper or something similar, but remember the coils in your speakers are made most of the time from the same simple enamelled wire.

So good luck!

It is just very easy. The point is, if you rev up your screw machine to 2000 and walk the 100 meter in 8.9 seconds, you'd be doing 40 meters in 8.9/100*40 seconds, right ?
(and have 60 net meters of wire). Haha.


But all you need to do is concentrate on the tension (whatever that is you want) and walk forward. But really, we (one person) do 15 meters (this is our practice because of the room size) in something like 2 minutes.

Peter

You will need much longer !

Sounds like a "crazy" contest. Like one of the northern traditions like "wife long throw" .... :) Maybe we could establish a tournament.

Us British men have a lot to learn about romance...

I'm kind of confused. Looking at the datasheet for an LM3875, the slew rate is given as only 11V/μs. These gainclones should sound terrible. How come they [apparently] don't?

(Just for reference, my Sanders Magtech amp has a slew rate of 500V/μs.)

Mani.

Count Japanese in for such a contest !

Mani - Slew rate mmmm how fast does it need to be I have not a clue - what does NOS1 produce Peter? all I know is that I am sitting here with a huge great smile on my face. Certainly it sounds fast like it never has done before.

Well I wish I had asked about wire twisting before I started - could have saved much time. I used a manually turning handle with the wires clamped in (my wife did that bit - I have to buy here a meal by way of compensation). Why did I not think of an electric screw driver? - simple when you say it. Anyway all I would add is that when I twisted eight 0.5mm wires together I passed the wires through the teeth of a comb so that they were an equal distance from each other - so then moving along the wire as it was twisting and feeding the wires into the twist using the comb (if that makes sense). That gives a nice even twist.

Hell wish I had thought of the electric screw driver though.

Also where do I enter the "wife long throw" contest sounds fun.

Paul

NOS1-USB does 650 on the outputs (NOS1-original does a 100 less). But this is over 10 times what it atually needs.

I had another post here briefly, but deleted it again. Maybe it was not appropriate (to delete it, haha).

This summarized, nothing tells that "our" Gainclones overhere are build around the LM3875.
See Bert's earlier post in this topic ...

Peter

To apply for the UGH-Lympics you have to download this:

http://www.abandonia.com/en/games/219/Caveman+Ugh-Lympics.html

And that:
http://www.dosbox.com/download.php?main=1

With dosbox you can install the game, start caveman and enter MateToss. I don't know exactly what to do. I click like wild on the arrow keys and on enter, then he throws his wife...those were the days...I have mixed up the idea of this old DOS game with the finnish "wife carying". LOL

Hi Peter,

I´d like to know if would be possible to put on the upgraded NOS1 four RCA/XLR outputs (total 8 outputs). I ask it because in case I take the route of the Gainclone I´d like to amplify the speakers with the Spectron monoblocks on the bass part and with a pair of Gainclone monoblocks on the mid/treble part of the speakers. That means four amps.

Thanks, Juan

Now Peter, could you please provide us with the two minute gainclone? I'm sure you've been whipping them up while you're wife is resting from her part in the wire factory.

A fine and funny thread (page 7 anyway) and informative too.

-Chris

Hi Juan,

I am sorry not to have answered your earlier same question. I was more up and down between jobs of uninstalling W7 RC yesterday, and was having some fun in here. Ok, here goes :

I don't know !
The question has been brought up before in another context, and then my answer was that I can't be sure how different the reactance will be from the both amps (for one channel), and thus how the "feedback" of that may influence the outputs themselves. So, physically it can be done allright, but I wouldn't even be able to test (and for sure not measure !) it myself. It would be a trial ...

You and me also talked about similar (but in again another context), and then I adviced to use (home made) Y plugs on the outputs (well, something like that is what's in my mind). This would come down to the same. You could even try the Y connector at the othe end (near the amps). At least this way it's not a destructive thing. Also, you can try it right now (if you have those 4 amps - I'm not sure).

If you recognize (I mean, right now) that it works, I can make it a fixed thing which will be somewhat better (avoids a couple of connections). If it then bothers afterall, it would come down to just not using those connections, and if it psychologically bothers you, you can cut the wires close to where they output from the gain stage.
Think 160-180 euro for this (which is mostly about the soldering of the bunch of wires to the now actually small connector pins inside).

Lastly, we must be careful. Because now I think of it ... it formally will need twice the current output to drive the interlinks. I don't say it won't work, but theoretically it will work less (or the allowed interlink length will be less).

Maybe by now we must wonder how Paul did it, because he must have done something similar ...

Peter

Thanks Peter, yes we were talking about the Y cables but in that case they were to be used with my Spectron amps in the monoblock configuration (this is basically a stereo amp  that can be bridged inside to be used as monoblock).

I understand that different reactance/feedback problems could appear using a combination of different amps. I don´t have more amps so I can´t test it, maybe someone else in the forum could do it and let us know if the use of two different pairs of monoblocks affect the sound.

I guess that if I want to test the Gainclone I´ll have to buy it or build it  :nea: and see if they work in combination with the Specton. Anyway, I´m very happy with Spectron and I feel quite lazy to start with this new path. The only temptation I have is the Gainclone price (if I could build it by myself).

Juan

Since I was measuring a DAC this morning anyway, I captured this screenshot. *This* is how I want to see a noise line, and not one spike else.

Notice this is measured from the DAC output, and not from the amp's output. But it really looks the same, with the difference of the gain - hence noise level highered by that gain.

If this line has spikes to begin with, or is not straight over the full frequency range as shown here, sound will be dead.

Peter

PS: For the knowledgeable : this is a normal 8K FFT, so no tricks or anything to lower the noise line (which can be dropped by 20dB easily with a deep FFT).

OK Peter, I'm dying to ask you this: would you see this level of performance from the output of your power amps? Actually, with your 115dB/W@1m speakers, I wouldn't be at all surprised if the answer is 'yes'...

Mani.

For a cheap gainclone offer:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=160590654250#ht_4487wt_1235

Hey Mani,

Although the answer should be Yes, this is not 100% honest. I mean, a few months back I never looked into it for a couple of years. I *had* to though, when things appeared to be wrong-sounding (and it was wrong, never mind for now why).

Do, today the answer would be Yes. This, while the honest answer from a few years back would be "impossible !". So here too, all is about the reference.

PS:
Please notice this is all in the context of a zillion thing wrong anyway. I'll be back on this soon (the Bass thing).

Hi Peter & Juan,


Into my GC amp there is just one pair of RCA as normal but inside I simply connected the amp inputs in parallel - so just the same as Peter's "Y" connector (but v short distance). OK not ideal because as Peter points out my poor old NOS1 has to provide twice the current and also sees the reactance of two GC inputs. I am sure that this is the way Nick connected his active set up and as Peter points out it is (maybe) not ideal. Peter has pointed out elsewhere in this thread he used a different arrangement with the LF unit GC taking the output of the HF unit GC as its input and a volume pot to control the output to the LF unit (well I am sure Peter explains it better elsewhere here). I am waiting a dark rainy evening to try this arrangement - so being in the UK that should not be too long ha ha.

Thinking about it there is an easy way to see if the above paralleled input to the GC from NOS1 adversely affects the sound. With the inputs from NOS1 connected together (as in my GC amp now and described above) connect the speakers as normal (ie not bi amp) to the outputs of one stereo pair of the GC amp and take listen. Then remove the link (paralleled input links) and take another listen - better still put a temporary switch to aid the comparison. The comparison is not perfect because one GC amp pair will not have an output load - but if a difference is heard then it would prove that paralleling the inputs is not ideal. When I get some enthusiasm I will try it over the next few days and report back. Too busy listening to music at the moment though  :)

P

I tired to check the inputimpedance of the gainclones, but it looks like they are at least higher then 100k.
As long as it is not changed by an extra inputresistor.
I guess that is high enough for the NOS1 to deliver into two parallel gainclones?

Hi Gerard  -  The input resistance is not a problem but possibly the "reactance" or different varying voltages appearing on the two inputs is - well maybe.

P

About the volume pot ...

There is a trick in order here ...

Although I never tried it without pot (would be useless as well because of too much gain), I am fairly sure that this pot will create anomalies similar to any volume. Now, one could say "hey, this is bass only, so don't worry too much". This by itself would be the wrong idea IMO, but never mind for now.

What happens in my situation is that the woofer -net- will be fed by low passed signal only. So, in my case the roll off starts at 250Hz or so, and beyond that (higher frequencies) all dies out. It is this which makes it the practice case of whatever anomalies the pot would show in the higher frequencies (it is always there where things start to go wild/strange/wong), are not there at all.

Of course any filter behind it (in speaker) would incur for the same, but it is still something to take into consideration. Not that things should be changed, but it may be good to know that IMO the pot here will be harmless (when not too poor).

What we should be worried about more is the low (50 or 60 Hz) mains frequencies which will be "visible". Without an inherent balanced design this will be difficult to avoid. But I think that anyone who starts a Gainclone project now, should especially concentrate on that, because it is there where the potential improvements are.

Peter

The sound quality of my GC's is discussed elsewhere but the level of mains hum with no music being played has been just a bit more than I liked (I could hear hum - just - 10 feet away). I thought it was not too bad given that the GC is on full volume all the time (ie no volume pot) with 98db sensitivity speakers. With all the systems I have had in the past if I turned the volume to maximum I would certainly hear hum and that was through low sensitivity speakers (88db typically) and these were big "name" systems. Also RF breakthrough etc.

Well for reasons I will post separately I decided to take apart and rebuild my amps and whilst doing so I created the most perfect "star" earth arrangement that I could. To put it in perspective though the old earth arrangements I really did think were good enough and if you look at the pictures earlier in this thread you will see that there is a 1.5mm2 bus bar arrangement connecting the ground points on each of the circuit boards and then earth points from regulators taken up to the bus bar.

Well now (and I will post pictures soon) I have taken a 2.5mm2 wire from each of the 4 circuit board ground connections separately to a single earth point (SEP) on the chasis. The main incoming earth is connected to the SEP and also each of the 4 regulator earth points is separately wired to the SEP. So there are 9 connections to the SEP. So what is the difference between the previous earth arrangement and the new one? - well electrically not much at all but in terms of mains hum a very big and surprising reduction. Now if I put my ear right next to (3cm away) the LF speaker cone I can hear hum but 1 - 2 feet away I can hardly hear anything. That is a big reduction. I always knew the importance of star earthing but I am surprised at the difference in hum. So if anyone out there is thinking of building a GC it is not possible to pay too much attention to earthing arrangements.

Also I am going to increase the heat sink capacity. Whilst I have not had any problems with the current arrangement they do tend to run a bit warm at extended very high volume. I will post pictures when complete for anyone who is interested.

P

gringrinmoregrin
(nose on keyboard)

........ but I have to say the heat sinks only get as hot as the hottest in NOS1 even on high volume so maybe not so much of a problem after all.

Peter  -  Hope your nose is OK!!

Hi everyone!

I have been recommended to try a Gainclone amp on my Avantgarde Duo speakers.

Do anybody know someone that can build one for me?

Interesting thread....

I looked at the web and there are several chips used for Gainclones:
- 1875
- 3875
- 3886
- 4780

Does anyone know what the sonic signature is of each of these?

For instance, would it be an idea to use for bi-amping 2 different kind of gainclones?
F.i. a 3886 for the bass section and a 3875 for mid/high?

Hi Bigear,

All I know is that the Peter Daniel 3875 version has been tested, modified and stripped down to the fewest components possible. Components have been removed to the extent that there is only one (3875) in the signal path. There are no caps / resistors in the signal path at all (just like NOS1 - well for NOS1 just one resistor I think). The other components have been carefully selected for best sound quality. So how important is the actual chip type? - I really don't know but the starting point for me would have to be the same layout and minimal components of the right quality before I would even start to consider comparisons. I really don't know the answer.

Not easy to get an equal comparison.

Regards

Paul

Hi Bigear and Paul,
On DIYaudio.com there is a lot of knowledge about the gainclones. But if you really want to know it's easy to build a gainclone and interchange the chips because mostly the're build on very small printboards.
But I think the real merit will be heard with all of them.
Since it is a clone of a design by 47-labs the components will not be the real gamebreaker. 47-labs uses some cheap components that can easily be bettered but that's the fun part of DIY.
And of course the power supply is very important!

The final amplifier layout with a change to the grounding arrangements hum is now down at very low levels http://www.phasure.com/index.php?topic=1844.msg18856#msg18856 and the heat sinks have been bolted to the case. Well maybe that was not strictly necessary but it was easy to do.

 

Well I won't show such a beautifull amp, but the sound is amazing! I replaced my Hypex UCD180 and Stoetkit 300B with double chinese Gainclones and it is really an improvement!
Thanks for the tip!
It's like the music fits in the environment like a picture photoshopped into another picture without seeing where one starts and the other stops.

Congratulations Gerard !

Nice one Gerard and welcome to the club!!

Scroobs

Hi Gerard,

did you get them from ebay?

Best,
   Quint

Hi Quint,

Yes the're from ebay, including the powersupply which I did not use yet.
Just replaced the old Hypex amps with the new amplifierboard and connected the wires. Added a second one in the metal box and screwed it to the side for cooling. Really simple!

Later I'll put them in a small box on the back of the speakers to save a lot of space.
BTW. No hum at all.

Gerard hi,

That's great congratulations ! That is now three of us with 300b SETs who have moved over now. Not that I have anything at all against 300b sets, but its amazing when you think of the relative cost of the gainclone amps and how they can sound driving suitable speakers.

KR,

Nick.

I have been considering changing my Gainclone amps and thought that I would post the idea.


To recap first, the amp is a bi amp Gainclone setup with options of direct input or volume control via TVCs. A further point is that the bass channel of my speaker system is active so  the system is tri-amped, the main point is that there is no bass power load on the gainclones.


The aim would be to take the gainclones off the mains and improve the power supply to the amp by using battery power. The hope would be to benefit from the air, transparency and natural tone of battery power, but to ensure that nothing is lost in terms of speed substance and dynamic slam.


I am still researching the exact approach (surprisingly I cannot find many posts talking about this) but the idea looks like this:


Batteries to give +- 19.2v rails, currently the amps mains supply's are running at 28v but with 103db sensitivity speaker I think 19.2v could give enough headroom. Not sure if the voltage drop could impact sound elsewhere though ?


Batteries will defiantly be LiFePo4 and most likely A123 brand, these things are fantastic things. Massive current delivery, superb transient current delivery and very, very quiet form a noise point of view.


I would like about 100 hours turned on with no music playing and about 20 hours playing at normal listening levels (quite loud :-) ). Also important is the ability to swop between battery and mains playback.


The main issue right now is finding a battery management system that will help me mange the batteries. LiFePo4's need clever and careful charging. I am prepared to have four separate BMSs in the case one for each battery /PSU rail. They will need to be capable of managing  a 19.2v (6 cell) 5.2Ah battery. The charge cycle on the BMS has to start automatically (eg I want to throw a play/charge switch and have charging start automatically.


No parts have has been bought yet, if it goes ahead it will be a winter project, it's still just an interesting option but I just have a feeling it might give some very good results.


If anyone has any thoughts and in particularly how to set up the BMS and where to get one that would be great.


Cheers,

Nick.

Hey Nick,

Before you dive into the obvious hussle of charging those battery packs, just know that my own GC's do not show any additional noise but the 21 or so dB gain which amplifies the input noise with exactly that. This doesn't necessarily count for your GC's, and all I want to say is : it can be done like this and normal mains power.
Well, you know my thinking.

Peter (not wanting to destroy your winter project !)

Peter hi,

Thanks for the information about the input and output noise values. A great amp performance even with mains  :)

I was thinking that given the low smoothing PSU capacitance used by gainclones, as the power consumption goes up that mains noise may start to rise. I know power supply noise rejection is very good but there must be a limit to what 1000uf or so of psu smoothing can do. There is also quite a lot out there about transformer size improving / regulated supplies improving aspects of the sound.

One question about the noise measurement, did you look at input / output noise with the amps at a reasonable working load ?

Best Nick,

Ps still searching for a practical charging solution and not that easy....

Well Nick, you got me thinking afterall ... (ok, that happens more often :))

Of course I tested the lot "under signal", and it is this situation I talked about before (though I don't where that was) : all what is added is a small 3rd harmonic, and further it is an exact copy of the DAC what I see (FFT wise). The difference is the gain (21dB in my case). However ...

What I did not incorporate in this little story (and didn't think about yesterday either), is that the sheer connecting of the amp to the mains adds noise to the signal, just because "noise" creeps in by other paths. So, this is not exactly mains noise, but paths being created along which the noise travels from somewhere else. In my (great) example it is always that 380V PWM controlled pump motor (heating system). And, although very well under control (better than -100dB) *that* "noise" sure becomes audible just at connecting the amp to the mains.
To keep in mind : this noise is hammering against the NOS1 just the same, but as a self contained device under my complete control, I was able to 100% reject it from there. Not so with the amps, or I'd have to rebuild them from scratch. But for example, if I, in my current situation, could connect the amps to PE, that pump noise easily adds a 20dB of its frequency (which is at 17.8KHz).

So, it would be fair to say that indeed a battery feed will help ...
Or not, because the inherent noise can't go anywhere.
As usual, this is not so easy.

Anyway, for sonic performance you also have to wonder - in advance I mean. Already with the NOS1 experiments it was clear that battery feed would not go without a plunge of caps. No power in the sound otherwise. Dead. Lifeless. And this is the few mA of a DAC !! (ok, 200+ mA).

You could also do some nice math on what happens with this pile of capacitance added (assumed it is necessary indeed), and what happens after switching on the amps. Pop goes the weezle and the batteries might be half empty instantly ...

Maybe you'd be better off with putting two 12V car batteries in series, but don't forget those caps.
Notice that this it is not uncommon of course to put forward car batteries, but what might be less common is the reason why to do it. Not because "hey it should be better", but because of not creating ground paths, plus those caps could be a new thing. But never forget : you can't get rid of inherent noise either, unless via the signal path (interlinks).

Peter

Hi Nick,

I read in DIYaudio.com that the higher the suplly voltage the more dynamic they sound. Think Peter Daniels uses 2x22V -> 2x32V to get a relaxed comfortable sound but people who prefer more action can go till 2x28V -> 2x38V max. (trafo AC/ rectified DC). So that's a lot of batteries!
If it's working out big time then I'll try it too, but not for now!
(Already have a car battery for my single tube preamp, if I forget to turn it off and charge it the battery is dead....)

BTW, for midd/high I shorted the inputcap since there is one already in the speaker and the preamp. So only one resiator and the chip.

@Gerard: These are the ones I bought, but directly from China. It seems now they are in stock in the UK, so that makes it even more easy. http://www.ebay.nl/itm/LM3886-Hi-End-AMP-Amplifier-DIY-KIT-68W-component-SET-/260836152041?pt=UK_AudioTVElectronics_HomeAudioHiFi_Amplifiers&hash=item3cbb0bd2e9 (http://www.ebay.nl/itm/LM3886-Hi-End-AMP-Amplifier-DIY-KIT-68W-component-SET-/260836152041?pt=UK_AudioTVElectronics_HomeAudioHiFi_Amplifiers&hash=item3cbb0bd2e9)

You could also try the Peter Daniels made gainclone, should even sound better but more euros, and not pre-built I think.
(See DIYaudio.com chipamps section)

Is the gain defined by the desing or by the chip of the chipamp? I am also looking for a low gain amplifier to reduce noise for high a efficiency speaker setup.
Greetings
Adrian

Hi Adrian - the gain is defined by the value of two resistors Rf and R3 in Peter Daniels kit and it will be similar in yours I am sure. You can change the gain by selecting different values for those resistors.

P

Hi Gerard - higher supply voltage would be interesting but you need to make sure you have 8 ohm speakers. A higher voltage and low impedance speakers can be a problem. 

P

Yes, Higher voltage with 4 ohm is a little waste of money, but I think the protection will keep it save. Or not?

Hi Nick

This guys experience is interesting you may have read it http://www.shine7.com/audio/gainclone.htm - he measured the DC rails and found up to 20% drop when driving the speakers hard with a lot of bass so he added a high current regulator and improved the sound a lot.

You would not have that problem but maybe others of us (like me for example) do? I will get my DVM out when I have a moment to check it.

Peter's point is interesting about noise though.

P

Hi Gerard - my recent experience was that protection on my GC did not work (with high offset input voltage) it fried and ended up toast. So now I am very nervous about invoking protection.

P

True. But notice the relation between the higher resistance (higher value Ohms on the resistors) and the thermal noise implied. So, there's also a relation between the supply (V) applied and what value of the resistors now need to "surpress" the gain.
But, sort out the way this works, because "feed back resistors" operate like the higher the resistance, the more gain you will have. Thus also (and for Nick) : the higher the supply, the lower the value of those resistors, the less (thermal) noise.

Regards,
Peter

Thanks Scoobius and Peter.
I am also playing with the idea to get one of these:
http://www.transcendentsound.com/Transcendent/Products.html

I once heard the single ended OTL with a fullrange horn. I could compare it to my EL34 ClassA amps I had back then. I remember it as an "that is it" experience. Very clear and extremly fast. Has anyone experiences with one of these amps?

The problem is, it is hard to design a speaker that works well with these amps (constant 8 ohm impedance and sensitivity>>90db)

Furthermore dont forget to look at the specsheet of the chipamp for minimum closed loop amplification if you plan to change the resistor ratio. For the gainclone chip this is about 20X (as I remember...), lower and you are asking for problems (instability).

Regards, Coen

I am looking for an amplifier for a friend and the Gainclones seem to be a good choice. But I do not want to build such an amp and it should not be expensive (200€-300€). Are there any ready made amps in that price range? I searched the web and found nothing. I had a test with a tripath design:
http://www.audiophonics.fr/poppulse-t150-mkii-amplificateur-classet-tripath-ta2022-p-3626.html

For the price you can not argue about the sound but I like my friends to have something better. My class A ss amp with tube pre amp is of course another leage and it would be nice to find something equally good or even better for small money. Considering what you say about the gain clones, they should also be at least one step ahead of the tripath designs, even if they are in the cheap price range.

- Maybe / maybe not - best to compare them in your system before making such a claim (I and others have made that comparison and been surprised). And of course using XX means that the pre amp can be removed and that is always going to be a substantial improvement.

I have not found any "cheap" ready built GainClones (and 200 - 300 Euro's is very cheap) and there is a reason. You would not be able to build one for that price. Just check out the price of good quality transformers cases and fittings. Then add on the hours labour and you will come up with a price that no longer seems so cheap. Peter Daniel sells a very good ready built GainClone and also there is the original GainCard - of course they are not cheap. And then again there is Bert who is making a GC "special" for Mani. I am sure there are others but I have not come across them. Best of luck.

I have compared the linked amp (tripath) to my amplifier in my system. With my old passive preamp the sound of the class a ss and passive pre was better. Now with my active pre, the difference shoud be rather bigger.
I suspect from what the people in this thread were saying that the gainclones are realy good and maybe better than what I have. That is why I am ineterested in this too.
Because it is a simple design like the tripath it should be possible to make an ready made amp that is not much more expensive than the mentioned one. For a non hifi freaks like my friend is, using blackgate caps, high quality transformers in an expensive enclosure would be a waste of money. But anyway, gainclones seem more popular in the realm of diy then in the mass market.
Removing of the pre amp and using xx volume control is not always an improvement. At least with the DAC I have tried it was not (Audio-GD ref7). With my new dac (Northstar Extremo) I did not try it, because there is a problem of "popps" when I use it with the audio-gd digital interface. When the fridge turns on for example, there can be a popping sound out of the speaker. It does not occur that often but when, it would be very loud without a pre.

I don't think a TriPath is so simple ... :)

A tripath chip is not simple itself. But you do not need a 1000w trafo and 100000uF of filtering and so on. That is what I ment. And that makes it cheap. Same is true for the gain clone if you look at the design of peter daniel for example.

Hey guys, listen up now. My 'super' gainclones will arrive here today!

Bert (BD-Design) built them for me, and from the pictures (posted with Bert's consent) you'll see he seems to have done a really great job. And Paul (Scroobius) yes, they really are Black Gate Ns - 8 per channel!!!

I just can't wait - actually, the last time I was this excited about hifi was just before my original NOS1 was due to arrive, around this time last year IIRC.

This post may be a little premature as I haven't even tried them (not here yet), but I wanted to give you all a heads up.

I wonder if I'm the first person to have tried driving electrostatics with gainclones?

Mani.

Hey Mani - they look familiar  :) excellent taste in enclosures I must say.  Any chance of a picture where we can see all three boxes with the tops off looking from the rear? On the other hand I suppose that the other box will be another amp identical to the amp we can see in the picture?.

Only one transformer ??? or does it have two secondary's one for each amp?

If the amp does not work out with the Quads you could always give those Blackgate N's to a good home (i.e. me) - I promise I would look after them for you he he  :).

Just out of interest what size (uF's) are the BG's?

P

+1 on the enclosures  :)


Mani,

They look fantastic, a great build by Bert and I bet the post cannot come fast enough   :)

Not just 8 Blackgates but 8 BG N types, that should really help the sound.

I think I can count 4 amps per channel, are we looking at bridged parallel for each channel with differential inputs ?

Very, very nice I understand why you want to hear them.

Best Nick.

I just wanted to say : XLR output (stupid !) Inputs eh ?
But you were just ahead of me.

Heeelllp
:)

UPS hasn't arrived yet :(


No, 4 boxes in total - 2x amps and 2x power supplies.


Yes, exactly.


XLR inputs only and balanced output at speaker terminals.

I really should have waited until I've at least tried them. It  could well be that my Quads don't 'like' being driven by an amp with balanced outputs. We'll see...

Mani.

PS. Nick/Paul, does this mean we all have gainclone amps housed in identical cases (sizes apart)? But I bet my amps are stronger than yours :arrogant:

Puh, count me in !

Bert, I wondered ...

It seems to me that the "preamp section" (the mid-high amp in my situation) can be fed with XLR, while the bass amp behind it can be normally SE. Am I right ?

Peter


PS: You are providing DC Offset meters in there of course ?
:swoon:

Mani, no contest   ;) with bridged parallel you should have power and current delivery,  just the thing for quads. I'm actually excited to hear the results, could be very very good indeed. Fingers crossed.

Mani - you obviously will need an experienced pair of (GainClone) ears to check these new amps out - I would be pleased to offer my services. I can also hold babies while listening but it has been a few years since I did that (mine is 6ft 6inches tall now).

Four boxes - it will be really interesting to hear your first review and as you say you could just be the very first quad owner with GC so could this be a world first scoop on the XX forum.

P

See ? you don't know Nick's car.

Well I do know Mani's car I still have the oil stains on my drive ha ha  :nea:

You want oil stains? I'll bring my wife's Golf diesel next time! Assuming I get another invitation that is...

Anyway, the BD-Design gainclone 'super amps' are here!!!

I've just set them up, run Densen De Magic and listened to a couple of tracks. It's actually quite difficult for me to compare these mono block amps directly to the stereo Sanders Magtech amp because I had to change all the interconnects and speaker cables also. I'm now using some anti-cable interconnects and very cheap 'n' cheerful single-strand enamelled copper speaker cable (the same sort of stuff used in the interconnects, but not twisted).

I'll hold off making any detailed comments until I've had a good listen. But I'll just say that until today, the Magtech was easily the best amp I'd ever heard with the Quads. And the BD-Design amps are in the same league at the very least.

Mani.

No problem as long as the opposite leg of the XLR input is showing the same input impedance to the source. or use a line transformer to do that but I think that you do not want those in your signal path anymore...  ;)


If the source is already having this equipped then why needing another? Besides that, the servo regulation cancels your DC on the input up to 15mV at least so hopefully it stays below that, otherwise the amp needs an input capacitor (Mani's amps are fully DC-coupled now).

Bert

Correction after reading back in time where I assume you are referring to the voltage drop at high power? Tests with 180Watt did not show any weird things so it is of no use to have those DC Offset meters with these amps.

Did you ever hear the difference between a regulated power supply and one that just delivers at the moment when asked without holding back?

Didn't I mention the overkill used for the Power Supply with 500VA transformers and about 85.000uF per channel?

Indeed you should wait a little and have the amps burn in for a few days...especially those BG's need to settle down to show their true potential.  :)

Glad to read that the amps arrived in one piece!

Bert

Hey Bert, thanks soooo much for putting these together for me. Already, they sound, really, really great. I strongly suspect that getting these amps will be on par with getting a NOS1... in terms of the best hifi choices I've ever made.

Thanks again.

Mani.

I was only kidding.


Not talking about the transformers, but what would be the 40000uF/ch in the NOS1 then ?
:)

Peter


PS: Mani, now I wonder, did the Offset as read at the NOS1 side change ?

Haha...

Well, I've been playing around with various things, trying to eliminate a tiny bit of noise that's coming through the speakers that wasn't there with the Magtech amp. Bert's been helping me via email. Anyway, I've just stumbled upon an old pair of XLR interconnects that I had hanging around in my basement (not used for I guess 8 years or so). They have an additional alloy jacket, that I think is connected to ground at one end. And you know what? The noise has completely gone.

But what's even more remarkable is that my DC offsets have gone from 5mV and 1mV to exactly 0mV and 0mV!!! And they remain there whether music is playing or not. Oh, and I've re-tried all the other XLR cables I have here - the speaker noise returns and the so do the nominal DC offsets.

:wtf:

Any explanations most welcome.

Mani.

PS. Amps sounding more and more wonderful every minute!

Same problem here, with Nordost unshielded cable I get some noise, RFI, hum through the gainclones, especially when the preamp is not ready for connection because the relay is open.
With shielded cables no problem and dead silent.
The pity is the Nordost sound more open.
Maybe A DIY-cable has to be made. :(

:wtf:
Yea, say that.

Noise gone ? I'd say signal gone as well. So, it would be my idea that you have shortcut the leg where the DC Meters are connected to (don't ask me which one that is (hot/minus), but it will be the same the RCA output works with.

This doesn't mean you won't have sound. Only 6dB less of it (but a bit depending on how BD's amps have been setup).

:bye:

For what it is worth mine are home made interconnects they are not shielded and have cotton covered silver wire loosely twisted  - no problems with interference I also tried long versions of the same with no interference problems.

Paul

I don't think it has anything to do with screening noise out - I think Peter is correct.

But I've checked the cables with my meter and can't find an issue with them - there are no shorts between pins 2 and 3 on either cable. However, as I suspected, the alloy jacket is indeed connected to pin 1. So in effect, the outer body of the XLR plugs at both ends are connected to ground. Somehow, this is causing a short between pins 2 an 3. So there's either an issue in the amps or in the NOS1.

Mani.

As Peter said wonder how Bert set up the grounding? - I changed mine to be the same as Peter's and Nick's so grounding in amp not connected to PE and grounding between left and right amps not connected. Noise levels went down to barely perceptible.

P

Maybe it needs more thinking, but at first glance I'd say it is not allowed to connect the ground pin to anything else than the virtual ground. Now it becomes vague for the workout, because you are connecting two virtual grounds that way (let the one be different than the other, and there comes the current). That is, if the BD is setup for its balanced input "floating" all the way. And if not (like signal ground is connected to the case or worse, PE), there might be another problem, because you'd be interfering with the midpoint of the signal.

Any voltage present between signal ground of the two devices (NOS1 / BD) ? (don't connect the interlinks)


Are you sure about this, or do you only think so because of the Offset being 0.00 ? Remember, that Offset is hooked to one of the pins only, so shorting that pin is enough to have the Offset at 0.00, while the other pin nicely will put out signal.

Thinking in a complete different direction :
When the balanced output of the NOS1 is not correctly connected to the balanced input on the other side, and one of the signal wires will sease letting through a signal, you'd have noise because of no possibility of cancelling out noise. This is a bit of a too long shot, because it would be similar to using RCA (SE), and at least in my case no noise comes from that. Now, however, it depends a lot on the shielding of the cable, and whether this is necessary in your situation. In my case I need that shield for sure (which is why I use plain coax).

Lastly, don't make yourself crazy by some cable which "cancels" noise and which further doesn't work at all (no music, or only half of it). Just use the normal cable and find the noise source. Let Paul advise you. :) :)
I think we can be sure that at Bert's place there was no noise, and otherwise he should tell you of course. But you can be sure that your amps were tested under the same condition (same NOS1, no preamp), so it can't be about things being different there. But mains setup ? you bet it will make a difference, so an advice from my side would be to try all the combinations with/without PE you can think of. You're also allowed to disconnect PE from the NOS1, if that brings the better solution, although I am sure *not* saying that it was tested OK without PE (this wasn't tested at all). It just will work without, and it is up to you whether you introduce more or a different kind of noise with it (which noa maybe can't be lead away).

HTH
Peter

It is because I fucked up and know how this issue can be fixed.

Could be but don't know if it helps with noise though, all was dead silent here (97dB speakers) but I do not use balanced cables (RCA).

PE is in the amps nowhere connected to the signal ground, I do not want dirty mains earth on my signal.

In my system the only mains earth (thick wire and very low resistance earth) is connected to the PC to get the dirt out from there to earth as quickly as possible without giving it a change to find another path to follow.

Bert

Thanks for the thoughts Peter. Bert's given me an idea that I'm going to try later this evening. Will report back later...

Mani.

Guys, when I can help (if at all) it is nice that I can. This doesn't mean though that suddenly people's mistakes must be on a forum only to please me or something. It is okay with me (it doesn't bother) but it looks strange maybe.

Everybody makes mistakes of course, sometimes. What about my cooking eh ? haha

Peter

Just a very quick post for now. My noise issues are solved. It was a simple matter of changing a wire from one pin to another on the XLR input.

Thanks to everyone who chipped in... but of course Bert who figured it out, and was actually on the phone to me just before I proceeded with the wire swap around.

The amps are totally silent and sound just wonderful. And the bass... what bass... I'll post more thoughts on sound quality later...

Mani.

Great news Mani - and those BG's have hardly even started to break in yet ......... it can only get better  :) :)

Paul Gerard, hi,

I'm going to build some of your speaker cables to give them a try. I have been reading back through the post and have a good sense of how to make them but I wonder if you could give the number of strands and wire gauge you are using. Also have you used any sheathing to protect the overall bundle ?

Paul if you have any suppliers for the wire etc in the UK that would be great.

Many thanks,

Nick.

Nick - Well first my apologies are due I have not had the opportunity to pop in to see you with my speaker cables and to have a listen to your duos. It would be really interesting to hear your system as it is the same as mine except for speakers.

You are in for an interesting time. I ended up with 4 x .8sq mm each speaker (ie 8 strands for stereo) for the bass units and 8 x .5 mm each speaker (or 16 strand for stereo) for tweeters. Connected in bi-wire configuration. They are very long runs of about 8m. They blew away my old "top end" copper speaker cables. I bought the wire from here  http://wires.co.uk/acatalog/cu_enam.html
and if I remember rightly had polyester coated wire. I suspect it does not matter what the coating is but if I was doing it again I would look at a coating that can be soldered because stripping the ends with a sharp stanley blade is a pain.
But first I would recommend you to lash up some runs of single core enamel coated wires of say 1.5 to 2.0 mm - I was blown away by the improvement in my system - if it does the same for you then it certainly be worth the time to wind up the above spec of wire. The wire is cheap enough so it is just time really.

How to wind them? check this out for a picture  http://www.phasure.com/index.php?topic=1575.msg17981#msg17981  hope this explains it if not clear let me know and I will try to explain.

As for outer sheath most of my cables have none - but near the ends at the speakers I put a sheath of techflex for cosmetic reasons - my gut feel is that techflex does not affect sound quality it is mostly air but you will only need it for cosmetic reasons transformer wire enamel coating is v tough and will withstand major battering methinks.

But these home made speaker wires were only the start of the story for me. I next tried silver cables borrowed from a friend and they were significantly better - BUT there was a question in my mind - did they sound a tad harsh? I think so but no time to properly test. But that was all blown away into insignificance when I installed the ByBee's it is difficult for me to put into words how important I think they are. They completely negated any difference between the copper and silver cables and moved the system to a whole new level.

The great thing is that transformer wire is so cheap so it is worth playing with if you have the time. 8m runs with the construction above is a lot of wire but the whole lot cost me just £30 including delivery and there is loads left.

All the best and I will be very interested to hear how you get on.

All the best

Paul

Paul hi,

I'm really looking forward to trying this. I cannot believe the costs  :).

When you say

"I ended up with 4 x .8sq mm each speaker (ie 8 strands for stereo) for the bass units and 8 x .5 mm each speaker (or 16 strand for stereo) for tweeters. Connected in bi-wire configuration."

Do you mean ?

4 live and 4 return wires per speaker at 1mm diameter (approx 0.8mm sq) for the base

And

8 live and 8 return wires per speaker at 0.8mm diameter (approx 0.5 mm sq) for the tweeters

It just that in the picture the gauge difference looks a bit bigger than the diference between 1mm and 0.8mm diameter wires.

(Keen to get the right wires with so much making up to do. The batteries for my drill are on charge  :) )

Best Nick.

Nick - am away today watching Wasps hopefully avoid relegation and oblivion will check tomorrow and let you know.

Apologies for the confusion. Checking back my original calculations and order was for 0.5mm dia for the HF units but the sticker on the wire reel incorrectly shows 0.8mm dia. I mic'd them and they are 0.5mm dia. Good spot Nick. So this is what is installed: -

For each LF unit 4 x 1.0mm (dia) wires to one speaker terminal and the same to the other speaker terminal SO 8 1.0mm dia wires in total for each LF unit.

For each HF unit 8 x 0.5mm (dia) wires to one terminal and the same to the other terminal SO 16 0.5mm (dia) wires in total for each HF unit.

According to my original calcs 4 x 1.0mm (dia) wire = 3.142 sq mm equivalent and 8 x 0.5mm = 1.57 sq mm equivalent.

More thinner wires for tweeters should be better but then winding them becomes even more of a pain.

Out of interest what is the construction of your current speaker cables? For me the initial quick lash up using bare 1.5mm dia electrical (1.76mm sq) wires (i.e. the earth stripped out of lighting grade electrical wire) gave a really big improvement and if you did the same and got a big improvement then for sure all the work involved in winding the lacquer coated wires will be worth it.

Looking forward to hearing how you get on.

All the best

Paul

Hi Paul,


I guess it will be obvious somewhere, but ... why ?
(don't forget the drill and to charge the batteries)

Peter

This is about the skin effect, electrons repel eachother and go to the outside of the conductor. That is with a perfect conductor, imperfect -real- ones with resistance, also conduct within the conductor depending on frequency. If the conductor is too thick, group delay can occur within the conductor. This is why it is recommended to use conductors less than 0.6 mm diameter in audio, at least for the audible spectrum. If you want a large conducting surface, use a lot of isolated strands.

(source Allan Wrights cable cookbook. A very good read).

Regards, Coen

Haha, thanks.


That's what I meant ...

Well replace "should" with "maybe" or "perhaps". But I just checked some notes I made originally and skin depth at 20kz in copper is around .5mm so that "should" or "maybe" is an ok size for the tweeter wires. On the other hand "maybe" we need to be able pass higher frequencies without skin effects in which case "maybe" or "perhaps" the wires if thinner "might" or "maybe" "could" possibly sound better. The only way to test it would be to try it "maybe". I will not be doing so because I spend my time listening to music nowadays ha ha.

Cheers

Paul

Paul,

Thanks for doing the checks on the wire sizes for me, I will go ahead and order the wire in the confident certainty that all of the theory is well and truly nailed  :) and my sore hands will not be in vain  :)

As Peter says the drill is charging and ready to go.

Current cable is this http://www.vandenhul.com/p_c05.aspx

Best,

Nick.

Interesting looking speaker cable - especially the carbon surrounding the conductor. So the carbon turns the radiated magnetic field back into electrical signal? - which could infer the carbon has magnetic properties mmm a thought worth pondering further. Also interesting that the cable effectively has a single central conductor - assuming the silver coating of the strands prevents any significant boundary effects between the strands.

My old speaker cable http://www.vhaudio.com/speakercables.html had PTFE coated 48 strands of wire. A lot of plastic which I suspected was why they did not perform as well as the lacquer coated home wound replacement. Electrically (R, C and L) the cable was up there with the best but they just did not match the lacquer cables.

Interesting that Van den Hul speaker cables do not seem to address skin effect - so he obviously does not consider it significantly problematic compared with other factors.

I suspect your VdH cables will not suffer so much from "plastic" problems and so maybe there will not be so much difference in your case. It will be interesting to see how you get on.

Maybe we ought to get Peter to pop round the corner and ask VdH about his cable theories?

P

Why not? They are made of many many strands of very thin carbonfibre. I have some stiff vdh internal speaker wire which is made of 5 conductors of silvered 0.6mm copper wire. Most metal vdh wire is quite flexible and made of multiple thin strands and silvered on the outside (better conductivity on the surface of the conductor). Flexibility is obviously important but I skin effect must have been taken into consideration.

If you take the trouble of asking this to him personally I am shure he will reply!

Well i agree with Scroobius that we can debate about the effect of the skin effect. My favourite Spica TC-50 cables where 2mm+ thick copper conductors (originally the core of a high power radar signal cable). Those don't fare well on my Altecs though.

Regards, Coen

I'm not sure whether you know, but the other day you practically ran over him. That's what Paul meant ...

Paul, Coen hi,

I am ashamed to say that the selection process for these cables amounted to seeing them in the corner of a hifi shop in Norwich at an exdemo knock down price. That was in about 1994 so it's taken a while to part me with any more money for replacement cables ( i just spent a truly cheapfi worthy £25 last nick on enamelled wire  :)  :)  :) )

In fairness the vdh cables give a detailed and well balanced presentation, i can see why they got good reviews back in the day.

I did try some audioquest cables years ago that really allowed the music to detach from the speakes (quad esl 63s at the time). I have a feeling that it would not be too hard to get this again, hopefully these enamelled wire cables will do just that.

Coen, I tried vdh first carbon interconnects again a long time ago. They had a really smooth (in the best sense) presentation as I remember.

Best,

Nick

Nope! Any othe (audio) celebreties in the neighbourhood? I inquired about a retipped phono catridge once with the vdh company. To my surprise he was on the phone personally with an elaborate anwswer the same evening!

Regards, Coen

Out from my driveway to the right, 80 meters on the left side.
I only found out after a couple of years living there. :blush1:

Peter - maybe you could arrange to have a music evening with VdH and compare systems?

P

A recent post by Paul in the "PC electrical isolation from DAC" thread reminded me that I really should get the 'super gainclones' that Bert built for me up and running again. I need to replace their transformers with a new pair that Bert sent me, as the current ones tend to hum too loudly (nothing else in my system hums at all). Humming aside, they sounded amazing with my Quads when they were connected. I'll report back once the new trannies are installed (probably next week once I have a bit of downtime from travelling).

Once everything is up and running, I'll also do a comparison between the super-gainclone monos and the Sauermann monos - should be interesting...

Mani.

Hi Mani - I think that would be very interesting but I have to say that for me I would want to start with a comparison between simple GC's i.e. single chip GC's (if I remember yours are paralleled and bridged) and Sauermann and with simple speakers lets say AN/E's for example then move on to Bert's amp with both sets of speakers.

Really this is just a thinly disguised excuse for another round of boy's with their toy's I would be very interested to hear the Sauermann.
 
P

Paul, I'm happy to bring the Sauermann monos and/or my 'super gainclone' monos over to your place if you're interested. Let's discuss over PM.

Meanwhile...

I mustered enough motivation to get the soldering iron out yesterday with the intention of changing the trannies in the gainclone power supplies to reduce the humming I'm currently getting. And then I remembered that I have a 'PurePower 2000' AC regenerator sitting gathering dust in my basement. This has a built in adjustment to reduce humming in trannies. So I thought I'd give this a go before powering up the soldering iron. And lo and behold, I found by using this adjustment I could reduce the humming to a level that was simply not audible a few centimeters away from the trannies (and in the dead of night too).

So, I replaced the Magtech in my home study/office system with the gainclones being fed off the PurePower and started listening. The sound was nice, but a bit bright and thin. Although the PurePower was being fed off the same circuit as the rest of the system, I thought I'd try powering the NOS1 off the PurePower too. There wasn't an improvement in sound though. I then thought I'd disconnect the PurePower totally from the wall socket and power the gainclones and NOS1 from the PurePower's internal battery. The brightness in the sound vanished immediately, replaced by a gorgeous 'organic' sound. But the PurePower had been sitting idle and unplugged for over a year and the battery was dissipating quickly.

Having replugged the PurePower into the wall socket, the brightness/thinness returned. So I thought hell, I'll try the PC off the PurePower too (with an isolation transformer between the PurePower and PC). And...

... WOW! I hit jackpot. With the PC, NOS1 and gainclones all powered off the PurePower, I'm finally getting a sound that is pretty much identical to that from the PurePower's battery. And I love it. I know this because I sat in my study/office just listening to music into the early hours of the morning - the first time in a year or so I've done this.

I'm getting a balance of microscopic detail and total smoothness that is just beautiful.

So in a nutshell, maybe I should in some way be thankful that my gainclones had buzzing trannies (here at least - they didn't buzz at Bert's place). I would never have tried the PurePower otherwise - I had decided that I would never use it for music duties but as a UPS for my work PC because I thought it just made the sound too bright when powered off the mains (I always loved its sound off battery). BUT, I never used it to power the whole system before, including the PC. I don't know, but maybe this inadvertently creates ground loops or something.

I'll post more thoughts of the gainclones once I've listened to and compared things in more depth. But for now I have to say that they're sounding very, very nice being fed directly from the NOS1 and powering my 'small' Quad 2805 speakers.

Mani.

Mani,

I don't think you ever saw it, but in my basement I have a 25Kg (sine) power regenerator as well. It sits there ready for action, but I never use it anymore. Ok, but why the story ?

Already at line 3 or yours I saw the same coming up that I experienced here : power regeneration - nice, but it won't stop any coming in mains disturbances over PE (or maybe the Neutral). So, undoubtedly that is what happened to you, but with the idea that two (or more) devices not on the same PE (/Neutral) explicitly create the problem (of potential difference).
The story is a bit similar to what I said maybe a year back or so : connect your PC to the nice dedicated (audio) mains source as well, or otherwise it's looking for trouble only. So, seems counterproductive (PC's perceivedbly throwing out noise) but I don't think it is (at least not net).

Anyway, recognizable all over ...
Peter


PS: Here too, at using the battery only (mains cord not in) all is fine. Connected to the mains - the thing is on separate PE (pin), but any distortion source will come through anyway (this is about my fameous PWM pump example).

A familiar story indeed - I posted quite some time ago that I originally had my PC on the house PE and the amp/NOS on the dedicated hifi PE. The thinking was to keep to PC hash away from the dedicated hifi PE. But moving the PC onto the dedicated hifi PE resulted in a dramatic improvement.

But I think that was in the days when my amp ground was connected to PE. Nowadays the amp ground is connected to NOS ground which in turn is connected to the PC ground (and PE? probably) so I am not sure what would happen now. The hifi PE is therefore only connected to the metal cases of NOS and the amp for safety reasons. So now if the PC was plugged into the house PE what would happen? the hifi PE and house PE would not be connected together (as they were before). But the hifi ground would become connected to the "dirty" house PE via the PC.

Out of interest how is the PE connected in your GC amp Mani?

P

Hi Paul, we all know that grounding and PE are a nightmare story. My understanding is that my amps are configured in exactly the same way as yours are and the NOS1 too, i.e. signal ground is not connected to PE. But as you say, who knows what happens in other components or the PC?


Generally I would characterize grounding issues as increasing the brightness, glare and edginess of the sound. It can sound more detailed at first, but with prolonged listening you realise that this is superficial detail that becomes fatiguing with time.

Does your 'dramatic improvement' match this?

Mani.

Mani - yes I would say that what I experienced is similar to what you have described. But given that your grounding arrangements with your electronic ground not connected to PE other than via the PC is it just a result of being connected to a "dirty" PE. Well earthing as Peter points out is so complex and is different in almost all installations that it is impossible to draw meaningful conclusions.

It is difficult to compare with my experience at that time because in my case I ended up (without realising at the time) connecting the PE's together. I will try again tonight and see what happens just out of interest. 

Peter's new Phase Alignment (PA) feature turned out to be a big challenge for me and I learned a few things about how to get the best out of a GainClone on the way. Thanks Peter for your help (big help)!!!.

First of all I could not use PA because my GC's do not have any blocking capacitors in the signal path. So the first challenge was to find blocking caps that did not affect sound quality too much. The second challenge was that I came to realise that there is a problem with GC's that I kind of knew about but did not realise how important it was and that is DC offset.

Blocking Caps
The GC's amplify DC (and oh boy do they!) if you stick 300mv in then a whopping 11volts appears at the output to the speakers (34db gain). Wow. Not many amplifiers do that they mostly ignore dc. So the only answer was blocking caps. Having tried a few caps I did not hold much hope of finding any any of good enough quality to put in the signal path. I tried some caps but they were not of good enough quality or I could not get the right values. To cut a long story short I found a quartet of Blackgate 22uF NX Types 6.3v (my GC is biamped). Some would say the best capacitor ever made. So I installed them and to my amazement I could hardly tell the difference with them installed versus no caps installed. Maybe a bit less distinct at the top (straight after install) but very very good indeed - and they can only get better with burn in. I can easily live with these BlackGates. I found this useful in getting some idea of the value of cap needed http://www.st-andrews.ac.uk/~jcgl/Scots_Guide/experiment/lowpass/lpf.html. Peter Daniel recommends no more than 100uF. My GC's have a 22k resistor between input and ground so the input impedance of the GC is 22k and that is the resistor value for the RC low pass filter. If you use one capacitor to feed two amps. (bi amp) then the resistor value is going to be 2 x 22k in parallel i.e. 11k.

GC DC Offset and Phase Alignment
So it turns out that DC offset appearing at the speakers affects phase alignment (it does elsewhere but that is another matter). The really bad thing is to have a difference in dc offset between speakers i.e. if there has to be dc offset then it should be the same for both speakers (and the offset should be -ve i.e. the speaker cone should be pushed back slightly). This is where it gets to be a challenge. There is an inherent dc offset derived within the GC's which exists with no input connected. But when you connect an input impedance it reduces (a lot) depending on that impedance. But then there is also an offset that is derived by amplifying the dc offset that exists on the output of the NOS1 (say 5-15mv). So if you have no dc blocking caps you can get a BIG dc offset at the output. In my case this was 380mv R and 120mv L (roughly) with no blocking caps and this was derived from the NOS1 output dc offset because of the gain of 34db applicable to dc. That is a HUGE offset problem and has a big impact on phase alignment between speakers. It is a problem I have had for a long time that I did not realise.

So today when I installed the blocking caps the dc offset went up because of effectively removing the NOS1 source impedance of 30ohms but went down because the GC could no longer see the NOS1 output dc offset. In the end I decided to rebuild my amps by selecting LM3875 chips from a batch of 12 to find those that matched. I ended up with -80mv L and -80mv R for bass units and -87mv L and -85mv for the tweets. But note that offset is potentially not an issue for the tweets because of the capacitor that is in the crossover for all HF units (no cap for bass units).

After final install that offset changed slightly because the surrounding components are slightly different in the final build compared with my test rig. But anyway I ended up with a few mv difference between drive units rather than the 150mv before. The difference in sound was immediately apparent the image between the speakers is clearer with fewer strange phase effects.

But then I switched on PA and WOW WOW well you will have read the reviews elsewhere.

The message is that if you build GC's then to get the very best out of them you really MUST match the offsets of the LM3875 (or whatever) chips to at the very least match the offsets on the outputs to each speaker. It is unlikely you will get very low offsets (but you may be lucky) but at least get them as close as possible left and right at the speakers. But most important is that you will have to match the LM3875's for your particular install (i.e. with or without blocking caps) in other words the same LM3875's will produce different offsets depending on whether blocking caps are installed and if none then depending on the output offset of your NOS1.

It took me a whole day to do all of this but I have to say it has been well worth it.

So you can tweak yourself completely crazy with this one. Best of luck!!!!

All the best

Paul

Nice Post Paul ,

From what I read it means that your NOS1 is working perfectly well about the new PA feature of XXHE ( plops , etc ) .

Stefano

Hi Stefan - my first extended listen with PA engaged was yesterday and I listened to many familiar albums - and that would be the way to describe the sound - engaging it was difficult to pull myself away and even my wife joined in making requests for her favourites.

Everything here seems to be working perfectly (I have to be careful saying that I don't want to invite trouble) since reloading my o/s. But I do get "plops" sometimes when an album starts playing and maybe that is as the dc offset is applied (I guess) if in unattended it happens once at the start of the album. Other than that everything is fine. I have not started playing with settings yet.

I will catch up on "plop posts" now ha ha.

Peter - the sound quality I get now is STUNNING it has to be better than I thought was possible when I set out to build this system - I mean WAY better. I still have to pinch myself that the sound I hear is coming from mere loudspeakers in a very odd shaped room. Thanks for your total dedication and hard work - it is certainly appreciated here.

All the best

Paul

I must play with setting tonight.

Hi Paul,
I also spend the Saturdaynight with XXHE/PA. Same experience here, even if I am currently using a temporary amp+preamp (blocking DC). I ordered a pair(+) of Gainclones, including caps. I expect these to be delivered by Bert soon.

Cheers
Eric

Hi Eric - Good to hear from you I hope things are good with you. Any chance that Bert could get the dc offset outputs to the speakers (at least the bass units) matched closely? Also if you can bi amp or more not sure I can remember how many drivers you have that is more than well worth doing.

Any idea what blocking caps Bert is using?

Anyway all the best with your GC's it will be interesting to hear how you get on.

All the best

Paul

Paul, is that because you are not allowed to drink a fine Whisky in the morning ? haha

Thanks for all the kind words. But I estimate I learned as much from your little project. But with your great summarize of it from your earlier post in here, I sure hope others can benefit from it as well. And let's say it is a completely logical thing, but who thinks of it ? okay, we did now, but it still took a few dozen emails, right ?

And so we go on and on.

Let me add -although a relative few will read this- that it is the Xth time that a single person (like you, but also others) shout outloud that sound again improved by 100% (the 100% is on my account) or even more (which is also on my account :)). Still it is so. I myself am amongst those persons and while people may claim something like "5 % improvement" (FWIW anyway of course) this is *really* 100% and more.

I think I can say that this is another "fork" of improvement, but why not. Point is (for those who can't understand sh*t of all this) : the further you are, the more easy it gets. The better SQ just becomes grabbable, and as I said about a year ago : it even becomes fairly easy to hear where the improvement should be (or where the remaining faults are). Phase Alignment is such a typicle improvement, although it sprung rom theory. But for those who are using it for a week or so by now ... go back and listen to what sheer distortion you were listening to all the time. Not that you could hear that before ...
All is about the reference.

Ok. So now I should check my own L/R offsets, right ? haha.

But maybe to emphasize what this (last larger post of Paul) is all about :
The fact that an amplifier blocks DC does not prohibit that it may have DC offset itself at the outputs. So, in the case of the GainClone chips this is the most easy to have, because the chips are not equal for it. This is a bit similar to the NOS1 which may exhibit a few mV difference between Left and Right just the same. However, while this few mV is neglectable, a few 10s of mV is not at all and this can happen at the output of amplifiers. Also non-GainClones;
Like Paul told, for the mid-high this won't be much of a problem, because the capacitors in the Crossovers (assumed they are there of course -> not with electronic crossovers !) will block the Offset from there. But not so with the woofers.

There is not so much you can do about it with any normal prebuild amplifier, unless you make a high pass at the speaker side that won't allow frequencies to pass under, say, 0.5 Hz. But this requires huge good quality capacitors (like 150,000 uF) and once you spent the money on them, it is still to see whether it works out for the better. But when you are trying out an amp you might want to buy, go out with your Voltmeters !

Super, Paul !
Peter

There is NO DC-offset at the outputs. A servo regulation makes it 0.


Duelunds  :prankster:

Bert

Interesting to see that the much maligned capacitor is playing a role again , in a place where at least a couple of things are assumed not so correctly
For example . It is incorrect to say that any amplifier blocks dc , aswell as calling capacitors Blocking DC .
Capacitors do Couple for DC .
 
b- a gain clone chip which has I believe at least 60 db (90db?) open loop gain , and some 30 or 40 db of gain is going to waste by the feedback loop and is it comparable to the distorsion of a capacitor ... I mean in Music?

c- we are talking a bit of the PA feature of XXHE , that *for now* can be far more offensive than the vaste majority of capacitors on the planet .
Offensive for its inherent problems ( for now - for someone .... ) about plops at start , or at stop putting up volume over the +zone in db.

and more : AP is not about L and R phase somewhere I did read ( if right of course) and now suddenly and with reason it is ok to predict that the dc offset of the amps should have same values to every channels , for those concerned about the maybe not neglectable 300mV .
.

Maybe I am missing the point .
When and why is it safe to use PA in XXHE judging with the Voltmeter?

Is it really DC ,something that let your woofer excurs for a centimeter when music Stops ? What can capacitors really do about that?


Stefano

There is one thing I forgot to mention I had 3 out of 4 amps "blow" 2 were driving bass units 1 was driving a capacitor protected tweet. So maybe a humongous capacitor on the output of a GC (with none on the input) may not be viable for GC users.

I agree with Peter SQ with PA engaged is a very big step forward. I could not rate it % terms and I would find it difficult to describe apart from "clearer" but anyway I do not need to go back & turn off PA to hear the difference it is as clear as night and day here. Nice one Peter.

It is a bit too early for a whisky for me but maybe later ha ha.

P

Well maybe that is telling us the GC is not good without the input being DC cap Coupled ....and maybe the enourmous gain of it looks toward wild instability if the input sees some of the source changes .... .
The output cap is another story .

stefano

Yes, true DC which will misalign the cone in one direction and even burning the coils (depending on the output power and DC Voltage) if its on there too long (getting overheated while not moving!) but that is the least of the problem...

A capacitor in series with the DC/AC signal blocks DC and passes AC (this AC is making your diagramps vibrate and accepted as making music or "wanted" noise).

Edit: Best spot for this protecting capacitor is at the input. The value can then be a lot lower due to the higher impedance as seen by the capacitor. This position also prevents the amp to become instable when a source outputting DC (should not be allowed if you ask me) is connected to the amp.

Bert

And what if the cap lets your amps response down to say 5Hz? Why do you think the only danger is pure DC ?
Again we must see capacitors as DC coupling ( and filters ) for some reasons of schematic and topologies , not only to block dc and let music play .


Stefano

What has this to do with protecting the woofer from DC?

5Hz is still AC (5 times per second) and not just leaning in one direction (DC) making no movement at all...

If your speakers can't handle 5Hz is something for a complete other topic and therefore not on topic in this topic...or even at this forum.

:offtopic:

Bert

well probably this is off topic , and I pass this . But , what have to do dc coupling or blocking or the woofer being capable of the 5Hz response with the PA feature ?


Stefano

Bert - servo and Duelunds nice

Stefano -

There is no instability in these GC's they have been rock solid. I have been using them dc coupled for a long time with 34db gain - no problems whatsoever and sounding wonderful. No instability at all and there is no reason for them to be unstable (well there could be but not here). The point is simply that the GC's amplify DC so put dc in and you get big dc out. That is all.

What is good though is that my prejudice against caps in the signal path seems to have been misplaced. But if caps are needed they have to be a bit special maybe.

P

Sounds good to me  ;)

With all respect, I did not bring up the 5Hz response, I guess you have to ask this to yourself? I only tried to explain that your 5Hz is still AC and has nothing to do with DC.

Bert

No prob , I am not taking any offense .

;-)

Outside of the main discussion, I sure am able to concur with Stefano. My take :

If we look at the current side of things, the average current at 5Hz we may see will be zero. If only the meter is slow enough. Otherwise it will jump up and down (eh, 10 times per second). Is that bad ? IT CAN"T BE. Why ? because we just "use" our speakers like that. Okay, maybe not at 5Hz but let's say at 30Hz. And as far as they go low enough of course (notice that the current may still be visible at 5Hz, never mind nothing moves anymore - not sure here). So, 5Hz or 30Hz or 100Hz, no difference. The difference should be in the cooling though. So, the higher the frequency, the more cooling.

Now suppose you feed the driver with DC (offset !). Cooling will be the same, but the average current will not. That will be higher. But :
Absolute current will be higher as well. That is, for that part the offset goes to.

Keep in mind : we are playing music. So this little story will never be about DC to the driver in rest (no AC is fed). This is related to the cooling again, which will be zero when we wouldn't play music.

What, net, remains is this :
In order to play withg offset as loud as we were used to without an offset applied, we will need to apply more current. For the peaks to the one side the offset goes to that is. Twice as much. So, all it comes down to is how much headroom the driver has with you playing on your comfortable max anyway. If that is on the (excursion) limits, no drop of DC is allowed. But if that is half of the (allowed !) excursion only, there is no problem. But the cooling problem remains. And you do NOT hear me say that things only get twice as hot when twice the current applies - this can well be much more.

But what it is about - and it will be here where I concur with Stefano, is that when a driver would be allowed to play 5Hz (no DC offset applied), all the time the diaphragm is at one side it will distort as much (but divided by 1.414 or so) as when the DC is applied and it plays "behind" all the time. For this we'd have to understand that no matter what lower frequency is playing, the higher ride on that so this is speaker driver behavior anyway when lower frequencies apply.

Nobody needs to agree with this, but this is what I think. And say.

Summarized, when a driver is capable of receiving say 20 Volts for allowed excursion, it can receive 10 volts for the excursion to one side only and nothing will distort more now if you only think in terms of 500Hz which needs to play nicely on a 30Hz "base" (wave). Peak current will be the same, but to one side only, BUT THE HEAT WILL BE MORE.


Well Stefano, you said "with reason" so this reads a bit contradictionary to me. But anyway, Phasure is called Phasure for a reason and actually all what I do is about phase as such. You know, ALL we hear is about phase only. Localisation, recognition of voices and instruments ... I try to turn this the other way around so to speak; *Create* localisation in air from the source we call speakers. This is in the software and this is in the NOS1. It takes a few years thoug, but slowly I'm getting there.

I will announce it more explicitly on the forum, but next month Bert and I are going to a show, and there we will demonstrate what can be achieved when these aspects are taken to that other level. Half of that is software + DAC; the other half is the speaker. It will really be something different ...

Btw Stefano, my GainClones add nothing to the "distortion level" (THD) of the DAC but the gain (for noise floor). Officially though (measured (eh, by me) including noise) it is 0.004%. Not the 0.1% you read in the datasheet ...

Ok, this was my take. All in secrecy of course and as said, outside the main discussion.

Regards,
Peter

Simply because it will displace the coil out of its center alignment and that will give extra non-linear distortion on its own and it will also limit its max linear excursion.

Think about a 6mm high coil in a 9mm thick magnet gap...move it 1,5mm with DC (not that much DC is needed to do that!) and the driver will not be linear as soon as it starts to move when playing music.

Just my 2 cts.

Bert

Not to debate this Bert (of course not). But all what follows from it is that speaker drivers are not allowed to have any lower frequency applied to them. So all what I say is that since we allow for that, applying real DC is the same (but with all the side notes from my previous post).

The better way would be to use a driver with spades of allowed excursion which just is not used (like say your subwoofer appliance).

So this little part summarized : we agree, but then we shouldn't allow those low frequencies to pass. That is just temporal DC ...

Peter

So Peter "Phasure" did not come from your secret Government work after all then? he he

P

Agreed but when is there 5Hz in the signal or even 10Hz. Perhaps rumble from a wobbly LP gives these ultra-low frequencies on a turntable but with digital sources?

Besides that, the level compared to higher frequencies is always a lot less in dB's and also highly reduced by most  speakers themselves (mechanically restricted by suspension or air load).

I do not see a problem with that theoretical 5Hz signal simply because it is not there or very low in gain; it will hurt less than using a capacitor trying to reduce these ultra-low frequencies (phase shifts already kicking in 4 octaves higher and ringing with very steep filters which should then be used anyway to be somewhat effective...).

Bert

Let's not forget that this is still outside the official discussion (eh, which was ? :)).

So I too didn't make up the 5Hz, but I do say that what counts for 5 Hz counts as well for 30Hz. It is only "less DC" but the theoretical problem is the same. Only less severe.


1.81Hz ? haha


Yea, something like that. But when high passed at 0.5Hz that leaves 4Hz and above to be okay I think.


Now, what was this all about ?
When there is this source device (take the NOS1 for the example) there's very minor DC. Let's say it can't get much lower for any DAC. Practice denotes that 5mV is about most. But common knowlegde always tells that 20mV is allowed.

In the mean time we audio geeks think that DC blocking is not the best. So we don't when we have the chance.
And THIS is the wrong idea and it was nicely written down by Paul. So with his gain of 34 - some call that very high, so we take 20 now, this is DC of 20 x 5mV = 100mV. Is this bad ? I say no. Notice that this is static DC, so it is there without music playing;
Since this is not about the NOS1 at all, and 20mV is well accepted, we now talk about 20x20 = 400mV. This already starts to look less nice.

The new thing is (I think) that nobody says that this 20mV should be there on both sides or otherwise it is way bad. So, this is what I now say and claim. For fun now we use +20mV and -20mV which will be as accepted. But now we talk about +400mV vs -400mV. Well, unless you know what you are doing to this regard, this will be connected like that to the L/R speakers. One plays behind with a tension which meets nature somewat (a slam goes forward) and the other works against nature (the slam is breaked by the tension of the diaphragm which already is in the wrong direction).

Do we expect good sound from this ?
I don't.

The story from practice is next that something like GC chips "produce" their own gain (but theoretically every amplifier can exhibit this) and this too isn't necessarily equal for L/R. This is to be added to the DC already there incurred by the input. And do it gets worse again.

If DC is blocked at the input then at least that part is eliminated, and it is the part which can be gained. Of course, 0 remains 0.

And so I say that DC coupled amplifiers are not allowed, unless you know that the source blocks DC.

It is not said that putting in some nice capacitor of the right value (to begin with) has a net better result. But think :
Suppose the phase anomalies can't be prevented (say it exhibits up to 200Hz), what do you think is better ... the same anomaly Left and Right or that other anomaly which will cancel out frequencies in unpredictable fashion because it is out of control ? I am sure I choose for the former.

It is way more complicated than this alone because most solutions (to random audio problems) compare like apples and oranges. The above is an example of that. One problem is eliminated, but another comes in place of that, and that other exhibits totally different. When this is carefully worked out, we can still choose for the best solution. And this is not a subjective thing. So, no way you'd want different phase behavior from left and right, or worse from bass vs. mid. Only if you subjectively (!) like spatious "music" you'd choose for the L/R difference. But this is because you know listen, and THAT is subject to ... subjectivity.

Once the solution is chosen to block DC, we can always work on the solution that exhibits the best phase behavior. And when this is not my 0.5Hz high pass (effectively 4Hz or so), then work as long as necessary to get it done anyway.
or
Well, you will see. Later (with regards to Paul ;)).

Peter

Hey Peter


Did you mean "dc offset" rather than "gain"?

P

Yes I did !

Apologies ...

I concur with Peter,

LF below the cutoff frequency of the loudspeaker is probably worse than DC. even with 1Volt DC across the terminals, with an 8 ohm (nominal, 7 Ohms DC resistance)  this translates to 140mWatt of power. This is not going to blow your amp on transients.

LF below the c.o. frequency is not well damped by the speaker (if at all). This means high excursions wich is bad for linearity when playing loud, limiting the peaks and not to forget gives rise to doppler effects, modulating the signal.

regards, Coen

Hi,

My system has too much gain and I wanted to ask for comments and advice about losing gain with the least impact on sound quality.

To connect my NOS to my gainclone amps, as I am using phase alignment, I have 5.6 uf capacitors on the inputs to my gainclone modules. The input capacitors work with the 22k ohm input load resistor of the gainclones to give a high pass filter with a -3db power point of 1.3 hz (optimal point for response down to 13hz). All good so far.

The problem I have it that in my system as a whole the 2v output level of the NOS combined with the 28db gain of the gainclones and 105db/m/w response of my speaker means that I have about 15db too much gain in the replay chain.

This means to set a reasonable playback level using only the XXHE software  volume control to generate a "loud" listening level I have to play at a -30db volume setting in XXHE. I think that the 32 bit based volume calculations and 24 bid DAC have the headroom to play at this level but the issue is that -30 db turns the 2 volt full scale NOS output into only 6mv full scale output from the NOS ! I can hear that at this signal amplitude by the time the signal has traversed interconnects and gainclone input caps, sound quality is being impacted. I have been using transformer volume controls (TVC's) to "soak up" some surplus gain but I am very keen to take these out of the system.

My question is dose anyone have any ideas about how I can lose about 15 db of signal level. Ideally I would do this after the signal has been passed through the 5.6uf input caps if the gainclones. The obvious way is to pass the signal through a potential divider with resisters chosen to lose the gain. I have been trying this but so far have not had great results as far as sound quality is concerned (this might just be the resistor values I have tried so far). Does anyone have any ideas on a good way to lose 15 db of gain whilst maintaining sound quality ?

Thanks for any suggestions.

Nick.

Hello Nick,

I do not have the values and numbers of the resistors for you at this moment but it is fairly easy to change the gain of the clone to lower values. If I remember correctly you can't go too low as the amplifier would become unstable but a fair range of adjustment is possible. Probably this was discussed on DIYaudio.
Just my 2 cts
Leo

Leo, I guess that doesn't help much because it seems (to me) that Nick is missing some point in the "Gainclone basics".

So Nick, assumed I'm correct on this (which feels almost impossible - just saying) the Gainclones operate with gain resistors. See the datasheet. So, adjust those - done.
And *then* what Leo told applies.

Does this help, or was it only stupid of me that I thought you overlooked the obvious ?

Regards,
Peter

Hey Nick - you can set your gain to whatever you need without affecting quality. The circuit diagram of your amps is below (I think!) and the gain you currently have will be:

Gain is  =1+Rf/R3, so 1+(22.1K/680)=33.5dB

You can change Rf and R3 to give a lower gain without affecting quality.

I would have thought best to change Rf and leave R3 at 680ohm. If Rf is the same spec i.e. Caddock as in the original kit then quality should not be affected.

Cheers

Paul

Nick,

You really can't lower the gain of the amps to any extend. Iirc the datasheet tells that below 20X (26dB) the chip becomes unstable. You win 7dB at best.

A additional resistor devider at the input is easily calculated but introduces some unwanted series resitance, noise and minute distortions. I would go for a low as possible impedance that the opt of the NOS allows (1K?).

Another option is the transformer route. My opinion is that is can be done at very high cost, but some may argue that it will never work.

Both are a bit like introducing a fixed pre-amp in the chain.

If loud is -30 dB you might consider an unity gain amp (a buffer). I don't know if these a commercially available.

Otoh I am shure lowering the gain somewhat will work. Peter demoes on very sensitive speakers (his own and Berts latest) and i've had read no reports on having too much gain.

Fyi I have similar gain issues with my tube amp and have not found another solution that to redesign the amp with lower gain.

Regards, Coen

Things are not completely comparable. IIRC my Gainclones are 30W and utilize a gain of 21dB. Can be 3dB less at least (I once had it like that).

Next I will play easily with -30dBFS (but -34.5 would be minimum by nature).

Lastly, nobody is really bothered by my playback level. That the room is some 290m3 does matter somewhat but not all that much (because of the directivity of the horns).

But where it all starts is my gain of 21 which is siginificantly less as the 38 I think I read from Nick's.

Peter

Leo, Paul, Coen, Peter,

Thanks all for your help with my gain question.

It's interesting that pointers are towards increasing feedback to reduce active gain and using passive attenuation, looking at it I think I may need to use both to drop 15db.

The gain factor of my gainclones 33.3 or 30.4db (thanks Paul). As Coen pointed out the LM3775 according the data sheet should not really go below 26db, so changing the feedback resistors could yield a 4db reduction. I'm worried that I may end up listening to a lot of negative feedback, but its easy to try so this could be option 1.

Option 2 is to apply passive attenuation on the inputs. The choice here would be to apply the attenuation before the input caps or after the caps and before the Gainclone modules. I guess after the caps would be best so the signal level is higher as it traverses the caps. The problem here is that I have just bought these rather nice (and expensive) mundorf silver gold oil caps which are the correct size for the standard gain clone module 22k input load. I like Coens suggestion about keeping the "input load" of a resistor potential divider as low as the NOS output can handle to keep series resistance of the input as low as possible. To lose 9 db on the input I could go with a potential divider of 820 and 270 ohms. The problem is at these values I would need input caps of 100uf a bit bigger than the new 5.6uf  mundorfs. Hummmm a bit to think about here.

Thanks guys for the help so far.

Nick.

You're welcome!

I would first try to increase feedback on the amp and assess then if you are really in need of more attennuation.
The way the LM works is that I think a little extra feedback will not hurt sq (au contraire).

I see no reason to be afraid of low signal levels through the caps. If you need the divider please put it before the cap, you can keep you precious gold cap for minimum distortion. AFAIK larger voltages across caps can cause audible minute deforming of the dielectricum, so I'd rather keep them small.

regards, Coen

Nick - this is a quote from Peter Daniel on the Audiosector forum about gain maybe As discussed above R1 will affect SQ but its easy enough to try it. Also I suppose that R1 at 22k may affect the L/C equation for the input caps but again easy enough to work out.

Quote
It's best to change the gain with R3, but I wouldn't go over 1k6 or so. If you need additional gain trimming, increase R1 (22k will give you -6dB gain reduction)
Unquote

Maybe 22k input R will not be ideal given the low output impedance of NOS1 - any comments Peter?

Am looking forward to hearing your system. After all the mods I have made to my system recently I have some spare amp boards (populated and burned in) I measured the quiescent offset voltages so if you measure the offset voltages on your amps we may (but we will need some luck) be able to find a combination where the Left Right offsets are reasonably well matched (at least for your bass units) which I assume (?) do not have capacitors in series with them in the crossover. I have matched my amps to within 5 mv offset for the bass units and it was worth doing (previously I had about 60mv offset difference) removing "confusion" in the stereo image and just sounded more coherent overall.

Paul

Coen,

Thanks again, I'll start with the feedback loop. I’ve been messing about with zero feedback valve amps for a few years before the gainlones came along so the idea of adding more feedback didn’t sit too well. It’s good to know that more feedback in a gainclone could be good for sound quality  :).

Also its interesting what you are saying about putting the passive potential divider before the caps.

Time to get some parts to give it a go.

Regards,

Nick.

Well, why not ? It almost looks like you want it less ? (knowing that 47K and up will be normal for main amps)
Anyway, less is OK just the same. But I'd stop at 1K to be on the safe side (NOS1 is ~33R). And otherwise you'd hear it on the too tinny sound.

Nick, by all means try to avoid any (more) resistor in the signal path.
Also, the application of a feedback resistor is a most normal thing for these amplification means. It would be true though that there's an optimum for THD (and R value). Well, that's what I think. And when it is, it should be in the datasheet.


I agree with this. So, change RF as long as you can, before touching anything else.

Peter

For anyone interested this is my latest Gainclone layout.

As you will see there are four amps for bi amping. Recently I rebuilt the amp modules selecting LM's to give small offset between channels (all in the interest of better phase alignment). I took the opportunity to simplify the layout of the amps by removing the aluminium heat sinks and bolting the LM's direct to the case. There is a little more internal wiring now but not much. The aluminium sinks were only there originally because I was lazy having used the Al sinks to breadboard test the amps first I just transferred them straight into the case. Much better without them.

One important thing I removed the plug and socket I used to have in the power cord that runs from the power supply to the amp. The reason is that if one of the power lines is lost to the amp 35V can appear across a speaker drive unit. That was a risk not worth taking so everything in the power delivered to the amps is hard wired.

The SQ nowadays still surprises me and I have to say when I look at the amps and think how much they cost to build I do smile from ear to ear as they "wipe the floor" with much much more expensive amps.

Hey Paul,

Thanks for posting the internal shots, we have ended up with very similar layout of the modules but in the next version I build am going to put the transformers in a separate case like yours. Good point about hard wiring the psu connections I think I will do that too.

Now i have to work out how to get a good layout with 8 modules and 4 transformers  :grazy:

Cheers,

Nick.

Paul,

Stupid question : are those transformers loose in there ? And if not, how do they stick ?
But if yes, try the sound when putting them on top of eachother (but have some pad in between them).

Regards,
Peter

Peter - Ah yes the traffo fixing well they are sitting on Blue Tack and you would be surprised at just how firmly they are in place. And NO FOR SURE I WOULD NOT CONSIDER courier transporting the amplifier anywhere with that arrangement (I thought I would say that before you ask ha ha). But they have been to visit Nick in my car like that without any problem. It was always (and still is) the intention to fit the securing bolts to the Traffo's but I just have not got round to it yet. I left them out because I was trying different locations for the traffos to see if I could hear any difference but I could not. I did not try one traffo on top of the other though and I will try that tonight.

So for everyone out there tempted to build a GC it is strongly recommended to fix your traffos in place with the supplied fixing kit.

I will try mounting one traffo on top of the other (just like in NOS1) but the amplifier case is not big enough to do that and get the lid back on. So I hope it makes no difference to SQ or I will be looking for a new case.

All the best

Paul

Nick - you mentioned using 4 traffo's and I guess that will be one traffo for each amp module. I considered doing just the same when I built mine and discussed it with Peter Daniel. He said that 2 traffos is a good idea (good for SQ for sure) but that 4 traffos would be a waste of resource and would not improve the SQ. In fact I seem to remember he mentioned that 4 traffos in his experience increases the chances of earth problems in a bi amped arrangement and he does not recommend it.

It was a while ago and you may wish to drop him an email to confirm that.

P

I am not sure why you put this text here, but do notice that the bolt which is to be there in the middle IS part of the whole "construction". It can even be dangerous to leave it out. Think of the bolt acting like a coil. And for example, it is not allowed to fix that bolt to both the bottom and the top of the chassis (not that you would do that, but that will be esepcially dangeous).

Peter

Paul,

Thanks for mentioning this, I was puzzled about how to tie the signal grounds using two transformers to power the modules for one channel (eg say left side speaker). Using PE it is possibly easy to establish a "good" sinal ground but poor for sound quality. In our setups with signal ground established by the PE back in the pc it could be more difficult to iron out any differences that two transformers in the same channel try to exert on the singnal ground.

I may just go for 2 x 500 va transformers to cater for the additional modules.

Cheers,

Nick.

Nick,

If this is going to be for a balanced setup, I think you will need separate secondaries for each "PSU" (plus separated from minus). Then still it may not work because the secundaries have to be isolated. I guess they normally are, but when not ... they should. With separate transformers for plus and minus it will always work.

Peter

I think Nick will be simply bi-amping but anyway my Airlinks had separate secondaries +22v  0v  0v  -22v  so would be good for balanced.

P

I jumped on the bandwagon too and build up my first gainclone ;). It's based on the integrated LM4780 chip with the hope that the specs of both amps are very near to each other. That works at least for the offset voltage that is exactly the same for both channels. It's done on the expense of channel separation but that's the decision. Input coupling capacitors are build up of 10 NOS WIMA MKS caps parallel soldered with a total value of 3,3 microFarad and the chip is connected with Dale instrument shunts. The power rails are fed by a  toroidal transformer build up with 4 Vishay Roederstein EYH Caps 6800 microFarad each, Elna Silmic 2 ROS caps 1000 MicroFarad and 100 MicroFarad Silimic 2 RFS near the chip. All signal cabling is silver (my wife is a part time silversmith  :)) and power cables are of copper. The chip is cooled by two stripes of massive copper. All is build in an aluminium housing made for electrical installation.

The system is burning in but shows it's potential by now.  8)

Georg

Hey Georg

Sounds great and built with some nice components. A couple of things that came to light here that may be be of interest.

I changed the caps next to my modules from 1000uf to 1500uf and was supprised at the improvement in balance and control of the sound.

Also I suspected field interference from one of my transformers which are R-Core types and currently in the same box as the gainclone modules. The transformer was turned 180 degrees and the sound was really improved.

Might be worth playing about with these variables whilst the soldering iron is hot

keep is posted on how you system sounds as it runs in.

Cheers,

Nick.

Hey Nick,

thanks for the hints  :) - the soldering iron will be hot for a while. I will change the caps after burn in. Eventually I can get some blackgates for a while, the next weeks will be thrilling!

Cheers,
Georg

Hi,

I finally got around to starting a gainclone MK II build and thought I'd post some pics of the build so far.

As you can see the modules and rectifiers are built (sourced from Peter Daniel) and I'm close to completing the case fit out. Next stage is wiring up then testing.

The final result will have balanced inputs and bridged and bi amped outputs, so it used 8 gainclone modules in all driving 4 output channels with each output having a pair of bridged modules. It's dual mono back as far as the mains input lead.

Really looking forwards to seeing how it all sounds.

Cheers,

Nick

Hey Nick - Looking great so far - those lumps of copper (?) look good!!  Balanced amps should be better from a noise point of view. When finished maybe you could bring them here and we could have another session.

I have spare 4 amp modules so I have all the parts (also the same amp cases) to build a balanced myself. Perhaps you could bring them here when you are finished and we could compare balanced versus the stock version?

P

Interesting pci card on the top picture!

How is the power going to enter the amplifier module? Theresoes not seem to be a lot of space left on the "backpanel"?

Nice wrk btw!

Regards, Coen

Hey Nick, nice !!

Maybe a possibly important tip for you :

We already know about the necessary matching for DC offset between the chips. Well, I think that when setup differentially this is even more important. Besides that, equality of the level (output) of both plus and minus will also be crucial. So :

Either DC Offset which is not the same, or the output level which is not the same, will create anomalies which may let fail everything. Not only you'd have distortion but also the noise cancellation will act strange (when both signals are not exactly equal).

When you can get it done, the improvement should be vast. At least it is totally easily measurable, so it should be audible just the same.

Jaleous Peter


PS: Do I see 8 secundaries in total in there ?

Hi Peter,

Thanks for the encouragement and the advise on offsets. I had been giving them some thought it looks like i have to setup a test harness to power up each module measure offsets and match pairs of modules.

Yes there are 8 seconderies overall, I was keen on having each output electricly a monoblock as far back as the transformer iron. The ratio of two modules to a supply seems to work very well so to keep to this and limit the transformer count to 2 I had these custom wound with four secondaries. They are 625va which keeps the ratio of 300 va for each pair of modules.

Of course this is just a load of theory until i can take a listen :dntknw:

Cheers
Nick.

Paul,

Im definathy up for a visit when the amp is finished. It would be great to try it though your system, provided it dosen't sound awful, there are a lot of assumptions in the design right now :)

Nick.

Coen hi,

You made me smile asking this, I have two options worked out and I'm trying to decide which to go for.

Option 1 is to run an umbilical into the back of the amp module. I left space  to mount a cable gland on one side of the speaker posts. Internally i was going to run the 16 ac power lines to the rectifiers through 15mm plumbing copper tube (earthed to PE ) and under the amp modules. The issue is the possibility of ac hum from the wiring so close to the modules and the cable run is long at 1m in total from transformers to rectifiers.

Option 2 is to run the ac psu wires through the side of the PSU and amp cases just behind the front panels. To do this i would link the two cases with bolts and spacers (bolts are in the pictures). This approach gives a very short and direct  run for the ac wiring away from the modules. If i take this option was planning to insulate the fixings that join the cases, I don't want case hum being transferred from the PSU Case to the Amp case, Iv had a real problem with case and transformer induction fields in my current one box amp. The two cases will be electricly connected but only via their binding points to PE.

Option 2 is my favourite at the moment.

Nick.

Is it a most stupid idea to put the rectifiers and all in the Supply case ?

IIRC I had that conversation with Peter Daniel and he recommends the rectifiers to be next to the amps. I think it was to keep the earthing arrangement inside the amp as short and simple as possible.

I have a very similar arrangement to Nicks (just 4 amps though not balanced  -  yet!) exactly the same boxes. I have an umbilical arrangement using the back panels (and it was a pain to install). If I was doing the same again I would probably opt for option 2.

P

Peter hi,

I had stayed away from putting the rectifies in with the transformers in part because of the reasons Paul gave regarding the earthing but also because of a problem I had with my current amp build.

In my current amp, the rectifier boards are close to the transformers also when I built the rectifier boards I fitted the 10µF capacitors that Peter Daniel supplies (they are optional). Well, I spotted that PauIs amp dosen't have the 10µf capacitors fitted so I snipped them out of my amp. What a difference it made to SQ, much better. I figured that the 10µF capacitors were either resonating very badly, which seems unlikely, or they were picking up hum field from the transformers. The clincher for the transformer hum theory was when I tried moving the physically large Mondorf input capacitors in my amp off of the aluminium case bottom (just placed them onto some 8mm foam) this gave a MASSIVE lift in SQ and fixed an oscillation problem when the amp was on without anything connected to the inputs.

So within reason I'm very keen to keep the transformers well isolated from the amps.

A long explanation but hopefully it may prove helpful if anyone is doing a build.

Cheers,

Nick.

EDIT I quoted the capacitor size incorrectly as 47uf above. Now changed to the correct value of 10uf.

Haha Nick. I figured that you would (no, could) come up with those 47s as they are related to the DC protection. Now, I was reluctant to say that you'd need output elcos inside the supply case while maintaining the input ones in the amp case.

But since you now bring it up this way, I thought to mention it anyway.

The proper earthing ? sure. But "as short as possible" - not sure. You can almost think like : when too long, make others more long (just create the path you want which goes over the shortest, same wire thickness assumed).

This too I didn't bring up, because I personally think you may encounter massive problems anyway. The oscillation and such.

In the (very) far end the two boxes may imply more problems than it solves. At least in my case (built by Bert and far from a PD design) I can't see things wrong, or it must be an added small third harmonic. NOS1 same story. The only thing important there as far as I can measure, is the output wires not being allowed to be close to the 230 itself (with a "sorry" for the position of the power inlet vs the outputs :)). However, there *are* these compartments of course.

Influence of the internal (9 and 12V) AC ? I never saw anything from that. So to that regard it should be OK to have the AC in there.

But you have a nice o'scope, so you can trial and see !

Peter

Peter - just out of interest is your amplifier balanced?

No Paul ...

Nick - I was just perusing the Peter Daniel forum discussing SQ and found the following quote from Peter (you may have seen it but I thought I would post it anyway).

"Please note, that if you have balanced source, the bridged GC will be better than regular one, especially when it comes to dynamics and bottom end extension. But you need speakers that are at least 6 ohm.

The bi-amping is also a big advantage, and the GC will sound only as good as the rest of your system, if it sounds bad, the fault is usually somewhere else"   ;-)

Initially when you posted this I have to say I had my doubts because each GC amp will be looking at 4ohm load but seems as if it should be OK. So now I am even more interested to hear your amp in my system because I am starting to eye up my soldering iron again.

Best
Paul

Hi Paul,

I had seen Peters Daniels comments some time ago when I started with the bi-amp build. Peter comments do sound very promising for bridged / bi-amped though don't they.

My speakers are quite hi nominal impedance, I think at least 8 ohms and could be higher IIRC. Although bridged modules see half the nominal impedance of the speaker they do only have to deliver only half the output voltage level compared to single ended. Also because our amps are bi-amped they must be seeing higher impediance than the overall nominal speaker impedance because they drive the individual speaker drivers which will have a higher individual impedance than the speaker overall.

My main concern is driving so many modules ("8") directly from the NOS1 outputs. In theory the loading of the bridged design per NOS1 output op amp phase should be the same as our current bi-amped builds. I guess that question will have to wait for listening tests.

I have made some progress with the amp, I took the option to join the two cases to route the transformer secondary’s to the rectifiers (option 2 above) and it looks very nice and neat. Next step is wiring up individual modules to pair them into low DC output offset pairs and then start final the wiring up. I think it may take quite a while to wire up but I am keen to take my time and get it nicely done. Have to say I am REALLY interested to hear the end result.

Cheers,

Nick.

Edit fix unwanted smiley

Nick - What spec of transformers did you order in the end? rating? any ES screen? any special windings?

Paul

Hi Paul

They are 625 va AirLinks with 4 x 22 v secondaries. They were custom wound, as you know standard they have 2 secondaries. I did not bother with the primary secondary shield or GOSS emi shield options, your AirLinks seem to work rather well without these options ;). The price for the custom wind was close to the standard 625 va with 2 secondaries.

Cheers,

Nick.

Hi I spent 18 hours wiring up the Mk II Bi-amped bridged gainclone amp this weekend, I was determined to get it finished.

IT LIVES !  :)

 :NY01:

I'll post some pictures of the build the finished amp together with thoughts about sound quality soon, but for now first imprisons are VERY good,, a game changer, and that is with just a couple of hours ruining in. Definitely worth all effort.

Time to sit down with a glass of wine and listen to some music   :)

Nick,

Looks like a drawing. So beautiful Nick ...

Thanks Peter,

It ended up looking a bit more "complex" when the input caps went into the centre between the amp modules, but the sound is in another league which has made the effort worth it  :).

Nick.

Nick - your amp is looking very nice!. So much so I have succumbed to temptation and have started ordering parts to build one myself!!. I probably have enough parts here to convert my existing amp into balanced mode (except transformers of course) but that would mean an unacceptable period of time without music. Also I think I have a buyer for the existing amp. so I should recover costs.

The layout will be slightly different (I just need to double check I can do it though) and I will be using Blackgate NX decoupling caps as I am using at the moment.

Looking forward to dusting off my soldering iron. 

Do you find the balanced amp generates significantly more heat?

Happy days!!

Paul

Paul hi,

Regards the heat it runs warmer but not that much. The playing at -30db out of the NOS (which is louder because of the balanced design compared to single ended) the amp just gets warn, say 40 deg C on the case sides. The copper heat sink arrangement in the Modu case seems to work very well so the chips are the same temp as the case.

The NOS1 appears to have no problem driving 8 amp modules. I think using your Black gates as input caps would be a good move, the Mondorfs I have are only 5.8uf and are driving two amp modules per capacitor so they are theoretically marginal for the -3db low pass point, but I have to say that listening the much stronger, snappier base is a big feature of the sound so the 5.8uf caps are ok for now.

I would be tempted to build new a new amp myself. It would be worth having a chat off line, I spent quite a while working out the details of the signal earth, heat sink and setup the case work / wiring. If you use the same layout I can help with some ideas and pointers on materials etc.

Besides the sound quality, the amp behaves in a very civilised way, absolutely no clicks or pops turning on and off and very very quiet (save for a small amount of transformer hum from one of the transformers, need to think if this is worth changing out the transformer for).

The amp has a very valve like warmth and presence but with detail and extension at both ends. I guarantee you will not regret building one for your system (except the hours and hours it takes ha ha  :) ).

Happy to pop down with it if you want to try it in your system.

Speak soon,

Nick.

Nice job Nick!


And I'd be happy to bring along my BD-Design monos if you'd like. Could be an interesting gathering...

Mani.

Mani - Nick I will check diary and get back to you later with possible dates.

Looking forward to it already.

Paul

Sounds like fun Mani, that would be great.

From the internal shots of Berts amps its looks like there are some very fine components in them. It will be really interesting to see how Berts parallel bridged design with input buffering (I think) compares to the Bii-amped bridged approach I have taken.

I am sure there will be some nice sounds  :)

Regards,

Nick.

PS could you bring any balanced interconnects along that you rate ? At the moment I am using a set I make up for the new amp which are loosely based on a VH Audio diy recipe that Paul put me onto, but I'm keen to hear other good cables.

At last the parts are here and I have now started to build a bridged biamped GainClone similar to Nick's. But there there will be some differences in layout. I will post pictures as I go along. Nick bought his balanced GainClone here yesterday and both Nick and I will post what happened during the day. Suffice to say my enthusiasm to build is even more than it was before now that I have heard it!!

The following is a picture of the amplifier so far!!

Nick and I have learnt from each others experiences building these GainClones and we will post of summary of all the things we consider to be important at the end of the build.

Can't wait to get my drill out and start filling those boxes!!.

Regards

Paul

Nervous times - settling on the layout of the amp is the "point of no return" when holes are being drilled in the cases. Anyway the supplier of my copper heat sink let me down and so I decided to go ahead without the copper and bolt the chips direct onto the case sides. That will make wiring later more more difficult but doable.

Just a little concerned that passing the ps wiring through the side of the case is not ideal because it is after all a heat sink. But there is little option really - I used 10mm OD rubber tubing to protect the wiring through the 13mm transit holes (the rubber expands when the wiring is pulled through). I had to use talcum powder to get the wiring through the rubber tubing as it was a tight fit but not too difficult really. The advantage is that the amplifier smells great!!.

Drilling big holes through the profiled sides of the case is not easy great care is needed.

By the way I drilled and then tapped the holes used to bolt the chips to the case side because it is difficult to get access to the back of the case side and I will want to remove those amp boards later to match offsets. Also I really like running a tap through a hole - it is just one of the jobs is very satisfying!

Regards

Paul

Power in and working - the transformers will shortly be bolted down Peter!!

All the hardware finished and started to wire up amp modules. Amp modules have been selected to minimise offset voltages at the speaker terminals. Lucky that four of the modules had offsets of 54 to 57mv so offset voltages should be very low of the order of 1 to 3mv (hopefully!!!) to the LF units. Offsets at the HF units not so important but even so will only be of the order 15-30mv but as there is a capacitor in the crossover that is not an issue.

Another super nice "drawing" appeared !

Nice job so far Paul. Super nice ...

Earthing is the next challenge....... watch this space!!!

Paul,

It's looking very nice, not far to go now ! Really looking forwards to having a listen to your finished amp. There are subtle differences in our builds, will they be audible I wonder ?

Cheers,

Nick.

Power lines wired up to amplifier boards and earthing wired in. Last step to come is to wire up speaker binding posts and XLR's. I have to say this has turned out to be a much bigger project than I had anticipated - if I had not heard Nick's amp at this stage I would have been wondering if the effort would have been worth it. At least I know the answer to that!!.

Hey Paul,

Assuming you caught some sleep last night - thus while you still have a chance : Make all of your output wires exactly as long. ;)

Peter

At last - another early morning start and the amp is finished. I totally underestimated the work - this was a big project eight amplifiers in one box is a lot of wiring and when you try to make the wiring neat the time increases a lot. A check of the offsets showed just +3mv and -4mv to the LF drivers. So how does it sound? well no surprise just like Nick's when he brought it here. The main difference is the sheer effortless presentation of the music, the sound is bigger stretching beyond the speakers bass extremely well controlled all clearly delineated. The difference with a balanced setup is not subtle at all. The only downside at the moment is that the sound has a slight rough edge but the amp has just been switched on and so needs some running in. Also the interconnects I made use transformer wire and I will be rebuilding them with fine silver wire which is what sounded best when Nick came here. My old setup used the fine silver interconnect and it makes a significant improvement to my system.

A big thanks to Nick for the inspiration to build this amp - the SQ is really top notch. Ros and I will shortly be going to visit Nick and I shall be taking the amp with me to compare with Nick's. That may sound odd but we have found it really beneficial to compare our amps it is the only way to iron out any small problems that we would not otherwise find. We will post more after that visit.

If you are thinking about building a balanced bi amped amp do not hesitate do it you will not regret it (except the amount of work perhaps).

If anyone wants to know more information do not hesitate to post - Nick and I have been through the "pain" and we would be happy to share our experiences if it helps.

Paul

PS - Oh just one thing - when running at high volume (-21db with my 98db sensitivity speakers) and with a highish ambient temperature the heat sinks on my amp get very hot ~ 45deg C so I will have to look at improving that. Nick does not have a problem with his 115db speakers. I am using the whole case of my amp as a heat sink. Each amp is now seeing 4ohms output Z so significantly more current being delivered to the speakers.

PPS - Also I started off with earthing arrangement as I had previously i.e. with L and R channel grounds not connected to each other and not connected to PE - the result was that there was some "mains hum" through the left speaker and less through the right speaker. I then connected L and R channel grounds to PE - the result? total silence no noise at all through either L or R speakers. So the grounding requirements for balanced amp is different - not totally surprising but the only way to find out is to experiment.

Watch this space for more info ..............

Hey Paul, remember this one ? http://www.phasure.com/index.php?topic=642.msg14360#msg14360 (bottom picture in that post)
I counted 75 pieces of cut wire there, and this was from the final assembly of one NOS1.

But your end product looks nicely tidey and neat !
Good result on the DC Offset as well (eh, how couldn't you hehe)


So, N&P Productions ... is it actually common these days to see GainClones in a balanced setup ?

Funny that I was discussing the very same with Nick recently and it appears that there are very few balanced / bridged GC's out there. Now that I have heard the difference I would only consider balanced. But of course the problem is that there are not so many source components that offer decent quality balanced outputs. So the only option is to install transformers and then of course you have to start spending a lot of money to get good quality - and even then you are putting something extra in the signal path. From where I am sitting (which is in front of my speakers listening to the Duke in mono ha ha) NOS1 with balanced output into balanced bi amped GC is very special indeed.

On the offset both Nick and I were lucky to be able to get such good offset results. But of course balanced configuration should always work out good because the net offset at the speaker terminals is the difference between the offsets of the two individual amps whereas in a normal config you are stuck with whatever is on the output of that amp.

Happy days - my amp is starting to relax a little now and losing some of the "edge" noted when first fired up.

It would be good to get someone else on board - is anyone out there thinking of balanced GC? if so let us know.

Cheers

Paul

I just re-checked mains hum levels in the dead quiet of morning and actually there is some mains hum through my speakers. It is at a low level but it is there with slightly more through the LH speaker. Just the same as Nick's amp when it was here but I cannot comment on the absolute levels (something we will check when I visit Nick soon).

So earthing now is with both channels connected to PE in a star. There is a potential loop with this because the USB input to NOS1 is connected to PE also (but what does this mean with balanced?). So I tried a number of different earth arrangements breaking that loop but the best arrangement is what I have now. It seems that the mains hum is probably not caused by earth loop but by hum breakthrough from the bigger nosier transformers.

From work on the previous amp and Peters input I know that any noise affects SQ because hum/noise rides on top of the signal (not under it) and modulates it. The next thing to try would be different layout inside the transformer box (one transformer on top of the other as in NOS1) and to locate it further away from the amplifier box. Big work and I am not sure I want to do that but for anyone starting a project like this it would be worth considering.

Paul,

Assuming this hum is 50Hz mains frequeny, I would consider the transformers to be too close to your further components. Actually they are in my view.
Point of course is that you can't move them away, hence you made it so nice and tidey that the supply box can't be moved away even for trying. I didn't look at Nick's at this time, but he may have the same problem. IOW :

Once you grant yourself lengthening the AC wires from the transformers (which should be harmless I'd say) you can do anything with that extra length. Does it help, then you can always eliminate one connection by resoldering all to the rectifiers. So, now cut it all just inside the amp box and insert 30-50cm there just to try. BUT you will know how hard it can be to "strip" this wire and solder (etc.) it, because of its coating. So on that matter maybe resolder at the recifiers end to begin with (which allows you to put all back in original state when it does not help).

In the mean time this will allow you to stack the transformers for trials.

A first and more easy thing to do might be laying around the wires from the switch (well, the blue/brown you have on top of the one transformer). Although I don't have the experience that such a thing is not allowed, I would never do that. But indeed, I would also never put the transformers next to eachother like this.

A last easy thing to try may be a sheet of mu-metal in between the boxes. So, detach them and put the sheet in between them. But have it twice as high as the box.
If you can get it, that is, and the internet is your best chance (then easy). Expensive stuff though.
Remember : mu-metal (or anything) does not filter; it can guide away (magnetic) radiation. Bounce it into another direction.

... if it is from transformer radiation at all ...

Peter

I don't know whether this can be of help :

I can't value your remark about the USB Input to the NOS1 being connected to PE. I mean, on the NOS1 end it is not at all. But if you mean via the PC, then you will be correct. But this is out of my reach of course.
But if you think this is of importance - disconnect the PC from PE.
I don't think it will help - up to creating other problems (DC Offset on the NOS1 end ?), but it is worth a try.

FWIW remember : I have my PC connected to the audio ring just the same, it is connected to PE while my GainClones are not and the NOS1 is not connected to PE as well. And, nothing in the NOS1 is connected to PE for ground; only the cabinet is. Still, connecting the cabinet to PE (hence connect the NOS1 to PE) makes differences all over the place (the cabinet is an antenna and whatever it catches now goes into PE).

Test all without your monitor's switching block connected to the mains (pull out the mains plug). Just to be sure.

Peter

A few thoughts, and a question.

Be careful isolating your pc from PE, beyond the obvious safety points, I did this a long time ago and found that the singnal ground in the pc rose to 79v as the smps drifted upwards. As soon as the pc was connected to PE via another component the voltage dropped but this would have been due to a ground loop. The other thing is that if a ground was connected to the pc (eg monitor d connector) it would give a very feeble spark as the voltage was pulled down to earth.  You pc smps many not behave the same but put a meter from the signal earth of the pc to PE before you try using a disconnected PE as a diagnostic.

I arrived at the star earth signal ground to PE set up in the balanced amp after trying a few permutations. Its a little counter intuitive but it gave the best results here.   

I think the gainclone hum I have here is due to an out of spec tranformer, it humed before it was put into the amp and I have an RMA from the supplier to have it replaced. Even with 103db speakers the hum is not bad and its actually as loud coming from the transformer inside the amp.

As I mentioned by mail I think  Paul, I would try reversing the mains phase of one transformer first. They are very close to one another and the flux runs in opposite directions at their closest point. Assuming the windings have the same direction reversing the phase will make the fields run in the same direction. This may work like stacking the transformers. It a quick arrangement to test and may help. Otherwise Im with Peter regarding the position  of the transformers.

Im planning to take a close look at the ground set up when my new oscilloscope arrives is a couple of weeks.

Peter,

Am I right saying that the NOS is signal ground is connected to PE on the interface via a 100k resistor and a cap in parallel with the resistor  (i'm unsure of the cap value) ?

These components sit between the usb cable shield connection (which routes off to the nos case ground) and the signal ground of the usb lead which in turn sets signal ground....I think.

Cheers nick

Hi Nick ... No ...
But if so, you did that yourself. Haha.

But don't get fooled by looking at the USB interface only;
At the cabinet's end the USB shield is disconnected from the chassis per the outwardly bended clip on the connector ...

So totally nothing is (or should be) connected to the chassis. Ok, PE itself. :)
Can easily be checked for course (multi meter).

Regards,
Peter

Whoops sometimes even the most obvious is not (but it should be). The top plates for both the amp box and the transformer box were still in their plastic wrapping never (yet) having been installed. So I unwrapped them and screwed them in place. The mains hum is lowered and is now the same from both speakers. Better for sure but not quite as low as with my previous amp.

The SQ is not quite where I want it yet - my previous amp was smoother but not as good in most other areas. I know just how much difference fine silver makes in the i/c (the RCA i/c's on my previous amp were fine silver and Nick's were great when we tried them with his amp) so that is the first thing to try. If that does not work then I will have to consider surgery (maybe) but it is much work to do.

P

Hi Peter,

Thanks for explaining. I measure 104k ohm from the usb shield connection point  on the NOS input (the port on the outside of the case) to [ EDIT  the NOS signal ground at the RCA output ground or the velaman psu ground]. This measurement is taken with mains cord, interconnects and usb cord all disconnected from the nos.

I know the whereabouts of the 104 k resistor an it is bypassed with a cap meaning that for ac, pe and signal ground are connected. I have not changed anything electricaly in the ground layout. I cannot see any point where the usb sheild is "lifted" in the cable between the nos port and the card.

I have just broken the shield pe (using tape) after the point that the pe wire runs back to the chassis but before the shield connection reaches the interface. What a difference to SQ.

 Is this the intended arrangement ?

Thanks,
Nick.

Ps 09-z9 sounds MIGHTY, MIGHTY  fine  :)

Uhm Nick,

Can you please show a photo of the USB chassis connector, from above, inside of the chassis ?

Regards,
Peter

Nick,

Reading your post better, I think we are talking about something else. Never mind that photo because I think you will find that clip is "bended outward" anyway.

Maybe better take this to email ?

Thanks ...
Peter

Peter hi

I had the same thought see your email in box, it looks a bit unusual in there ;)

Nick

I mentioned in previous posts that the new balanced GainClone had taken a big step forward in SQ. But I still had a few reservations namely a slight unevenness in the mid to upper registers and a slight cold hardness in the sound.

Whilst I would never be going back to my old amp I still hankered after its refinement.

Well today I received some fine silver wire in the post. I set to with my soldering iron and replaced the enameled copper wire in the i/c with fine silver.

The result is a complete removal of any doubts I had about my system the SQ is back where I want it  - very refined top end and the slight course unevenness in the mid to highs is completely gone. Mmmmmm I was missing this.

I have to say I am still surprised by how much difference silver makes. My interconnects are only about 10cm long. So why the difference? Surely the electrical properties remain the same as the actual construction of the i/c has not changed. The resistance of silver is not much difference to copper and over the short lengths I doubt that is a factor.

I still think silver / gold etc work their magic at a quantum level. Who knows? - what I do know is that I am totally happy with the SQ now.

 :)

Paul

Here is the actual result of my last weeks effort. The system is build up on commercially available pcbs, assembled and wired as written in my first post.

Nothing special but I tried to get analog in and analog out as near as possible. Therefore the copper rails and the chip are mounted at the bottom of the cabinet.

The pcb at the left side of the transformer is a softstart unit which is not necessary at this time and unused.

Because I'm using very rare Philips Alnico drivers there is no chance to do without a coupling capacitor. I tried a lot (had two well broken in black gates for lease from a friend, which showed the direction) and ended up with a satisfying solution: A (new old stock ) combination of Roederstein KP1832 and NSF Styroflex capacitors.

Georg

Georg hi,

The build looks great ! And an interesting story about the input caps, defiantly best to be safe there  you certainly don't want any accidents with your drivers.

How is the sound ?

Best,

Nick.

Hi Nick,

that was he biggest challenge and hopefully I have not to repeat the process too often ;). After trying oil caps, mica, foil, even vacuum caps and other nice stuff I ended up always with a sonic compromise. Some types work best for rock and jazz others better for classical music. The actual combination is suitable for the long term, works with all types of music and has a delicate and colorfull character.   

Georg

P.S.: Did not try very interesting but very expensive solutions like Duelund or Thel Mica Caps (4,7 MicroFarad!)

Here comes another "Gainclone". Seen on the photo are basic building blocks of bridged LM3886 amp.
Interesting (domestic) project using 'External Feedback' like it is used in any (serious) measurement instrument.
With this concept, speaker cable influence on sound is diminished. Another advantage would be using such an amp
with servo controlled woofer(s) - there are few on the market today.

(http://i851.photobucket.com/albums/ab79/giokka/LM3886_ExtFB_zps7e8b78a5.jpg)

After trying out a lot of things with 0.9z-9b and Windows 8 the system seems to be very stable and the time has come to ban the input capacitor. For the case of accidents it's easily connected again. Additionally I connected in parallel with the buffering capacitor a styroflex capacitor on the amp board which works beautifully for the heights.

Georg

That's true. But did you calculate-in the FACT that when accidents happen all your GC chips will have been smoked and that you need to replace a couple of speaker drivers ?

NOT ADVISED.

Peter

Georg hi,

I tend to agree with Peter. In addition I find the sound tends to get better with input caps. Without is very pure but with seems to improve balance of the sound. I prefer with.

Paul and I are using blackgate n types. In my case I use type nx in 47uf size. There is a guy selling them on ebay from japan at the moment. They are expensive but very very good on the input of gainclones.

Cheers,

Nick.

Nik.d hi,

Interesting looking boards, how is your build coming along ?

Regards,

Nick.

Thanks Peter, thanks Nick,

it's easily switchable. So I can try a new hardware config or software version with input capacitor at first  ;). I found the Blackgate seller in Japan and another source in Europe and will order a pair.

Georg

Hi Georg - as the unfortunate owner of a couple of "smoked" drive units I would not go without caps now. The BGN's are superb and they have very little impact on SQ. Best of luck with your project.

Paul

Hi Nick

Sorry for late reply, spent 'long weekend' on Danube river with friends. A nice one. I'm a bit slow, actually wanted somewhat better basic components and ordered
(and finally received) Bayschlag 0.1% resistors and some Roederstein elcos. Transformers are ready. Need only good 4.7 uF polyprop cap and final decision: to build
separate cases or one as dual mono. This because I already use two sub (plate) amps. Adding separates near mid/hi section would eliminate long speaker cables.
Old dilemma: long spkr cables or long IC's... Esp. when amp is constructed to minimize (spkr) cable influence :)

Brgds,
George

George hi,

That is a chestnut, long interconnects v speaker cables.

I found that having transformers in the same case as the amp modules really hit sound quality. My current design is two box but to split is amps in one box and transformers in the other. The amp performs much better like this.

If the 4.7uf caps you are looking for are for input decoupling it might be worth hunting out some 22uf or 47uf Black gate nx types (see earlier post). Paul and later Myself have had very nice results with them. I preferred them to mondorf silver gold oil caps, wish I had known earlier !

Cheers,

Nick.

Nick hi,

Yes, am looking for good input cap. Reading nice things on ClarityCap MR series PP's but guess what: seems I'l be able to get few
BG-NX 22uF/6,3V caps as well. Seems my old friend (passed away last Summer) left behind him some good audio parts, not only
couple of unfinished GM70 power amps. His book was (still is) nice reading on tube audio gear.

Brgds,
George

That is sad to hear about your friend George.
GM70s look like a serious tube, 1000v b+, your friend must really have known his way around valves. What book did he write your?

Regards,

Nick.

Book was collection of ideas & examples on building tube audio pre/power amplifiers, with god examples of various 'schools'.
One of rare books well written in local language. His 6L6 integrated was positively reviewed by Jean Hiraga in (think) 'Revue
Du Son'. Had cooperation w/ 'Teresonic' as well. And above all, loved motorboats and Danube river.
Some pic from his book presentation (https://www.dropbox.com/s/48qwhecg9yh35ih/N_V.jpg)

Hello fellow noise hunters. I thought I would post on my continuing quest to get noise in my system to absolutely the minimum possible.

As described elsewhere I got a very significant reduction in noise resulting in improved sound quality by separately connecting the cases of my amp, PC and NOS1 back to the earth rod in the garden. NOS1 and amp cases are NOT connected to electronic ground. I have to say I was very surprised at the level of improvement this brought about (but see below as this may be related to the problem found below).

But still although noise levels were now low there was still some noise I could hear with ear next to speaker cone albeit at a fairly low level.

My amp layout is shown here http://www.phasure.com/index.php?topic=1575.msg26709#msg26709    there was some discussion at the time about moving the transformer box away from the amp to potentially reduce noise. So this weekend I spend many hours doing just that. I moved the transformer box 2 feet away from the amp box and guess what no difference at all in the noise levels. Ah well something learnt.

Later that night I was lying in bed thinking that I must have missed something. And I had !! - because then I thought that whilst I was listening to the speaker cones I was missing the point  that actually I was hearing more noise from the actual transformer box in the conservatory than from the speaker. Surely the transformers should not hum that loudly!!.  This thought was exemplified by the fact that the character of the sound I was hearing from the speaker cone was exactly the same as I was hearing from the transformers.

Mmmmm so there was only one conclusion the transformers were louder than they should be because they were next to each other in the box and their EM was interacting. So I removed one transformer from the box yesterday morning and hey presto the noise from the transformers AND the speakers lowered.

So then I removed both transformers from the box (and well away from each other say 2 feet) and the noise level went UP slightly. So I tried them in all sorts of positions (including stacked Peter) and the noise level changed each time.

But then the really strange thing was that putting both transformers back in the box but in a different position resulted in the lowest noise of all!!!.

So then I spent hours rebuilding the amp to end up with exactly the same amp setup that I originally had BUT with the transformers diagonally opposite each other in the box but with very low hum levels and lower than in any other configuration (including 2 feet apart on the carpet). But then it gets better because putting the lids back on and back in the final setup on the hi-fi stand lowered the noise even further. In fact I really struggle to hear any noise at all from the speakers (OK I know they are only 92db sensitivity) even with my ear right on the cone and I mean right on the soft dome in the centre of the speaker oh and also with a very quiet house.

This is all basic stuff which really I knew before but had overlooked in my enthusiasm to build the amp. The moral of the story is when building the amp in the first place check all positions for the transformers to find where they hum the least it is really really important to do that especially with the big transformers I now have. And make sure you only use high quality quiet transformers.

I have been listening to strings and opera and I have never heard this level of clarity - stress free crystal clear clarity - in my system before.

A lot of work this weekend but well worth it. I cannot imagine what else I could do now to get noise levels lower but I would always be open to suggestions.

And also having got the noise levels so low I can now really appreciate just how good this bridged bi amped gainclone amp is.

Paul

Congratulations Paul !

And since you're asking anyway ...
I am fairly confident I am overlooking something now, but why do you use two trafo's in the first place ?
I forgot a bit, but looking at your picture again, did you put two (differential) channels in there ?
If so, mono's would be the solution ...
I don't think you'd gain on anything but a cabinet and a power inlet plus switch and fuse. Or ?

And if you have a wire connected to the shield of the trafo's, just don't connect it to anything (for trials).

Regards,
Peter

And Paul, one more other thing I wanted to tell you, but forgot :

I don't think any torodial transformer should acoustically hum in the first place.

Unless you have DC offset on your mains.

Yea, now what. Another can of worms ?

And I dare to add : when your transformer hums because of DC Offset, you will have a very hard time in discerning the same noise from the speaker (which really is there - watch it !).

Assuming you have a sort of isolated situation of your audio chain, hence not all that many devices can imply the DC offset, investigate your PC's PSU ...
(hence switch off the PC and and listen again).

HTH,
Peter

Hi Peter,


Peter Daniel recommends 2 trafo's for bi amp operation. Each of my trafo's has 4 x secondary's by the way and had to be specially wound as 4 secondaries is not standard. One trafo would have needed 8 secondaries!!!!!


There are four "amplifiers" but each amplifier comprises 2 x amps bridged so in total eight amplifiers (hence 8 power supplies needed)! And oh boy they draw much more current than in non bridged operation. Heat dissipation is MUCH higher than before.

I did not have screens in my trafo's (again after discussion with PD) and based on the results I now get they probably would not make any difference (Nick has screens in his).


Right!! and mine don't NOW but they sure did before when they were located closer together in the box presumably because the EM fields of the trafos interacted. But just moving a couple of cm apart (in the right direction) and they are quiet now. Stacking them in the box probably would have been best but that would have needed a very tall box!!.

Paul