USB Cables... again

[manisandher]

Hi Coen, I'll back to you once I've tried a few things.

Meanwhile, I've just taken my office music PC into my main listening room and given it a go. Firstly, the 'USB noise' is identical. With a regular USB cable, I can hear the crackling noise. This increases whenever I move the mouse. It's at a level pretty much identical to that of Le Monster. When I use the new USB cable with only the data line connected, the noise is totally eliminated.

I use identical XX settings in both PCs - both have powerful CPUs and 16GB of fast RAM. But I much prefer the sound of Le Monster. This is probably down to the fact that it has a HDD and my office music PC has an SDD. In any event, it looks like Le Monster's mobo is absolutely fine (hope you're reading Brian) and also that my long USB run is absolutely fine.

No doubt it's my amp that's allowing me to hear the noise coming through the USB cable. As I've said before, it's a strange design - DC-coupled, with the ground connected to its PE. This complicates matters of course. I've attached a couple of circuit diagrams for anyone who's interested (good luck with them - they're totally beyond me!).

Mani.

[CoenP]

Just got some more ramblings in my fingertips...

I realise I forgot one important thing and that is the whole point of earth referencing is to NOT have currents flow through them that are generated by the circuitry; it is just creating a "silent" point for your circuit towards all other voltages are referenced. they don't change their relative values. It is what it is: a reference. This is where the refrencing the usb at two points goes astray: the groundrefs 2 and 3 are able to and will carry a signal.

Than there is the matter of radio energy that has a potential to the earth and as such can induce currents in conducting objects (wires, casings etc) to earth, closing the loop from the antenna to ground. So here the ground ref has a completely different function: getting the radio energy out of the circuit (or let it stay in the box... ). For this you only have to ground the casing that contains of the circuitry.

So my common sense says: or wire 1 or wire 2, no 3 and no connection between the case and circuit of the poweramp would be optimal.

Regards, Coen

[CoenP]

Hi Mani,

Hi Coen, I'll back to you once I've tried a few things.

Meanwhile, I've just taken my office music PC into my main listening room and given it a go. Firstly, the 'USB noise' is identical. With a regular USB cable, I can hear the crackling noise. This increases whenever I move the mouse. It's at a level pretty much identical to that of Le Monster. When I use the new USB cable with only the data line connected, the noise is totally eliminated.

This sounds awfully familiar. I had a similar experice some time ago, but I never changed the stock USB cable. I could hear everything from the pc, mouse, hdd, cpu with distinct noise patterns. Annoyingly loud with horns. Afaik i toyed around with pc powercables (with/without PE connected, different powerstrips) and with the cabling in my amp. It is solved but not by usb but either by amp rewiring or powercable optimisation or new power supply(?). Have to remember a lttle harder how I did it...

Afaik the usb powerwire isn't connected to anything in the NOS1, except maybe the spike protection chip at the input of the USB. I'm not shure about that, but if correct noise from there can enter the circuitry.

I've been studying the Berning a couple of years ago. A very interesting, very, very clever design. But you have to realise it works like an AM  radio with the output transformer as sending and receiving antenna. The whole high voltage is modulated at a high frequency at the 300B side. Allways wondered how this works out for RF.

Regards, Coen

[juanpmar]

3) The USB noise has only now become an issue because until very recently I've been using balanced connections between the NOS1 and power amp. With balanced connections, there is absolutely no USB noise, irrespective of USB cable type or length.

Hi Mani, with balance connections you don´t have that noise but any difference in SQ with the Ridge Street Enopias USB cable versus the stock cable using in both cases xlr connections?

Juan

[CoenP]

Hi,

The 5V of the USB cable is internally protected by the ESD protection chip in the NOS1. So it is connected. I can imagine some mechnisms that noise from the 5V line will be entering the NOS circuit, but those will be fairly small.

I toyed around with powercables, powerstrips en earthing (as far as possible here), but the system remains silent in all cases, that is the PC noise (if present) is swamped by the noise of the amp that is allready quite low.

In the time I had a flawed psu that distroyed -according to my measurements and theory- the internal NOS1 ESD protection chip. So I switched PSU to Seasonic X-400 and got a repair from Peter (actually Cisca!). Can't remember ever having any noise problems after that. That is with the regular USB cable.

My hunch is that the noise on the usb 5V is able to enter your NOS1 circuitry as a result of damage. In my view this noise problem could point to a half blown ESD protection chip; the result of one experiment too many (and I also use a tube amp...). I could very well be that my power connection arrangement was the culrpit of the blown ESD chip rather than the PC PSU. Since I am only partially familiar with the internals of the NOS1, it is hard to reason how such a thing could happen and wether it happened to you.

Regards, Coen

[PeterSt]

Maybe a bright guy is able to solve this puzzle... Peter? Anyone?

Not me. Your outlays (also following the post I just quoted from) are quite thorough. It's only that no real rules can be applied because we are not on control of the devices themselves (and how they connect to PE, up to even needing that for some).

USB is just a nasty thing, and I myself don't understand it. This is about how it is organized in the PC which I think is "a standard" and how it runs its gnd over PE as far as I can see.
When the PC is not connected to PE (Protective Earth - your third pin on the mains cord) this is never for the better. Not in my case (all kinds of noise pickup occur, generally only measurable but can also turn into audible noise).

[ I added all sorts of explanations about connections, but it got too complicated so I removed those ]

Now, back to the "Not me" I started out with;

So I have a new set of amps and after some twiggling I was able to have it completely noise free (I didn't measure that, but audibly is enough for me, for now).
In the mean time I am measuring loudspeakers and for that it needs another USB connection to the amplifiers (to program the DSP in there for the bass drivers). So what I have is an audio PC with USB to the NOS1 and USB to the amp. There's another connection through Ethernet to another PC; this is inherently isolated.
No problem anywhere that I can see.

Then I changed the DSP programming from the audio PC to that other PC, because that is more convenient for me. The only thing changed was removing the additional USB connection from the audio PC to the amp, which now changed to a connection from that other PC to the amp.
Both PCs are in the same mains outlet (which is also where all the audio is connected to).

Now I receive a buzzing noise (something like over half of a second) just before music starts, plus when the monitor switches off or on I hear it through the speakers (a small tick). This is just already the "soft" switch off/on implied by software - not the hardware button.
The buzzing noise I also had when there was audible noise through the speakers. I don't know about the ticks. So, solved that noise (XOver filter pickup that was), but now the same is back by my different "USB path".

And now you think that I know how to solve this ?
I can't even reason out where this buzzing noise comes from. Looks like something is hammering on the monitor and tells "shut off man !" and it keeps on doing that until it doesn't respond anymore. The tick too feels like coming from just software usage.
Anyway I can test this by not letting switch off the monitor (I am thinking just now). But still, if that is the source, I would not know a single means to solve it.

And might Brian read this : Brian, that buzz could well be the same you reported about.

I don't think mouse movement causes any audibilities in my situation. However, I am using PS/2 connections for that which is of course on purpose. Mani, you ?

One thing should be clear : this whole sh*t is so complicated that it only lets solve itself by trial and error. A little base knowlegde is a good idea but in the end not enough.
And when all is solved you should regard yourself lucky. The very best thing though would be to have that reference, like KNOWING that your situation can be noise and tick etc. free. Like in my situation. Without that - like years ago - I would not even start looking. "Belongs to it" I would think.

Peter

[PeterSt]

it is hard to reason how such a thing could happen and wether it happened to you.

I don't think Mani is telling that he can't revert to his old situation (with balanced interlinks). So, as far as I read it, a change of topology now lets this happen.

Another thing is that any real balanced situation ("differential setup") like what would happen with the NOS1 connected to the Sauermann amp, forces DC Offset to be 0, hence forces a good ground reference. This may lead (thoug a long shot) to forcing the USB reference into the better just the same.
Ok, must stop here, because otherwise I'm again trying to make a too complicated story full with hallf-truths and uncertainties.

Peter

[PeterSt]

About the ESD protection (I wanted to reply to that yesterday, but forgot) :

So, the 5V USB power is only used for that part of the USB interface. It's a bit of an odd solution perhaps, but as soon as the ESD protection chip is "passed" on the PCB, the USB (5V) power is cut off and goes nowhere. 3.3V is self-generated on the board.
In the mean time, the chip itself should never blow, although indeed that happened to you Coen (and is still the only case we know of).

Peter

[manisandher]

I would rather experiment what the difference is between referencing at the pc ground at the plug versus referencing at the usb side.

OK, FWIW... I've played around with the USB grounding connections under my control, namely 2 and 3 in my earlier diagram. With only the data line of the USB cable connected to the NOS1, here's what I'm hearing:

1) 2 and 3 both disconnected

Very bright, thin sound. Lot's of HF stuff going on, e.g. can't hear the initial strike of cymbals as it's drowned out by too much 'glazed' shimmer. Too much sibilance on voices. Actually, voices don't sound real - there's no three-dimensionality to their edges; they sound like two-dimensional cardboard cutouts. General background is rarely totally 'black' - there seems to be a constant reverberation of some sort or other going on. This leads to the sense of 'extra detail', but I think it's false detail.

Interestingly, I had the music just stop playing mid-track on one occasion - the audio device was lost. Coincidence? Well, it's never happened before when I've had both 2 and 3 connected.

2) 2 connected, 3 disconnected

Better. Things are generally calmer. But there's still an edge to the sound. Leading edges seem to be unnaturally amplified. I start tensing up as I'm listening because I know there will be something screaming out of the speakers as soon as a leading edge (be it a vocal, a piano, an electric guitar, or whatever) comes my way. It's almost as if the music is being unnaturally modulated by the leading edges.

3) 2 and 3 connected

My preference. This simply sounds the most realistic to me. The brightness has been all but totally eliminated, and yet there's plenty of inner detail in the sound, e.g. I can hear the 'rasp' of a sax or the texture of a cello. Voices? Well, well recorded stuff is incredibly realistic - I was shocked at how good Cobain sounds during quieter pieces on my MFSL CD of Nirvana. Is it perfect? No, I don't think so. There's still the last vestiges of sibilance sometimes, but this could be down to the recordings, or even perhaps my mobo.

And to reiterate, all of this is with my 'new' cable using just the data line.

Mani.

[manisandher]

This is where the refrencing the usb at two points goes astray: the groundrefs 2 and 3 are able to and will carry a signal.

Coen, my feeling is that this is a good thing. What we want to do is reduce the return current in the USB cable, as this will only increase the ground plane noise. With 2 and 3 connected, the return current has a much lower impedance return path via the thick (6mm) earth wires.

But hey, this is pure speculation on my part.

Mani.

[manisandher]

This sounds awfully familiar. I had a similar experice some time ago, but I never changed the stock USB cable. I could hear everything from the pc, mouse, hdd, cpu with distinct noise patterns. Annoyingly loud with horns.

I don't think mouse movement causes any audibilities in my situation. However, I am using PS/2 connections for that which is of course on purpose. Mani, you ?

Yep, I'm using PS/2 for both the mouse and keyboard (using 2x 5m PS/2 extension cables, which don't seem to be affecting performance). But moving the mouse is clearly audible along with the more general 'crackling' noise. Of course this is only the case with the USB power line connected and when using the Berning amp (with all its idiosyncrasies). There's nothing, nada, when the power line is disconnected.

Mani.

[Scroobius]

Mani,

Not sure I understand your earthing arrangements.

1. Are the yellow lines actual wires? for example I would expect only one earth wire connected to the PC in the basement. All sorts of "things" inside the PC will be connected to the one earth connection including the USB shield.

2. Similarly do you really have two earth wires (3 & 4) connected to NOS1? again I would expect only one earth wire connected via the mains plug and it would be connected internally to the case of NOS1 only (and not the ground of the electronics board).

Just an observation it seems from the Berning cct diagram that the electronic earth is connected to protective earth. Often there is an earth break resistor in valve amps but that does seem to be the case with yours. What this does mean is that with your system you will always have at least one earth loop. Because Berning amp electronics ground is connected to the earth rod and also connected to NOS1 electronic earth (via i/c) which is then connected back to earth rod via the USB shield. Unavoidable in your system. But if the earthing is as shown in the diagram then there are other earth loops. What happens when you connect / disconnect wire 3 would be totally unpredictable.

Cheers

Paul

[manisandher]

I've been studying the Berning a couple of years ago. A very interesting, very, very clever design. But you have to realise it works like an AM  radio with the output transformer as sending and receiving antenna. The whole high voltage is modulated at a high frequency at the 300B side. Allways wondered how this works out for RF.

I'll put a longer post in my 'Testing a few amps' thread. But for now I'll just say that the Berning sounds as interesting as it looks on paper. Absolutely no softness at either end of the bandwidth. And a sound stage that goes back behind the speakers for miles. The sense of recording space is incredible, especially on classical recordings. I think this is one good test of 'real' resolution. What's this down to? Its SET design? Its single output device design? Or just the fact that tubes have no hysteresis affects? I don't know. But I really like what it's doing to the sound.

Better than the Sauermann, BD-Design or Sanders amps I have? Hmmm... "Different", I'd say. These push-pull SS amps all sound way more dynamic and 'in your face' than the Berning. Also, there's no way the Berning can energise the room in the way these others can. And the BD-Design monos have a clarity that is hard to believe - better than anything else I've heard. BUT... I just think that there's a certain magic with the Berning and the 114dB horns. My hypothesis is that a horn needs to be driven single-ended and not push-pull to sound most natural...

Mani.

[manisandher]

Hi Mani, with balance connections you don´t have that noise but any difference in SQ with the Ridge Street Enopias USB cable versus the stock cable using in both cases xlr connections?

This is more difficult to test as it would require taking the Berning out and replacing with one of my other amps, and I don't really want to do this just now. But in my office system, I use a fully balanced setup - NOS1 going into active Genelec satellites/sub. From what I remember, the Enopias reduced some of the brightness of the printer cable even in this balanced setup. Could be a good thing, or a bad thing. Will need to try again and get back to you.

Mani.

[manisandher]

My hunch is that the noise on the usb 5V is able to enter your NOS1 circuitry as a result of damage.

Well, this is easily tested - I have two NOS1s! I'll take the one from my office and try it in my main room.

Mani.