Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[mattkhan]

I have a rack containing all my av equipment as well as my main file server. Recently I have noticed that PC noise is being poetaster through my sub amp. It is clearly system noise as it various with computer activity. I have determined that the noise must be from the server not any component in the audio chain. The two devices are not v far apart in the rack and they are powered by the same mains feed.

Doors anyone have any tips on the best ways to tackle this?

[mwillems]

I have a rack containing all my av equipment as well as my main file server. Recently I have noticed that PC noise is being poetaster through my sub amp. It is clearly system noise as it various with computer activity. I have determined that the noise must be from the server not any component in the audio chain. The two devices are not v far apart in the rack and they are powered by the same mains feed.

Doors anyone have any tips on the best ways to tackle this?

I've had this issue in the past (repeatedly and at length) and I've only ever solved it in three ways: 

1) Change how the PC is connected to the DAC.  Some DACs will pick up CPU noise because they're inadequately electrically isolated from the PC.  I've had several soundcards that would pick up CPU noise.  In the case of internal cards, this can sometimes be helped by changing which power supply molex they're hooked up to.  In all seriousness I had one that was hooked to a molex in parallel with several fans, and the noise dropped dramatically when I connected it directly to the PSU.  For USB DACs, try a different port, or connecting through a USB hub.  Sometimes it can make a serious difference.

2)  But in some cases, you'll just need to change the DAC for one with better electrical isolation.  That's what I ultimately did and never looked back as my current DAC doesn't much care what it's hooked up to.  Hooked to the same computer that made me miserable with other interfaces, it picks up no noise whatsoever. 

3)  But if that's not a viable option, the final solution is passive line level attenuators.  Put some attenuation between the DAC and the Amp inputs and you'll more or less eliminate quiet noise, and significantly reduce louder noise.  You may lose maximum volume, but my experience is that most amps are fully driven well below the maximum output of most DACs, so in that case you won't even lose volume.  A balanced transmission line between the DAC and the AMP will also help. 

[dean70]

1. You could try a ferrite bead on the power cords of the PC and DAC.
2. Try a grounding strap/wire from the PC expansion card back-plane bracket to the PSU mounting screws using spade connectors
3. RFI/EMI shield around the sound card / usb card (if applicable)
4. Ferrite bead on the PSU molex power connector to sound card / usb interface (if applicable)

[mattkhan]

Thanks for the suggestions. A few more details to clarify exactly what the setup is

The signal chain in question is

HTPC -> firewire -> RME FireFace 800 -> Marantz AV7005 -> SpeakerPower SP1-6000 sub amp -> sub

The PC that is introducing the noise is not in this playback chain, it is the file server which is connected to the HTPC via gigabit ethernet and which serves media content to the HTPC. i.e. I can physically turn off the HTPC & the RME FF and put the Marantz prepro into standby but the noise remains. I haven't tried physically unplugging those devices though yet.

I think, but am not certain, that the problem was introduced when I changed my gain structure to avoid the need to attenuate all the main channels digitally (to account for LFE) & instead I now raise that back up later in the analogue chain. The net result of this is that the gain dial on the sub amp is now about 8dB higher, i.e. the attenuation it applies on the front end is now smaller. This means the SW signal coming out of the RME/Marantz is now weaker than it was before given the same dBFS signal.

I'm not sure if this extra detail makes any of the suggestion more or less applicable here. It seems like restructuring gain, so that the signal is hotter into the amp,  should do it but seems like it would be preferable to tackle the noise at the source instead.

[mwillems]

The PC that is introducing the noise is not in this playback chain, it is the file server which is connected to the HTPC via gigabit ethernet and which serves media content to the HTPC. i.e. I can physically turn off the HTPC & the RME FF and put the Marantz prepro into standby but the noise remains. I haven't tried physically unplugging those devices though yet.

I would take some additional steps to isolate before I got too much farther down the road.  I would start by unplugging the inputs to the sub amp (as in the rca or xlr input plugs), then unplugging the inputs to the marantz, then the fireface, etc. until the noise presents.  I'd also test with everything on, but with the server's ethernet cable unplugged (or with the server entirely off).  It would be unfortunate to spend a lot of time troubleshooting the server if it turned out that the noise was coming from a different source. 

I say all this because if the server is only connected to the HTPC, and the noise continues with the HTPC entirely off, that makes me skeptical that the server is actually the problem (or suggests that the noise is propagating in an unusual way).  It's possible, but I've never personally encountered (or heard of) noise propagating via ethernet, much less when intervening stages are not powered.  So unless it's propagating via powerline or RF pickup (which is less common, but not impossible), the server seems like an unlikely candidate.
And if it is a powerline or RF issue, then you've got a completely different set of problems (my advice above would not help either one, although dean's advice might).

Severing the connection at various stages will tell you if the server is "communicating" through unusual channels.

I think, but am not certain, that the problem was introduced when I changed my gain structure to avoid the need to attenuate all the main channels digitally (to account for LFE) & instead I now raise that back up later in the analogue chain. The net result of this is that the gain dial on the sub amp is now about 8dB higher, i.e. the attenuation it applies on the front end is now smaller. This means the SW signal coming out of the RME/Marantz is now weaker than it was before given the same dBFS signal.

I'm not sure if this extra detail makes any of the suggestion more or less applicable here. It seems like restructuring gain, so that the signal is hotter into the amp,  should do it but seems like it would be preferable to tackle the noise at the source instead.

I think that's significant as it suggests that the noise has likely been present previously, but inaudible.  As for tackling the noise at its source, if it is the server, the way to tackle it will depend entirely on how it's propagating. 

[mattkhan]

To be clear, as I wasn't earlier!, there is one mains feed to entire rack and the server & sub amp are plugged into the same power strip.

[meep]

I've had this exact problem which I managed to solve.

Recently I added an older MOTU 828MkII to my HTPC. This is a multi-channel firewire audio interface to replace my m-audio delta PCI card. I also added a Parasound 5 channel amp driving front l/r speakers (bi-amped)

When I connected it all together, I could hear distinct electronic noise in the speakers. The noise seemed to match disk access activity - when the HD light flickered, the noise was in synch.

It turned out it was my PSU. I'd first used a PICO PSU which was super noisy. I replaced that with a Corsair CX750 and, with my ear to the psu, I could hear the exact same noise inside.

I replaced that with an old spare PSU and the noise was eliminated.

hope that helps.

Peter