Hello both, Thanks for your replies.
Fortunately I found a way to crank up the noise and bring it a bit under my control. By accident I found out that putting some strain on the Nginx http server increases the tone significantly. It doesn't seem to matter whether the data comes from an externally powered sata drive or from memory, noise is all the same. Some other observations: copying via scp to /dev/null or the external drive does not introduce/increase the noise. With the help of a dB meter I think I have localised the source. It seems to come from the nearest capacitor to the CPU (FP; 02AG; 8 2 1; 2.5). For clarity, here is an image (note the arrow): http://temp.muffins.diskstation.me:8080/soekris.jpg @ED. Assuming I indeed got the source of my troubles, how would I get rid of this noise? Alternatively: is this a sign of a failing capacitor and is it better to be replaced? Kind regards, Maarten On 25/01/16 07:04, gavrik peterson wrote: > I have done a fair amount of design work with switch-mode power > circuits. Nearly all low power regulators operate at frequencies that > are well beyond the audio range. Despite this, it is possible for > these circuits to emit sounds that we can hear. This usually happens > when the circuit is not operating contentiously. One reason might be a > design defect that allows the voltage control (or perhaps current > control) loop to be unstable. Another possible reason, is that some > modern-switch mode regulator devices change their mode of operation > when operated at low output current. That is, they go into a mode > where they operated in a series of burst of normal (high frequency > switching) operation. This is done to make the regulator more > efficient when operating at low current and to save power. > > In both cases, there is a possibility that increasing the load current > would cause the sound to disappear, as Ed Fochler suggested. > > The best image of the net6501 I can find does not allow me to read > device part numbers. If anyone knows which device is creating the > noise, is it possible to read a part number on it? > > I know from experience, that identifying the source of noise can be > difficult. One way is to use a small microphone as a probe to find the > source. Another possibility is to somehow couple a hose to your ear > and use the other end as a probe. The most likely source is one of the > inductors. > > Gavrik Peterson > > On Sun, Jan 24, 2016 at 4:55 PM, ED Fochler <[email protected] > <mailto:[email protected]>> wrote: > > Soekris boards use at least one DC to DC transformer. These are > notorious for creating whine, even when they are spec’d in such a > way that audible whine would seem impossible. they also react > differently under load. The only one I can find listed is for the > pci slot, so you may want to try adding load to that (insert a > card) and see if that changes or eliminates the whine. > > Otherwise, you’re in sound dampening territory. High pitch whines > are easy to squelch if you can locate them. I can’t hear anything > from my boards, so I can’t help you in a more practical a manner. > > ED. > > > > > > On 2016, Jan 21, at 12:05 PM, Muffinman <[email protected] > <mailto:[email protected]>> wrote: > > > > Hello, > > > > My NET6501 runs quite well except for a very annoying > high-pitched tone > > originating from the board (sometimes). Since I intend to use this > > computer in my hifi setup, well, it's a bit problematic. > > > > I've got an external 12v power supply some distance away from > the board, > > so that isn't creating the noise. In addition, I've tried two > different > > 12v power supplies, both with the same effect. I don't see a > relation > > with CPU load (though the processor has an easy job anyways), or any > > relation at all. Sometimes the tone is there, gets more intense, and > > then it's gone to a low level (or perhaps gone). > > > > Has anyone got experience with this (annoying) feature. Also > important: > > does anyone have a (possible) solution? > > > > Kind regards, Maarten > > _______________________________________________ > > Soekris-tech mailing list > > [email protected] > <mailto:[email protected]> > > http://lists.soekris.com/mailman/listinfo/soekris-tech > > _______________________________________________ > Soekris-tech mailing list > [email protected] <mailto:[email protected]> > http://lists.soekris.com/mailman/listinfo/soekris-tech > > > > > _______________________________________________ > Soekris-tech mailing list > [email protected] > http://lists.soekris.com/mailman/listinfo/soekris-tech
_______________________________________________ Soekris-tech mailing list [email protected] http://lists.soekris.com/mailman/listinfo/soekris-tech
