Very low level hum and how to get rid of it

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
ilya
Posts: 88
Joined: Tue Dec 29, 2009 2:10 pm
Location: Russia
Contact:

Very low level hum and how to get rid of it

Post by ilya »

Hello gents.

I've finished one particular build and I'm getting a very tiny amount of 100Hz hum. When I say "tiny" I mean just that - I have to boost the input of my soundcard 60dB to hear it (I'm injecting the signal in a universal input that can take both line and mic levels). I wanted to figure out if I can get rid of it completely.

I've looked at my scope and output noise measures approximately +/- 4mV p-to-p, approx. 1.1mV RMS; when I put both probes on the ground (not literally, of course :) ) I see +/2 mV of noise.

I've attached the pic of the output I'm seeing.

I'm sure this is not placement dependant. I removed the power trafo out of the chassis and nothing changed. I've moved the wiring and got no change as well.

So, what do you think? Worth digging deeper or leave it as it is? Where should I start looking?
Attachments
acc.jpeg
acc.jpeg (102.84 KiB) Viewed 15254 times
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Very low level hum and how to get rid of it

Post by JR. »

It is difficult to give circuit design advice with no idea of the circuit, but that never stopped me rambling on before.

I will ASSume your mains power is 50 Hz. !00 Hz is the frequency of full wave rectified power supply ripple.

The fact that moving the transformer doesn't change it suggests it not being picked up from the transformer's magnetic field, but is getting in from the power supply voltage or ground scheme.

Are DC voltages clean?

Are your audio signals treated differentially to and from sundry grounds?

The hum could be sneaking in from several different directions.

Good luck.

JR
Cancel the "cancel culture", do not support mob hatred.
ilya
Posts: 88
Joined: Tue Dec 29, 2009 2:10 pm
Location: Russia
Contact:

Re: Very low level hum and how to get rid of it

Post by ilya »

Well,
Basically, this is a front end for some console modules.

Yes, my mains is 50Hz, so 100Hz really sounds like a ripple thing.

Audio signals are treated respectfully, as well as grounds. I think I found the possible culprit. I've measured DC voltages and my positive rail shows 0V AC, and negative rail shows 0.002V AV.

Here's how they look on the scope.

Positive rail:
pos.jpeg
pos.jpeg (93.39 KiB) Viewed 15224 times
Negative rail:
neg.jpeg
neg.jpeg (98.47 KiB) Viewed 15224 times
Now, PSU has LM317 in both pos and neg rails. They are essentialy the same regarding cap and res values. Looks like I need to dig into the neg rail deeper...
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Very low level hum and how to get rid of it

Post by JR. »

2mV is not bad but the + rail looks better.

The PCB layout around power supplies is pretty critical. Think of PCB traces as having resistance, because they do.

The loop from rectifier to reservoir caps and back to common will have high AC current, looking similar to that waveform.

What is important is where the ground leg of the neg regulator is attached since that provides the 0V reference for the regulator.

Since the plus regulator output looks OK and the neg regulator has some ripple look at where the ground pin connects (should be same as the plus regulator that is clean).

Another issue that can cause ripple voltage in a regulated output is if the unregulated voltage is marginal and regulator is dropping out of regulation at bottom of ripple voltage dips.

That waveform looks more like charging current to me, so check regulator ground pin connection layout first.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Very low level hum and how to get rid of it

Post by mediatechnology »

ilya - I'd look exactly where JR pointed you.

I had a similar problem many years ago in a layout where filter capacitor ripple current was flowing through a trace that had a regulator referenced somewhere between the transformer CT and the capacitor. Moving the regulator ground fixed it.

This is also the symptom developed in gear where a filter capacitor has aged enough and lost value (or wasn't big enough to begin with) to the point that the unregulated supply ripple is causing regulator drop-out.
ilya
Posts: 88
Joined: Tue Dec 29, 2009 2:10 pm
Location: Russia
Contact:

Re: Very low level hum and how to get rid of it

Post by ilya »

Ok. J.R. nailed it.
Wayne, thanks for chiming in.

It was indeed a layout issue. Snice negative rail LM317 output was connected to GND and its GND pin was referenced to -V, I checked where that connection was physically made. Stupidly enough, the output was taken directly from the rectifier bridge. Yes, this is the same branch, but I was stupid enough to take -V not from the smoothing cap (C33p on schematic), but right from the bridge (circled spots on the picture).

Cut the trace and redid the connection from the right point and voila! Both rails look exactly the same. And that annoying low level hum disappeared completely!

Thanks a lot, gents, for your help and insight. That was an interesting and educational experience.

As a side note, is it possible to see the differential waveform on the 2-input scope without differential probe? I was playing with transformer balanced outputs and I can't see the correct waveform on my scope for some reason (with electronic balanced outputs everything's fine). Any thoughts?
Attachments
psu.JPG
psu.JPG (35.45 KiB) Viewed 15196 times
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Very low level hum and how to get rid of it

Post by mediatechnology »

As a side note, is it possible to see the differential waveform on the 2-input scope without differential probe?
Yes, absolutely. Use A+B mode and invert B (or whichever channel offfers it.)

My regulator issue at the time was the LM7915's ground having the filter cap after it so it sounds like exactly the same thing.
tubegeek
Posts: 25
Joined: Mon Mar 18, 2013 12:36 pm
Location: Brooklyn NY

Re: Very low level hum and how to get rid of it

Post by tubegeek »

In the context of high-voltage supplies, point-to-point wired, for vacuum tube circuits, the advice I have been given is: connect the power transformer center tap directly to the first filter cap after the rectifier. Connect the audio circuit ground directly to the last filter cap in the smoothing network.

This removes the high current rectifier pulses from the audio ground wire, preventing its nonzero resistance from developing a noise current from that particular source.

This is basically the same thing on a smaller scale physically, but possibly even higher currents, depending on what is being powered.
ilya
Posts: 88
Joined: Tue Dec 29, 2009 2:10 pm
Location: Russia
Contact:

Re: Very low level hum and how to get rid of it

Post by ilya »

I've got another strange behaviour.
Every once in a while on the positive rail I've got the following going on:
pos.jpeg
pos.jpeg (95.26 KiB) Viewed 15156 times
Is it LM317 oscillating or what? I can't find out what causes this. It seems that moving the reg leads sometimes helps, but I'm not so sure.

The regs are mounted off board. The leads are approximately 10-15cm. Any thoughts?
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Very low level hum and how to get rid of it

Post by JR. »

That looks like dropping out of regulation. Consider making reservoir cap larger to reduce ripple voltage as easier and cheaper fix than buying bigger transformer.

or not...

JR
Cancel the "cancel culture", do not support mob hatred.
Post Reply