How to measure/trigger/view low level 60/120hz hum?

Fusion916

Member
2012-02-27 5:02 am
Do you use an oscilloscope or a measuring mic to measure the inherent 60/120hz hum from your amp? Without anything hooked up at the input, the output of the amp has a very small hum that is only audible if you put your ear up to the speaker. I'm trying to quantize the hum level and maybe do something about it (wire placement/isolation/etc), but my oscilloscope can't even trigger it.

Is a better way to measure this with a measuring mic? What is your technique?
 

Fusion916

Member
2012-02-27 5:02 am
You can measure it with the amp input shorted and a scope on the output.
You will need to put the scope on a low voltage range to pick it up enough to trigger the scope.

Hum problems are usually down to bad ground routing and/or poor decoupling.

The best way is a star ground at the smoothing capacitors.

Yes that what I'm doing right now, my oscope can't seem to trigger noise that low. If i apply a signal (even a small one, like 50mV), it can trigger fine.
 
If you have to put your head right up to the speaker to hear the hum then I don't think you have much of a problem.

You can try earthing the zero volt line, that might help.
Or even lifting it, sometimes earth loops can cause problems.

I did have slight problem wit ha class AB amplifier with hum with no input signal and that turned out to be poor decoupling between input and output parts of the amplifier.
I put a 33R resistor in series with the B+ and B- lines from the output power supply to the input supply.
It would need a decoupling capacitor after the 33r.
 
+1 I invested $200 odd in an emu sound card and fft software and have not looked back. I've reduced my system noise floor and distortion considerably by being able to measure changes in earthing, psu and bias setup.
Have you the time to give us a tutorial on how to connect the emu?

I downloaded a set up from this Forum. I have had no sound ever since. The software file changed some setting and I have never been able to undo that change. even reinstalling has not worked.
 

ctrlx

Member
2012-07-12 10:20 pm
Set timebase to around 2mS, set amplitude to minimum say 5mV/div, turn the trigger variable knob to come off "auto" to allow manual control of triggering.
Depending on the earthing of scope/amp weather or not to 0V scope probe to allow clean image.
 

discrete

Member
2010-09-11 8:39 pm
Best is to FFT on a 24bit audio card so to see the spread of harmonic noise in spikes across the bands.

This is the way to do it.
Best I can see with my 'scope is around 0.2 mV.

The attachments shows the output of a power amplifier, with a zero signal at the input.
The 60 Hz + harmonics peaks visible are 110 dB down (re.1 V).
This is about 3 uV.
The noise floor of the amplifier is better than 120 dB down, or about 1 uV.

The brown curve is the emu USB interface used for the measurements, wired in loop-back (basically the noise floor of the 'test equipment').
 

Attachments

  • 00 fft blame L, 0V, 10ohm, ifbox-0dB_cr.png
    00 fft blame L, 0V, 10ohm, ifbox-0dB_cr.png
    37.8 KB · Views: 94