technical noise measurement question

[EC8010]

Engage smug mode. I have a DPO4034 oscilloscope with 10Mpt memory per channel at work, so I can use the FFT to plot 1/f spectra. The white bit can be done quite handily with a dedicated audio test set plus HP filter. One day I'll get a thermocouple meter.

[jackinnj]

you mean "thermopile" meter perhaps. The Fluke 8920 runs around $20 on EBay and is every bit as good as the HP 3403.

I have to use a much less elegant scope for looking at noise -- TEK 5223 with 5A22N -- I purchased it from Medtronic via EBay for a brass farthing -- everything was in cal.

You can also use a Keithley nano-voltmeter -- these have a provision for attaching a recorder so the various DC levels are really noise -- as in the case of the TEK plug-ins you have to discharge the capacitor attached to the unit prior to measurement so that you don't over load the meter.

As well, I think that it is important to "know" noise both along the ITU-486 curve as Dolby explained in his AES article -- as well as the impulse noise which you can measure with a peak reading meter or the envelope function on your scope.

[EC8010]

We call them thermocouple meters over here; a tiny heating element driven by the noise in thermal contact with a thermocouple to measure the heat. What really hurts is that years ago I passed up the opportunity for a video thermocouple meter. It would have done very nicely for noise...

I looked at a Keithley nanovoltmeter, but it didn't seem to have the required bandwidth. perhaps you've seen a model I didn't see?

486? Is that different, or did you mean 468? My audio test set (not quite a brass farthing, but close, at £47.50) can apply 468-2. I hadn't thought of using envelope as a means of catching noise peaks. The audio test set has a post-filter oscilloscope output (I use it for distortion).

[jackinnj]

the Keithley 181 is a DC meter -- you aren't going to get an instantaneous measurement but you can catch the readings on the 2mV scale after the thing is zero'd and integrate them. It's kind of like catching a bunch of butterflies with a net by waving the net at fixed intervals.

I am sure that some physicist has figured out how to devise a rectifier for nano's.

The Keithley 181 is so sensitive that if you put warm up the measurement leads even by holding them in your hand for a few seconds the meter reads off the thermo-electric effect.

[Juergen Knoop]

Originally Posted by jackinnj

to get an idea of the noise which is relevant to listening you should download the ITU CCIR ARM-486 paper

I have read now the CCIR-468 complementing DIN 45405. Those features a quasi-peak rectifier.
I'm still curious about the arm proposal. There seems to be a 'vanished' natsemi app-note about that.
Sadly I got no answers back then:
Noise Measurement AN-180
Regards

[EC8010]

AN-180 "RMS Converters and Their Applications" is in the 1991 Linear Applications Handbook. I've just been reading it and it doesn't seem to be all that useful. It's mostly about how to do RMS conversion (which is what you'd expect from the title), where you might want to do it, and how jolly super their LH0091 RMS conversion chip is.

[Juergen Knoop]

aah thank you!
Regards