Create 'known' signal distortion

[Tony_G]

Hi All,

I have an old 331A Distortion Analyzer that I'd like to check out - Ideally I'd use a signal with a known distortion but I can't seem to find any documentation on how to get my arbitrary waveform generator to do it.

I can take a sine wave, modulate it with the noise signal and get either a noisy amplitude or frequency but I can't seem to find a wave to get it to distort the sine itself. There has to be some math somewhere that I just can't find that says "Square/Triangle/Ramp etc signal of amplitude X, frequency Y and other properties A & B gives an equivalent sinewave of Foo % distortion".

Any thoughts?

TonyG

[TimFox]

Any of the usual engineering handbooks should have tables of the Fourier series components for typical periodic waveforms, including symmetric triangle and square waves, asymmetric ramps and pulses, etc.
It's a good mathematical exercise to convert these components to THD.  Of course, the usual function-generator outputs will have relatively high THD.

[DmitryL]

I have an old 331A Distortion Analyzer that I'd like to check out - Ideally I'd use a signal with a known distortion but I can't seem to find any documentation on how to get my arbitrary waveform generator to do it.

I can take a sine wave, modulate it with the noise signal and get either a noisy amplitude or frequency but I can't seem to find a wave to get it to distort the sine itself. There has to be some math somewhere that I just can't find that says "Square/Triangle/Ramp etc signal of amplitude X, frequency Y and other properties A & B gives an equivalent sinewave of Foo % distortion".

If you take a look to THD definition, you will find out that it is just enough to have 1 extra harmonic to produce a given THD. Sum of 2 sine waves should be enough. If it is enough to test the equipment I don't know

[Marco]

A quick google tells me you use a high purity sine generator, clip the sine, attenuate the clipped signal and add to original sine.

[Tony_G]

Thanks all - I'm either having a senior moment or it's just a lack of coffee this morning.

I did a bunch of searching but it never occurred to me to try "calibrate THD" or to just go crack open the old books.

Thanks everyone for setting me straight.

TonyG

[oldway]

You must not expect high precision measurements with HP331A because it is not a true rms measurement.

[Tony_G]

Really just trying to bring it back into spec but i'd like to confirm real operation above just the 'meter should show 0.4% or less distortion here' that the performance tests in the manual show.

[TimFox]

If you need true-RMS indication for large THD, you can take the analog output (after nulling the fundamental) and feed it into a true-RMS audio meter.

[oldway]

If you use a true rms multimeter, it must have an "audio" bandwith as the Brymen BM869.

A tip: if you don't have a low distortion generator, you may calculate in first aproximation the distortion of your amplifier of this manner:

Mesured THD = THD of the generator + THD of the amplifier
Then, THD of the amplifier = measured THD - THD of the generator.

[Conrad Hoffman]

Cordell's site is down right now, but you might be able to use a similar procedure to what he did with his design, if you can download his THD build article. Just a matter of applying a known amount of second harmonic frequency with a single sine generator. Nothing fancy required, not even a particularly good generator.

[Tony_G]

Thanks all, appreciate the info and guidance.