Measuring THD (Total Harmonic Distortion)

Dave Lowther

Joined Sep 8, 2016
224
In building a sine wave oscillator the question is how pure is it? I need to get a piece of test equipment to measure THD, preferably in percentage. The best I have found so far is this:

https://www.amazon.com/gp/product/B07KC6X51M/ref=ox_sc_saved_image_10?smid=A18RFE6ZVD9PX6&psc=1

Which I really don't like . Any other suggestions?
It depends how low the THD is that you want to measure and how much you are prepared to spend.
I use this software "Multi Instrument" (which is free for 21 days) https://www.virtins.com/multi-instrument.shtml with a USB sound card. Of course if the sound card has more THD than your oscillator then all you will be able to deduce is that your oscillator THD is < your sound card THD. I bought a Roland Rubix 22 because of its 24 bit / 192 kHz capability, it's good up to about 80 kHz. It wasn't difficult to build a 1kHz spot frequency oscillator with about 0.0003% THD.
 

Audioguru again

Joined Oct 21, 2019
6,674
Years ago I made a sinewave oscillator with digital steps, then used switched-capacitor lowpass filter ICs to create very low distortion.

I made a distortion analyser using the same switched capacitor ICs and same clock as for the oscillator.
I compared mine to a very expensive analyser and the measurements were close.
 

tautech

Joined Oct 8, 2019
383
And you already have a scope with quite reasonable FFT capabilities that will show you a lot......if you know how to properly drive it.
This vid from Alan of Tek gives you all the basics:

However with your Siglent scope Wendy you need display more waveform cycles than is normal with other DSO's as Siglent memory management is more like the LeCroy WYSIWYG.
 

bassbindevil

Joined Jan 23, 2014
824
Mass market computer sound cards can achieve performance comparable to that HP analyzer. I picked up an Audigy 2 ZS for $5 at a computer recycler, but I expect the newer X-Fi cards are even better.
https://audio.rightmark.org/products/rmaa.shtml
Here's someone testing the output of a Blu-ray player; note that Rightmark shows you the spectrum, you don't have to sweep a filter and call out the numbers to your assistant who records them on graph paper.
https://www.jensign.com/RMAA/BDP83/index.html
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
1% would do, I may just settle for displaying the FFT from my DSO.If I ever figure out how to bring up that display.
 

AnalogKid

Joined Aug 1, 2013
10,987
Measuring the harmonic distortion in a signal, and measuring how much harmonic distortion a circuit introduces into a signal, are two different things. The first is a precision subtraction of a theoretically perfect signal from the signal under test, while the second is a precision subtraction of a signal from itself, before and after the circuit under test. What is your ultimate goal?

BTW, I think that old HP unit is the model we had back in my TV days. Watching it tune itself in Auto mode was very cool.

ak
 
Top