How to measure THD at home? (1 Viewer)

[MingL]

Anyone knows how to measure THD using a RS SPL meter and a pc-based audio spectrum analyser?

Thinking of measuring THD and reporting how it really affects bass reproduction.

 

[TimForman]

I'd be surprised if this is possible. I've only heard of THD being measured using THD analyzers or shaker systems as an output of an amplifier circuit. As far as I know this is an electrical measurement not an acoustic sound wave measurement.

 

[MingL]

My asking came from the fact that subs distort when they pushed hard.

When they distort, some of the distortion arises from the mechanical stress the driver contributes. The electronics will also introduce but surely some distortion comes from the driver too. Driver distortion should also introduce unwanted harmonics too, right?

THD could also be a means to measure the total distortion arising from electrical and mechanical non-linearity. We shall assume that the microphone's non-linearity is compensated for.

 

[Edward J M]

Hi Ming:

You've may have read I've been thinking about the same testing ability.

This spectra software program looks interesting, the price is right, and it offers auto compensation for the FR of your mic and your sound card. The financial risk is tolerable if it doesn't work.

http://www.trueaudio.com/rta_abt1.htm

This was lifted from the Q&A portion of the website. I particularly like the table showing how the levels of the woofer-created harmonics correlate into a distortion percentage.

Regards,

Ed

Q: I'm thinking about purchasing the full package primarily for testing loudspeakers. Do you have any other insight as to how this package would serve me for testing loudspeakers. Can I use it to measure a loudspeakers distortion? You can use TrueRTA to measure loudspeaker frequency response and also evaluate the distortion performance of loudspeakers. For distortion testing you drive the speaker with a sine wave and directly read the levels of each distortion component in the measured spectrum. For example, a distortion component which is 20 dB below the level of the test tone constitutes 10% distortion. Here are some other levels and the corresponding distortion level:

dB Below Test Tone % Distortion
-5 dB 56.2%
-10 dB 31.6%
-15 dB 17.8%
-20 dB 10.0%
-25 dB 5.62%
-30 dB 3.16%
-35 dB 1.78%
-40 dB 1.00%
-45 dB .562%
-50 dB .316%
-55 dB .178%
-60 dB .100%
-65 dB .056%
-70 dB .032%
-75 dB .018%
-80 dB .01%
-85 dB .0056%
-90 dB .0032%
-95 dB .0018%
-100 dB .001%
-105 dB .00056%
-110 dB .00032%
-115 dB .00018%
-120 dB .0001%

 

[TimForman]

Would you be performing these tests in an anechoic room?

 

[MingL]

No anechoic chamber to do that sort of THD testing. Perhaps quasi-anechoic?

But it would be useful to approx reconiseable certain levels of THD by ear. I mean, we always ask ourselves" How do we know the 20hz tone is pure?"

Bass tones are more tolerable to more distortion, to say about 10%. But how does one know what 10% THD sounds like.

 

[Joel X]

Would outside qualify as quasi-anechoic?

 

[TimForman]

Would outside qualify as quasi-anechoic?

Sure, unless you live next to the airport.

So, in the event you're successful at getting accurate readings how will you use them to improve the sound of your system?

 

[MingL]

Nah, not my idea to improve the sound by getting the THD.

Instead, i hope to be able to understand and correlate THD to the actual sound. Sometimes, even tweaking the channel levels, BFD levels and sub gain, THD can possibly reduce.

But just a wild guess, since I'd really like to know how THD affects my bass and to what level is recogniseable.

 

[Michael R Price]

Ming, Spectraplus has a THD measuring function. Just plug a microphone into your computer and run the spectral analysis, go to the Utilities menu and select the THD option. It won't be very accurate unless you have a decent sound card and microphone and you test with low ambient noise; although at bass frequencies it should be easier to get decent readings. You could try testing close-miked to the driver and port to reduce room frequency response effects on the apparent THD.

 

[TimForman]

Just trying to satisfy a little intellectual curiosity, huh? Excellent! Please share what you find out.

 

[Joel X]

Or at least share the experience of getting to the data (software learning curve, what microphone you end up using, outside/inside...).

Instead, i hope to be able to understand and correlate THD to the actual sound.

It will be a bit tough to describe how differing intensities and resonance of distortion sound...

 

[AndyDC]

I've tried using spectraplus to measure distortion and found it was pretty tricky, more so than frequency response. The main thing for me was that background noise is a much bigger problem than for frequency response. After all, if your background noise is 20db lower than the testing frequency, that's not a big deal for frequency response (you'd be off by less than a decibel, I believe) but of course that's a lot of distortion, 10% I guess.

 

[Michael R Price]

Andy,

Right. My friend tried to do a quick THD measurement on his speakers, in a somewhat noisy room with a laptop and RS SPL meter as a mic, to see how much the bass distorted with volume. Playing a tone at low level (and presumably low distortion), Spectraplus quoted the distortion at 11%. When the volume was increased, the measurement changed to 18%. Not a very useful reading, but I assume it could be done with a good measuring setup.

 

[MingL]

Just recently downloaded Spectraplus and did some simple sub THD measurements on my SVS.

I'm not sure if this is the correct way, but here's how I hooked it up.

RS SPL mic -> RCA out (line level/mic level?) -> Notebook (on batts) mic-in. My notebook didn't have a line in. Ambient noise below unmeasureable threshold (

 

[AndyDC]

From memory, I'm pretty sure that the meter is a line out, not a mic out. I don't know what happens when you get that wrong, though . . .

 

[MingL]

Would there be any prob feeding the line-out from the SPL meter to the mic-in of the notebook in terms of accuracy?

Or should I get a Creative Extigy to get a proper line-in input?

 

[Joel X]

I snagged this from the spectra plus site

"For laptop users we are recommending the ST190 USB sound card from Sound Technology Inc.
This provides a professional quality sound system using the USB (Universal Serial Bus) interface.
For more information on this device go to http://www.soundtechnology.com/buster.html.
"

I suppose it is more accurate than a Soundblaster...