anyone built an amplifier Distortion measurement circuit - Page 2

Bob Cordell · 2nd May 2010, 10:31 PM

Quote:

Originally Posted by steevo

Has anyone built a distortion meter and signal genrator to measure
amplifier distortion accuratley. I have been looking around the web for circuit designs and in electronics magazines and cant seem to find anything.
Have any of the popular magazines ever published a distortion measrement
instrument. Has anyone come across any.

Cheers
Steevo

Hi Steevo,

You can find the original construction article for my THD analyzer (published in Audio) on my web site at Cordell Audio: Home Page. However, as mentioned, it is a difficult build. A good distortion analyzer is quite complex. BTW, the oscillator portion of my THD analyzer is less difficult to build and provides very low distortion.

Alas, all of that was about 30 years ago (time flies), before PC's and soundcards. It is a lot easier to start out with a good sound card and some software. I use a Juli@ soundcard, although I still also use my original distortion analyzer.

Cheers,
Bob

Pano · 3rd May 2010, 04:04 AM

With a good card like the M-Audio you can do a lot. Some folks have even built custom front ends for the card.
Plus you get all the cool stuff like spectrum analysis which is much more valuable than THD.
These days, it's really the way to go.

Samuel Groner · 3rd May 2010, 08:29 AM

Quote:

If possible I would like to measure down to 0.01% distortion at 500 Hz, 1 kHz, 5 kHz, 10 kHz, 15 kHz and 20 kHz.

As the others said: that's easily done nowadays by a decent interface and some clever software. Analogue oscillator/analyser is needed if you want to get well below -100 dB and/or above 50 kHz or so.

Samuel

Bob Cordell · 3rd May 2010, 01:42 PM

Quote:

Originally Posted by Samuel Groner

As the others said: that's easily done nowadays by a decent interface and some clever software. Analogue oscillator/analyser is needed if you want to get well below -100 dB and/or above 50 kHz or so.

Samuel

This latter is a very good point. Doing a decent job of measuring 20 kHz THD requires measurement BW out to 200 kHz. Few soundcards, do very well above about 50 kHz, even when sampling at 192 kHz.

Bob

EUVL · 3rd May 2010, 03:55 PM

A living example of Bob's design :

My implementation of the Cordell Distortion Analyser

Patrick

Ian Finch · 05:18 PM

Hi guys,
I'm faced with a similar measurement need and I looked about for economical solutions. The professional PC cards packages are excellent, of course, and I wish I could afford one etc.
I thought of a more flexible approach, since I only infrequently need to use it, though it has a flaw; no software for harmonic analysis.
I have this start, though.
1) cheap local kit DSP sig. gen. kit with <.001% distortion.
2) Audio Dac (existing gear) but kit OK (optical links preferred)
3) 16-bit resolution PC scope (Picotech or Handyscope etc.)
4) Usual low inductance loads for things like power amplifiers.
Since I like to measure harmonic content up to, say the ninth, calculating this is real
tedious. I'm not disagreeing with Bob's assessments here, just suggesting that it will suffice for my DIY needs.
I have had the gear working, but it is tedious to read off harmonic peaks to calculate THD. It should be possible in software to do this rapidly from the FFT plot, you might expect. Anyway, I feel confident of measuremets down to .002% or better. I hope to verify this in the coming months.
So, what do you think of this under 500 UK new gear solution?
What hopes for upgrading PICO'S software? (I have asked)
Any (polite) suggestions? Does it help or suggest something better for your own needs?
regards I

pdp11 · 3rd May 2010, 05:46 PM

I had buided two oscillator circuit,
one based on a wien bridge circuit (inspired by a linear application note or the lm1007 datasheet, i don't remember) stabilazied with 4 small lamps. I can control the frequncy with a uC by a doble digital potentiometer and the selection of the capacitor.
I use this generator as source
the other is a vco inspired by a national semiconductor application note and use this as a frequency source for an AM demodulator of the signal.
The voltage control is generate by a dac connected to the uC.
The uC can adjust in real time both the frequency generators comparing signals with their cristal.
I'm not satisfied at all for the solution, the first generator works well, but the the vco/am block doesn't a clean work because measure more distortion than the reality.
I wish to replace both demodulator and vco, but now i have enough time to follow this project.

3rd May 2010, 04:04 AM	#12
Pano diyAudio Moderator Join Date: Oct 2004 Location: North Carolina Blog Entries: 4	With a good card like the M-Audio you can do a lot. Some folks have even built custom front ends for the card. Plus you get all the cool stuff like spectrum analysis which is much more valuable than THD. These days, it's really the way to go. __________________ Take the Speaker Voltage Test!

3rd May 2010, 05:03 PM	#16
Ian Finch diyAudio Member Join Date: Apr 2010 Location: Coffs Harbour, on the east coast	Hi guys, I'm faced with a similar measurement need and I looked about for economical solutions. The professional PC cards packages are excellent, of course, and I wish I could afford one etc. I thought of a more flexible approach, since I only infrequently need to use it, though it has a flaw; no software for harmonic analysis. I have this start, though. 1) cheap local kit DSP sig. gen. kit with <.001% distortion. 2) Audio Dac (existing gear) but kit OK (optical links preferred) 3) 16-bit resolution PC scope (Picotech or Handyscope etc.) 4) Usual low inductance loads for things like power amplifiers. Since I like to measure harmonic content up to, say the ninth, calculating this is real tedious. I'm not disagreeing with Bob's assessments here, just suggesting that it will suffice for my DIY needs. I have had the gear working, but it is tedious to read off harmonic peaks to calculate THD. It should be possible in software to do this rapidly from the FFT plot, you might expect. Anyway, I feel confident of measuremets down to .002% or better. I hope to verify this in the coming months. So, what do you think of this under 500 UK new gear solution? What hopes for upgrading PICO'S software? (I have asked) Any (polite) suggestions? Does it help or suggest something better for your own needs? regards I __________________ regards Last edited by Ian Finch; 3rd May 2010 at 05:18 PM. Reason: OOps , sorry Bob

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Distortion measurement on Mac	peranders	Software Tools	8	19th April 2010 08:04 PM
Measurement of speaker distortion	PMA	Multi-Way	15	3rd November 2009 06:08 AM
Who built the RED CIRCUIT 3-5 Watt class-A amplifier?	rmgvs	Solid State	1	2nd November 2005 09:52 PM
distortion measurement	djdamix	Solid State	1	20th August 2004 07:11 AM
distortion measurement	audiobot	Tubes / Valves	9	23rd January 2004 12:14 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

3rd May 2010, 03:55 PM	#15
EUVL diyAudio Member Join Date: Oct 2003	A living example of Bob's design : My implementation of the Cordell Distortion Analyser Patrick

3rd May 2010, 05:46 PM	#17
pdp11 diyAudio Member Join Date: Apr 2010	I had buided two oscillator circuit, one based on a wien bridge circuit (inspired by a linear application note or the lm1007 datasheet, i don't remember) stabilazied with 4 small lamps. I can control the frequncy with a uC by a doble digital potentiometer and the selection of the capacitor. I use this generator as source the other is a vco inspired by a national semiconductor application note and use this as a frequency source for an AM demodulator of the signal. The voltage control is generate by a dac connected to the uC. The uC can adjust in real time both the frequency generators comparing signals with their cristal. I'm not satisfied at all for the solution, the first generator works well, but the the vco/am block doesn't a clean work because measure more distortion than the reality. I wish to replace both demodulator and vco, but now i have enough time to follow this project.