Easier to build a simple distortion analyzer.
I will try to make it my next project after building Cheff's UGS and UP amp.
Humm, will have to get an excellent A/D converter, don't think my Audigy EX will be up to the job, suggestions anyone.
Nelson describes a VERRY simple way to look at an amplifier's distortion
on page 12 of the pdf " Build a Mosfet Citation 12 " found on the Passdiy
web page. It is a good read so I won't spoil it for you by describing it and
it sure isn't perfect but talk about easy! Perhaps Nelson would tell us if
it's ok to use to find this "sweet spot" with.
on page 12 of the pdf " Build a Mosfet Citation 12 " found on the Passdiy
web page. It is a good read so I won't spoil it for you by describing it and
it sure isn't perfect but talk about easy! Perhaps Nelson would tell us if
it's ok to use to find this "sweet spot" with.
Thank you Woody for the tip, I looked at the article, very interesting, and many thanks to the Master
for all your work.
Probably this will not be good enough for some of your
lower distortion circuits, but it's better than nothing.
Also, it only reveals distortion from current at 0 volts, so it's
really applicable to biasing push-pull output stages, not
finding sweet spots.
A good quality audio interface is the CardDeluxe: Digital Audio Labs Home Page
And a great software program is SpectraPlus: SpectraPLUS - Audio Spectrum Analyzer
I have been using the SpectraPlus for years and it really is a great value and
super easy to use.
The easiest would be an simple op amp based IM analyzer.
I am going to build my F5, and I want to find that sweet spot!
Has anyone built (or looked at) Richard Crawford's "1 PPM IM Distortion Analyzer" (AudioXpress, March 2004)? I'm looking for something that won't break the bank, but something I can use for a while. The total cost was quoted at $150 (in 2004) including the board which is still available from AX, but excluding knobs, connectors, etc (and you use your DMM for the display). It's claimed to be able to measure IM to 0.0001%. It measures the difference frequency at ~800 Hz and uses various test tones: 1) 19.2 kHz & 20 kHz for 2nd order harmonic, 2) 9.6 kHz & 20 kHz for 3rd order, and 3) 9.6 kHz, 10.4 kHz & 20 kHz (all at same amplitude) for 2nd plus 3rd order.
Unfortunately I'm really new at this and don't always completely understand everything I'm reading. Also, I don't know if $150 is reasonable for what this does. I hate that!
Any help or advise would be very much appreciated!
Thanks,
Looney
Fore biasing my AB amps I have always used my ears, which seem to work fine
Small changes to one channel at a time
Its easy to detect which channel is better
I have found it usually ends at a point where DC on output is most stable, maybe not at the lowest value, but where its most steady
And I like heatsink on both channels to be equally hot
Needless to say I do it while playing music
I could imagine it might be a dangerous approach with classA
Small changes to one channel at a time
Its easy to detect which channel is better
I have found it usually ends at a point where DC on output is most stable, maybe not at the lowest value, but where its most steady
And I like heatsink on both channels to be equally hot
Needless to say I do it while playing music
I could imagine it might be a dangerous approach with classA
Last edited:
I would like to second Itsmee's motion for a list or table of approximate "sweet spot" operating points for popular parts. Especially 2sk170,2sj74 fets, typical "audio" bjts (ztx450,550; bc550,560 etc). Also, some sort of mention regarding the range of deviation.
I did a little reverse engineering and I would like to see if I'm on the right track; I went back and looked at the F5 and F4 schematics. I am guessing the jfets are optimal around Ids=6-8mA and Vds ~20v (Since I'm sure Mr. Pass would have gone through the trouble to find the "sweet spot" for his own products!) Is that about right for the "bl" parts?
I did a little reverse engineering and I would like to see if I'm on the right track; I went back and looked at the F5 and F4 schematics. I am guessing the jfets are optimal around Ids=6-8mA and Vds ~20v (Since I'm sure Mr. Pass would have gone through the trouble to find the "sweet spot" for his own products!) Is that about right for the "bl" parts?
Well, for the JFETs you might read the JFET BOZ thread. They play with the Supply Voltage alot looking for it...
http://www.diyaudio.com/forums/pass-labs/103050-jfet-boz.html
http://www.diyaudio.com/forums/pass-labs/103050-jfet-boz.html
Thanks tinitus,
Great idea of using one channel as a reference while tweeking the other. I will definitely try it.
Looney
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.