Bob Cordell Interview: Error Correction - Page 368 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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
Reply
 
Thread Tools Search this Thread
Old 23rd January 2009, 11:04 AM   #3671
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Default Re: Re: Re: Integrator options

Quote:
Originally posted by traderbam


Amusing emoticon sequence!

I suppose one consideration is summer input differential voltage range. With an op-amp, the input difference is going to be very tiny. With the PFB buffer, the input difference will be as large as the voltage swing to the plant...because the buffer only has a gain of 1. So one needs a very linear summer in the face of a relatively wide differential input voltage swing.
That's one reason I used a CCII in my paX amp. Open loop G=1 buffer yet very small diff input signal.

Jan Didden
__________________
If you don't change your beliefs, your life will be like this forever. Is that good news? - W. S. Maugham
Check out Linear Audio!
  Reply With Quote
Old 23rd January 2009, 11:05 AM   #3672
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Quote:
Originally posted by Edmond Stuart
Thx

Edmond,

This is what I got with Mathcad trying to get some insight in Hec as to stability. Not very sophisticated, but maybe it is usefull as a start.
Note that I have left the input summer at =1, and varied the fb summer b between 0.9 and 1.1, and K=0.95.
K, b and a are simple first order things with the same -3dB freq.

What I would like to do is vary another variable, like the ratio between -3dB frequencies and produce a 3-d plot like a sheet, but so far I haven't been able to coax Mathcad into doing it....

Comments?

Jan Didden
Attached Files
File Type: pdf clg mag revised.pdf (24.0 KB, 87 views)
__________________
If you don't change your beliefs, your life will be like this forever. Is that good news? - W. S. Maugham
Check out Linear Audio!
  Reply With Quote
Old 23rd January 2009, 11:25 AM   #3673
ingrast is offline ingrast  Uruguay
diyAudio Member
 
Join Date: Sep 2004
Location: Montevideo
Default Re: Re: Re: Integrator options

Quote:
Originally posted by traderbam


..I suppose one consideration is summer input differential voltage range. With an op-amp, the input difference is going to be very tiny. ...
In the OpAmp implementation of the unity gain summing node, you must resort to a network of resistors to make for the gain an summing ratios as required, the tiny differential voltage you refer being the virtual ground, external voltages substantial.

I actually computed, simulated and built an amplifier, with everything checking as expected. Stability could be coped for with basic lead compensation both for a mosfet Szilaky output (posted somewhere here) and for a LM3886 as plant, obtaining excellent results both in measurements and listening (Mosfets doing quite better than the LM3886 and both better than standard LM3886 configuration). Measurement results are also posted somewhere here.

As said, just stripping all the local feedback should have yielded the same results, but did not bother to put it into practice except in simulation.

Actually I have been happily listening for years to my homebuilt electronic crossover 2 way active boxes totalling 4 mosfet output amplifiers of the early type (EC).


Rodolfo
  Reply With Quote
Old 23rd January 2009, 12:22 PM   #3674
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Hi Jan.

Looks interesting, but I'm afraid there's still a lot of work to do.
I would suggest the following:
Since the OPS has probably the lowest cut-off frequency and there is little room to change that, it might be interesting to see what happens with a higher ft of S1 and/or S2. Somewhere there must be an optimum with regard to distortion reduction, overall BW and peaking. I realize it's a lot of work, but worth the effort.

BTW, though the peaks are not that large, I'm not really happy with them. If the system is embedded in a complete amp, it might cause troubles.


Good luck,
Edmond.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote
Old 23rd January 2009, 12:55 PM   #3675
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Quote:
Originally posted by Edmond Stuart
Hi Jan.

Looks interesting, but I'm afraid there's still a lot of work to do.
I would suggest the following:
Since the OPS has probably the lowest cut-off frequency and there is little room to change that, it might be interesting to see what happens with a higher ft of S1 and/or S2. Somewhere there must be an optimum with regard to distortion reduction, overall BW and peaking. I realize it's a lot of work, but worth the effort.

BTW, though the peaks are not that large, I'm not really happy with them. If the system is embedded in a complete amp, it might cause troubles.
Good luck,
Edmond.
Well, indeed, if the plant's Fc is (much) lower than that of a or b, peaking increases. If the reverse is true, also. Peaking appears minimal with all Fc's equal.
One solution is to lower a and/or b fc, but that also lowers the error correction of course.
But I'll get there, eventually. I need to look what happens if a, b and K are more realistic with multiple poles etc.
But it's just a matter of imagination and then soldering it all together. Easy, right ?

Jan Didden
__________________
If you don't change your beliefs, your life will be like this forever. Is that good news? - W. S. Maugham
Check out Linear Audio!
  Reply With Quote
Old 24th January 2009, 09:47 AM   #3676
diyAudio Member
 
traderbam's Avatar
 
Join Date: Jan 2002
Location: Earth
Default Re: Re: Re: Re: Integrator options

Quote:
Originally posted by janneman


That's one reason I used a CCII in my paX amp. Open loop G=1 buffer yet very small diff input signal.

Jan Didden
Yes, the input voltage difference is kept small by using a current FB op-amp like the AD844. It also helps that the system gain is only 1. You could have used the AD844 in conventional forward gain mode. Did you try this?
  Reply With Quote
Old 24th January 2009, 10:01 AM   #3677
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Default Re: Re: Re: Re: Re: Integrator options

Quote:
Originally posted by traderbam


Yes, the input voltage difference is kept small by using a current FB op-amp like the AD844. It also helps that the system gain is only 1. You could have used the AD844 in conventional forward gain mode. Did you try this?

No I didn't. I didn't want to get into discussions that it did what it did because of some 'hidden' high feedback in the summers ....

Jna Didden
__________________
If you don't change your beliefs, your life will be like this forever. Is that good news? - W. S. Maugham
Check out Linear Audio!
  Reply With Quote
Old 24th January 2009, 06:24 PM   #3678
diyAudio Member
 
traderbam's Avatar
 
Join Date: Jan 2002
Location: Earth
Consider these two circuits. Fig. A is a simplified PAX (I believe). Fig. B is a conventional NFB circuit. The components within the dotted box represent the AD844.
Attached Images
File Type: jpg pax hec vs cfb.jpg (48.7 KB, 537 views)
  Reply With Quote
Old 25th January 2009, 09:51 AM   #3679
diyAudio Member
 
traderbam's Avatar
 
Join Date: Jan 2002
Location: Earth
I modified Fig B to correctly include a FB resistor and the 150pF input capacitor.

I present these to facilitate a comparative discussion. Fig 1 is Jan's "PAX implementation of HEC" and Fig 2 is a functionally similar circuit that does not use a PFB loop. Showing the two in stark contrast may help with clarity of understanding.

I am interested in your opinions about the pros and cons of each circuit. What would cause you to favour one over the other?

Brian
Attached Images
File Type: jpg pax hec vs cfb.jpg (49.0 KB, 498 views)
  Reply With Quote
Old 25th January 2009, 10:54 AM   #3680
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Nice! I like it when everything is made as simple as possible, but not simpler, to quote Mr. Einstein (I think).

I don't intuitively see immediately what the performance difference is between the two. C2/C4 being as large as they are, I would *suspect* but don't kill me for it, that the performance is pretty similar.

That C2/C4 at 150pF is determining the final ec in the paX amp. On my test bench I had it working with just 36pF, and the ec was at least 5x better, but I couldn't get it unconditionally stable with cap loads, so in the interest of stability margin I went for 150pF.

I guess you simmed these, can you show us the beef please?
As shown I guess it is stable without C2/C4; how does it look without these caps?

Jan Didden
__________________
If you don't change your beliefs, your life will be like this forever. Is that good news? - W. S. Maugham
Check out Linear Audio!
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 05:45 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2