Super Regulator, collecting the facts - Page 16 - diyAudio
Go Back   Home > Forums > Amplifiers > Power Supplies

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 22nd October 2003, 10:23 PM   #151
diyAudio Member
 
john curl's Avatar
 
Join Date: Jul 2003
Location: berkeley ca
Well folks, isn't it great to have Walt Jung communicating directly with us?
A few more comments: "The Art of Electronics" is one of the best general/ beginning books in linear design. I have it here, and gave one to my business partner, Bob Crump (degree in psychology), for his birthday. If someday I am not available to answer a technical question, he can refer to his book. I, especially thought that the section on low noise design was very advanced, even today.
I used to have #18 'Vacuum tube amplifiers', before the firestorm. I might just try to find it again. I remember learning about the 'White follower' in that edition.
Another book that I might recommend for beginners especially:

The 'Active Filter Cookbook' by Don Lancaster Sams, isbn 0-672-21168-8
from 1975, or even earlier. It saved my tail almost 30 years ago.
  Reply With Quote
Old 23rd October 2003, 06:16 PM   #152
EUVL is offline EUVL  Europe
diyAudio Member
 
Join Date: Oct 2003
Default Audio Electronics 4/2000

Dear Mr. Jung,

A bit late perhaps, but I would still appreciate a copy of the article. My email address is ybpkwan@yahoo.com.

I have also experimented with one of your regulator designs, not the usual EDN version, but that from Electronics Design Analog Application Issue of June 1997, with a few modifications :

-- Replace D1, D3+D4 by two sets of LM336 5V in series with a green LED to get 7.5V. I used LM336 instead of LM329 because I like the idea of being able to adjust the voltage by connecting a 10k Trimpot across the diode.

-- D2 is red LED

-- Q1 is D44H11; Q2,3 are 2n2907's; U1 is AD797 amongst others

-- R6 is adjusted to allow 2mA through D1

-- Additional 7.5V ZD between emitter of Q3 and ground to lower dissipation of latter

-- L1 included but at 330uH

The rest is pretty much the same as in the article, and I have followed the grounding scheme, etc . straightly. With an OP37 as U1, there were no stability problems and the noise level is beyond the resolution of my oscilloscope. But I did experience some stability problems with AD797, and even worst with FET opamps like OPA604 or OPA637.

In the end I solved the problem by adding an additional RC filter between Vout and pin 7 of U1, plus a current source (J508) to replace R6. But I still do not fully understand the cause of the problem (I think LM336 is quite sensitive to PS noise), and wonder whether what I did was the best way to address the issue.

Your comments therefore most appreciated.

Best regards,
Patrick
  Reply With Quote
Old 24th October 2003, 04:52 PM   #153
WaltJ is offline WaltJ  United States
diyAudio Member
 
WaltJ's Avatar
 
Join Date: Aug 2003
Location: USA
Quote:
I have also experimented with one of your regulator designs, not the usual EDN version, but that from Electronics Design Analog Application Issue of June 1997, with a few modifications :
The article in question can be found at http://home.comcast.net/~walt-jung/w...g_Circuits.pdf
Quote:
-- Replace D1, D3+D4 by two sets of LM336 5V in series with a green LED to get 7.5V. I used LM336 instead of LM329 because I like the idea of being able to adjust the voltage by connecting a 10k Trimpot across the diode.
The D1 change is functionally OK, but if working at low voltages, LEDs work better for D3/4. I assume you *aren't* doing 5V. A *high* voltage for D1 is better, as it means less gain from the error amp. Plus, an LM329 is more quiet than an LM336.
Quote:
-- D2 is red LED
OK if R7 is scaled like in the 4/2000 article (249 ohms).
Quote:
-- Q1 is D44H11; Q2,3 are 2n2907's; U1 is AD797 amongst others
Q1 change is OK, *PN2907* is preferred, actually.
Swapping op amps isn't encouraged in such a finely tuned system, unless you are fuly prepared to troubleshoot. The AD825 of the 4/2000 article seems to be much improved as to stability issues. See caveats and tips in that article on stability.
Quote:
-- R6 is adjusted to allow 2mA through D1
OK...
Quote:
-- Additional 7.5V ZD between emitter of Q3 and ground to lower dissipation of latter
I don't understand why this is necessary, or how you mean it is connected. Q3 won't overdissipate, unless you are making very high voltages??? How much current is coming from Q2?
Quote:
-- L1 included but at 330uH
OK
Quote:
The rest is pretty much the same as in the article, and I have followed the grounding scheme, etc . straightly. With an OP37 as U1, there were no stability problems and the noise level is beyond the resolution of my oscilloscope. But I did experience some stability problems with AD797, and even worst with FET opamps like OPA604 or OPA637.
Pretty much the same? I don't think so! I repeat, see caveats on stability! FYI, OP37 and OPA637 are *NOT* unity-gain stable!
Quote:
In the end I solved the problem by adding an additional RC filter between Vout and pin 7 of U1, plus a current source (J508) to replace R6. But I still do not fully understand the cause of the problem (I think LM336 is quite sensitive to PS noise), and wonder whether what I did was the best way to address the issue.
The pin 7 (or 4) RC decoupling can't help but help stability, but should not really be necessary. I really can't give you a concrete answer here, as so many things have changed. I recommend adapting what you can towards the 4/2000 article.

All that said, some comments on this editing system at DIYAudio. It stink! I doubt that I'll answer any more mesages where you can't quote a big block of text, can't spell check, etc. This is near torture!

Thanks for your interest, Patrick, and good luck with things.

wj
  Reply With Quote
Old 24th October 2003, 09:21 PM   #154
EUVL is offline EUVL  Europe
diyAudio Member
 
Join Date: Oct 2003
Default Version June 1997

Dear Mr. Jung,

-- I assume you *aren't* doing 5V.

I was trying to do 15V.

-- OK if R7 is scaled like in the 4/2000 article (249 ohms).

I was still using 50R, as I tried to see if I can drive 500mA load, which is probably why Q3 got too warm.

-- The AD825 of the 4/2000 article seems to be much improved as to stability issues. See caveats and tips in that article on stability.
quote:

-- how you mean it is connected.

My typing error. I connected the ZD between Grd and Q3 collector, in an attempt to spread the dissipation of Q3 (R7 at 50R, see above). Vin was 18V, Vout 15V.

--I recommend adapting what you can towards the 4/2000 article.

I'll try the 2000 circuit in due course. But as you mentioned in your direct email, there are benefits of the June 1997 circuit. And I did it as an experiment just to try things out.

And again many thnaks for your comments.

Patrick
  Reply With Quote
Old 26th October 2003, 12:03 AM   #155
diyAudio Retiree
 
Join Date: Oct 2002
Location: Spain or the pueblo of Los Angeles
Smile Jung Super Regulator, collecting the parts

I have finished the positive regulator using the kit I got from ALW. I made few changes that seemed good ideas in theory, but I have not done listening test to compare with the stock version of the regulator. I cut the trace to the op amp supply pin and added a series 10 ohm resistor before the 10uF Oscon decoupling cap. This was to not have two high Q caps in parallel and to decouple the op amp from the bootstrapped supply at high frequencies where the op am has very low PSRR. The increased impedance for the op amps supply is not an issue since the driver between the output BJT and the op amp is a low capacitance mosfet. The mosfet is a very high impedance for the op amp to drive at audio frequencies and should not wiggle it's supply despite the 10 ohm resistance. The higher gate to source voltage of the mosfet allows the use of a LED, with it's much lower noise, for the level shifting. A correctly RC bypassed LM329 might be an even better choice. The current source and pass transistors are 100 MHz Toshiba BJTs and the regulator appears to be very stable. I would use a Hitachi C1775 and A872 for the current sources if I had some on hand. I replaced the 10K resistor bias for the current source LED with a jfet current source. This should result in about a 45dB improvement in LED voltage change with respect to input voltage change. Replacing the 10K with two 5K resistors with a cap to ground might be an even better approach. The feedback and voltage reference filter resistors have been greatly increased in value to allow the use of F-Dyne polypropylene and foil caps. Lead free 4% silver solder was used for the whole board. Will probably try a pre-regulator later on.

Andy did a fantastic job on the PCBs and I look forward to finishing them and listening to them this week. I hope he and Walt both, will forgive me for messing with an already excellent design, but it seems to be a rather incurable habit of mine at this point.

Fred



No, I will not answer questions about this on the forum, Mr. Pand S. I might discuss the rational for the changes via Email to interested individuals, but really don't want to get in that level of detail on the forum.
Attached Images
File Type: jpg andyreg.jpg (49.1 KB, 1273 views)
  Reply With Quote
Old 26th October 2003, 12:36 AM   #156
diyAudio Member
 
Join Date: Sep 2002
Location: Sweden
Fred,

nice work. I would have been surprised if you had not tweaked
it.

Anyway, on a first brief read-through I think you argue well
for the reasons for your tweaks, and I am not so sure any
further clarification should be necessary.

It looks very professinoal, by the way, but then, you are
a professional.
  Reply With Quote
Old 26th October 2003, 04:00 AM   #157
grataku is offline grataku  United States
diyAudio Member
 
Join Date: Dec 2000
Location: -
Very good Fred, thanks for the description of what you did.
As X-ster says you description seems fairly complete.
If your purpose was not to provide help to peranders you might have failed.
At this point anyone that can't understand what you said shouldn't be messing with the circuit in the first place.
If it's not too much trouble let us know your sonic impressions.
  Reply With Quote
Old 26th October 2003, 05:31 AM   #158
diyAudio Retiree
 
Join Date: Oct 2002
Location: Spain or the pueblo of Los Angeles
Default A picture may not not beat a thousand words

"At this point anyone that can't understand what you said shouldn't be messing with the circuit in the first place."

I was joking about refusing to help anyone. But as you point out, changes of this nature require an intimate knowledge of the circuit. I really don't think most people should make changes this extensive. Despite the description, there are about a half a dozen questions one should be asking before attempting these mods. There is more to this than I have described, and I would not attempt this on the basis of only what I have written, it is insufficient information. I do think I add the LM329 level shifting circuit but it will require a little Spice modeling to do correctly.

As you can see Andy's stock version is much nicer looking than my modified version and is shown with the DIP op amp package option.
Attached Images
File Type: jpg awlstock.jpg (58.2 KB, 1165 views)
  Reply With Quote
Old 26th October 2003, 06:37 AM   #159
Electrons are yellow and more is better!
diyAudio Member
 
peranders's Avatar
 
Join Date: Apr 2002
Location: Göteborg, Sweden
Blog Entries: 4
Nice soldering work Fred. Not very often you post any DIY'ish of your own.
__________________
/Per-Anders (my first name) or P-A as my friends call me
Super Regulator SSR03 Group buy
  Reply With Quote
Old 26th October 2003, 09:23 AM   #160
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Default Mods to ALW's Jung reg

Fred,

On the replacement of the LED current source with a JFET current source, I second that idea. I tried it out some time ago and it did improve the line rejection by a factor of more than 10. I also had trouble the right JFET CS, I settled for a 6mA one, most others I found were lower current, and with high loads the pass transistor ran out of base current (was a D44H11 I believe). But it also ruins your dropout voltage, which of course may not be an issue if you have enough input DC at hand.

Jan Didden
  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 04:46 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