Yet another HV regulator - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

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 3rd July 2015, 02:42 PM   #1
popilin is offline popilin  Argentina
diyAudio Member
 
popilin's Avatar
 
Join Date: Oct 2010
Location: Traslasierra
Default Yet another HV regulator

This is only a simulation, this circuit was intended to be a "raw supply" for a CCS plus a shunt regulator for sensitive valve circuits.

Rectifiers are EZ80, nevertheless in the simulation are used 6X4 because they are the most similar models that I have.

The results are too good to be true, so any criticism is welcome.

Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.

Added a soft-start for valve happiness

Click the image to open in full size.
Attached Images
File Type: jpg ScreenShot001.JPG (94.7 KB, 1332 views)
File Type: jpg ScreenShot002.JPG (121.6 KB, 756 views)
File Type: jpg ScreenShot003.JPG (121.6 KB, 698 views)
File Type: jpg ScreenShot004.JPG (73.3 KB, 692 views)
__________________
I do not suffer from insanity, I enjoy every minute of it. - Edgar Allan Poe
He has the most who is most content with the least. - Diogenes of Sinope
  Reply With Quote
Old 3rd July 2015, 06:36 PM   #2
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
ARC has developed a whole series of solid state regulators with floating op-amp based error amplifiers - every time something happens on the load side those op-amps generally get fried. I would recommend looking at ways to reference the error amp to the ground side of the world so that in real world malfunction situations the op-amp will not get fried. Start up behavior may be problematic with circuits implementing what I have suggested.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 3rd July 2015, 07:13 PM   #3
Elvee is offline Elvee  Belgium
diyAudio Member
 
Elvee's Avatar
 
Join Date: Sep 2006
Quote:
Originally Posted by popilin View Post

The results are too good to be true, so any criticism is welcome.
You are right: such figures are too good to be true, and questioning them is a sensible engineering practice.

I have published a number of HV regulator designs, some being conceptually very similar to your approach (and most of them tested for real), and my figures are nowhere near yours.

Here is the one I think is the most resembling: it uses an opamp and essentially identical principles. Some details differ, the supply of the opamp for example is a real one, not a floating spice source, but these details will mostly have a second (or higher) order effect.

CCS + Zeners or resistors for 450V reference

So why do I arrive at a mere 96dB ripple rejection instead of 150dB (the output impedance is 110µΩ vs 80µΩ, thus not that far off)?

The answer could be the opamp: you use a 5534, somewhat better than the 358, but that is probably not the root cause.

Your circuit includes no explicit frequency compensations at all, which for high OL gain circuit is an important omission, and that's probably where most of the difference lies. In order to have a stable and working circuit, you need to dumb down the idealistic version, and this has a cost in terms of raw performance

That said, you could also have hit something, because the floating supply allows a kind of bootstrapping helping the PSRR.

This certainly deserves some in-depth investigation, which I didn't carry out.
If you post the .asc of your circuit, this will allow easier scrutiny.

Note that the circuit I gave as an example has been tested and found working with some minor adjustments mainly related to latch-up and similar issues, and I have also posted similar ones in the same vein, you can make a search in the power supply section.
__________________
. .Circlophone your life !!!! . .
♫♪ My little cheap Circlophone© ♫♪
  Reply With Quote
Old 3rd July 2015, 09:43 PM   #4
popilin is offline popilin  Argentina
diyAudio Member
 
popilin's Avatar
 
Join Date: Oct 2010
Location: Traslasierra
Quote:
Originally Posted by kevinkr View Post
ARC has developed a whole series of solid state regulators with floating op-amp based error amplifiers - every time something happens on the load side those op-amps generally get fried. I would recommend looking at ways to reference the error amp to the ground side of the world so that in real world malfunction situations the op-amp will not get fried. Start up behavior may be problematic with circuits implementing what I have suggested.
Hi Kevin, nice to see you around here.

You are absolutely right, floating regulators are some faith act.

Nevertheless, I always used another circuit, with self-powered Op-Amp, much more suicide-style than this with no problems for years, ofcourse with lighter and almost constant loads as SRPP based preamps.

Start-up problems are solved in this case, but my fear is about the MOSFET, I want to use the IRF820 for its low capacitances.

Let's cross fingers.
__________________
I do not suffer from insanity, I enjoy every minute of it. - Edgar Allan Poe
He has the most who is most content with the least. - Diogenes of Sinope
  Reply With Quote
Old 3rd July 2015, 10:40 PM   #5
popilin is offline popilin  Argentina
diyAudio Member
 
popilin's Avatar
 
Join Date: Oct 2010
Location: Traslasierra
Hi Elvee, thanks for your reply

Quote:
Originally Posted by Elvee View Post
You are right: such figures are too good to be true, and questioning them is a sensible engineering practice.

I have published a number of HV regulator designs, some being conceptually very similar to your approach (and most of them tested for real), and my figures are nowhere near yours.

Here is the one I think is the most resembling: it uses an opamp and essentially identical principles. Some details differ, the supply of the opamp for example is a real one, not a floating spice source, but these details will mostly have a second (or higher) order effect.

CCS + Zeners or resistors for 450V reference

So why do I arrive at a mere 96dB ripple rejection instead of 150dB (the output impedance is 110µΩ vs 80µΩ, thus not that far off)?

The answer could be the opamp: you use a 5534, somewhat better than the 358, but that is probably not the root cause.
Rats! Mathematical models are treacherous, especially those of MOSFETs, the simulation is very sensitive to that.

Nevertheless I had checked the model for IRF820 and the result is quite good, corresponding with IR datasheet at least until Vds = 50V

The NE5534 is a great Op-Amp, and it is very cheap.

Quote:
Originally Posted by Elvee View Post
Your circuit includes no explicit frequency compensations at all, which for high OL gain circuit is an important omission, and that's probably where most of the difference lies. In order to have a stable and working circuit, you need to dumb down the idealistic version, and this has a cost in terms of raw performance

That said, you could also have hit something, because the floating supply allows a kind of bootstrapping helping the PSRR.

This certainly deserves some in-depth investigation, which I didn't carry out.
If you post the .asc of your circuit, this will allow easier scrutiny.
In simulations the circuit does not oscillates, the problem is that in real world I have only an old 20MHz scope, simulations are my silver bullet.

Into the .zip attachment, the .asc, thanks in advance if you find any problem.

Quote:
Originally Posted by Elvee View Post
Note that the circuit I gave as an example has been tested and found working with some minor adjustments mainly related to latch-up and similar issues, and I have also posted similar ones in the same vein, you can make a search in the power supply section.
Thanks again, I will take a look more carefully.

BTW, the auxiliary PSU

Click the image to open in full size.


Click the image to open in full size.
Attached Images
File Type: jpg ScreenShot001.JPG (108.0 KB, 594 views)
File Type: jpg ScreenShot002.JPG (120.6 KB, 587 views)
Attached Files
File Type: zip +B_Series.zip (3.7 KB, 11 views)
__________________
I do not suffer from insanity, I enjoy every minute of it. - Edgar Allan Poe
He has the most who is most content with the least. - Diogenes of Sinope
  Reply With Quote
Old 3rd July 2015, 11:32 PM   #6
diyAudio Senior Member
 
fdegrove's Avatar
 
Join Date: Aug 2002
Location: Belgium
Hi,

Quote:
Originally Posted by kevinkr View Post
ARC has developed a whole series of solid state regulators with floating op-amp based error amplifiers - every time something happens on the load side those op-amps generally get fried. I would recommend looking at ways to reference the error amp to the ground side of the world so that in real world malfunction situations the op-amp will not get fried. Start up behavior may be problematic with circuits implementing what I have suggested.
That plus the fact that it never works reliably in the long run and sounds SS anyway.

Who wants to count the number of ARC units, Counterpoint or whatever (mostly)magical US based designs that failed in always the same way?

Sometimes I wonder, did this fail on purpose or where they that stupid? I suspect the former......

Cheers,
__________________
Frank
  Reply With Quote
Old 3rd July 2015, 11:49 PM   #7
popilin is offline popilin  Argentina
diyAudio Member
 
popilin's Avatar
 
Join Date: Oct 2010
Location: Traslasierra
Hi Frank, nice to see you around here!

What is your opinion about my humble regulator?

Do you think it would be enough and forget the CCS plus shunt regulator?
__________________
I do not suffer from insanity, I enjoy every minute of it. - Edgar Allan Poe
He has the most who is most content with the least. - Diogenes of Sinope
  Reply With Quote
Old 4th July 2015, 05:03 PM   #8
Elvee is offline Elvee  Belgium
diyAudio Member
 
Elvee's Avatar
 
Join Date: Sep 2006
I made some checks on your sim.

The unbelievably good performances have a number of causes: one is the MOSfet model, with a reasonably realistic one, there is a 15~20dB drop in PSRR.

Another one is the opamp model: even with the LT1115, in principle vastly superior to the 5534, there is a drop of another 30dB (BTW, with the LT1115 in transient analyzis, it oscillates)

And finally, there is the floating supply of the opamp: it also helps a great deal.

However, in the real world, it would be difficult to make a supply sufficiently floating to keep the performances at that level (except with a battery)

For example, here is the rejection of a 1µA stimulus injected as a leakage: the effect at 50Hz is -75dB, much worse than what the input ripple can do.

Currents of this level are easily caused by parasitic capacitances to the transformer, AC wiring, etc

That said, the circuit does show promises, and it would be interesting to develop a working version tackling the issues identified: for example, use another opamp to bootstrap the supply of the regulating one. This would give the benefit of the floating supply, without the problems associated with it
Attached Images
File Type: png Pop1.png (82.9 KB, 102 views)
__________________
. .Circlophone your life !!!! . .
♫♪ My little cheap Circlophone© ♫♪
  Reply With Quote
Old 4th July 2015, 07:34 PM   #9
popilin is offline popilin  Argentina
diyAudio Member
 
popilin's Avatar
 
Join Date: Oct 2010
Location: Traslasierra
Quote:
Originally Posted by Elvee View Post
I made some checks on your sim.
Thank you very much for your time and dedication.

Quote:
Originally Posted by Elvee View Post
The unbelievably good performances have a number of causes: one is the MOSfet model, with a reasonably realistic one, there is a 15~20dB drop in PSRR.
As I said before, my model fit reasonably well with IR datasheet

Click the image to open in full size.

See for yourself

http://www.google.com.ar/url?sa=t&rc...96952980,d.eXY

However your "reasonably realistic one" does not

Click the image to open in full size.

See for yourself

http://www.google.com.ar/url?sa=t&rc...96952980,d.eXY

Quote:
Originally Posted by Elvee View Post
Another one is the opamp model: even with the LT1115, in principle vastly superior to the 5534, there is a drop of another 30dB (BTW, with the LT1115 in transient analyzis, it oscillates)
Little detail here: LT1115 is NOT for single supply operation

See for yourself

http://www.google.com.ar/url?sa=t&rc...96952980,d.eXY

Quote:
Originally Posted by Elvee View Post
And finally, there is the floating supply of the opamp: it also helps a great deal.

However, in the real world, it would be difficult to make a supply sufficiently floating to keep the performances at that level (except with a battery)

For example, here is the rejection of a 1µA stimulus injected as a leakage: the effect at 50Hz is -75dB, much worse than what the input ripple can do.

Currents of this level are easily caused by parasitic capacitances to the transformer, AC wiring, etc
After finish the match Argentina vs Chile, I will post the complete circuit with real world auxiliary supply, without fanciful additions.

Quote:
Originally Posted by Elvee View Post
That said, the circuit does show promises, and it would be interesting to develop a working version tackling the issues identified: for example, use another opamp to bootstrap the supply of the regulating one. This would give the benefit of the floating supply, without the problems associated with it
It is far easy to ruin a reasonably good design than improve a bad one.
Attached Images
File Type: jpg IRF820.JPG (77.7 KB, 528 views)
File Type: jpg STP8NM60.JPG (78.5 KB, 528 views)
__________________
I do not suffer from insanity, I enjoy every minute of it. - Edgar Allan Poe
He has the most who is most content with the least. - Diogenes of Sinope
  Reply With Quote
Old 4th July 2015, 10:02 PM   #10
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Quote:
Originally Posted by popilin View Post
<snip>
It is far easy to ruin a reasonably good design than improve a bad one.
This, unfortunately seems to be true.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  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


Similar Threads
Thread Thread Starter Forum Replies Last Post
HV shunt regulator iko Power Supplies 8 27th September 2011 03:24 PM
Please check my first HV regulator diego6339 Power Supplies 4 26th March 2011 09:42 PM
HV Regulator rsumperl Tubes / Valves 4 18th September 2010 07:20 AM
HV Regulator habsrock93 Tubes / Valves 3 21st October 2008 07:15 AM
PCB for HV Regulator jackinnj Power Supplies 13 5th February 2006 03:20 PM


New To Site? Need Help?

All times are GMT. The time now is 02:12 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright ©1999-2015 diyAudio

Content Relevant URLs by vBSEO 3.3.2