High voltage, high current super regulator, possible group buy

[peranders]

The super regulator as it was presented had limited data because of the design but nothing says that it can be done for any voltage and any current but if the requirements are more than 36 volts out and 1.5 A (1 A in real life) a redesign is necessary. There are a half dozen of avaible pcb's out there. The latest one comes from Taiwan but this one and noone else is rather well suited for higher power.

My idea was to design a regulator for

* 5-100 V out, 3 A min
* Normal 36 volts opamps, speed goal AD825 without oscillations
* LM431, LM329 and similar references
* LM338 preregulator, maybe
* Overvoltage protection for the LM338 and SR
* Short circuit protection (more advanced than usual, low loss)
* Maybe rectifier bridge and smoothing on the same pcb
* Fuse for incoming power
* Hole mounted parts mainly but possible to use SO08 opamps also
* Nice looking pcb
* Mainly "normal" parts, high industrial quality

Time frame: First half of 2006

At the moment I haven't dived into possible design problems concerning using opamps with less supply voltage than the output voltage.

I may drive this project according to my group buy guide lines.

How do you think it sounds?

[analog_sa]

Sounds very good. Is it too much of a stretch to ask for 250-300v tube-friendly variant? What are the issues with higher voltages? Just the size of electrolytics/resistors and the minimum track clearance?

[jackinnj]

i think you overstate the difficulty of building a hv regulator -- if it's opamp based you lift the "4" pin off the ground with a zener diode, protect the inputs with clamp diodes, etc., etc. use higher voltage components -- take a look at the Audio Research SP-10 which uses localized HV regulators for the phono and amplification stages.

in hv you are dealing with necessarily noisy thermionic devices anyway...

jack

[agent.5]

Originally posted by analog_sa
Sounds very good. Is it too much of a stretch to ask for 250-300v tube-friendly variant? What are the issues with higher voltages? Just the size of electrolytics/resistors and the minimum track clearance?

http://www.tubecad.com/2005/January/blog0031.htm

I actually think a 300V tube version is quite easy to make. I am working on a 2000VDC shunt reg using the same circuit.

[analog_sa]

take a look at the Audio Research SP-10 which uses localized HV regulators for the phono and amplification stages.

The Jung regulator is way more sophisticated than the SP-10. It would be great if it becomes available as a high quality PCB suitable for high voltage as similarly to an SP-10 i feel each stage should get it's own regulator.

I actually think a 300V tube version is quite easy to make.

I am currently using something similar with good results. Still, building 6 of these critters may become a bit tiresome.

[peranders]

Originally posted by analog_sa
It would be great if it becomes available as a high quality PCB suitable for high voltage...

My limit for voltage is around 100 volts.

[peranders]

Originally posted by jackinnj
i think you overstate the difficulty of building a hv regulator -- if it's opamp based you lift the "4" pin off the ground with a zener diode, protect the inputs with clamp diodes, etc., etc. use higher voltage components -- take a look at the Audio Research SP-10 which uses localized HV regulators for the phono and amplification stages.

in hv you are dealing with necessarily noisy thermionic devices anyway...

Making a wideband device which I plan the regulator to be is not easy. Compare our Taiwanese friend which has made a nice looking pcb but it's impossible to get it stable with AD825. He has made a minor error about the grounding and the bandwidth is heavily reduced. This may not be a problem if bandwidth wasn't the issue.

So it may be easy, it may also be difficult....

[blip1882]

Hey I was just wondering about a high-current super regulator!

Anyway, I'd be definitely be interested depending on final parts costs etc.

Also, if higher voltages became the sticky-wicket... I'd probably be interested in just a high-current version...

[peranders]

What is your application for "high current"?

[analog_sa]

our Taiwanese friend which has made a nice looking pcb but it's impossible to get it stable with AD825.

I had absolutely no problems with stability with 825.