Generic supply for detectors

Things at the limits.

Re: Generic supply for detectors

Postby Jerry » Tue May 31, 2011 2:00 am

If you want to do up some boards I use the dorkbotpdx.org circuit board fab program. You get three boards for $5 a square inch. A 2"x2" board would cost $20 and that would get you three boards. Hard to beat for little runs. And they are really nice boards. Double layer, plated through, gold flash, soldermask and silkscreening. routed to shape and size as well. http://www.dorkbotpdx.org/wiki/pcb_order
Jerry
 
Posts: 573
Joined: Sun Jul 18, 2010 12:07 am
Location: Beaverton, OR

Re: Generic supply for detectors

Postby Joe Jarski » Mon Jun 27, 2011 6:56 pm

Here's my first crack at the PCB for the generic HVPS. It has a 2 stage multiplier and is putting out ~1500V from 4.2V input. I had the PCBs made by the place that Jerry suggested above. There are a few things to improve upon, one being that some of the plated through holes were a little tight for the components that I have. In the same batch I also got the N-detector preamp PCBs that I will follow up with under that thread shortly.
Attachments
HVPS-PCB.jpg
CCFL HVPS
User avatar
Joe Jarski
 
Posts: 231
Joined: Thu Sep 16, 2010 8:37 pm
Location: SouthEast Michigan

Re: Generic supply for detectors

Postby Doug Coulter » Mon Jun 27, 2011 7:31 pm

Nice! I always run into that too tight holes thing, to the point where AP Circuits now insists you spec the net hole size rather than drill size -- their different processes depending on which you order "fill in the holes" a different amount with plating.

This looks real good for proportional tubes and other things that don't draw a ton of current or care too terribly much about noise on the HV (though this should be nicely quiet) -- it will serve almost all purposes, and the extra stage of multiplication can be used to run the CCFL on low volts, which makes it draw less quiescent current.

I'll have to put out my design mods for phototubes soon, and maybe we can get a board for that too. Since PT's are so doggone sensitive to output volts and noise (if they are 2 wire types), for that I add an opamp to close the loop around the whole thing for regulation, and tailor the volt multiplier accordingly. I've got one PT setup that draws nearly 5 watts - so you can't have extra stages and run the CCFL on low volts, it runs out of power(!).

Gee, I might have to buy some of these myself. Save me a lot of hassle around here, I must've wired 20 by hand so far.

FWIW, the hv diodes I have in stock are longer than those shown. For stuff like this it's usually best to make the board a little bigger -- save the "scrunching" for specific product designs.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Generic supply for detectors

Postby Joe Jarski » Mon Jun 27, 2011 8:46 pm

In the next version I'll spread stuff out on the output side so that those longer diodes can be used along with larger caps. That should make it easier for everyone to find parts that wants to use it, and more flexible. I can remove the holes that I put in for the regulator protection diodes and adjustment pin cap too - since those aren't used.

It would probably be too much to put both versions on one board and then pick and choose which version you want when you build it, you think?
User avatar
Joe Jarski
 
Posts: 231
Joined: Thu Sep 16, 2010 8:37 pm
Location: SouthEast Michigan

Re: Generic supply for detectors

Postby Doug Coulter » Mon Jun 27, 2011 9:04 pm

That sounds good. Yeah, we should just do a separate one for "high power, high regulation", I think. It'll necessarily be bigger to support the opamp and larger output caps, no need to make the simple one big -- more useful as is, and the small disc caps on this model are fine, standard and gettable forever. The "blue block" caps are only required when the current demand is high and the ripple needs to be really low. That model would just be a simple doubler anyway to get the output current as good as possible, since some phototubes really eat the power in the divider chain.
My large ones are "only" about 4 megs - at 2kv (tops, usually use less volts) that's about a watt, but it seems like more when you see how hot the socket full of R's gets.

I have built one of those already, and it works great, I just have to pop the box top and see what I did and write it down. Closing that loop was slightly tricky due to the delay (phase shift) in the CCFL and output filter. The bench model I put a 10 turn pot on and it's real reproducible and nicely stiff for things like that. I scored about 100 of those nice 10t pots surplus....so I have a supply of them that won't break the bank.

The diotech diodes are .55" body and need about .65" hole spacing unless we want to stand them on end (not that I haven't forced them into narrower spacings, but it looks ugly). Keeping them flat is good from a noise standpoint so they don't act like antennas for the 50kz on them and radiate it all over the place. We'll want to fashion a box for this anyway, metal for shielding, they're real noise makers. Or find something off the shelf. I've been known to use metal flashing tape to make a plastic box act like a metal one re shielding.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Previous

Return to High Voltage, High Power

Who is online

Users browsing this forum: No registered users and 5 guests