Looking for a moderately high voltage power supply (200-500V)

[DaJMasta]

I've got a couple good bench supplies for my home lab but occasionally run into something requiring a higher voltage than they can supply (60V in series) that I want to be able to test out of circuit or as a module and I currently have no means of doing so.  That said, I also want to avoid having extremely high voltages available on the bench for the safety of myself, whoever may walk in and push a button without me knowing, and for the sake of my more sensitive equipment/probes.

I'm looking for a supply that can supply a maximum of say, 200V-500V DC, and I won't need any more than 100mA, maybe with a minimum of 50mA or so.  I'd like current and voltage monitoring (preferably dual display) and constant current and constant voltage mode operation, like a standard bench supply.  There are some options (like the SRS PS310/325/350) which would otherwise fit the bill... but I think for the aforementioned safety concerns I don't want to be able to actually generate 1kV+ on the output.  Are there supplies designed for this high-but-not-too-high range?  Are there supplies that can generate more but can be hardware or software locked to a 500V or so limit?


Because the warnings are likely inevitable, I know that even the low end of what I'm looking for is potentially lethal and needs to be handled and setup carefully when in use or even powered up.  I am just looking for another precaution built into the supply (limited maximum output) to reduce the chance of a terrible outcome in the event of a fault or accident.

[Performa01]

I have an ancient Rhode & Schwarz NGK 280.
300V/200mA, 10-turn pots and analog instruments for voltage and current, of course CC and CV opertion and even sense terminals (inherited from its low voltage / high current siblings):

https://www.radiomuseum.org/r/rohde_stromversorgungsgeraet_dc_power_supply_ngk_serie.html

Great unit, I'm very happy with it.
 
There even appears to be a NOS unit for sale:

http://www.messtechnik-colberg.de/common/stromversorgung/geraete/w7de40e0e22270038ae8f26f9b70cd55

[Gyro]

A variac, an isolating transformer (or even 120-240 isolated step-up) and a voltage multiplier?

Crude but effective for that sort of current.

[anotherlin]

You may want to try the HP 6209B, can be easily found on ebay and lower priced ones go under $150.
However, voltage is limited to 320V, but it fits your current need at 100mA.
Completely analog and fully discrete!

[TurboTom]

If you don't mind importing from China (AliExpress), HSPY (Bejing Hanshengpuyuan Technology Co. LTD) may have suitable units. Their advantage is that they are fairly affordable and very small size. They may require some "touching up", though. Have a look here if you like.

[Zucca]

https://www.ebay-kleinanzeigen.de/s-anzeige/bio-rad-powerpac-3000-elektrophoresenetzteil-0-3000v/1269204384-168-6457

I can collect it and ship it to your home in USA if you want.

[Gyro]

https://www.ebay-kleinanzeigen.de/s-anzeige/bio-rad-powerpac-3000-elektrophoresenetzteil-0-3000v/1269204384-168-6457

Electrophoresis power supplies are very cheap, however they are not exactly the same as a lab power supply. I have a similar 3kV unit, and if it doesn't sense a minimum current on turn on it will disable the output. Also if the negative terminal is tied to neutral they will shut down. So they have their limitations.

I have an Amersham EPS-301 Electrophoresis supply, 0-300V, 0-400mA. That one has a fully floating output but regulation is VERY slow, major undershoots and overshoots on load change. Something I intend to work on when I have time!

[TimFox]

I have two -hp- 6209 power supplies which are invaluable for tube project development.
I had a problem with one unit, which worked nicely until I connected the negative binding post to the ground post and it went into current limit.
A surprising fault:  the original design had a rubber grommet in contact with a high-voltage trace on the circuit board, at one of the screws that mounts the PCB to the structure.  Over time, the grommet had carbonized into a small lump of charcoal, and would conduct to the chassis.  I replaced that grommet with a piece of teflon and it now works fine.

[unknownparticle]

I was going to suggest looking for a Farnell E350 but then I noticed you are in the USA!  So, keep an eye open for a Heathkit unit, I don't recall the model reference off hand but they appear on Ebay now and then. Good unit with a valve (tube) output stage and regulated.

[DaJMasta]

Thanks for the tips, the HP 6209B checks all the boxes except for the dual display, the HSPY-400 isn't one I would have really considered on my own given the unknown-to-me brand and the potential for danger in the supply... but it seems to be built the right way (maybe with the exception of the clearance mentioned), it seems like a bunch of people have been using them, and the rated specs seem to be quite good.

I think so far the HSPY is probably leading, since the price between that and a used 6209B is about the same.  As for applications, it is mostly tube stuff!  Biasing geiger tubes and photomultipliers, testing nixies, and a voltage source for high resistance measurements and similar, so the low current is plenty.

[james_s]

I have a Heathkit IP-32 that works quite well for this sort of thing, it's certainly not modern though. A friend of mine built a clone of it that he uses for working with tubes.

[DaJMasta]

Looked around a bit more, ended up making an offer on a supply that was accepted (sort of to my surprise) - splurged a bit and got a Kikusui PMC500 (500V, 100mA)

If anyone else is looking for a similar kind of supply, I hadn't thought of them but Kikusui has a bunch of well rated options.  For earlier models, there's the PAB350, similar to the other analog display supplies here but with a current and voltage meter and 200mA output capacity (and the going rate seems to be about that of the HP 6209B.)  Then they seem to get into the hundreds of volts range with most of their subsequent power supply lines, usually with a ~200V, ~350V, and sometimes 500V options with relatively low max output currents in the small form factor, then higher current options in larger footprints (as you'd expect.)


In any case, they're similar with the load regulation and ripple specs to the 6209B and others, but have units with revisions that are much more recent, up to their current supply ranges.

[anotherlin]

That Kikisui seems indeed the perfect choice for your need.
Looks like a HP 6209B, fully analog, but with digital read back.

If you can, I'd be very interested if you can do a teardown of it.
In particular, what's the "architecture". The HP 6209B is mid-60s design, with no opamp, fully discrete.
Yet it has a two stage regulation (SCR then classic linear), so it was somewhat an advanced design at that time.

[DaJMasta]

Showed up this week in good condition, a little teardown, then:

Very easy to tear down, you undo two screws on the top and the side of the unit folds off and pulls away - beyond that, you've gotta start pulling parts because it's packed in behind a bunch of things.

No fans involved, just two large heatsinks, one has two mosfets and a thermal switch, the frontmost one has just one mosfet and a thermal switch.

Doesn't appear to have any regulation of the high voltage, just a switcher of some sort, only linear regulators I could find were the top front section which are just powering the electronics on the front panel.

A pair of nice sounding relays click away when turning up the voltage.

Looks like the input voltage switching is done by changing connectors on the back of the input board rather than a differently wound transformer, which is handy.

Output reads fine with no significant AC voltage with a multimeter (the meter reads about half a volt high, but the front panel shows no decimals), don't have a proper high voltage probe to scope it and see and haven't done load testing, but it looks clean and I doubt Kikusui is going to be fudging their output specifications.

[BravoV]

Thanks for the teardown photos. 

I got an offer locally for a 350V model, relatively cheap, though I don't need it. 

Just curious, overall how do you score it ?

[DaJMasta]

Not sure I've really done the testing to score it, but as an impression: it's a very straightforward, clean design for a supply.  Sort of no frills in a lot of ways, minimal front panel and I haven't looked into the remote control options (though I believe there are some), but you just push the power button and it's ready to go, not a lot of input can be made because it's just very simple to use if you've used a lab supply before.  The inside seems clean and well built, and the design seems to be an elegant solution in its simplicity.  Not really sure if the overshoot/noise/input capacitance/dynamic load responsiveness (50us sounds fast, though) specifications are groundbreaking and don't have all the means to properly test it, but it certainly seems to suit any of my needs.

Oh and for some reason it struck me as being smaller than I expected.  As deep as a longer benchtop unit, but the footprint is much smaller since it's pretty narrow.


I said before I thought it was a switcher, but I think I'm just misunderstanding the design - a 50W switcher is not going to have that transformer.  I remember the transistor on the front heatsink is a mosfet, but maybe the two on the rear heatsink are configured to regulate rather than switch (maybe two PMOS and the one NMOS in the front?), I don't see an obvious variable regulator IC to drive them, though (maybe the measurement and feedback control happen on the front panel board?)... and it's packed in well enough I don't think I'll be pulling the transformer to investigate any time soon :p

[BravoV]

Thanks for the brief insight, appreciate it. 

It does support remotely controlling it, I've read the manual that is available online, it has remote voltage & current adjustment thru ext. voltage references or resistance, and the remote ON/OFF too.

Looking forward to hear more, if you happen to measure it's performance during the real use of it.

[BravoV]

Post here as reference , did similar tear down on my recently scored Kikusui PMC350-0.2A

-> https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/msg2866746/#msg2866746