Commercial SMPS series wired outputs

What is the procedure for wiring two commercial SMPS's in series to double the output voltage?

You used to be able to do this easily with transformer supplies, but, as I understand, the SMPS's have a built-in sensing circuit.

Thank you for any advice.

Bill Stuart

Reply to
Bill Stuart
Loading thread data ...

I've never seen a need for that how ever, if the commons are isolated from each other, and you have a small load on each output, I don't see why you can't do it..

this most likely also means, you can't tie the cases together depending on the design.

+....-+....-
Reply to
Jamie

Many of them have the minus output connected to chassis ground, so you'll short the "lower" one out through the ground connections.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

Depends on whether the supply was designed for that. A "well designed" supply should work. But it was probably never in the spec and an average designer won't think to worry about it or management won't pay the extra cost for a fix.

Steady state is not the problem. Problems happen in startup/shutdown/current limit. When one drives the other negative, smoke can ensue...depending on what the designer was smoking when he tested the design.

You can also get some wonderful limit-cycle oscillations when foldback current limits are in series. If it's driving the robot arm at your car assembly plant, you've got trouble.

Sometimes, you can put reverse protection diodes across the supplies to keep 'em from blowing up, but don't count on it without looking at the circuit and UNDERSTANDING how it works. And while "not blowing up" is good, it doesn't guarantee anything about the stuff it's driving.

And even if it does work today, you can't depend on it working tomorrow. All it takes is some purchasing manager at the supply vendor to enhance his bonus by removing 7-cents from the BOM. Still meets specs, ship it! The design team was disbanded years ago, so there's nobody there to complain. Or maybe YOUR purchasing agent sources an "equivalent supply" at lower cost. Seen it happen WAY too often in the real world. You don't find out about it until you have a thousand systems in the field and your customer suing you because his customer ran the auto assembly plant and somebody was killed. Ok, so I exaggerate, but you get the point.

Bottom line, if it ain't in the spec, don't count on it. The fact that you're even considering it is proof that designers do dumb things...those OTHER designers... ;-)

Reply to
mike

I was about to string eight of the MeanWell 48-volt, 3 amp switchers in series to get about 400 volts. That seemed about the easiest way to get to that output from 120-240 volts. They're only about $44 each. But the project got cancelled, or rather the customer did.

I figured that a reverse protection diode per supply was about all I'd need.

John

Reply to
John Larkin

With the combined output voltage in several hundred volts, there is going to be a significant AC+DC voltage stress between the primary and secondary side (including any feedback arrangements), so check these specifications for constant use (not just the short time mandatory isolation voltage test).

Reply to
Paul Keinanen

Mobile amateur radio transceivers typically require 13.8-14.4 V to give the rated output. For mains operation, some have used three identical computer power supplies by wiring the +5 V outputs directly in series to get 15 V and then using a series diode to drop the output voltage around 14 V.

Other than removing any secondary side connections from the chassis and mains ground, these seems to work without problems.

Reply to
Paul Keinanen

With such high combined voltages, look for the voltage ratings of bypass capacitors from the DC side 0 V or positive voltage to mains ground.

While one would not expect to find such capacitors, if the device did not originally pass the EMC test, some vendors might try various tricks randomly to get through the EMC tests and add capacitors and ferrites at random places. I would not be surprised to find capacitors with quite low voltage ratings from the DC side to the mains ground:-).

Reply to
Paul Keinanen

Fortunately, I've only ever seen proper Y type caps (those fat blue ceramic disks, sometimes an MKP/MKT). Those are all rated 250 or 275VAC, but they're really 1000-2000VDC. The films are lower (630-1000V) since they break down gracefully (self healing), but the ceramics have to take the whole thing else they just go *Phut*. (I have thinner ceramics rated for

3kV, also in the blue epoxy, so I wouldn't be worried about them.)

As long as the outputs behave, I wouldn't worry about putting a few hundred volts in series.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Reply to
Tim Williams

I hope you are right, but since the Far-East manufacturers that might encounter EMC problems might also be quite cost conscious, they might get any capacitor with the requested capacitance (but not necessary the required impedance or stray inductance and operating voltage :-).

Reply to
Paul Keinanen

You really should say (c) Joerg when you say *Phut*.

John

Reply to
John Larkin

I recall seeing regular capacitors instead of the required Y capacitors. So beware...

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
 Click to see the full signature
Reply to
Nico Coesel

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.