Hi - so at work we are looking at buying a DC 200V power supply. Would need to be maybe 50W or so. We have someohuge 48V DC power supplies. Is there any way to up that to about 200V DC? I've seen dc to dc converters, but all those that I've seen have been for smaller voltages and smaller currents. Hmmm... I suppose we could just wire 4 in series, could we not? Not sure we have 4 of them though, and that would be kinda clunky. Are there any better solutions than that?
I suppose you could have some sort of SMPS setup, with 48v going into four or five inductors, fed through switched power transistors, and each fed into a capacitor. The capcitors would be connected in series to create a bank, and the 200V DC would be tapped from teh top of the bank. Four stanges would give you 192V-odd, so if you need closer to 200v, you could have 5 stages, with a regulator before the final output.
Four power supplies in series is best. No offense, but you shouldn't be dealing with SMPS stuff if you had to ask this much. SMPS can lead to destructive results for the unwary.
If this is for a workplace, Health & Safety aspects are important. That said, you'd only need three 48V transformers if your existing units aren't smoothed. After rectifying & smoothing, 48V RMS goes to 48 x 1.414 = 67.8V DC, less about 2V rectifier drop. Important thing is to wire all the secondaries in phase, or use 3 rectifier bridges and series the DC outputs. Big 80-100V caps are also easier to find than big 200V+ ones.
I really ought to flame you for being such an arrogant prick - but I'm in a good mood, and it's nearly Christmas, so I'll be nice. First of all, I have no idea what kind of power supplies they are, so I don't see why you are making assumptions about that. If I had to make a guess I'd say that they are NOT switching, but again I'm not positive about that. And no kidding - of course high voltage can do damage. Why should I not ask a question about this? Better safe than sorry, eh? I remember my electronics teacher in highschool told us NOT to wire PSUs in series, but then again he was an idiot (a nice idiot, mind you, but an idiot none the less) so normally I take the opposite of what he said to be true. And anyways, I'm fairly sure that we don't even have 4 of em, so I don't think putting them in series is a viable option. We have 48V DC power supplies. I doubt that stealing their transformers would be ok with my professor. This is for a research project at my university. Much of what we do I expect doesn't follow normal guidelines.
Check and see if the power supplies are isolated/floating or not. Because if they are, then it is perfectley safe to series them. If each - is grounded, than it isn't. If it isn't floating, you could get a bunch of beefy power transistors, and set up some capacitors so that you charge them in parallel, and discharge them in series. Make it 2 seperate banks of caps, that way it alternates, one charging one discharging, than it switchs. Of course you would need beefy output caps to smooth it. So you would need lots of caps. Lots and lots of caps. The good thing is that it would be hard to make one of these so that it runs out of control and gives you really high voltages. (192 V at absolute maximum if you use 4 stages, still have to add in losses from the switching transistors) At 250 mA (50W) it shouldn't be too bad, but at higher currents, expect lots of voltage sag. Of you can get AC, you could check out a Cockroft-Walton voltage multiplier. Normally they are for (really) high voltage but it should work fine with a lower voltage. In fact a switching capacitor DC design would be similar to the Cockroft-Walton except that the Cockroft-Walton dosen't need transistors, just diodes. Of course this probably will be expensive for the caps, so you would still probably be better off finding a transformer and making a 200v supply, or buying one.
If your floating isolation is on the edge and you can choose your ground at the output make it +96V (2x48V) and -96V (2x48V). That way you reduce the stress on the 48V supplies.
I don't care about your flames, since it's your problem anyways. I didn't mean to assume that the power supplies you had were SMPS, but rather, that nohope4me recommended an advanced solution that assumed a lot about you, the thread starter. He left out a lot of specifics that could cause problems if they are ignored. SMPS is dangerous stuff, so my post was merely a warning to tread carefully if you use his solution. Perhaps I should have specifically acknowledged nohope4me's post, but it was right above mine and mentioned switching-mode stuff while yours didn't. Okay, finally, why is SMPS dangerous? The high voltages are one thing, but an unstable SMPS setup can result by simply picking the wrong inductance/capacitance/resistance/part number. Instability can lead to excessive current draw, thermal runaway, and a spectacular explosion of chips and hot shards all over the place. It's more than just voltages.
Why not go simple? Use 240v AC, run it through several diodes to take it down a bit, then regulate and smooth it, just like a normal linear supply. You only need 250mA anyway, right? Or even run it through a transformer. Dunno where you can get one with such a low ratio, and capable of 240->200.. but itd be a lot easyer than messing with SMPS. Edit: now I see your in usa, which is going to be a pita for this. use a 1:2 transformer? = 280v dc, and use 80 diodes (or a better method) to reduce it down to 200...
Or build a voltage doubler from a few diodes and caps, and lower the voltage with a zener. Only downside is that you're not isolated from the mains, and you may need some phat capacitors to smooth out the rectified AC.
