12v to 300v @ 1200W

I did kind of the reverse of that for a school project years ago; synchronous rectified 0-6V at up to, 50 or 100A, I forget. Single turn copper strap secondary, PP. A proportionally larger core, at somewhat higher frequency, will be more than happy with an application like yours in PP.

I wouldn't go half bridge on this, 100A is a lot to ask from a massive wad of capacitors.

SiC schottky sounds good, if expensive; junction (high speed) diodes should be as good if the frequency isn't insane (under 200kHz, say).

Tim

-- Seven Transistor Labs Electrical Engineering Consultation Website:

I'm thinking push-pull or half-bridge.

I'm leaning towards the half-bridge because it will make transformer easier to construct. I might be able to use one turn of copper or the primary and probably 55 or so turns on the secondary. Maybe some SiC schottkys on the output. The unit only has to run for a few seconds at a time. I figure I can put some small sharing resistors across the split capacitor bus to keep flux from walking up on the core.

Any thoughts?

Bob

ng near 1200w on the input side when efficiency taken in.

er to construct. I might be able to use one turn of copper or the primary a nd probably 55 or so turns on the secondary. Maybe some SiC schottkys on t he output. The unit only has to run for a few seconds at a time. I figure I can put some small sharing resistors across the split capacitor bus to k eep flux from walking up on the core.

Commercial or 1 off? Cheapest or best? Quality or cost cutting? etc.

NT

ng near 1200w on the input side when efficiency taken in.

er to construct. I might be able to use one turn of copper or the primary a nd probably 55 or so turns on the secondary. Maybe some SiC schottkys on t he output. The unit only has to run for a few seconds at a time. I figure I can put some small sharing resistors across the split capacitor bus to k eep flux from walking up on the core.

Thanks Tim.

I was leaning towards half-bridge because one turn would be easy for primar y. If i go push pull then I have to go put two turns on the primary and sp lit them. Not the end of the world - I could probably wind it. I was thin king leakage would be less with one primary turn and one secondary layer di rectly on top. Volume is a concern, so I will run at least 200kHz. Figure d that would be as slow as I would go.

I recommend push-pull with a 'dead zone' to keep from walking up the BH Curve.

I know, more on the transformer, but simpler on the overall gotcha's.

Else you could spend a lot of time debugging when you sound like you simply need the function.

And cycle by cycle current mode control too.

piglet

That's a lot of current to be switching at high frequency. So EMC compliance is of no concern?

If you don't have, get femm 4.2 to help you 'design' that transformer.

Yeah, there is a very decadent solution: Buy a big camping inverter for the European market (the cheaper modified sine variety) and bypass the output bridge :-)

Hard to beat the price if you only need one

-Lasse

Yes, but if the OP really needs 1200W continuous duty he should spend 10 bucks more and get a 2000W or higher version. Often the specs on such converters are on the ... ahem ... optimistic side.

However, one must keep in mind that many such converters have no isolation between 12V side and HV side.

true marketing watts ;)

in the initial post OP said it only needed to run for a few seconds at a time

-Lasse

Sure, so maybe use instead a motor/generator pair! Any good used-auto-parts emporium can get you a starter motor and a dead generator, that you rebuild with higher voltage rectifiers. Couple with step-up pulleys as appropriate. Some welding setups use this combination, I believe...

If you only need one, and don't care so much about the price:

Tripp Lite makes some 12 V DC to 240 V AC inverters that might fit the bill.

Buy a switching power supply with 240 V AC input (which usually means about 340 V DC on the transformer primary) and a 12 V DC, 75 A output, and then steal the transformer from it.

Matt Roberds

On a sunny day (Tue, 1 Apr 2014 06:27:43 +0000 (UTC)) it happened snipped-for-privacy@att.net wrote in :

How about doing that charge 25 caps in parallel, and then put those in series thing. Does not seem impossible, you would need an extra 300V cap for charge transfer, some small inductors (25) to limit charge current to 3A... batch of MOSFETS, driver, no big transformers, efficiency? MIT....

ng near 1200w on the input side when efficiency taken in.

er to construct. I might be able to use one turn of copper or the primary a nd probably 55 or so turns on the secondary. Maybe some SiC schottkys on t he output. The unit only has to run for a few seconds at a time. I figure I can put some small sharing resistors across the split capacitor bus to k eep flux from walking up on the core.

Thank you all for the input. I'm probably going to do push-pull. Sorry I can't get into the particulars. My client doesn't want me spilling my guts on the internet. I especially liked the comments regarding pulling apart an inverter. The Tripp lite 12v to 240v might be worth looking into.

regards, Bob

You might also take a look at a 1200 watt or greater 12V to 240 VAC off the shelf inverter. Rectified output of that is not quite but almost what you are looking for and might be good for at least an idea of what they do. Output would be 100 Hz or 120 Hz of course full wave rectified.

boB

At lower voltages a push-pull topology is preferable with only one Vce drop while a (half)bridge arrangement always has two Vce drops.