12V->60V @3-4A SMPS design

You could start of here

martin

Do you need isolation? Do you care about weight/size? What is the actual load? Are there EMI compliance requirements? Do you have airflow? Paul Mathews

Polyphase flyback would be interesting. There are controller chips around nowadays. It devolves to, say, three 80 watt boost converters.

But what about using standard bricks, Vicor or something, with some opamps or whatever for charging control?

John

No. The inputs are isolated- we're using individual Lambda supplies from line voltage to 12VDC.

No, within reason.

Approximately 20 to 60V at 0.1 (?) to 4A, programmable by a micro. Since it's charging batteries, current control is the ultimate goal. A trickle charge capability is desirable.

No, within reason (the chargers can't interact to the point where performance is affected). There is some commoning of one side of the output lines, so a synchronized setup would be preferable so that current and temperature could be measured without interaction. That may be just a detail-- if it's designed in from the start.

Yes. I have to assume that the environment may not be air conditioned and could be fairly hot (eg. Africa) or fairly cold (eg. northern Canada).

Best regards, Spehro Pefhany

Interesting idea.. the voltages are off standard, but might be close enough for "turn down" capability. Last brick I used had a pretty simple op-amp/resistor setup for changing the output voltage. I don't want custom- they have this great web-based quotation setup for custom voltages, but IME, a real quote never comes back from it... maybe it's just a black hole for gathering data. ;-)

Best regards, Spehro Pefhany

OK, the obvious question: why 120 to 12 to 60?

John

Presumably you have size or efficiency requirements that preclude plugging together a ready made mains inverter and a ready made battery charger?

I would be reluctant to bother doing a design for two dozen units. Plenty of inverters and 48V 3-stage battery chargers available cheap, delivered direct from Hong Kong or Taiwan.

The closest thing I found off the shelf with a two minute google search is this;

I'm guessing you want to fast charge a nominal 48V stack of lead acid batterys. 53V is enough for float charging. It's probably possible to change a resistor and get a few more volts out.

Bob

I have a universal charger for laptops... changing the connector changes the FB resistors to set the supply voltage.

...Jim Thompson

My personal favorite for high power levels - to several kW - is an H-bridge driven with a phase-shift controller. But these designs can be a bit cantankerous, and might require rework or several passes before production. Once working well, they are champs, however.

We know that there are (were?) lots of > 300W off-line flyback PC power supplies, made in huge quantities. So they can't be too bad.

But to reduce 12V current ripple, I was going to suggest using two high-frequency flyback ICs with two small transformers, but synced out-of-phase with each other. I'm not sure which IC would be the easiest to sync that way, Spef, would you like some suggestions?

Someone else may suggest a sepic converter.

Automotive class D subwoofer amp with DC isolation removed? ;-)

LTC makes polyphase controllers, and now maybe TI, too.

At 5:1 boost, it can be done with inductors, so no custom magnetics.

John

What I'd consider WRT to input ripple current is to simply eliminate it.

A 50%+50% duty cycle push-pull stepping up the voltage to something like

Ripple currents are reasonable everywhere.

Since you don't need isolation, and boost ratio is not too high, I'd consider boost, or boost with tapped inductor, if you want to keep it simple. If you don't go too high in frequency, you can use a powdered iron core. The most important component choice will be the fast recovery rectifier. Paul Mathews

They mostly don't exist in this voltage range, not that I can find anyway, and the ones I've tried don't work very well and have algorithms that have problems.

Yes, it's not my first choice, that's for sure.

If I could find something industrial strength at an acceptable price, I'd buy it.

That's interesting. Maximum voltage is a bit on the low side, and so is output power, so it's not quite there.

They're NiMH, I want control over the algorithm- thermistor as well as voltage depression (and discharge) and prefereably the ability to change battery technology with approximately the same voltage (currently 33 cells and 16 cells per pack). I also have to deal with possible charger-charger interaction. Lead acid is definitely not in the cards, but one of the Li technologies might be, if it can be proven safe.

Best regards, Spehro Pefhany

To re-use some existing supplies and because they should be isolated, and because I don't think it's worth trying to make the off-line part with the safety issues for a handful of pieces. I certainly could use

Best regards, Spehro Pefhany

Several years ago, I had a conversation with Laszlo Balogh (TI/ Unitrode) about why they don't push the full-bridge phase control chips much anymore (UC3875/UCC3895), and he admitted that it was because there are too many pitfalls in the design of those converters, including problems with flux-walking due to drive asymmetries in certain conditions (power-up/down, brownout, overload) and various MOSFET failure modes associated with that control method. There are lots of interesting papers in the literature about how to overcome these problems, but I'd caution anyone thinking about putting something together in a short time. If you do go that route, start out by at least including a capacitor in series with the primary. Paul Mathews

Doesn't current mode pretty much get rid of those problems?

Best regards, Spehro Pefhany

Hmm.. nasty, nasty peak currents, >100A.

Best regards, Spehro Pefhany

Have you considered push pull. No fancy high side gate drivers to deal with. Full(two quadrant)flux swing on the core.