need a 5kw DC - DC mutiplier, please help, have some ideas!

[electrokute]

I need a 5kw DC - DC mutiplier. The input is 12V with approx 5kW-hr batteries, and the output is 48V min. 120A max continuous used for BLDC motor control. I have found circuits that are high voltage 300-600W, but only about 10A. I looked over half bridge designs, daisy chain of voltage doublers, etc., but this is the closest thing I could find (60V, 8A)



I also have found a mosfet that is 60V, 270A N Ch, Internal rectifier 1RFB3006PBF

Also I was looking at another circuit, a half bridge using a Renesis RJH 60F7DPK, with output VceSat 90V (1.37V@50A), but I'm not sure if voltage will be a problem



I was looking for a relatively simple, cheap solution that would have a response of 1mS.

Any suggestions to fix the above circuits or other would be very helpful.

 

[mtwieg]

For such a power level, you should stick with a full bridge converter. And you should realize that you're going to have to draw >500A from your source, which will require tremendous cabling or busbar, and extremely heavy winding for the primary of the transformer. And forget about using anything other than MOSFETs for the primary switches, otherwise your efficiency will be terrible.

 

[schmitt trigger]

Agree that the power level is unrealistic for a 12 volt battery.

Your current draw will be 500 to 600 amps, and every milliohm of resistance equals 250 to 360 watts of loss.

 

[BradtheRad]

I agree with the above replies.

Is there a reason you cannot hook up the batteries in series, and obtain 48V that way?

Or if not that, then you might get by with several boost converters in parallel (interleaved). They would send output to a capacitor (or bank of capacitors), to power your load.

The alternative is a transformer wound with hefty wire, of a size which is several inches on each side.

 

[electrokute]

I agree with the above replies.

Is there a reason you cannot hook up the batteries in series, and obtain 48V that way?

Or if not that, then you might get by with several boost converters in parallel (interleaved). They would send output to a capacitor (or bank of capacitors), to power your load.

The alternative is a transformer wound with hefty wire, of a size which is several inches on each side.

ok, I wrote it quickly. Actually the batteries in series would be 24V supply voltage. The problem is I cannot fit 4 large lead acid batteries in a small package with any kind of battery capacity in W-hr's. Also motors tend to like higher voltage from everything I read (back emf, etc) to get good rpm's. I am ok with the transformer, as long as it doesn't have huge losses with it

- - - Updated - - -

to be clear, are you talking of the transformer size of a few inches in profile (I hope you are not talking about wire size!!!!)

 

[BradtheRad]

to be clear, are you talking of the transformer size of a few inches in profile (I hope you are not talking about wire size!!!!)

Ah, well, of course it's the transformer size. But the wire will need to be '0000' or at least '000', to carry the current safely.

This simulation shows how an H-bridge would drive a step-up transformer. It's the method that wastes the least power.



Switches Q2 and Q3 close at the opposite times from Q1 and Q4. In real life these will be mosfets.

Your OP mentioned an activation time of 1 mS. This simulation is snapped at 3.46 mSec after power-up.

Where there is a resistor marked 'load', this would include a diode bridge and smoothing capacitors, to provide 48 VCD to your motor.

Over 250 A is drawn from your battery bank. A high amount such as this will create spikes when switched back and forth through the transformer. To reduce such problems will require some ingenuity.

 

[BradtheRad]

For cost reduction an autotransformer could do the job. The primary and secondary have one winding in common.

If it will be operated for short periods of time, then you can get by with a thinner gauge wire.