Using electro-plpating rectifier as a low-voltage high-current power supply

[stube40]

We've been looking for a low-voltage high-current power supply for our lab to put between 200A and 1000A through a superconducting coil with inductance of 50mH. Total resistance of the load will be around 0.1 Ohm.

The only thing I've been able to find so far is an electro-plating rectifier. Although it's called a rectifier, as far as I can see it's actually a low-voltage high-current PSU.

The manual (attached) is not very informative. It is a 3-phase input. Switchable DC output from 0-15V and 0-1000Amps. It is based on IGBTs. Power factor of 0.95, 2% ripple. It has either constant current or constant voltage operation.

It's very expensive ($12500!!!) and I'm worried about buying it and then it not doing what we require (or, even worse, we break it because we're mis-using it).

Can anyone shed any light on whether we're going doing the wrong track or not?

 

[mtwieg]

1000 amps DC is a huge amount, so you'll likely need something highly specialized like this. And that price tag really isn't so bad, considering its performance.

Is this for NMR applications, by any chance? Are you just charging a magnet, or does it need to be powered continuously?

 

[stube40]

Hi. No, it's not NMR. We are just performing some research tests on new types of superconducting coils.

It will be powered continuously in relatively short bursts (maybe a few minutes at a time).

 

[mtwieg]

Okay, had to ask.

Another possibility is to get hold of an old gradient amplifier from a clinical MRI scanner. Probably won't be cheap, but cheap is probably out of the question for you...

Also I hope you're planning for safety. 25kJ is a lot of energy to store in a coil, especially if it decides to spontaneously warm up a bit...

 

[Genomerics]

Here's a mad idea..



See. What you do is set up a 'flyback converter'.

Given you have access to superconducting wire the transformer should be a snip.

Now all you do is switch on M1 until you get 10Amps then switch it off and switch on M2 and you get your 1000A.

Quite simple really.

Did I miss anything?

 

[stube40]

Here's a mad idea..



See. What you do is set up a 'flyback converter'.

Given you have access to superconducting wire the transformer should be a snip.

Now all you do is switch on M1 until you get 10Amps then switch it off and switch on M2 and you get your 1000A.

Quite simple really.

Did I miss anything?

Great idea! I dont think you're missing anything at all, in theory at least. In practice it will be difficult and expensive to make the super-conducting transformer. Also, the control will be non-trivial and need a linear opto-coupler for either voltage/current sensing feedback.

Probably quite a project involved, but a really interesting project!