Oh no I m sorry. A typing mistake. Its 2.63 ohm.

Then why 50V?
You only need 26.3V to pass 10A through 2.63 ohms.

 

Using PWM on a 48 H magnet coil doesn't work too well.

With that much inductance, maintaining a stable current won't be too much of a problem. The magnet itself has a nearly 20 second time constant.

You will want a high-current, high-precision four-point resistor. These usually look like a bar bent in a serpentine shape held off of a heat shielding base board. There are four terminals -- two large terminals for the current and two smaller ones inside the other two for voltage taps. Typically they are intended for 100 mV at full current and 10 A is a common full-scale current, so a 0.1 Ω current sense resistor. They aren't outrageously expensive.

You want your power supply to be voltage-controlled and then you want the control voltage to be the voltage across the sense resistor.

One thing you want to be DAMN sure about with that much inductance -- make sure that there is NO WAY for that magnet to become open-circuited when it has ANY current flowing in it. Put suitable protection circuits permanently across the terminals. There are several ways to do this. One is to put enough power diodes in series/parallel so that it takes more voltage to make them conduct than you plan to ever apply to the magnet. Put these in both directions so that it doesn't matter which way you charge the magnet. If you don't and someone disconnects it while it is operating that magnet will light them up, even if it has to generate a few million volts to do it.

 

I would really love to see those coils.

 

Then why 50V?
You only need 26.3V to pass 10A through 2.63 ohms.

They may not need the full 50 V, but they also don't want it to take forever to charge the magnet.

Say they are at 9 A headed for 10 A. That's ~24 V across the resistance. If they only had, say, 30 V to charge it with, that gives them only 6 V to charge the inductance with, or about 125 mA/s. Depending on what they are trying to do, that may just not be good enough. But it might. They should look at what they really NEED, and then give themselves a modest margin.

 

I would really love to see those coils.

Haven't you ever seen a Helmholtz coil?

The superconducting 10 tesla split-pair (pseudo-Helmholtz coal) that I worked a lot with at NIST was 47 H and 110 A at full field. If I recall, the magnet itself weighed about forty pounds, give or take. It was heavy, but you could lift it by yourself without too much problem. It fit within a volume of roughly one cubic foot.

The non-superconducting split pair was a much different story. It never got used (we called it the doorstop magnet) because it was huge (a couple of tons, probably), required serious cooling, and only produced about 2 tesla between the pole pieces. I have no idea what its inductance or full scale current was. The superconducting magnets made it obsolete a couple years before I started there.

 

I've only seen the small ones. Never saw the supers. But I don't think I've seen a multi H coil.

 

It's actually not to hard to get into the multihenry range, even on an air-core coil.

A solenoid that is about a foot long and half a foot in diameter only needs a few thousand turns.

 

Then why 50V?
You only need 26.3V to pass 10A through 2.63 ohms.

The extra voltage is so the pass transistors serve as a room heater in winter.