overloading wirewound resistors

On a sunny day (Sun, 01 Mar 2009 10:15:54 -0800) it happened John Larkin wrote in :

How about some wire in a bucket of water?

I really want to do a water-cooled linear MRI gradient driver, but I don't yet have a sponsor. In some imaging experiments, gradient coil noise is a limiting factor, and the big H-bridge switchers are noisy.

1/20 KHz = 50 microseconds. 1 ppm == 50 picoseconds. You can see the problem.

John

Are these test loads where you just HAVE to dissipate THAT much waste energy or what? Is there no way to avoid that?

Wouldn't your safety and product liability issue make it unwise to turn the equipment into a friggin' TOASTER?

MRI: Magnetic Resonant Imaging?

A few hundred watts are peanuts for MRI imaging. Liquid Nitrogen should be available for the main coil, if you can tap into the system. For a cheaper solution, I would just wrap some coffee pot heating elements on copper pipes and recirculate it in water. Alternatively, you can just tap in the coffee pot and make coffee all the time.

So it has to be done linear?

what power/voltage/current is involved?

-Lasse

e ?

But WHY is it this energy has to be wasted?

Is there no way to avoid wasting this energy?

Yes - when I was ~12 I used to enjoy putting those across car batteries. They glow red hot and you can see the incandescent wire through the glass IIRC.

They seemed to work OK afterwards :)

--

John Devereux

You only need to pulse the gradient field for a short time. So, the average energy wasted is not significant, compared to the super- conducting magnetic coil.

Yes.

Several amps of several hundred volts for several milli-seconds.

I've made very high dissipation resistors with a stack of folded stainless steel plates bolted together :

------

Increase the voltage and you can have fun like this:

Dave.

One thing to watch: vitreous enamel coatings conduct at high temperatures, ISTR a caveat from Welwyn, or somebody, about this. Touching resistors might be a problem.

A coil of nichrome would need some support against sagging. Say wind it on a ceramic or silica tube like they do with heating elements. How much inductance can you tolerate? You might have to do an Ayrton-Perry sort of thing.

"High temp" solder isn't that high temp (around 300 Celsius). Silver solder might work, otherwise brazing or spot welding to a steel or brass plate. Nichrome is a pig to solder, you need a very aggressive flux.

Isn't there enough room for a TAP1000 series resistor? or won't the heatsink cope?

Another line of inquiry might be motor braking resistors. They make some quite compact ones, now.

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
                                             (Stephen Leacock)

Amplifying on that, electric water heaters have about

10 ohm immersible heat elements, good for 5 kW. You can grab two from any rusted-out unit. Everyone should have these in their junk box...

If the OP is talking about MRI excitation coil drivers, the duty cycle is very short. Yes, you need to sink several hundred watts for a couple of milliseconds. But relaxation times are of the order of 500 milliseconds. Adding idle times for mechanical motions, loadings and unloadings. The cooling system only needs to handle a few watts of average energy. I am having flash back of NMR requirements of 20 years ago, before nuclear is a dirty word. If he hired me to build the system, I would have told me what to expect.

On a sunny day (Sun, 01 Mar 2009 11:24:21 -0800) it happened John Larkin wrote in :

In very precision head positioning in quadruplex video recorders several servos were working on top of each other. I could imagine a PWM system, when locked in range, being supplemented by either a smaller second one, or a second smaller analog one to work away the ripple.

As far as heavy duty resistors go, I have a 2 kW heater with fan in my workshop, for very cold days. Computer controlled of course :-) About 8.7 A at 230 V, so 25.3 Ohm. I think the thing was about 25 $, but I am willing to sell it to you as scientific equipment for 1000 $. :-)

I think a ceramic housing is a good start. The ohmite resistor's housing looks like epoxy and it doesn't seem to like high temperatures.

Mounting is going to be a real adventure. Perhaps its best to leave the legs as long a possible so very little heat can be transferred and use large pads with copper pours to take the heat coming from the legs away.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
                     "If it doesn\'t fit, use a bigger hammer!"
--------------------------------------------------------------

I meant, "told him what to expect". Anyway, my $10 coffee pot should be able to handle it. Just integrate the coffee maker into the MRI machine. But it might be difficult to get the authority to approve the coffee in the imaging room.

^ hundred

Some of the imaging waveforms are ghastly, and the water-cooled gradient coils can take all the power you can give them.

John

Imaging waveforms can be very complex, and have high duty cycles. The gradient coils are water-cooled and are not superconductive.

The power resistors in question is in my "boost" circuit, which recharges/overcharges the power supply caps between pulses, so users have extra voltage to make fast leading edges (to drive lots of L di/dt) for pulse-type waveforms. The higher duty cycle imaging waveforms are stressing the charging resistors past what we designed for. When we designed this amp, we were thinking more in terms of analytical NMR, which does tend to use low duty cycle pulses.

ftp://jjlarkin.lmi.net/L500_Front.jpg

John