100ns Pulsed Magnetic Field Coil feasibility?

Thread Starter

Twigg

Joined Jul 25, 2015
23
Hi all,

In the lab where I work we have a potential niche application for a Helmholtz coil pair that can be swept from 0 to a peak value between 100G and 1000G over a 100ns pulse. The smallest size the coil can be is 1" in both diameter and height. Using AWG 20 wire and the smallest possible coil dimensions, this would mean a peak current of ~2.3 amps.

The issue we're worried about is the impedance of the coil making it impossible to pulse that quickly. I know that with the right lumped passive elements we can impedance match to the coil to make it pass the lowest harmonic of the pulse with no problem, but it would still disperse the higher harmonics and the pulse would not work. Are there practical workarounds or techniques for pulsing reactive loads with an appreciable amount of power?

Thanks in advance!
 

MrAl

Joined Jun 17, 2014
7,659
Hello there,

In theory, you can ramp up an inductor current at any rate you wish. You just have to have the voltage to do it and most likely some control circuit to control that. That's actually how some stepper motor driver circuits work. The stepper motor has some inductance so the current which causes the force for movement has to be high enough to get it to turn correctly under load, so a higher than normal voltage is applied and the current is monitored, and once the current gets to the right level it is PWM'd to keep it constant for the duration of the pulse.
As a side topic, this is also a little interesting because we could create a near perfect buck converter with this idea in mind if we had a nearly unlimited voltage source to work with (a theoretical topic which illustrates a part of controllability).

To get yours up to the right current in the required time could take an appreciable voltage level which you would have to calculate based on the inductance using (to start):
V=L*di/dt
or:
di=V*dt/L

This means knowing the inductance, but you could run some tests with a given voltage V and see how fast it gets up to current di using some test equipment like a scope and current sense resistor or something.

Note that in the above last formula, if we have V=10 we get a certain rise in current over time dt, but if we increase that to V=100 we get 10 times that speed of increase. That's the basic idea.

The other part of this, the limiting part, is the voltage rating of the wire insulation. If we try to go too high we may exceed that rating, so the coil voltage rating has to be thought about also so that the coil turns do not arc over.

The final question to be answered then with a particular coil is do we exceed the voltage rating of the coil before we reach the desired rate of rise of current. If we do, then it wont work with that particular coil and we might have to design a better coil.

Of course every coil has some self resonance, so we would also hope we reach the target current level in the required time and dont get too much oscillation in the process. A better driver may help here, having a low impedance output for both high and low levels. A different coil of course might help too.

You might also want to note that there is no such thing as zero ramp time because that would take an infinitely high voltage, only near zero ramp time relative to the application where we find a ramp time that is acceptable. This means you have to decide on a ramp time that will work for the application but is definitely not zero.
 
Last edited:

GopherT

Joined Nov 23, 2012
8,012
Hi all,

In the lab where I work we have a potential niche application for a Helmholtz coil pair that can be swept from 0 to a peak value between 100G and 1000G over a 100ns pulse. The smallest size the coil can be is 1" in both diameter and height. Using AWG 20 wire and the smallest possible coil dimensions, this would mean a peak current of ~2.3 amps.

The issue we're worried about is the impedance of the coil making it impossible to pulse that quickly. I know that with the right lumped passive elements we can impedance match to the coil to make it pass the lowest harmonic of the pulse with no problem, but it would still disperse the higher harmonics and the pulse would not work. Are there practical workarounds or techniques for pulsing reactive loads with an appreciable amount of power?

Thanks in advance!
With a current of only 2.3 amps, you cannot get to a 1000 Gauss field in the 1 cubic inch volume for two coils. It will take about 13 minimum (likely 20 - my calculator was not capable of multilayer wrapped solenoid calculations).

A 20 gauge wire can easily handle 20 amps if you are pulsing at low duty cycle.
 
Top