How would I go about calculating and designing a circuit that can produce an axial magnetic field greater than 20 Tesla for about 1.4 μs (full width at half maximum). The dimensions of the solenoid must meet the constraint of a cylinder with length=5mm and radius=0.5mm. The risetime of the pulse must be less than or equal to 1μs.

This lab seems like a LOT to digest... Anyways, I figured it would make most sense to use a capacitive energy storage device. How would I go about beginning to develop a design for this, assuming that this needs to be buildable with real life components, and not just have a 2000A current source powering a solenoid with a little switch :^) Any pointers on what formulas to start looking at, terms to start googling would help. I tried to solve this as a physics problem, but I'm not getting much information from that. I assume this requires a more intricate usage of 1st and 2nd Order circuits.

Thanks in advance for the help! :P

  • \$\begingroup\$ You've specified the radius of the cylinder. Over what area within that radius must the 20 T be produced? How will the measurement be made? \$\endgroup\$
    – jonk
    Commented May 12, 2020 at 2:02
  • \$\begingroup\$ Marx generator? Supercooled magnet? \$\endgroup\$ Commented May 12, 2020 at 2:05
  • 1
    \$\begingroup\$ I'm thinking that the area which the 20T field must act upon is within the inner diameter of the solenoid. Not outside the solenoid. @jonk \$\endgroup\$
    – graphpaper
    Commented May 12, 2020 at 2:18
  • \$\begingroup\$ You can easily calculate the magnetic field of a solenoid. The problem is 20 Tesla is a lot. For this not to be trivial you must be expected to take heating into consideration. But thats a whole problem itself, especially given how small the thing is supposed to be. You will probably want superconductors, but thats not something many labs are going to get into. \$\endgroup\$
    – Matt
    Commented May 12, 2020 at 2:19
  • \$\begingroup\$ What makes you think it is possible.? Do the math. Compute the concentric force vs current \$\endgroup\$ Commented May 12, 2020 at 2:21

1 Answer 1


over 0.5 millimeter, you can only wrap the tiniest of wire gauge.

Give copper of size a METER has thermal time constant 9600 seconds,
of size 0.1 meter is 96 seconds,
of size 0.01 meter is 0.96 seconds
of size 0.001 meter (1 millimeter( is 0.0096 seconds (9.6 milliseconds)
of size 0.0001 meter (100 microns, or 4X that of bondwires used in silicon packages)
is 0.000096 seconds or 9.6 microseconds,

then the thermal duration 9.6uS >> 1uS suggests you CAN build something that thermally MIGHT survive long enough.

You can now examine the mechanical stresses (sudden acceleration) of gold wires (why not) only 25 micron in diameter.

And examine the rate of heat generation, to determine if the many? turns of gold wire will vaporize, or that the surface merely will ablate away because SKIN EFFECT will keep the heat on the surface.

Have fun.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.