I need help switching an active inductor controller using inductive reactance to control the current of two feeds. The circuit I have built uses make-before-break switching just like a rotating brush making contact with two contacts at a time for non-sparking and continuous current flow. Below is the main device and circuit.

enter image description here

What I need is a switching circuit that can change the positions of a contact, just like the brush rotation or movement. When started, the voltage will be 100 V but will climb to 140 to 160 V suppressing the starting voltage. In the picture above, I want the brush to be replaced with a high-side switching circuit. I know it can be done but how?

The switching circuit I have now is a Teensy 4.0 on a custom 8-chip shift register board that works very well. I just need to marry it to a transistor board to switch the active inductor controller. This will change the inductive reactance and thus the current flow of two separate feeds 180° from each other.

Answering comments:
Yes I will be replacing the moving brush with 60 taps across the top of the active inductor controller. This is just like an AC variac except there is no sine wave just DC, so I have to continuously move the contact to achieve on going inductive reactance according to Farday in 1831. What this does by changing the amount of windings on either side of the brush, I am actually changing the inductance which is the opposition to current flow which gives the DC, AC-like qualities thus controlling the current flow with inductive reactance of both feeds 180 degrees out ie... one increasing , one decreasing.

The high-side driver and transistor switches will be connected to and switched using a Teensy and a custom shift register board I designed myself. It is on this YouTube video.

So with this connected to the high-side driver switching at the 60 taps on the controller, I can achieve my goal of replacing a rotating brush with electronic switching. Please keep in mind the Teensy is 3.3 volts but the shift registers are at 5 volts, so the high side driver transistor has to be able to be switched with 5 volts yet handle voltages around 100 starting volts to a running voltage of upwards of 160 volts.

I hope this clarifies things a little, but if it does not I can give further information.

enter image description here

  • 1
    \$\begingroup\$ You might want to add more detail about what you're doing. It might be clear to some, but I can't really understand what you're asking for here. You want to electronically replace a moving contact/brush? Or you want some sort of actuator to move your contact? \$\endgroup\$
    – John D
    Aug 5 at 22:48
  • \$\begingroup\$ Is this a mysterios inductive "rheostat"? Do you want to tap the coil at multiple points and feed in a DC current at one of them? \$\endgroup\$
    – Jens
    Aug 6 at 1:35
  • \$\begingroup\$ Marathonman - Hi, I appreciate you trying to give the background to your question, but IMHO the specific question is getting lost in all the text. You saif: "What I need is a switching circuit that can change the positions of a contact". But later you seem to say you already have a circuit which changes the equivalent of the positions of the contact: "a Teensy 4.0 on a custom 8-chip shift register board that works very well" & what you need is actually "a transistor board to switch the active inductor controller". That only needs to do the switching, not decide when to switch - yes? \$\endgroup\$
    – SamGibson
    Aug 7 at 16:46
  • 1
    \$\begingroup\$ (continued) Adding a schematic of what you already have, with an empty block showing what you need help with, and explaining the relevant connections between that block and the existing design, would be helpful to clarify things for readers IMHO. Isn't a high-voltage transistor all you need? What have you already considered as your solution, and why weren't they suitable? Thanks. \$\endgroup\$
    – SamGibson
    Aug 7 at 16:47
  • \$\begingroup\$ Yes, that is what I need is an transistor board that connects in place of LED's in the Video. I have tried a high side driver from Roam but it was a failure and burnt out. I have 400 watt IGBT's but i think they are to slow for what I need. I have also been looking at the STP46NF30 transistor with a high side driver from Skyworks which I think would work well and a 15 volt power supply to compliment it.. I will work on the schematic and post it very soon. \$\endgroup\$ Aug 7 at 18:39

1 Answer 1


I am afraid, that you will be very disappointed about the result later, but if you want to give it a run, ok.

You applied an AC voltage in the range of 100 V at the inductor tap. The AC current, that flew was manageable, because the impedance of the inductor was suitable for the used frequency. So far so good.

If you feed DC in this inductor, the amount of current is defined by the bare copper resistance of the wire, which will probably be below 1 ohm. You can verify this easily. If you apply 100 V DC there, the current will be at least 100 A consuming 10 kW power. This will immediately blow everything, absolutely worth to take a video of the moment.

So forget the 100 V and define a manageable current you can provide using a variable DC supply. Your coil looks like 10A without melting (just a guess), 1 ohm DC resistance assumed, we talk about a variable supply with 0-10V and 10A limit for the first test. Test this and check the coil temperature, it will receive 100 W.

Now the switching part:
It is expensive and much more hassle to switch the positive side of the supply ($300 for 60 taps). I think you can swap the polarity of this setup since the problem is symmetric. Whether there are two magnetic north meeting at the tap or two magnetic south, is not relevant for the effects.

The driver circuit for one tap could look like this:


simulate this circuit – Schematic created using CircuitLab

  • Be sure that the supply voltage of the shift register is 5 V, GT105N10F will not turn on properly at 3.3 V.
  • C1 is a protection against the feedback of high voltage gradients at the coil (Miller effect).
  • You need a large cooling aluminum bar, where you mount all the MOSFETS.

In the video you use a toroid core. This has a closed magnetic loop and the field will be very different compared to a linear coil. Whatever you want to achieve with this experiment, it will not work with a toroid. At least not the way you think.


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