0
\$\begingroup\$

In the circuit below L1 is a shunt coil rated at 115 VDC with DC resistance of 300 ohms. I need to use the same coil in an application with wide voltage 130-1000VDC from a bridge rectifier. I will apply a 30 ms pulse to the MOSFET. This means the coil won't see the high voltage for a long time. My thought is that if the pulse is that quickly the coil wouldn't get damaged. I'm thinking now how I choose the rating of the FET and what additional protection I would need to add. What do you think of adding a resistor in series with the coil to share the voltage stress and choose a 800V FET with a TVS diode to clamp some overvoltage between drain and source?

I'd appreciate your suggestions. I'll work on the math to calculate FET peak power.

enter image description here

Proposed schematic: enter image description here

Possible MOSFET: IPN80R2K0P7ATMA1

\$\endgroup\$
11
  • \$\begingroup\$ use isolated gate driver IC \$\endgroup\$ Mar 5 '20 at 17:06
  • 2
    \$\begingroup\$ If it’s rated for 115 volts why would the coil enamel insulation be rated for over 1 kV? \$\endgroup\$
    – Andy aka
    Mar 5 '20 at 17:11
  • 1
    \$\begingroup\$ To expand on @Andyaka comment, most magnet wire is only 300-600V rated. Can the coil handle this extra voltage? \$\endgroup\$
    – Aaron
    Mar 5 '20 at 17:16
  • \$\begingroup\$ Using a gate driver would be overkill. I won't be switching at "X" frequency. It's just a pulse not often. A simple opto would do the job (I think) \$\endgroup\$
    – user115094
    Mar 5 '20 at 17:52
  • 1
    \$\begingroup\$ The current through the coil can't change instantaneously, so when you first turn on the MOSFET you will have zero current. This means zero voltage across the resistor. This means full voltage across the coil. My suggestion on R2 is don't do it. \$\endgroup\$ Mar 5 '20 at 18:19
2
\$\begingroup\$

The FET needs to be able to withstand the FULL working voltage (plus safety margin).
When the FET is off, it will see the full 1000V on its drain.
I'd pick a FET around 1200V or so.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Did you (and the upvoter) read the part about te resistor and clamp as well? That makes the statement "The FET needs to be able to withstand the FULL working voltage ... 1000V" likely incorrect \$\endgroup\$
    – Huisman
    Mar 5 '20 at 17:29
  • \$\begingroup\$ @Huisman Hmm, you may be right. The diagram doesn't match the paragraph. \$\endgroup\$
    – Aaron
    Mar 5 '20 at 17:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy