1
\$\begingroup\$

Suppose I want to switch AC. I want to control (allow, prevent, or interrupt) current flow in either direction. SCRs won't do because I want to interrupt current, and I don't recall seeing many IGBTs or FETs that lack an antiparallel diode. Those few I have seen have been low-power and didn't have much of a spec for reverse-blocking.

Are there transistors or other solid-state devices that will do what I want? Or should I perhaps use two IGBTs in series with the directions reversed?

Voltages I'm interested in are on the order of 600VAC. Currents are 10-200 amps. Frequencies are 1-20 kHz.

\$\endgroup\$
5
  • 1
    \$\begingroup\$ put a diode bridge in series (the AC leads) with a MOSFET on the other side of the bridge (DC leads). Getting a good gate drive source might be a bit of a challenge. \$\endgroup\$ Feb 14 '14 at 20:23
  • 3
    \$\begingroup\$ 600V @ 200A = 120kW, what are you trying to do, drive a train? \$\endgroup\$
    – jippie
    Feb 14 '14 at 20:29
  • \$\begingroup\$ @jippie My customers have asked for such things before! \$\endgroup\$ Feb 14 '14 at 21:23
  • \$\begingroup\$ @WoutervanOoijen Could you draw a schematic of that? \$\endgroup\$ Feb 14 '14 at 21:36
  • 1
    \$\begingroup\$ @StephenCollings Here is a schematic. Take a look at tahmidmc.blogspot.gr/2012/11/… \$\endgroup\$
    – alexan_e
    Feb 14 '14 at 22:34
2
\$\begingroup\$

Yes, hefty back-to-back MOSFETs (maybe) or IGBTs should do what you want. Driving them would require isolated gate drive circuits. The complexity will vary a bit depending on how quickly you want them to switch. IGBTs tend to switch a lot more lethargically, but are more suited for the ~1200V PIV type operation required in this application.

SCRs (and triacs) as, you point out, have to wait for a (current) zero crossing, and the high frequency operation you speak of (20kHz) means that they would tend to stay on due to the high dv/dt at the zero crossing.

enter image description here

\$\endgroup\$
1
\$\begingroup\$

Try two fets like this: -

enter image description here

The opto is a bonus but getting the opto to switch at 20kHz might be a problem.

The way it works is that both MOSFETs can be activated by the common gate voltage (relative to the common source) and this switches the switch on. When the gate voltage is reduced to zero, both FETs are off and their respective parasitic diodes are in opposition therefore the circuit turns off.

The switch contact shown in the diagram is superfluous to the understanding and practical realization of this circuit.

I haven't checked but I'm presuming that FETs capable of withstanding voltages in excess of 850V pk are possible. Getting the current and the voltage may be trickier so you could opt to make parallel banks of FETs.

\$\endgroup\$

Your Answer

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

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