Electronics > Projects, Designs, and Technical Stuff

Bidirectional isolated switch circuit. Pitfalls?


I have been looking into different kinds of solid state switches to replace relays in an application with 24V DC or AC signaling. While looking into different circuits i came across bidirectional IGBT switches based on using a diode bridge configuration. https://www.researchgate.net/figure/Diode-bridge-bi-directional-switch-configuration_fig5_258448391

I exchanged this for an NPN optocoupler with an additional transistor for increased current handling and a quick simulation of the circuit (https://tinyurl.com/ydnp2p99) seems to show that the concept works.
 I'm kind of surprised to not have come across something similar before which makes me feel like I might be missing potential pitfalls? One obvious problem is the 2x diode and NPN voltage drop but this would no be a deal breaker in many applications.

It will be interesting to hear your opinions on the circuit? Have you done something similar or do you see any potential problems with it?

David Hess:
In the past they used the same circuit to allow an SCR or bipolar transistor to control AC before TRAICs became available.  Provisions need to be made for the optocoupler output transistor to handle high voltages.

I have used something similar with a triac too. At 230 V (maybe 220 at that time) the drop at the diodes is not so bad.

For 24 V I would consider photo-mos switches or ready made solid state relays. IGBTs are not really the right choice at low voltage.

Terry Bites:
You can use a MOSFET instead of the trickier IGBT, they're are a bit crap at low voltages. You only get the dissipation benefits at higher voltages. It will almost cetainly be cheaper.
Beware, leakage currents are a much higher than a relay will be, shockingly so.
But then you need a serious relay to break DC. At least you can hear it turn on when you're not looking.


[0] Message Index

There was an error while thanking
Go to full version