Testing An IGBT Based Solid State 40A DC Relay

[efficientPV]

I have wanted to test one for a while to see how fast they are. Listing said 10ms to turn on and off. This is not an anti bounce delay to prevent false switching. It takes 1.3ms before it starts to turn on. Full on takes about 10.4ms. Off you could call immediate. Like any IGBT, expect close to a 2V drop across the device. If you stay under 5A, it can be about a third of a volt and will rise quickly with current. Count on these producing a lot of heat and needing a really big heatsink. One YT video put two in parallel after a single one derated 30% burnt out. Two in parallel seems like a good idea given the saturation voltage might drop. Turn on input voltage starts at about 2V. I was surprised to see there was not any hysteresis on the input. Just before it turns on fully at 2V it begins to conduct. I had 16V across it with a couple amps of current. I had hoped that an analog voltage sensing input like a zener and resistor might be able to turn it on and off. This high resistance state around turn on would quickly burn it out. As a general rule no switching faster than 15Hz.

 

[zanydroid]

For your DIY projects, what would be a good use case of this kind of (and quality of?) SSR vs a contractor or other kinds of SSR?

 

[efficientPV]

I can't think of anything I would use this for and paying more won't likely get better results. It would be nice if there was a brick which would switch fast and not produce a lot of heat. Nothing out there I have seen will do this. It could be easy by adding a third terminal to the input for power. I get that most want the simplicity of screw terminals. Having to self power only using two input terminals makes anything slow.

 

[Solar Guppy]

What is the part number?, what are you using for a gate driver?

Not at all typical switch speeds for IGBT unless you have a very weak date drive signal, not high enough gate voltage, typical specs would be in single digit micro seconds for large high current ones I've looked at.

 

[efficientPV]

IGBT Based Solid State 40A DC Relay​

Not an individual IGBT. And yes, these all have very weak drivers. Takes a lot of photons to charge up gate capacitance.

 

[fourtytwo]

Fascinating how crap some off the shelf products are, I see it's got as Ch*nese Engineering badge on it
A zero cross switched Triac would out perform that with no fear of burning or part turn-on assuming it had an adequate heatsink.

Where did you find it Ali-Express or Ebay ?

EDIT 19:38 Ooops again!! I should have noticed DC........sorry

Well better off with a genuine branded product from a recognized distributor anyway rather than some unbranded fakery IMOP

 

[SeaGal]

A zero cross switched Triac would out perform that with no fear of burning or part turn-on assuming it had an adequate heatsink.

How would you turn a Triac off - he's referring to a DC solid state relay?

 

[efficientPV]

These do solve a need for those only comfortable with a screwdriver. I do think it is a odd choice to use an IGBT for under 200V applications. There don't seem to be many if any FET type bricks. I don't think many fully understand how much heat these produce or that data sheet specs for the device do not represent practical currents it can be used at. That the device can be maintained in an undefined state was curious. The lower saturation voltage under 5A makes them quite useful. This came from another big online seller. I think they all have to have about the same characteristics. At least these are the issues one has to be aware of.

 

[Warpspeed]

Depends what you want to use it for I suppose.
Probably at its best with high dc voltages, and slow infrequent switching.

Always possible to place one of these in parallel with a heavy mechanical relay contact.

Turn on the IGBT, and the relay pulls in a few tens of milliseconds after the IGBT is fully on.
That way you don't get the two volt drop under load, or any contact splat on closure.
Turn off the relay, wait sufficiently until the contacts are fully wide open, then turn off the IGBT.

No sparking on contact closure, no arcing on contact opening, and extremely low conduction losses through the contacts at high current.
The IGBT should also run stone cold so minimal heat sinking should be required.
And the contacts should last forever without any pitting or burning.

This might be useful for something like extreme inrush dc motor starting.