Detecting a bad IC [Optocoupler Gate Driver]

Thread Starter

Xavier Pacheco Paulino

Joined Oct 21, 2015
728
Hello,

Is there a way of detecting a bad gate driver like the one shown below? Using a test circuit or a meter?


The point is that I have many boards which have six of these drivers. Specifically, they are 3 phase motor drivers. They are like VFDs. The console throws an error when one of these drivers fails. But I don't know which one. I usually replace all of them and the problem is solved. But these drivers sometimes are "expensive". That's why I would like to detect the faulty one.

What do you suggest?
 

danadak

Joined Mar 10, 2018
4,057
Do you have an oscilloscope ? Look at waveforms on pins. You
have to be careful with probe ground lead to make sure you are
not shorting AC power.

https://www.tek.com/document/application-note/three-facets-floating-measurement-solutions

https://electronics.stackexchange.com/questions/98656/using-oscilloscope-safely-with-ac-mains

https://www.eevblog.com/forum/beginners/mains-and-the-oscilloscope/

If not an ohmmeter, set to diode range to see if Input LED good
forward and basically open or high ohms reverse leads.

Then look at collector to emitter output for shorts. Remember to
short leads together to measure lead R, then subtract that from
any readings you make.

Regards, Dana.
 

Alec_t

Joined Sep 17, 2013
14,280
The console throws an error when one of these drivers fails.
What is the usual failure mode? Have you done any autopsies on the failed drivers?
Depending on what other components are connected to the pins, testing with the IC in circuit could be inconclusive.
With the circuit powered down, a simple diode test might show up a faulty LED.
To test the output section, voltmeter/scope checks with the circuit powered up might suffice.
 

Thread Starter

Xavier Pacheco Paulino

Joined Oct 21, 2015
728
What is the usual failure mode? Have you done any autopsies on the failed drivers?
Depending on what other components are connected to the pins, testing with the IC in circuit could be inconclusive.
With the circuit powered down, a simple diode test might show up a faulty LED.
To test the output section, voltmeter/scope checks with the circuit powered up might suffice.
The board is from a control box for an industrial sewing machine. When one these drivers fails, and I intend to start the motor, the console throws an error meaning that the motor can't start. It's hard to take measurements when the circuit is being powered, since it's 220VAC, and the nature and structure of the machine makes it harder. I do not have an oscilloscope by now to take better measurements.

I have checked the diodes and they seem okay as far as I concerned. The voltage drop is about 1.5 -2V. The output is not shorted either. As I said previously, the problem is only solved when I replace the six drivers (that's why it's expensive, and more expensive when you have several machines with the same failure). I'm sure that not all the drivers are faulty. Because of that, I wanted to identify the bad one, so the repair cost is less expensive.

The problem is kind of weird. I think I'll need the oscilloscope to get a better insight.
 
Top