Thyristor Checking

[gsjhand]

Hello Everybody
I have a question regarding checking thyristors. What is the correct way to check a thyristor? Recently I got problem in a DC drive and I have to replace the thyristor board. On checking I got ohm readings as shown in attached table, but not sure what is the ideal ohm value for a thyristor? DC drive is of 640 Volts and 540 amps. Tyhristors are EUPEC- TT250N16KOF.

 

[Skogsgurra]

The correct way is to apply a voltage in series with a current limiting resistor. The easiest way is to use mains voltage and connect the thyristor in series with an incandescent lamp.

A thyristor in good shape shall not light the lamp. Applying a gate gurrent (around 100 mA, negative to K1 and positive to G) shall light the lamp. The lamp shall not burn brightly, since the current is half-wave rectified. Remove the gate current and the lamp shall go out.

PLEASE NOTE: You will be dealing with lethal voltages. Use insulating gloves and do not work on your own!

If you really want to characterize your thyristors, you can use a special thyristor check instrument. But that is seldom needed.

The values in your table look good. Not much to worry about. They show the normal spread. Don't think any of your thyristors is faulty - it usually shows in gate resistance either being very low or very high.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[LionelHutz]

Yes, your values look OK.

The correct way is to apply a voltage and measure the leakage current, then apply a gate voltage and measure at what level it turns on. Both of these will be in the specification for the device.

Typically, the gate resistance will measure shorted or open or the SCR will measure shorted anode to cathode if it is bad. The gate resistance is usually between about 4-40ohms.

 

[gsjhand]

But what about the one which is giving me 34 Mega Ohms. I was thinking that this resistance might be too high for SCR to Fire..

 

[Marke]

Hello gsjhand

The gate include a silicon junction so there is a threshold voltage. If the source voltage for you resistance measurement is too low, the resistance will appear to be high.
The resistance of a given gate cathode with change depending on the actual meter used. The good old analog meter virtually always works, but some digital do not.

Best regards,

Mark Empson
L M Photonics Ltd

 

[ScottyUK]

What are you measuring the resiatnce with? If you're using a multimeter then the readings don't really have much meaning because it's not an ohmic connection. If you are using a Megger to prove the A-K blocking then the readings may have some meaning. Firstly, choose a test voltage less than the rated blocking voltage. A low reading will show up a bad thyristor. The thyristor should block in both directions when there is no gate signal. Depending on how big the thyristor is I would expect better than 1M[Ω] in both directions, and only that low on a really big device. If this is a hockey puck thyristor don't forget to clamp it in the heatsink before testing it, otherwise the contact to the die won't be reliable.

I've used the method Skogsgurra suggests although my supply was a big 110V DC source loaded with a bank of 500W halogen floodlamps. Blocking in the forward and reverse directions was proven by a Megger test, then conduction and gate firing was tested with a pulse of about 100mA or so. Worked very well. Also scared one of the managers who walked into the makeshift test area only to be blasted by a few kW of floodlamps as I triggered the gate. Highly entertaining for those of us who knew what was going on!


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Skogsgurra]

Mark is right - it is a PN junction. And should show the typical non-linearity.

Out of curiosity, I checked a medium size thyristor (a couple of hundred amps in a six pulse bridge) and got the results in the attached pdf.

I remember that I wondered about this before. The characteristic is much more linear than one would think.



Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[gsjhand]

As far as checking thyristor with voltage and lamp method, my boss is not going to allow me for that. So possibility of that type of bench test is limited. The only other method remaining for me was to check the resistance. Multimeter which I was using for measument was Fluke Digital Multimeter. As everybody said it might me affecting my results but I have not cross checked it with analog meter. Would be interesting to see the results with analog meter.

 

[LionelHutz]

The turn on is controlled by the gate which is the G1 or G2 measurements. They look OK for an ohmmeter reading.

I missed it before but device #2 AK-K is likely bad if it is only 8.9 ohms. I hope you meant 8.9M??