SCR Power control for heating

[condern1]

Condern1
I am looking for some tips in troubleshooting a system I am working on consisting of a 3 phase 480 volt line conntected to a 3 phase fired scr power controller with command signals of 4 to 20 ma from a temp controller. The output is sent to the primary of a 110 kva transformer where voltage is dropped from 480 to 107volts. This 107 volts is fed to a delta heating unit where each element is 36.7 kw at 107 volts. The total FLA is around 132 amps. The SCR controller has a RC snubber amd MOV protection and is fuse protected. The unit also has RMS current regulation and RMS current Limit as well as Overcurrent Trip Auto Reset. The problem I am seeing is that the unit heats up and for some unknown reason the SCR power controller stops running and trips out on overcurrent and never resets itself automatically. I have monitored the voltage and current of the lines and see nothing to cause a trip. The RMS current limit is set at 100% of the units rating and the overcurrent trip is set close to 200% of the RMS current limit. When system shuts down I am still seeing a command signal. I have 3 units and I switched the units and the problem did not follow the hardware. I would like to hear if anyone has some ideas,
Thank you
Condern1

 

[jraef]

jraef,
The explanation of the snubber circuit being a result of the problem is a good one , but how do you explain that the unit has no problem in another line . I would also like your explanation of why the same symptom effects whatever controller I put in this one line and are not effected in the other 2

I guess I was not tracking the "following the controller" issue correctly, forgive me, I am preparing for an interview for a promotion and my brain is not able to handle the stress.

So that just eliminates the possibility of a failed snubber, but some other source of noise spikes peculiar to that location could be doing the same thing in terms of causing misfires of the SCRs. Skogsgura's suggestion of a much better look at what is going on is the right approach, because by moving the controller to another location and having it work, then also putting a known good controller in to this location and having it fail is proof that it is something local. With no better test equipment, your only other choice is to methodically remove other components of the overall system until the problem goes away. The Advanced Random Scientific Wildly Inaccurate Process of Elimination (ARSWIPE) method.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington