Hello,

I am working on a large project where I have about 60 SCR controllers powered by single phase 208V and each 25A rated. They are all located in the same cabinet and wired from the same three phase power distribution. Each SCR wired as such: the SCR leg is fused with a semiconductor fuse and the reference line has a circuit breaker. The reference line is also connected to one side of the load and the SCR output to the other.

Each SCR powers a separate heater in a chamber that is in rough vacuum. There are several types of heater including graphite and others.

The problem is that occasionally I will get a “short” and that short would cause many other circuit breakers to trip in other circuits and also some fuses open. The short would also create issues in adjacent tools that are powered from a separate power drop. The short has not been explained yet either.

So far here is what I am going to do:
Add an isolation transformer to the heater/ scr cabinet.
Add ferrite cores to the SCR output wires.
Check all wiring for temperature created shorts.

My questions are:

Why are so many circuit breakers tripping from problem in just one circuit?
Why are much less fuses than circuit breakers opening under this event? (last time I had about 30 CBs tripping and only 5 fuses opening)
Why is this problem propagating to tools that only share the facilities electrical panel?

Any help would be greatly appreciated.

Regards,

 

Are the breakers made by Federal Pacific by any chance?

 

Nope - They are ABB S201U series, UL489

 

Are the fuses slo-blo? And are they rated slower than the breakers? (Im guessing yes as they didn't blow before the breakers popped)

As for the whole kit and kabootle tripping, When 1 trips, it prouces a spike that causes all the others to trip. You may want to use a capacitor or a transient voltage suppresor or MOV to the lines before the breakers.
http://www.littelfuse.com/tvss.html

http://www.apttvss.com/

 

Fuses are super fast acting "semiconductor" type P/N HJS30.

 

How is the transient created? I would expect a voltage drop under short. And I thought that breakers trip on current only...

 

When the inductive load suddenly switches off, the quick change, or short causes a spike in current.

I would check your earth ground. If the spike is not being fed to ground, it will drive up your neutral and cause the multi-breaker trip.

 

The heater elements / assembly are not grounded as they are isolated in the chamber. The SCRs are grounded via the panel they are mounted to - Rittal galvanized steel. The main ground conductor is connected to the panel with a "big" lug. What else shoul I check? Am I missing something fundamental? How can I bring the spike safely back without tripping the rest of the circuit?

BTW - thank you very much for your help!!

 

Here is a datasheet for the SCRs: http://www.cascade.net/.%5CRKC%5CSSN%20series%20SSR%5CSSNP/SSNP_Brochure.pdf

I am using SSNP-25F.

 

There is a thread going on right now dealing with testing the earth ground rod.

The lugs and ground wire should be connected in the main panel. The ground buss in the main panel should run OUTSIDE to a ground rod that is 6 or more feet long, buried in the earth.

You may consider running an additional rod. Also, a three phase, phase monitor and surge protector should give you protection and an additional ground lug to run to earth. The better phase monitors will have data-loggers that give second by second (or faster) reports of what happened with the power and when.

I suspect, the shared ground/neutral bar in the mains panel may have either a bad connection to the outdoor ground rod, or is too small.

Here is the thread with instructions on checking the ground rod.:
http://forum.allaboutcircuits.com/showthread.php?t=38579

Post 19 has a good test. But read the whole thread.

 

This doesn't make sence. The load is resistive and there should be no back EMF or overvoltage conditions.
Even so, I cant see how grounding is an issue and there is no neutral conductor, it's single phase 208V. I can see enduced magnetic field from high current surges that couples to adjacent wires. That is why I was going to install ferrites. But why tools "far away" falt also?? This is still a mistery to me...

 

If the single phase loads are spread over the three supply phases, is it possible a serious fault on one phase will 'offset' the neutral and cause a high voltage spike on one or both of the other phases.

That would in turn cause a current surge on the loads from those phases, without a fault being present.

The actual fault could be another load in the same cabinet or a different circuit off the same distribution source.

Edit: I missed the last post; you say there is no neutral - do you mean the loads are across two phases? That's a different setup to Single phase (to Neutral) and different possible faults.