high inrush current during turning off

[mash98]

Dear Guys I am working on Magnetic particle inspection machine where the maximum out put current is 3000 ampere at approximately on 15 volts ac while the input voltages are 400 volts two wire.The same is being done with a transformer where the prim voltage is 400 and the output voltages are 15 volts we are controlling the input of the transformer voltages with an SCR controller( having two SCRs in reverse parallel) which in turn control the output voltages and current.I am having no problem when I control the output current gradually that is from 0-3000 max & from 3000-0 but when I abruptly interrupt the current( which is the process requirement) even at 40% the input breaker trips the rating of breaker is 250 ampere which seems sufficient.Kindly advice.

Regards

 

[jraef]

Investigate the concept of "inductive kickback". You probably don't have a snubber circuit across the SCRs. You need one, for the purpose you just discovered. By the way if you have done this much, you may have already compromised those SCRs as well.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[sreid]

jraef is correct. SCRs have a maximum dv/dt rating. If this is exceeded, the SCR can Rate Fire [turn on from the slewing voltage].

 

[jraef]

I should have mentioned, if you DO have a snubber across the SCRs, it could be that it has fried. It's typically an RC network and the resistor sometimes opens, especially if there were a lot of line side spikes from some other cause, i.e. distant lightning, grid switch issues etc. I even had a rash of them fry at one plant that I traced back to transition spikes from a large Wye-Delta motor starter.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[OperaHouse]

A company I used to work for manufactured RC snubbers and the most common failure was the capacitor failing from an open weld. We didn't buy the best capacitors and that should be a lesson for you. If the resistors are failing the network isn't rated properly for the currents you are seeing. Get a cap that is rated for a couple amps at 10khz if you want it to last.

 

[mash98]

There is an snubber circuit on the SCRs but later I have not checked if they are intact

 

[mash98]

I have a similar machine in my plant and a couple of weeks ago an SCR module was burnt on that machine and I replaced the same module I am using in the above machine & there is no such issue.So I checked its snubber circuit may be we have not calculated it incorrectly.

 

[mash98]

I have a similar machine in my plant and a couple of weeks ago an SCR module was burnt on that machine and I replaced the same module I am using in the above machine & there is no such issue.So I checked its snubber circuit may be we have not calculated it correctly.

 

[mash98]

Should we include a line reactor to limit the current spikes?

 

[mash98]

Dear guys The problem have been resolved by changing a minor alteration in the control circuit earlier we were interrupting the ready command of the phase control unit which in turn interrupt the output currrent.And now we are interrupting the reference command of the drive which has resolved the issue.
Thanks