Capacitor Bank failure mode?

Status
Not open for further replies.

iwire

Moderator
Staff member
Location
Massachusetts
I got sent to check out a breaker that was "taped off"

RocheBrosEaston1.jpg


supplying a capacitor bank.

RocheBrosEaston2.jpg


I have almost no info at all other then the customer wants to get this back on line. The conductors had been removed from the breaker and taped up.

The wire way showed signs of a heavy ground fault at the lock nuts

RocheBrosEaston3.jpg


and the hardware holding it together.

RocheBrosEaston4.jpg


The 480 volt circuit to this capacitor was supplied by a 400 amp breaker installed in a 2000 amp distribution section of service gear located very close to the transformer. The conductors where 500 Kcmil and a 1/0 EGC.

I tested the conductors, all drove my mega meter to it's limits (they tested fine) I then pulled them out and looked them over and they looked fine.

So that said, it must be a ground fault at the capacitor bank but that shows clear at 250 volts to ground and there is no sign of a flash inside the unit.

RocheBrosEaston5.jpg


Here is the units info

RocheBrosEaston6.jpg


Is it normal for a capacitor to fault to ground then show clear?

I am open to any ideas at all, I do not feel like putting power to this unit again without knowing what caused the ground fault.
 

zog

Senior Member
Location
Charlotte, NC
Good to see you are working again Bob.

Did you check the resistance of the internal discharge resistors? How about discharge time? For this unit the residual voltage of the caps should be reduced to less than 50V in less than 1 minute.

Did you measure the capacitance of all terminal combonations?
 

mivey

Senior Member
Any leaks? Are all the connections tight? Maybe they had a harmonics issue.

While you could verify the capacitance using a voltage test, you would have to energize the bank.

Other than what zog indicated, I would not think there are many field tests you can run without having the proper test equipment.

add: I guess you could check for balanced capacitor currents/voltages. Also you could check that the current is not excessive (could be an indicator of harmonics).
 
Last edited:
I got sent to check out a breaker that was "taped off"

supplying a capacitor bank.

I have almost no info at all other then the customer wants to get this back on line. The conductors had been removed from the breaker and taped up.

The wire way showed signs of a heavy ground fault at the lock nuts

and the hardware holding it together.

The 480 volt circuit to this capacitor was supplied by a 400 amp breaker installed in a 2000 amp distribution section of service gear located very close to the transformer. The conductors where 500 Kcmil and a 1/0 EGC.

I tested the conductors, all drove my mega meter to it's limits (they tested fine) I then pulled them out and looked them over and they looked fine.

So that said, it must be a ground fault at the capacitor bank but that shows clear at 250 volts to ground and there is no sign of a flash inside the unit.

Here is the units info

Is it normal for a capacitor to fault to ground then show clear?

I am open to any ideas at all, I do not feel like putting power to this unit again without knowing what caused the ground fault.

Did you have a bonding wire between the panel and the cap enclosure?
I think you want to megger at 1000V.
Sounds like an arcing fault due to switching surge or harmonic ringing.
I don't like constant value capacitors, could cause harmonics, ringing and overvoltage at low loads and no inductive contributors.
What a heck does it supposed to accomplish in this service anyway?
 

LarryFine

Master Electrician Electric Contractor Richmond VA
Location
Henrico County, VA
Occupation
Electrical Contractor
Bob, random thoughts (as if I have other kinds):

Could it be that a cap was internally shorted to its case? Look for arcing to the enclosure beneath each cap's case.

Are you sure the arcings are from a one-time event and not over time? I know they look like it, but you never know.

There may have been a short to the raceway outside of the cap enclosure, so check all of each conductor for nicks.
 

steve66

Senior Member
Location
Illinois
Occupation
Engineer
are those fuses at the top? if so, i assume you checked them.

Do you have anything that can actually measure the capacitance?

Edit: it looks like it might have blown fuse indicators on it.
 

mxslick

Senior Member
Location
SE Idaho
First off, Steve's right, it has blown fuse indicators, follow the red wires to the front panel.

Second, Bob look very closely at the capacitor on the top right: Those marks on top of the case make me wary. I'll bet if you looked at the sides or under that unit you'll find a blowout.

Also, are any of the cap's tops bulging? Even a slight bulge means an internal fault has happened.

My suggestion would be to install a PQM with recording function for a week, and if the PF stays within desired limits, abandon the cap bank. If the PF needs correction, find out the correct value needed and replace the cap bank.

Edited to add: Based on the look of the top of the cap bank can, it could also have been a one-time fault caused by liquid entry..I'd still do the PQM to see if the caps are even needed before reconnecting.
 
Last edited:

iwire

Moderator
Staff member
Location
Massachusetts
Thanks, keep them coming I can't stay long right now.

But I can tell you the caps are desired as there are a lot of motor loads here.
 

mxslick

Senior Member
Location
SE Idaho
Thanks, keep them coming I can't stay long right now.

But I can tell you the caps are desired as there are a lot of motor loads here.

Bob, are those loads continuous? Or do several motors cycle on and off at random? If they cycle on and off a lot, I'd agree with Laszlo when he said this:

weressl said:
I don't like constant value capacitors, could cause harmonics, ringing and overvoltage at low loads and no inductive contributors.

Any idea how long the bank was online and in use before it failed or was disconnected?

My electrical "bible", the "Standard Handbook For Electrical Engineers" (McGraw-Hill, Fink and Beatty authors) discourages the use of single-point power factor caps in multi-motor applications, preferring correction at each motor or group of motors (MCC perhaps?)
 

markstg

Senior Member
Location
Big Easy
Cap good or bad

Cap good or bad

As a capacitor is an open circuit to DC, when you put a DC megger on each capacitor, with others disconected (discharge resistors removed), I think you should see your meter go from some low value to a large value (open circuit). If the meter stays low the capacitor under test is shorted. Just a thought.
 

peter d

Senior Member
Location
New England
Bob, are those loads continuous? Or do several motors cycle on and off at random? If they cycle on and off a lot, I'd agree with Laszlo when he said this:

The gear serves a supermarket, so there is a large continuous load on the service. The many motors in the refrigerated cases do run continuosly, although the large compressors cycle on and off.
 

LarryFine

Master Electrician Electric Contractor Richmond VA
Location
Henrico County, VA
Occupation
Electrical Contractor
Lets say one or two of the caps are shorted...what would happen if Bob reengerized the capacitors?
A fuse would blow, or the breaker would trip again.

It might help to slowly ramp up the voltage with a Variac, while monitoring the currents.


Added: Do we know whether the breaker tripped or was only manually shut off?
 

iwire

Moderator
Staff member
Location
Massachusetts
Added: Do we know whether the breaker tripped or was only manually shut off?

I assume it tripped, no one would be around to shut it down.

Got home a bit late, I will answer all questions after I chill for a bit.

But as far as the need for these units, it is not really relevant.

Talk about the need for these if you want but the customer wants it and they will get this one or another. :cool:
 

steve66

Senior Member
Location
Illinois
Occupation
Engineer
I would have expected a fuse to blow before tripping a circuit breaker.

Just a guess, but I bet those are fast blow fuses whose purpose is shut off the power to a shorted cap before the arc flash gets too big.
 
Status
Not open for further replies.
Top