MCC Bucket Failure - Electrical Fault 1

[milwaukeebob]

Hello, first time user looking for some guidance. I am stationed at a combined-cycle power plant. An MCC which services our cooling tower experienced a major electrical fault during the start of one of the fan starters. The MCC bucket is no longer usable and needs to be replaced. The following are the relevant details so far:

(1) The fault occurred immediately when the fan starter received a start command.
(2) The fault appears to have originated on the load side of the bucket breaker prior to the contactor.
(3) The breaker did not trip when the fault occurred. Breaker was found in the CLOSED position.
(4) No unusual events occured prior to start of the fan.
(5) PI testing of the motor was initiated.Test result = 4.8
(6) Insulation resistance test of MCC bus was initiated. Test results = phase A (40 Gohms), phase B (22 Gohms), phase C (40 Gohms).
(7) Winding resistance to ground of motor was initiated. Test results ~ 30 Mohm for all phases.
(8) Except for the damaged bucket, the rest of the MCC line-up was returned to service and there have been no further incidences.
(9) The MCC and related equipment have been in service for approximately 2 years.

It is unclear what the root-cause of the failure is at the moment. Any experience with these type of failures and what we should be looking for? Thanks.

 

[PWR]

Damage was probably more severe than expected because the molded case breaker/protector did not open.

It sounds like you have a small physical area to inspect to determine the cause. The only conductors in the "zone" of failure are the main power leads between disconnect and contactor and the CPT leads.

Do you see any conductor material that appears to have arcing damage? Do you see any grounded metal that displays such damage? Is there any arcing damage on the small conductors on the primary of the CPT? Is there any evidence of overheating or arcing at a connection point of the power conductors?

 

[rbulsara]

Too many variables to begin listing them here. The most helpful will be MCC manufactrurers. They should be and will be interested in analyzing this type of faulures. Please use their assistance. Take lot of pictures, digital if you can.

Some questions I would have:'

1. How old is the installation?
2. Was this first time starting of the starter/motor?
3. Sounds like there was a phase to phase short which flashed over. That could be result of a bad installtion or poor maintenace or old failing component.

4.The MCCB/MCP may be bad as it did not open. You may want to check if in fact it was a breaker or motor protector (molded case switch with magnetic element) and not a plain molded case switch. As mentioned above, it may not be cause of the problem, it could certainly have limit the damage.

 

[busbar]

Was the local breaker thermal-magnetic [MCCB] or instantaneous-only [MCP]?

On the basis of personnel and equipment safety, a primary-current injection test of the MCCB should be considered. {Wouldn’t hurt to test to motor-overload relay in the process} and/or have the breaker evaluated by its manufacturer.

 

[dpc]

An arcing fault at 480/277V can persist for a long time without tripping an upstream circuit breaker, especially if it is magnetic-trip only.

It is often difficult to tell where a fault actually originates since the arc will migrate away from the initial source.

Since the contactor was apparently just trying to close when the fault occurred, I would be highly suspicious that the fault initiated in, or just downstream of the contactor.

Do you ground fault protection upstream - if so what were the settings of the ground fault device?

What finally cleared the fault?

 

[milwaukeebob]

Thanks to all those who responded. I apologize for the delay in responding to all your additional questions. I'm going to try to respond to all the additional questions in this one reply but before I do, I have some additional information:

I finally had a look at the circuit diagrams for the starter and found something I had never seen before. The contactor coil (42) is on the primary side of the CPT. There is another control relay on the secondary side of CPT which is energized once a command is received from the plant DCS. Auxiliary contacts on this control relay are used on the primary side circuit of the CPT to energize (and de-energize) the contactor coil. Both fuses that protect the contactor coil where found blown - appears like there was a phase-to-phase short in this area. Seems to me that this is a poor design but someone tells me this is common to size 5 starters and up. In this configuration, any failure of the coil could result in these type faults. Normally I have seen all control components (including the contactor itself) on the 120 Vac side of the CPT. Am I missing something?

To PWR:

Yes, damage was severe. The entire MCC bucket needs to be replaced. Using a standard fluke, resistance checks were performed on all phases of the breaker in the OPEN and CLOSED position. Phase B element is completely gone. Phase A & C appear to be OK. I can’t see if there is any arcing damage and nothing else stood out to me other then the general damage to the bucket.

To RBULSARA:

I agree. We already have engaged the OEM (Siemens-Furnas) and taken many digital pictures. The bucket itself will be sent back to them for a detailed root-cause analysis. The installation is approximately 2 years old. The MCC has been in operation for about 2 years and this particular starter/motor had been cycled numerous times. The breaker used is a Siemens model # JXD63H400 with an adjustable instantaneous trip only (MCP).

To BUSBAR:

As indicated above, the breaker used is a Siemens model # JXD63H400 with an adjustable instantaneous trip only (MCP). The bucket has been subjected to severe damage and is not reusable. It will be sent back to the OEM for a root-cause analysis.

To DPC:

I agree. It has been difficult to tell where the fault actually originated. There is alot of damage and shrapnel inside the bucket. We do have ground fault protection upstream on the 480 Vac unit substation feeding this particular MCC but after further investigation, we have found no evidence of a ground fault or the resulting DCS alarm. It appears as if the fault was eventually cleared through the contactor coil fuses blowing and B phase on the MCP.

 

[PWR]

From the information you have provided thus far, the failure may well have originated in the MCP.

Was the unit examined with an infrared viewer in the past 6 months or so? If a high internal resistance was developing in B phase of the MCP it might well have given some warning during a scan.

I hope you're manufacturer makes an honest effort to determine the root cause. Unfortunately, it has been my experience that manufacturers often do not put much effort into such investigations. What effort they do put forth is frequently directed toward finding a cause that does not reflect poorly on themselves.

 

[MakeItSpin]

milwaukeebob:

If I understand correctly what you're asking :

It is common to see large (Size 5 and up) contactor coils operating at primary voltage (480) using the 120 volt coil relay as a dry contact. I took a current reading on a Size 5 120 volt contactor once and saw 7 amps. That's a big chunk of the (usually 10) amps available in a typical 120v control circuit wired with #16. I'm not saying there's not a safer way to do it, but there's a lot of them out there.

 

[jraef]

Don't worry about the 480V coil issue. It is very normal. At 120V, the coil inrush current is extremely high on a size 5 contactor so many manufacturers opt for the "interposing relay" method as you described. Otherwise it may have required a 1kVA control power transformer to pull that puppy in, and it would take up too much real estate.

I agree that is sounds like a failure at the MCP, but remote diagnosis without research and hands-on investigation is relatively worthless. It could have been a bad connection to the busbar, a bad termination at the breaker or line side of the contactor, a nick in the conductor insulation, a spider web and sufficient condensing moisutre or a host of other possibilities.

If you suspect that Siemens is going to protect themselves at your expense, have in analyzed by an independant Electrical Engineer experienced in forensics. They are available in most areas.

"Venditori de oleum-vipera non vigere excordis populi"

 

[busbar]

If you have similar components in service and other failures could cause significant production outages or personnel exposure, third-party forensics may be desirable over investigation by the OEM.

 

[milwaukeebob]

Thanks again to all responders.

To PWR:

I'm also starting to think the fault originated in the MCP. Unfortunately no, the MCC has not been examined with an infrared camera since it was installed 2 years ago. We were planning to do this during our fall outage this year. You are correct that this is a good condition assessment tool. I also agree with your feelings regarding the OEM performing the root cause analysis. We are trying to pursue this with an unbiased third-party.

To MAKEITSPIN:

Thanks. You are correct. I should have realized that. It was a long day yesterday and my brain was at half power.

To JRAEF:

As MakeItSpin also indicated, this is common practice. Very good point. I just wasn't thinking. You are correct about having an unbiased third-party evaluate the damaged bucket.

To BUSBAR:

Agreed. We are proceeding with immediate thermo-graphic imaging of the rest of our MCC line-ups. We got lucky this time because we only lost 1 of the 12 cooling tower fans. This did not result in a combustion or steam turbine trip but the next failure could.

Thanks again to everyone for their support. You individuals are a credit to our industry.