Unexplained tripping of circuit breaker 1

[BK32]

Hi All,

I need some help in troubleshooting an unexplained tripping of an MCCB. This breaker is installed in the main distribution board on an 440V, 60Hz, 3 ph, 3 wire system.

The breaker is a cutler hammer (eaton) AQB-L250 type breaker with a 175A plug. The instantaneous trip setting is Lo (these type of breakers have either Hi or Lo settings which correspond to a numerical pickup depending on the plug rating installed).

This circuit breaker feeds the control panel for an AC chiller plant. The control panel contains DOL motor starters for two (2) off 63kW compressors, two (2) off 5kW chilled water circulating pumps and two (2) off VFD driven 5.5kW sea water circulating pumps.

The AC Chiller plant had been operating with one compressor, one chilled water circulating pump and one seawater circulating pump online. This had been the case for a number of hours. Then a fault occurred somewhere else on the ships power distribution network which resulted in a black out scenario (I.e all supplies to the ship were tripped. Note at the time the ship was alongside and being fed from shore supply connection). In any event, this was not a result of any faults within the AC chiller plant.

Based on information from the contractor conducting the stw and commissioning of the system, in the 24 hrs or so following restoration of power this AQB-L250 circuit breaker tripped approximately 5-6 times. The current draw recorded on the circuit was approx.100A on 2 phases and 70A on the other phase.

They stated that on each occasion all pumps had been started and were running normally. It was also noted that the protection devices provided in each of the starter circuits did not trip.

Given the cumulative load was significantly less than the breaker thermal pickup of 175A, I do not see that the breaker could have legitimately tripped on a thermal overload event. Also, I do not believe that the breaker is fitted with either a shunt or uv trip.

Could anyone provide some other advice and what to look at? At this stage it is a bit of a mystery to me. The only obvious conclusion appears to be that the trip unit on the breaker is failing (nuisance tripping).

Thanks in advance for your help.

 

[itsmoked]

Any chance the breaker is responding to the considerable imbalance? I suggest you hunt it down since you describe all balanced 3 phase loads, yet the sum is not.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[unclebob]

Check for a loose connection near or inside the breaker. Sometimes the trip unit is not bolted with enough torque in the breaker, causing overheating.

 

[waross]

If possible check the voltage drop across each pole of the breaker. Should be in the millivolt range and all poles should be close to equal. Quickest way to check for problem connections in the breaker.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[BK32]

Thanks for all the helpful replies.

I will check the tightness of connections at the breaker.

Also, should the imbalance be traced through to a three phase induction motor feeder, noting an imbalance of approx 30% any advice on what may be a typical cause? Could it be improper alignment, bearing wear, or possibly even a winding failure (internal winding failure)?

 

[waross]

1/ If you are checking voltage drops across the breaker, try to position your meter probes on the surface of the cables rather than on the breaker terminals. That will include the cable to breaker connection in the test.
2/ The most common cause by far is unbalanced supply voltages. If the voltage on one supply phase is 5% or 6% low it may account for the current unbalance. The next possibility is a failing winding. A failing winding won't last more than several minutes before complete failure. If several motors are running, one motor may have an open winding, but with an open winding it probably would not start. It would be running hot and the over load relay would trip on differential current.
Bearings or misalignment would load all three phases together. Such problems generally progress to complete mechanical failure before there is a noticeable effect on the current.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[JG2828]

Personally, I would rent or buy a 3 phase power recorder and connect it to the breaker output asap. That way you can see exactly what is happening (3 phase voltage and current) when the breaker trips.
The Fluke 1750 is a good choice. They can be rented with whatever current range CT Loops you need. Voltage inputs can connect directly to 480v.

https://www.trs-rentelco.com/Specs-Manuals/FLU_1750_Spec.pdf

 

[Zogzog]

Have the breaker tested via primary injection, that will tell you if you have a breaker issue or something else. What is inrush current when the chiller starts? That could easily be tripping the breaker on INST.

 

[BK32]

Again thanks for all your response.

I believe that all the motor driven pumps start up without tripping off the short circuit release. Initially it appeared to be tripping off via thermal release protection.

Measuring the Millville drop across the breaker poles will be a challenge as they are rear connected and only accessible through the rear. It would also involve isolating the board as the switchboard construction contains little or no internal form of segregation between functional units. But if it has to be done, then I will get it done.

I am also led to believe that this breaker was primary injection tested only a few months ago, I will try to track down the results of that test.

I will also check supply imbalance.

Thanks for all your help this far.

 

[waross]

A primary injection test is great. However the breaker may have to be reconnected to the test set. That means that the working connections are completed after the test and are not included in the test.
The milliVolt drop test may not be as pure as the primary injection but in many cases may be more useful in locating actual problems.
While others may be generating a beautiful spread sheet of very accurate values, I prefer to spend my time locating the source of the problem. I care more about markedly different values than about very accurate values.
The upstream connection of the voltage drop test may be made at the load side of an unused or lightly loaded breaker. Anywhere that the phases may be accessed safely.
The downstream connection may be made at the panel on the equipment skid.
As long as the breaker and the connections to the breaker are included.
The breaker must be under load while the test is conducted.
If the voltage drop measurements are close to equal, the test has been passed.
If one phase shows a marked increase in voltage drop then work to isolate the source of the voltage drop.
The breaker may have passed the injection test and may again pass the test, but if a connection made after the test was poorly made and is generating heat, the heat may be the cause of the breaker tripping. An overall voltage drop test will detect some issues that will not be seen by an injection test.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

So, with a 175A plug and "Lo" what is the instantaneous trip current? Just curious. It could be low enough the 63kW motors trip it during starting but I'm not spending the time to look your breaker data up.

Did you measure phase-ground voltages? Your chiller panel could still be running with a ground fault if you have a floating power system which is typical for a ship. Then, the breaker gets tripped every time another piece of equipment with a ground fault on another phase gets started or energized since that completes a phase-phase fault. The breaker with the shortest trip time will go first.

 

[unclebob]

Data here shows LO setting = 650 A.

 

[BK32]

Waross,

I will have a look at both options upstream and downstream. I guess the assumption is that I am expecting that all other connections on all three phases are good (e.g assuming that there are no loose connections on the lightly loaded breaker. Also, generally speaking, if I use a 4 probe ductor, is there a need to have the load running?

LionelHutz, it is possible that there may be phase to ground faults on the system. Normally the insulation monitoring system on the main switchboard would signal something, I will check the phase to earth impedances. If there is a P-E fault it may appear to be intermittent as the other equipment with the P-E fault may rarely energise.

Also, regarding the startup thus far there have been no instances of tripping during startup of the 63kW motor. I believe that the instantaneous pickup is set high enough to avoid tripping on start up.

Thanks again for your suggestions.

 

[waross]

Hi BK32;
The voltage drop test that I am suggesting is the poor man's ductor or the red neck ductor. Instead of injecting a test current, use the load current and just measure voltage drops with a multi meter. Generally fast and easy for trouble shooting. You are just interested in phases that show a much greater voltage drop than the other phases.
You can often do this test with a multi meter in less time than it takes to get the ductor and cables out of the carrying case.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Skogsgurra]

Still not found the problem? After a couple of days? It can't be that much complicated.

I guess that you have spares? Change the MCCB and get done with that part. If problem gone - OK. If not, then connect a recorder (a fast one) and see what is happening.

In the first case, you may be able to live on without knowing exactly what the problem was. If you cannot, then take the MCCB apart and find the reason. Use angle grinder, with care, if needed. Personally, I really wouldn't mind what the actual problem is.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[PHovnanian]

Are all those pump loads 3 phase? If so, I'd chase down the source of that imbalance. Current readings on each phase of each motor branch circuit should be taken.

It's possible that one of the motors has a phase winding that is going bad. High resistance results in an imbalance. But due to an intermittent connection or pump load change, the motor slows or even stalls and draws locked rotor current.

You say the motor starter protection did not trip. Does the motor overload sense all three or only a single phase?

 

[BK32]

Hi All,

Unfortunately, I have not been able to do much yet, as there are a number of other works being conducted within the space. I have conducted a phase-earth megger check with the Chiller Switchboard isolated at the incomer (i.e. just the suspect breaker at the main switchboard and feeder cable in circuit), but all appears to be fine (no P-E fault present).

Not likely to get much done before Easter, so will hopefully have better luck next week.

Skogsgurra, if I could just replace the breaker and confirm that the issue was gone that would be great. it is just that this is one of the main distribution boards and will require a black-out in order to effect the replacement. This is normally just conducted outside normal hours, however, at this stage there is a lot of work being done outside hours, so if I can take steps to isolate the fault to the breaker first, then I will do that to make sure I have a definitive reason to black the ship out.

Thanks all for your assistance, I will let you know next week how it goes.

 

[Skogsgurra]

Yes, I understand that you want to make sure before you shut down. But the second suggestion doesn't need any shut down.

A recorder that samples 1000 - 10 000 Sa/s, three current clamps, three voltage channels and perhaps one or two channels for trip signals is all you need.
Then set trigger to catch a breaker trip and read what caused the trip once it happens. Much easier than playing guessing games.

Finally, if there is no excessive heat or any other signs of bad connections or faulty motors, then the likelihood that the problem lies elsewhere is very limited. So, changing MCCB is still your best option. Even if you don't like it.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

If this breaker has a replaceable trip unit it may be well to inspect and/or replace the trip unit.
However, if the breaker passed a recent ductor test then the trip unit is probably in good shape.
Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Skogsgurra]

Good shape in the meaning that it trips when it should, yes. But it doesn't say much about nuisance tripping. That may still be a problem that cannot be ruled out because it passed the ductor test.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.