tripping motor 2

[Shaaban Emad]

hello
A 3ph 900kw 6.6kv 94A fire water pump.it has tripped when attempt to start by 50 and 51N protection (fault current 3500A, tripping time 0.05s )
The motor insulation is good and the rotor is free ,the CT was tested.it has re-set and start again it is consuming 60A.
what would be the possibly reason?

 

[edison123]

Poor trip settings. Malfunctioning relays. Failing CT's. Take your pick.

Muthu

http://www.edison.co.in

 

[Shaaban Emad]

Hello
It was proofed that the current consumption was high by voltage drops in network at the instant of the tripping

 

[waross]

Is there any sign of a flashover at the starter?

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

 

[Skogsgurra]

Is it a asynchronous motor? Or does it have external excitation? Slip rings or rotating transformer and such.

The latter is sensitive to load torque when starting and switching to DC excitation. Had a case with a (somewhat larger) motor that tripped during start and then it tripped again, weeks later. I visited the place and recorded what was happening. The recording showed heavy torque variations during start and then it tripped. I had them tear down the pump to see what was causing it. They found half a check valve flap that somtimes interfered with the rotor.
We also had violent voltage variations when that happened.

Check valve part removed and check valve repaired. Works since then. Must have been at least ten years ago.



Gunnar Englund

http://www.gke.org

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

 

[controlsdude]

Its best to separate any mechanical problem from the motor. That way you can focus on mechanical problems versus the motor itself.

 

[jraef]

The N designation means it is on the Neutral CT. Do you have a Neutral CT on a motor? Or was this on the transformer protection relay? If it was on the transformer relay, it’s not likely a motor problem.

Can you adequately describe what type of motor it is, how it is being started and the entire sequence of events leading up to the trip? For example, was a start attempted, immediately aborted, then re-started right away? Things like that are helpful in enabling us to help you, we are not there and guessing is tedious.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[davidbeach]

Actually, per C37.2, 51N is a residual connection, not a directly measured value. No end of confusion there.

 

[Shaaban Emad]

Hello
thank you all for your intersts
the motor is direct on line induction motor,the ct is core balance ct at the main cable ,the pump was separated and the motor restarted normally then the pump recouped and started normally as well .
the rotor wasn't locked

 

[Skogsgurra]

All well, then?

What was the problem?

Gunnar Englund

http://www.gke.org

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

 

[Shaaban Emad]

"Skogsgurra (Electrical)3 Nov 18 13:39
All well, then?
What was the problem?"

we still not found the reason of this large current value ,
how can we record the motor starting torque to tell that there was issue with pump load ?

 

[Skogsgurra]

Hello Shaaban,
It will be difficult to record something that doesn't seem to happen any more.

What you can do is to arrange a "mouse trap" that trigs on extraordinary high currents. It should also record voltages and speed of the motor. Plus, the trip signal. At least.

With that arrangement, you will have all data and information needed if the trip occurs again.

You probably have all the signals needed, except for the speed signal. A good substitute for a tacho generator is a pulse pick up that senses the key on the motor shaft (if accessible). You can also put a strong neodymium magnet on the shaft and pick up the field with a reed element. For that you need a voltage and a pull-up resistor. That will give you one pulse for each turn and that is suffient if all you want to know is if the motor turns or not. No pulse=stand-still and 25 pulses/second is 1500 RPM etcera.

If there are no CT:s in the system, then you need to use current clamps that can handle the expected current - probably a lot more than rated current's peak value. I use DC clamps for that so one doesn't get fooled by the DC transient at switch-on. That DC transient usually saturates an AC clamp and that is something you should avoid.

What recorder? There are plenty to choose from. Small and fast units like the Pico Scope family has more than adequate BW and many have quite a lot of channels. Like eight. Which seems to suffice in this case. You can use any recorder with BW better than 1 kHz.

Only current in one phase shown.




Gunnar Englund

http://www.gke.org

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

 

[flexoprinting]

Excelent idea Gunner, I am going to check the specs on a prox sensor and try that. We are currently renting a Fluke 1750,
a very nice recorder and were told it cost $1500.00 US new. It would be easy to supply 24V to prox and use one channel for that. Thanks

Chuck

 

[controlsdude]

Since its a fire pump. Don't you have all those header pipes coming out where you could close and open valves to make a circular loop for the pump? Since you now know electrically its good and have a plan to analysis the problem. That was you could say this area is ok and go from there.