Hi all,

Recently, I came across a 350kW 415V Low Speed Submissible Pump (high inertia load) with FLC 788A but failed to start using a softstarter
set at 70% Voltage and 3.5 times starting current.
The supply is from a 22/0.415kV 750kVA (ie. 1000A) Transformer with 6% impedance.

The scenerio is like this:

The supply is only 1000A from the authority. The external overcurrent relay was defeated so that it cannot trip the main 1600A ACB during starting. But the ACB electronic relay was set much higher but trip at about 4100A. Parties involved claimed that the supply is insufficient, pump is jammed or the problem is with the softstarter. Now one party wanted to verify the status of the pump by shorting the softstarter and on testing on no-load, failed (pump didn't run) with the current exceeding 14000A and the ACB flashover.

Can anyone say what's the problem? I estimated the starting current for full-load to be about 4 x 788A = 3152A which gives about 3152/1000A x 6% = 18.9% volt drop which I believe is way above manufacturer's voltage tolerance of 10% drop, meaning starting torque from the motor is too low to start the motor. I think a 1500kVA transformer at 6% impedance will work as it gives about 9.46% drop. But I reckon that if the 750kVA transformer was used to start the pump without the softstarter on no-load it should work though the final requirement is still on load. But why did the no-load test also failed?

If a VSD is used, what will be its rating and what will be the minimum rating for the transformer size at 6% impedance? Can anyone explain with calculations. Thanks!

Hello mark03

Welcome to the forum.

I would expect a submersible pump to be a low inertia load.The diameter is small and this usually keeps the inertia low.

I have found situations where the pump has become jammed due to sand or pebbles in the impeller housing. If the current is steady and balanced across all phases, and the pump has operated previously, then it is possible that some material has been drawn into the pump. I have cleared this before by reversing the pump. There is a risk in doing this, as it is possible to unscrew the impeller off the shaft in some pumps, but if the pump is jammed, it has to come out of the well if you can not clear it, so loosening the impeller is worth the risk.

The pump has a current rating of 788Amps. Typically the Locked Rotor Current of a submersible pump is in the order or 500% - 600% of the rated curent. If the motor is connected to full voltage with no voltage drop, I would expect to see a start current of around 4000 - 4500 Amps.The impedance of this supply would cause a significant voltage drop under full voltage start conditions, so I would expect the maximum current to be lower than this.
If the current is 14000A, then there is a problem with the motor windings, or the connections to the windings.
A pump of this size will probably have multiple tails that need to be connected in parallel, for example two star connected windings that must be parallel connected. If these cables are not correctly identified, then the connections can be wrong and applying voltage can damage the windings.
If the motor is wound as a double star, and the windings are not correctly connected, then it will behave like a short circuited transformer and the winding currents will be very high and the windings can overheat and fail.

Assuming that the motor is a parallel star stator, I would separate the two windings (I expect that all the tails have been brought up to a terminal box), and measure the insulation of each winding to earth, and to the other winding. If there is a low test between windings, or to earth, it would indicate that the windings are damaged.

Best regards,

Thanks Mark! Though I did not see the pumps, I now have been told that they are all new pumps and installed for flood mitigation. They have been installed a year ago without testing and commissioning until some 2 weeks ago. I have been told that a pump installer will be acquiring a gen. set to test the pumps.

I've asked some mechanical engineers about the problem and they also concur with your suspicions about the pump and electrical connections. I will be waiting for more information. Well, the forum is very good for all of us and I hope there are others who can contribute and replying to some of the other queries of mine in the beginning.

Best regards

Hi Marke03,

I haven`t much to add with regards to Mark`s excellent reply. However, I witnessed a similar scenario some couple of years ago when a 185 k W submersible pump was installed but only commissioned some 12 months later. The pump was jammed and after a few unsuccessful attempts the pump was removed and the thrust bearing was found damaged and the motor was beyond repairs. The manufacture claimed that the assembly was jammed due to the fact that the pump was left idle too long.

Regards.

Bob

Hi Bob

That is a new one on me!!

I have heard of thrust bearing damage by operation of the motor below pumping speed. It is very important to accelerate the motor up to pumping speed in less than three seconds. With a VSD, this requires a fast acceleration up to a minimum speed and a slower acceleration rate from there. The deceleration rate from minimum speed to zero must also be faster than three seconds.

Best regards,