1000HP Deep Well Water Pump 2

[gokulkrish2]

Hi All,

We have a deep well water pump in our site and we have had more than 5 failures on it now. I am given the task of finding the root cause for the failures and give recommendations of what can be done to prevent future failures.

Here is the Pump Information:

1000 Horse Power Centrilift 725 HMI UT Motor
29 Stage HC3500 Centrilift Pump
4 Bag Sealing Section 6.75 OD

4160v Switch Gear with 469 Multilin Power Management Relay

Load information

Avg Discharge Flows of 1350 GPM
Avg Water Temperature of 95 Degrees Fahrenheit

Background:

I talked with the Electrical Supervisor and he thinks that the pump is being started lot many times ( 8 to 9 starts per day ) and that is the reason it fails. He argues that the pump fills the store water tank quick and so the level sensor shuts the pump. He recommends we do a lake and pump water into that and store so that we have less starts on that pump.

Other argument is that the inrush on the pump would be too high. Right now the pump is started with a DOL starter. Some recommend that we go with a VFD or a Soft starter to start and hence reduce the inrush and increase the number of starts. I think this is good than the lake option.

Also, do anyone have a recommendation of how to measure the sub transient and transient inrush current values of the pump during the start? I want to measure and see how much is the current rise. Wouldnt that help in the appropriation for VFD or soft start?

I know a little about pumps and hence whatever you guys give would be a valuable peice of information for me.

Thanks.

gokul

 

[waross]

What is the failure mode of the pump? Does it look like overheating or insulation failure? Is there a possibility of water ingress? What is the static water pressure on the pump when it is not pumping?
Is the pump ever started when it is back-spinning? This may be the most probable cause of burnout, but proper setting of the Multilin will give a lot of protection.
The Multilin should give you more information than you need.
In a healthy motor the sub transient and transient currents are to a great extent determined by motor design. Of more interest is the duration of the starting currents.
Have you done any calculations on the starting voltage drop at the motor terminals. The long conductors to the motor, although adequate for running current may be undersized for the starting current. Too much voltage drop when starting may lead to extended start time that may lead to motor burnout.
It should be easy to determine the starting time from the Multilin. Contact the pump or motor manufacturer for their reocmmendations for maximum allowable stating times.

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

 

[jraef]

I think Bill hit the nail on the head. The first step is to determine the failure mode of the motor and/or pump. Anything else is just blind conjecture.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[waross]

Thanks jraef; One of those rare days when a couple of synapses connect in the old brain. It doesn't always happen.

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

 

[Marke]

I agree with Bill, until we have some information on the failure mode, it is not possible to give meaningful comment.

You could refer to the manufacturer in regard to the number of starts, DOL starting etc, but I would suggest that if you have evenly spaced starts, 8 - 9 starts per day is not an issue, neither is DOL starting.
I would be more concerned about insufficient water level above the pump or severely imbalanced or distorted supply.

Best regards,
Mark.

Mark Empson
L M Photonics Ltd

 

[MikeHalloran]

You might want to replace the broken compound gage in the riser and watch what happens to the riser pressure as the pump cycles.



Mike Halloran
Pembroke Pines, FL, USA

 

[rmw]

How deep is the pump? 95 degF for ground water sounds high unless it is some type of geothermal situation. There might be a clue there.

rmw

 

[itsmoked]

That's what I was thinking. I suspect the water is being heated by 1000hp.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[rmw]

Kieth, you are reading my mind. If the groundwater is the temperature I think it should be, that is a lot of temp rise for the water.

rmw

 

[waross]

My calculations show that at 75% overall efficiency the hole would be over 2000 Feet deep. Will someone double check my numbers, please?

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

 

[Marke]

Hi Bill

The numbers do not seem to add up unless it is a very deep well.
I wonder if the problem is related to the pump running backwards? that would explain the low flow and high water temperature.
Many of the submersible pumps that I work with have a maximum winding temperature of around 35 - 45 degrees C (95 - 115 degrees C) so if the water is that temperature at source, it would not give much cooling to the motor!

Typically, a 200 KW motor at around 100 meters will deliver in the order of 90 liters per second with a head above the surface of around 5 Bar. - total pumping head around 100 Meters of water.

Best regards,
Mark.

Mark Empson
L M Photonics Ltd

 

[electricpete]

Will someone double check my numbers, please?

Would at least like to know the current draw first.

Is it outdoor pump? Do failures tend to occur more during rainstorm? If yes moisture intrusion should be considered among possibilities.

Also consider turn insulation / surge failure if failures occur T-storms, system transients or during motor start.

8-9 starts per day every day seems to me on the high-side for 1000hp motor. Especially if 2-pole motor, and if starts are clumped together during a high demand period without adequate cooling. Also especially if motor runs for a very short time after start (motors cool a lot quicker when running after start then when standing after start). What is the rough duration between installation and failure? Are there any nuisance trips from overload? Do you happen to have winding temperautre indication? That would not be definitive, but would help assess the severity of the starting duty. Review of relay to determine time to start is good suggestion - good qualitative indicator of the severity of each start. If you have voltage issues or rotor degradation (*) prolonging time to start, it'll show up there.

* Repetitive starting of large motor can certainly damage rotor over time. To check health of rotor, do a high resolution FFT of the current waveform looking for pole pass sidebands around line frequency. Also look for current oscillations. Other checks possible in the shop.

If the starting duty turns to be an issue, I think installation of soft starter is a dubious solution. The total heat generated during reduced voltage start is at least as high as DOL start (and can be higher). The only good things are reduction in winding forces and the heat is spread over longer period of time... I have heard soft start is effective for some centrifuge type loads where start is prolonged to 5 minutes or so.... but I haven't heard of that strategy for centrifugal pump.

On the other hand, vfd with smoothly ramped frequency does significantly reduce heat generated during each start.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Clarification in bold:

Especially if 2-pole motor, and if starts are clumped together during a high demand period without adequate time between starts. Also especially if motor runs for a very short time after start (motors cool a lot quicker when running after start then when standing after start).

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Marke]

Hi Pete

deep well water pump

I suspect that it is a submersible, so water cooled.

Best regards,
Mark

Mark Empson
L M Photonics Ltd

 

[gokulkrish2]

Hi All,

everyone should pardon me for not answering your questions. I am closely following your replies and trying to collect all the information. I will get all the answers and reply to guys. Thank you so much for the responses and the time you guys took.

gokul

 

[gokulkrish2]

Hi all,

Here are the answers for some of the questions.....

What is the failure mode of the pump during all 5 events?
Three motors have failed due to end coil failure. Likely cause is from frequent starting and stopping of pump and starting current.

Does it look like overheating or insulation failure?
Overheating is not evident

Is there a possibility of water ingress? Is it a submersible pump?
We have not had any water ingress. It is a submersible pump

What is the static water pressure on the pump when it is not pumping?
Answer yet to be found out

Is the pump ever started when it is back-spinning?
Pump has 30 min lock out timer

Have any calculations on the starting voltage drop at the motor terminals
yes, voltage drop is minimal. Less than 5 %

What is the pump or motor manufacturer recommendations for maximum allowable stating times
Nothing specific. But recommends to reduce the no of starts by keeping the pump running longer, install vfd

Is the unbalance of currents at the motor terminals measured?
Multilin has not tripped from load imbalance. It is set to trip at 20% and provide alarm at 15%
How much number of starts per hour is allowed by the manufacturer?
Manufacture recommends 1-2 starts a day

How much time it takes for the pump to fill in the storage tank?
Less than two hours

How much deep is the pump setup?
1840 feet deep and the water level is 660 feet deep

I will try to find more answers and will post is as soon as i get them

thanks,
gokul

 

[waross]

We have not had any water ingress. It is a submersible pump

What is the static water pressure on the pump when it is not pumping?
Answer yet to be found out

If the water is not getting into the pump, this is not too important, but you have answered the question.

How much deep is the pump setup?
1840 feet deep and the water level is 660 feet deep

1840 ft. - 660 ft. = 1180 feet

Overheating may involve the entire winding, and the portion inside the core may be hotter than the end turns.
You may have high voltage transients damaging the end turns.
Are you using a vacuum contactor?

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

 

[electricpete]

Is the pump ever started when it is back-spinning?Pump has 30 min lock out timer

Backspinning might occur due to leaking foot valve which allows water to flow backward through the pump.

What is the pump or motor manufacturer recommendations for maximum allowable stating timesNothing specific.

There should be at a minimum motor time current curves showing damage region. One possible step is verify your starting times are below those, which is a necessary but not sufficient condition for reliable operation.

Manufacture recommends 1-2 starts a day

Sounds reasonable to me although I don't work with submersibles. I see the motor bears the same company name as the pump. You'd think the conflict between manufacturer recommendation and application demands would have come to light during design/specification stage




=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

One thing to add... if large motor is subject to excessive starting, I would more likely expect rotor problems than stator problems.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[LionelHutz]

No sign of overheating would mean the end turn damage is due to magnetic forces during the motor start?

If so, then a soft-starter should help. You would just need to limit the inrush and lower the forces in the motor.

Ask the manufacturer if there is any limits on the ramp-up time. Some submersible pumps use a water bearing which lifts the rotor/impeller assembly and they must reach a minimum rpm fairly quickly.