Cooling tower Pump movement during startup

[svkd]

Guys,

In our plant during startup ,one of Cooling tower circulation pump created abnormal sound,it ran for 5 seconds only and after stop, The Pump was found Shifted on its base around 3mm towards discharge side.We try to turn the shaft but it was jammed ,after dismantling ,both case wear ring and impeller ring had rubbing marks.Do any one of you come across this situation of pump movement?Please share your views.

 

[racookpe1978]

Show photos of the outside of the pump - it "shifted on its base" (how was it fastened and grouted to the concrete) and the inside - where the impeller jammed.

You will need to replace the pump, base and foundation bolts - if they didn't hold the first time, they won't hold the second time. Motor ok?

The why? Don't know enough yet.

 

[svkd]

Hello racookpe1978 , attached picture of pump movement,I don't have inside picture but there was rubbing marks on case wear ring and pump wear ring.Needs to find the reason for pump movement as well as jamming of wear ring.What happens first ,Since Pump has sleeve bearings so It seems wear rings rubbed first and cause pump movement?

 

[Artisi]

No where near enough information, it could have been a hydraulic problem or could be an alignment problem, an installation problem, or a pump assembly problem. Far too many variables to make any real comment at this stage.

What pump style?

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[svkd]

Artisi
Its between bearing cooling tower pump , Alignment was checked before startup and found good,Assembly was tested during factory test and results were good ,

1. Installation Problem ? What you can think are the problems that can lead to pump movement and wear ring rubbing
2. Hydraulic Problem? During factory performance test the results were good. What you can think are the hydraulic reasons at site that can cause pump movement and wear ring rubbing

I feel that the base bolts looseness leads to pump movement and pump movement leads to abnormal shaft deflection and leads to wear ring rub and pump jammed.

Guys , What do you think is happens first wear ring rub OR Pump movement?

 

[Artisi]

Did you physically check / witness alignment during installation?

Was alignment done before or after the unit was bolted into the pipework?

Hydraulic problem, was the pump completely primed / full of water at start-up.


Why would the hold down bolts be loose? You are sure alignment was correct but doubt hold down bolts?

Is the unit baseplate mounted?
If baseplate mounted and alignment wasn't correct, the baseplate could have been bowed / hogged / twisted.

Not the first time a 100% guaranteed installation wasn't correct, of course the installation team will always say it was done properly.

Bear in mind, you are there with the unit in front of you, while we are having to guess what is likely to have happened. Some decent photo's never go amiss, also what size unit is it 50 kW or 500kW?!


It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[svkd]

Artisi ,

This installation is 1500 KW pump ,mounted on base frame.
I am not saying the base bolts were completely loose but may be less torqued so cause pump movement during startup.My guys had witnessed the installation & am sure that alignment ,pipe stress and base frame level was good. Also during startup the pump startup procedure was followed,priming was done.Pictures are attached in previous post at top in reply to racookpe1978.

We have commissioned 5 pumps already ,this is 6th one and had faced this problem of pump movement.
Just want to know experts opinions on pump movement and wear ring rubbing. Is it possible the pump wear ring jammed first and that leads to pump movement?

 

[Artisi]

I would say the pump 'jsmmed' during start up and with 1500kW being applied to a jammed pump,think you are lucky it only stalled the unit.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[svkd]

Artisi , There are no observation of galling on the wear rings.Is it possible that the pump movement occurs first that leads to wear ring contact ? What you say about this?

 

[Artisi]

What would cause pump movement. Wouldn't think that if the bolts weren't torqued sufficiently it would result in casing or shaft deflection. If it were me, I would still be looking at misalignment or pipe load. But of course I'm not privy to the installation documentation or have the benefit of first hand knowledge.
Sorry I can't be more helpful.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[TheTick]

Stuxnet!

 

[Strong]

The photo appears to show one pump foot. Does the pump foot about 180-degrees away show similar "movement" on the opposite side that would indicate it was caused by torque? Could it be that pump was intentionally moved during/after installation after the baseplate was painted? Did anyone measure the base bolt torque to verify if any bolts were actually loose? Have you considered the amount of force required to overcome the base friction from the of the pump and motor weight, bolt tension, and hydraulic downward force? If the pump impeller jammed, then I would expect shaft shearing from the applied operating and inertial torque of the motor. If the shaft did not shear or no other major internal damage occurred, then I certainly would not expect the pump to move on its base. I also do not see any way for pump base to move due to normal startup and operating forces and somehow cause internal damage. Pump impeller wear ring rubs can occur on new pump if there is incorrect bearing or wear ring clearance or excessive vibrations from impeller unbalance or shaft misalignment; possibly combined with shaft or housing resonance.

Walt

 

[svkd]

Strong , All Pump feet had similar movement towards pump discharge side.It only moved during the startup before that it was on its position.My guys had witnessed the installation & am sure that alignment ,pipe stress and base frame level was good. Also during startup the pump startup procedure was followed,priming was done.

At the moment we are still analyzing the root cause for pump movement.What do you think?Is pump movement due to insufficient torque on base bolts can cause wear ring rubbing?

We have not checked the torque of the base bolts ,will do now.Thanks for this.

Also ,I would appreciate if you could explain how to calculate the amount of force required to overcome the base friction from the of the pump and motor weight, bolt tension, and hydraulic downward force?

 

[Artisi]

Did the whole base assembly move towards the discharge or only the pump, did it return to its installed position after shut-down?
What position where the pump/ pipe flanges relative to each other and their original position on disassembly after failure?
Did the pump move first or did the clash of w/rings cause the movement???????

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[svkd]

Artisi, Only pump move towards the discharge and it did not return back to its installed position after shutdown.
Below is description of the event:
''Pump ran for 5 seconds only and then stopped. Pump was found shifted on its base approx. 2-3 mm towards discharge side. During coupling spacer removal, they found that pump is jammed, hence they try to make free by installing clamp on Pump Hub. But they failed to move the pump shaft.''

In our case ,There is no need to disconnect the piping to inspect the pump internals ,As pump is split case design ,so Only the top casing is lifted up to remove the rotor to check for jamming.After dis assembly,both case and impeller wear ring were observed with rubbing marks ,other parts such as coupling ,bearings,seal were found good.

Did the pump move first or did the clash of w/rings cause the movement??????? This is want I want to understand.

 

[Artisi]

I think there is a very good reason to disconnect the pump from the pipework, you need to check flange/flange alignment.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Tmoose]

Hi svkd,

Sounds like there is a bit more thoughtful investigation required for a "root cause."

Do you have the results for the actual as-found bolt tightness? I think an actual measurement of lb-ft etc is required. A mill wright's report of "they are tight" is not enough.
Whether the bolts are found "tight" or not I would also test that each bolt is truly clamping the pump foot, and not bottoming in the tapped hole etc.
One test is to install each bolt hand tight back in it's pump foot without any washers . It should stop when it makes contact with the pump foot. If it stops with space under the bolt head, there is real danger that the bolt is bottoming in the hole before actually clamping.

You said "After dis assembly,both case and impeller wear ring were observed with rubbing marks ".
- How far around the circumference of the seal ID and OD are the rub marks?
- Were measurements for wear ring runout and case clearance made (and recorded) ?

I'd want to test the other pumps' bolts( one bolt at a time ) for measured tightness and depth clearance too.

regards,

Dan T