Loose Impeller Nuts 1

[HkyJky]

I have a duplex sewage pump station rated at 1,200 gpm pumping against 250 feet of head. Over the past year, both pumps have had the impeller nuts loosen to the point where the impellers dropped and contacted the wear rings on the volutes. Manufacturer stated that impeller nuts are lock nuts and are installed with LocTite. After reinstalling, the problem reoccurred. The pumps are controlled by VFDs which slowly ramp the speed up/down. When viewed from above, they rotate in a clockwise manner. The nuts are right-hand threads. What are the forces acting against the nuts? When the motor is slowng down against all that head, it would seem to me that it would want to tighten the nut. We have confirmed that the check valves are seating properly with no backflow from the foce main.

 

[IntroToPumps]

Have you confirmed that the pumps are actually rotating clockwise? If two of the motor leads are switched the pump will rotate counter-clockwise causing the impeller nuts to unscrew.

 

[JJPellin]

I assume that the impeller is key driven on the shaft. The nut tightens with rotation. Even if the check valve were to stick open and the impeller would spin backwards, the torque on the impeller would still be in the same direction, but the nut would be loosening with rotating. This nut should not unscrew. The relevant forces could include gravity which would encourage the nut to unscrew. Vibration from run-out, imbalance, or cavitation could work the nut loose and then gravity helps it walk off the shaft. The VFD could be exciting a torsional resonance at some speed within the operating speed range. If the impeller fit to the shaft is loose, this could create more potential for movement that results in the nut coming loose. I would recommend all of the following:

Calculate the torsional resonance of the pump rotor assembly. Determine if this is within the run speed range.
Measure the impeller to shaft fit and consider tightening this fit if it is exceptionally loose.
Make sure your mechanics are using correct practices with regard to cleaning, preparation and application of the thread-locker.
Consider modifying the design to include a set-screw, tab washer, cotter pin, Nord-Lock washer or other locking mechanism.
What style of pump is this? A drawing of the impeller mounting would be very helpful.


Johnny Pellin

 

[1gibson]

An impeller nut should have threads that tighten during normal operation. You will reverse the torque to some degree if the motor goes from driving the pump (constant speed operation, or increasing speed) to acting as a brake on the pump (decreasing speed.)

What if your slow ramp up/down is really a lot of very fast but small steps? Inertia of the impeller as you brake with the motor will change from one side of a key being loaded, to the other side being loaded. Do that for enough cycles, and I can see a nut coming loose very slowly. 0.005" slop in the key/keyway fit may be enough to loosen the nut a small fraction of a turn at a time. Any evidence of fretting between impeller and shaft?

I'm trying to think of a good example but coming up short. Say you're sitting on a barstool with a back, and you want to rotate yourself relative to the seat so your back is in a more comfortable place. You might slowly start spinning one way and then abruptly swing your arms to change direction quickly, so you slip a bit on the seat. My guess is that is happening to the impeller nut every time you command a reduction in speed from the motor. And it can only loosen itself, it won't tighten itself.





Or maybe they just forgot the Loctite!

 

[LittleInch]

Is this a vertical shaft horizontal impellor unit? I.e. is the nut actually holding the impellor vertically in position?

Does the impellor itself move vertically under load?

Maybe the impetuous vibrate ever so slightly?

WE need a diagram really.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[bimr]

For sewage pumps, it is common that impellers may be directly connected to the motor shaft with a slip fit and key, and then secured with an impeller nut.

Appears to be an assembly problem. Is the supplier using an impact wrench?

When installing impeller nuts, DO NOT use an impact wrench. Use of impact wrenches has proven to damage the impeller washers, impellers, and impeller shafts. Proper tightening procedure is to bring it snug tight, and then tighten it to the next available cotter pin hole in shaft and notch in the castle nut. Then install stainless steel cotter pin.

http://www.darley.com/documents/guides/pumps/accessory_repair/Basic_Assembling_Instructions

http://www.eng-tips.com/viewthread.cfm?qid=177887

You would obtain a better response if you posted the type of pump.

 

[Artisi]

Any problem in previous years?
How is impeller fitted, parallel shaft key or taper?

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.)

 

[HkyJky]

Thank you all for the great responses. I have attached drawings as some have suggested (Will it accept two??)

The manufacturer has not heard of any other reports of this happening with any of there other installations. The pump is a 150 hp Yeomans Dry Pit Submersible, series 9235

JJPellin: I am investigating your Nord-Loc washer suggestion. I believe this will solve it if the videoes on their site are representative.

1gibson: I think the slight changes in inertia when the VFD changes speed also come into play. In reality, this station probably should have had a soft-start rather than a VFD. The head is so great that we don't move any water until the speed is well over 90 percent - variable speed is impossible under these conditions. Also, the slop in the keyway fit could be an issue leading to the loosening. We will be pulling the pump tomorrow to inspect.

bimr: I wish we had a cotter pin arrangement! I think the Nord-Loc washers might do, though

 

[bimr]

Sounds like there is a maintenance procedural problem. Impeller mounting hardware is designed for one time use only. New hardware shall be installed during each assembly.

Is the manufacturer the one providing service/

 

[LittleInch]

Is it me or are the bearing surfaces for what I assume is a nut and washer on a cut down threaded part of the shaft rather small? Any small error in casting or machining of the impellor means that there is the potential for only a very small area for contact. Is there actually any of the impellor sticking out beyond the end of the threaded part of the shaft? i.e. is the nut really just tightening on the end of the shaft and not on the impellor casting?

Has the contact point on the impellor been damaged as a result of the previous loosening?

As your pump speeds up ( how long does that take?) I would imagine a fair bit of vibration until you actually start to flow some liquid at >90% max speed.

Photos would be great if you're pulling it out for inspection.


Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MikeHalloran]

If they're using Loctite on a nut whose locking mechanism involves plastic, that may be a problem; some Loctites attack some plastics.

Photos could be very helpful.


Mike Halloran
Pembroke Pines, FL, USA