Vertical pumps

[Texark]

What is the best way to determine what frequency you can expect to see excited when a spider bearing wears out on a vertical turbine?

Thanks,
Texark

 

[Strong]

Typically vibration amplitude increases at 1x shaft speed and also at natural frequency(s) of pump column (below base plate). May also see increase in vibration at motor-pedestal natural frequency(s).

Walt

 

[Texark]

Thanks Strong, that is very helpful information, but I guess what I am really asking is what is the best way to field test such a pump to determine what the shaft whip frequency is?

 

[Strong]

Most vertical pumps are fixed speed and do not have direct access to pump shaft and shaft bearings. Removing the pump or dewatering the sump to install sensors is often not practical. A diver can sometimes install casing accelerometer(s) depending on water clarity and safety issues. The best vibration measurements that can be made in this circumstance is at the upper (NDE) motor bearing. Various types of vibration measurements at this location can be made. Can you can provide more/specific information about the type of pump and the problem you are trying to solve? If you have access to shaft bearings, then either proximity probes detecting shaft motion or accelerometers on bearing housing can be considered. It also depends on motivation (time & money) whether measurements are for Condition Monitoring (Predictive Maintenance), fault diagnostics (possible legal actions) or for R&D to improve product.

Walt

 

[Texark]

Hi Strong,

Thanks for your reply.
A scenario would be this: Large sub-synchronous peak in data taken on the motor (pump shafts & bearings are inaccessible, as you stated). This peak is suspected to be shaft whip due to worn or missing spiders. What is the best way to verify this before recommending the expense of pulling the pump?

 

[spciesla]

Can you do a bump or FRF test on the motor to see if the sub-synchronous peak coincides with the reed frequency of the structure?

 

[Strong]

Texark,
Is there an identical pump nearby for making comparison vibration measurements? How long has the pump been in service? Does the water being pumped have a lot of silt or sand or other abrasive material? Do the bearings have clean water flush?
I would conduct an impact test with impulse (load cell) hammer; instead of an ordinary hammer. The hammer size/weight with extender mass would depend on motor size (typically 2-lb to 12-lb sledge). The impact tip should be soft rubber (grey) to get low frequency energy. The impact and response should be in two directions at top motor bearing housing (parallel and transverse to pump discharge). A low frequency accelerometer with high sensitivity (500 mv/g) is helpful for low vibration levels. I conducted a similar test on two 600-hp 720-rpm vertical circulating water pumps (10,700-b motor and 13,190-lb pump)a few weeks ago with a 2-lb impulse hammer and extender mass. Fortunately I did not need the 12-lb sledge! Good vibration test (operating and shutdown) data can be very helpful for making the decision to schedule and remove pump.

Walter Strong

https://www.linkedin.com