Centrifugal Pump Failure Mechanism on Dead Head Operation

[MEKENG]

Dear All,
I am a new engineer with mechanical background. I work in power plant.

Problem Description
We have a vertical centrifugal pump for cooling water for condenser purpose. The centrifugal pump flow is 9500 m3/hr, with 600 rpm. One day, operation team operate the pump and due to some reason, the control was not working and make the pump run for his dead head pressure for around 5-10 minutes. We check the amperage of operation and after 7 minutes it's decrease significantly and we assume the the shaft is broken.

Since the size is quit big, we dismantle the pump in the workshop and found that the shaft is broken.

Since I am new in this work, is there anybody can explain how is the mechanism of this failure?

So far, I can summarize as below. Is there anyone can help me to make a reasonable mechanism of failure of this pump?

When pump run in dead head condition, the fluid temperature will increase into vapor state. When it become vapor, then it any bushings or mechanical seals in the pump can heat to the point that they can either crack or melting. Then ????? (I have no idea how it will happen). In our case, the end result is fracture of shaft on the lowest bearing area. Anyone can help me to answer the failure mechanism on this case?

Thank You;
MEKENG

 

[LittleInch]

MEKENG,

We're pretty good on this site at trying to help people, but we're not possessed of extra sensory perception.

Therefore to assist further we really need some data and drawings

So yes flow, but not head or power
Pump curve might help
Vertical centrifugal can be many things so a drawing would help
Bearing lubrication type and method?
Any issue with this unit before?

I must say operating 5-10 mins at no flow shouldn't really cause a shaft failure unless it was either too small or already damaged.

Pictures of the shaft breakage would help
As would damage report on the bearings

Mechanism of the failure? No idea based on what you've provided so far.

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

 

[Rputvin]

What brand and model is the pump? I'm just curious as to see a pump capable of 9500 m3/hr (41,827 gpm).

If the shaft or impeller had any binding you could have hit a locked rotor condition, and it sounds like this is a big enough pump/motor that could likely break things easily if you had some issue internally.

Without knowing the design of the pump, seals, motor, controls, and how the shaft is coupled, it would only be aimless conjecture to comment on the mechanism of failure. I would think the controls safety (motor controller, fusing/circuit breakers, overload relays, etc) would have tripped well before enough torque was applied, but once again I have no way to know what the motor is capable of producing, what the safeties are setup for, and what kind of shaft is on the end.