Tripping Centrifugal Pump

What info is available in the documentation?

To use the pump curve you would need the RPM

Determine if the head is the total head (suction hight , static head plus friction losses)

the length of the lines are required and the losses through fittings should be considered.

An approximate Q can then be determined.

Rather contact an engineer (pump manufactures may also have one or more lying around)

A worn out impeller would usually pump less water and use less power - may not cause tripping.

The original design may have been for both pumps running at the same time and may require throttling when running alone. (operating at a lower pressure move the duty point to the right and raise the power requirements.

The bearings may be damaged.

Thanks Hendrik

The systems Technician from whom I can get the required flow rate is not around, he's attending a course. However, I have tried to get this new info:

Motor: Power=30kW RPM= 1465

De-aerator working pressure=20kPa

Treated water and surplus condensate temperature = 85 degrees celcius

I think I still need the required to know the required flow rate. I'd like to know what you think?

Once the pump curve has been found, make sure this information for this and all the other pumps is held in some central place where it can be found more easily.

Since you will be pumping hot condensate, is the condensate itself under atmospheric pressure? positive pressure? vacuum as a result of surface condenser?

All the conditions of the water being pumped need to be factored in, otherwise, you will cavitate the impeller of the pump, and damage it, and possibly trip out the pump due to overampere conditions.

How much real elevation exists between points in the system (from pump suction to the DA, for example). You have to factor all this into the final calculation of dynamic head on your pump, to know where it should rest on the pump curve.

By the way, what precisely do you mean, "that kind of pump they don't make any more"?

Seems to be some confusion about pipe size. You stated the pipe at the discharge was 150mm, then reduced (?)to 200 mm extending to the DA....is that right, or do you want to change stories?

Usually there is no need to choke the size of the discharge pipe, so as long as it is large enough in the first place, you should be good to go without tripping. If there is an admittance valve to the DA, you need to make sure also that this size has not been changed by anyone, and that it operates correctly.

Contact a couple of pump suppliers and answer the questions they will ask you and let them do the work for you.

Also, get on the phone to the manufacturer of this particular pump and ask for details of the latest version that might replace it.

• Recently, the repeated tripping of a centrifugal pump operating on sewage sludge was found to be a piece of blue polypropylene rope about 4m long, wrapped around the impeller shaft. After isolation, about 15min work with a penknife was sufficient to remove the offending object.

Thank you very much guys. I appreciate your help.

On a quick glance, this seems out of balance. For the given head and kW, the corresponding flow (~500 m³/h) would be too much for the pump inlet and outlet sizes to handle.

See if you can find out more data for this system; that would be helpful. At the moment, you are right; there is not enough info except for rough rule-of-thumb guesses.

Hi Nelton,

I am now retired but based on my 33 years experience as a pump specialist, I know that one of the causes of overloading in centrifugal pumps is the accidental installation of an impeller that is larger in diameter than the original impeller. All curves of centrifugal pumps show performance curves of a pump with different impeller diameters. The impeller diameters are chosen to furnish different flow rates at different heads. We used to remove the impellers from stock pumps and cut the diameter down to the correct size. Remember that impeller diameter is important when furnishing replacement impellers or replacement pumps. Good luck.

Thank you very much gentleman.

Listen to lyn, post 3. We build many process skids here per year. Basically ALL of them pump preconditioned water or treated water through a plant. We NEVER size the pumps anymore. We lay that on the manufacturer or his representative. The only real failure to size correctly, that I recall, was when we let the customer's engineering staff size one for us. That we have NEVER done again. The pump manufacturer knows best.

Now there is some good thinking, partner. That is why in the industry we have a slogan that goes: Fire! Ready! Aim! = Miss

Ready! Aim! Fire! works a lot better.

Somewhere in the files should be the packet provided when that pump was received. In it should be the pump curve either as tested or as originally designed and proven. Look at that curve and determine whether there were any changes, impeller diameter as mentioned before, and see where it lies on the pump curve. You should look for a pump with comparable operating properties, especially in a condensate service, you need to look at the NPSH Required to avoid cavitation. There should be a point marked on that curve showing the original design point, the TDH and the NPSHA as originally specified. Please don't tell us that someone threw out those files or that no one knows where they are. That would be too embarrassing for your company. It sounds like there were 2 sets of conditions originally planned for, one with a smaller throughput so that the one pump was undersized to eliminate excessive throttling. Either that or one of the original pumps has already been replaced by a different size. In which case, why wouldn't you just buy the same pump as a replacement.