How much margin to allow for internal wear?

[KernOily]

Good morning guys. Another question. Man I am taking over this forum.

How much margin on head do you recommend, or have you seen, to allow for the wear of internal clearances in a centrifugal pump?

The references I've seen (Karassik et al.) recommend 8%. My application is a split-case multistage BFW pump of about 3800' TDH, so 8% would be a lot of additional head to burn up across a control valve until the pump wears... The RFQ package is already out, so if I need to add a margin, now is the time to do it. I just don't want to oversize these things.

Thanks guys!
Pete

 

[BigInch]

Wait a minute, you're the guy with the 50% spare. Some credit could be given for the spare capacity, unless you're worried about the accuracy of the hydraulic predictions of flow and head loss, which I'd really only be concerned about if it were a much longer pipe. I'd say not much extra head is warranted, given the spare unit you have. The other alternative may be to buy a bit extra head and run with it to the generators without a pinch on the control valve. Are the generators maxed out at 3800 ft?

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[KernOily]

Thanks BigInch.

"Wait a minute, you're the guy with the 50% spare."

...and the fancy control system, which is getting fancier by the minute.

To answer your question, 3800' TDH is the load to the pump with the generators at full load - the 'rated' condition, as some like to say. This is what they have to deliver. With this TDH, I am burning up 10 to 15 psid across the inlet rate controllers to the generators.

I have 20% factor of safety in my friction loss calcs. The lines are not too long - several hundred feet.

Thanks!
Pete

 

[Artisi]

20% safety factor on friction loss calcs seems an over-kill and you are wanting to add a further margin because of possible wear on the pump unit. Seems that you are over-estimating what is actually required.



Phitsanulok
Thailand

 

[BigInch]

Artisi, you're right. I'd cut the hydraulic error margin allowance to 10% at normal load, possibly dropping to 5% at max load. 20% is for cases where you don't know the final configuration or diameter of the piping yet, you don't exactly know what maximum flow you will have or want, you have a bad configuration with a lot of bends, fittings, valves, to increase the minor losses, not a well defined high point elevation, you want to pump a range of different products, have some 2-phase flows, wax caking or dirt settlement, non-newtonian flow, i.e. some particular problem you want to CYA.... or ..(?) you have no confidence in the pump curves or hydraulic calcs (?).



BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[Artisi]

The old problem of everyone adding a little bit "just in case" ie, the process guys,the designer, the specification writer, the sales enginner, the application engineer, at the end of the day the pump is 1 size too big and the motor selection is 2 frame sizes up.
Of course, it's all the fault of the pump manufacturer - cause he doesn't know what he is doing.

Phitsanulok
Thailand

 

[BigInch]

Then they want to fix it with a VSD!

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[Artisi]

-- or post the problem in Eng-Tips looking for a miracle cure to the problem.

Phitsanulok
Thailand