Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

2x100% or 3x50% pumps in power plants

Status
Not open for further replies.

MechMad1

Mechanical
May 23, 2012
24
Hi!
When deciding which configuration of pumps to use in a power plant (feedwater, condensate, cooling systems, ...), I always have the same doubt, 2x100% or 3x50%.
As far as I know, 3x50% configuration are normally more expensive than 2x100%, but allows you to use just one pump if the duty is lower than normal.

Is there other reasons to use one or the other configuration? Maybe it depends on the system (feedwater pumps, cooling water pumps, ...)

Thank you all!
 
Replies continue below

Recommended for you

From a plant O&M perspective I rather see 3x 100%: one with maintenance, one about to break, one on standby. [wink]

Failing that I'd probably favour 3x 50% pumps if 50% was a useable flow which allowed plant operation at reduced output. Balance-of-plant drives tend to be on shared services with the capability to affect more than one generating unit, so the greater the redundancy the better. BoP drives tend be located where there's sufficient space for multiple drives, whereas space on the machine tends to be at a premium.
 
Depending on your flow, but if one pump trips in a 2 x 50%, then at least you maintain some flow, maybe up to 75%, while you start your standby, whereas for a 1 x 100, it all stops dead.

There's also the issue of starting currents being lower, so all in all, 3x 50% is often a better solution unless your duty is very fixed and doesn't mind stopping if it trips.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way
 
It's a reliability thing, or partial credit, so to speak, for maintaining some flow.
Simplifying a little, if each of your pumps have a failure figure of 10% of the time spent being repaired, then a system with 2 x 100% pumps (one standby) does not have a reliability of 100% because since there is still, theoretically at least, a 10% chance that each pump will be out of operation at any given minute, hence there is some small chance that both will be out of operation at the same time. If one is a standby unit, then it just might not turn on when you want it to.

To quantify that, each is assumed to be active 90% of any time period, we'll take a one day period in which I have to deliver 100,000 BBLS of oil. With one pump, theoretically, I will only be able to deliver 90,000 BBLS because of the 10% downtime it will have in any given time period. If both pumps can be assumed to be in a failed state 10% of the time independently of one another, the time that both will be in the failed state is 0.10 x 0.10 = 0.01, or 1% of the time, so 1.00 - 0.01 = 0.99, or 99% of the time we can assume at least one pump will be in operation and I will have the capability to deliver 99,000 BBLS instead of 90,000 BBLS.
2 x 100% pumps @ $100/BBL = $900,000 better than 1 x 100% pump per day of downtime

A three x 50% system, without rigorously doing the math, would get you something like
NOT the probability of all three being in the failed state at the same time = 0.1 * 0.1 * 0.1 = 0.001 or only 0.1% of the time, meaning 100% - that, or about 99.9% of the time that at least 2 are in the operating condition. Therefore you could run at least 2 50% pumps 99.9% of the time
Expected delivery = 99.9% of 2 * 50% x 100,000 BBLS = 99,900 BBLS * $100/BBL = $9,990,000 revenue.
3 x 50% pumps appear to be around $90,000 better/day of downtime than 2 x 100% pumps.

I'm a little rusty on that, but it's gotta be reasonably close.
The other advantage is you could get reasonable control of flow rate (say within +/- 10%) of any flow rate setting you might want within the higher half of typical operating ranges... without using a VFD.



I hate Windowz 8!!!!
 
And just what is wrong with VFDs...? [poke]

Us sparkies have to make a living too! [lol]
 
Very Fiddly Devices [soapbox]

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.)
 
Nothing's wrong with VFDs, but if I don't need a white elephant, or a gray one for that matter, I don't buy one.

I hate Windowz 8!!!!
 
Yes. I think you have confirmed my thoughts.

Thanks!
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor