NPSHa Margin for Condensate Pump 5

[pootong]

Is there any recommendation on NPSH available margin over NPSH required for condensate pump in power plant? Some designer give no margin, what's going to happen when the plant is in operation?

 

[Montemayor]

pootong:

If you don't have a "comfortable" margin between the NPSHa and the NPSHr when you're trying to pump a saturated liquid, you're going to have pumping problems: the pump will loose prime and fail to pump. Since you don't tell us, I don't know why your designer would give no margin except that it is foolish and dumb not to do so.

I've never had my pump designs fail due to NPSHa problems and I've pumped a lot of saturated liquids in 45 years - from the cryogenic liquids to steam condensate. My designs always stipulate:

NPSHa > NPSHr + 5 ft fluid
or,
NPSHa > NPSHr * 1.35

whichever is larger.

There's a big difference between gasification or loosing prime and "cavitation". "Cavitation" if frequently a misnomer applied to gasification and loss of prime.

 

[25362]

The reason may be that the selected pump is intended to operate in the break, aka on "submergence" control.

Such a pump must be carefully designed (stage pressures and rpm) and is supposed to cavitate at all flow rates. However, this cavitation is not considered damaging because the shock waves created by bubble collapsing don't have enough intensity to become harmful.

I recommend you read the subject of condensate pump regulation of Chapter 9.5: "Steam Power Plants" in the Pump Habdbook by Karassik et al., McGraw-Hill.

 

[rmw]

I was taught, (and successfully used for many saturated fluid applications over 35 years-I am not quite as senior as Art) to use the velocity limit of 6 fps max in the piping of saturated fluids.

On more than a few occasions, when phisical limitations prevented doing what Art suggested (for example; deaerator could not be raised to increase NPSHa) increasing the suction pipe size to get the velocity below 6 fps would give the desired result.

rmw

 

[pootong]

Thank you all for valuable recommendation, what I found is the designer give very small margin. Let say, condensate pump NPSHR is 1.00 meter and margin is 0.39 meter that means NPSHA is 1.39 meter only. I think they wish to aviod lowering condensate pump below grade to save excavation and underground structure cost. If the designer use this criteria NPSHa > NPSHr + 5 ft fluid, NPSHA will be 1 + 1.524 (5 ft) meter = 2.524 meter which is much different in construction cost. Is there any possibility of pump damage if the designer use the figure that satisfy only NPSHa > NPSHr * 1.35.

 

[rmw]

If you can tightly control your process to be very steady state at the condition that the NPSHr is 1 meter, you might get away with only .39 meter. If, however, there is an up-set, and a control valve downstream of this pump opens up and asks the pump to pump out farther on its curve (where the NPSHr curve is rising) then you may exceed your .39m margin, and flash (or cavitate as some would say) the pump.

Is there any way you can flow limit the pump so that it does not exceed its NPSHr margin?

rmw

 

[Scipio]

The thing to bear in mind with that margin is they're a guideline only, and vary from source to source, but one thing to bear in mind is when a pump NPSHR is determined, it's usually based on the point where the pump experiences a 3% drop in differential head due to cavitation. So when NPSHA = NPSHR, the pump has already been cavitating for awhile, the point of that margin is to keep the NPSHA well above that point. Similar to Montemayor, I always require 0.6 to 1.0 meter margin minimum between the two, more if the pump is going to be operated at flows above rated like rmw mentions. At 0.39 meters, there's a chance the pump could actually be operating with very slight cavitation already. You have to be careful with some designers, who don't really understand what NPSHR is and don't see why they can't set up a system where NPSHA = NPSHR.

 

[rmw]

When I learned all about this as a youngster, I was a design engineer for a process that handled 3 separate saturated fluids. The height, and therefore much of the cost of the product we made was determined by how much NPSHR plus margin our pumps required. Therefore we tested the pumps we used under lab conditions.

What we found out was that most manufacturers pump curves understated the NPSHr curve. That is to say if they said it needed 6 feet, it usually needed about 7.

We knew whose curves were accurate, and whose weren't. Unfortunately, the ones that weren't were from some of the biggest names in the business.

Some of the products we made went on military submarines and surface vessels. Finding out that the pump curve lied during sea trials was not cool.

Be suspicious of pump curves until you know. Part of the allowance that the responders to your thread are recommending is for that proviso also.

rmw

 

[pumpkind]

Sorry guys,
I am a bit late on this one - however maybe I can help a bit :
When due to the Projects constraints there is an unsatisfactory difference between NPSHA and NPSHR ( and this issue was extensively tackled by previous postings)there are few other things that might be considered :
1. Working to minimise NPSHR :
a. Consider fitting the pump with an Inducer.
b. Ask for pump special manufacturing attention - by controlling during the manufacturing process few aspects such as quality of liquid exposed surfaces ( both for casing and impeller ), underfilling of the impeller , cleaning eye of impeller, etc, the actual NPSHR figure can be usually improved. Any reputable pump manufacturer has some tricks in his sleeve for special applications.
2. Always ask for a NPSHR Test before taking delivery of the pump. This will protect you from underperforming pumps.