Estimate NSHr for a pump

[Cheetos]

Hello, I inherited a Flowserve pump without any information about the pump.

FLowserve WM inline centrifugal pump
Size: 2x3x12
Head:230'
Capacity:50
RPM: 3505

I've asked Flowserve twice about the pump curve and NSHr, but didn't get any response. I'm trying to design a water tank to create enough head to prevent cavitation. Does anyone have experience with pump this size that can give me a ball park estimation on NSHr? Most likely I'll run this on a VFD, so it's going to be 1500-3600 rpm range. Thanks.

 

[LittleInch]

This is the right forum, but you can't double post so please delete your other post.

You don't give a unit for flow but I'm guessing 50 gpm? It's not a particularly high flow or high head pump. Single impellor?

NPSHR is a test issue, but assuming the pump is close to BEP, its unlikely to be higher than 15 ft or 5m.

Lower speed will reduce this a lot, but at 3600 I would aim for that figure.

I don't understand the bit about designing a water tank.... Can you elaborate.

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

 

[Artisi]

there is more involved than NPSHr,so assuming you can't locate a curve the next best thing is to give more detail.
water source (tank) flooded inlet - give water height above the inlet or if suction lift - give the water level below the inlet.
Pumping distance from source to delivery point, pipe sizes and elevation to delivery point.

Unless you have a long delivery line or a very high deliver point, 230 ft is probably over-kill making it a problem.



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.)

 

[Cheetos]

My guess is 50 GPM. It didn't say anything on the nameplate.
It is a single impeller.
I want to have a water tank on the inlet side to provide enough water column to make sure the pump inlet has about 10% margin over NSHR. Since I'm not a pump guy, I don't have a concept for NSHR for this pump size. I think no more than 15' or 5m is a good starting point for me. Thanks.

 

[Cheetos]

The tank will be above the inlet and I have room to adjust the distance between the tank to the pump inlet. I'll try to keep the distance short to minimize the pressure loss.

 

[LittleInch]

Remember NPSHR is absolute head.

So unless your water is at high temperature or your pipe long, if the water level is above pump inlet you should be OK.

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

 

[RVAmeche]

In all likely hood, if you have a couple feed of head on the inlet (via the tank or supply source) and its atmospheric pressure or above, your pump is fine. Barring needlessly complex piping/clogged strainers/etc.

 

[Cheetos]

Thanks guys for your help!

 

[1503-44]

There are somewhat loose relationships between NPSHR and suction specific speed.
NSS can be calculated from pump parameters. If you can do that number, I'll try to correlate that to NPSHR.

https://www.pumpsandsystems.com/basics-suction-specific-speed

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[georgeverghese]

Assuming cold water, try to keep max. suction lift at < 5-6 metres.( minimum normal water level to pump centre line elevation)
Foot valves usually leak, so you'll have to manually prime the suction line every time the pump is started.
Water usually accumulates biological growth that will bung up suction strainer, foot valve, so dope the water with hypochlorite as required.
Min flow recycle at fixed speed would be approx 50/3 = 16gpm. On a VFD, min flow would depend on the speed turndown.

 

[Artisi]

with the pump capable of 230ft, if you have low discharge head and short pipe run, the pump will run out on its curved and cavitate due to NPSHr exceeding NPSHa.


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.)

 

[Cheetos]

Thanks for all the replies. After reading up a little bit on NPSH calculation, I gave it a try. I assume the water level is 6' above the pump and I got around 27' of head. Just want to run it by you guys to see if I did this correctly.

 

[LittleInch]

Almost....

You need to consider the worst case scenario of the lowest level in the tank, not the highest. If 6 ft is the minimum then fine, but think about what happens with less water in.

Your Hfr is incorrect. What you have calculated is the equivalent LENGTH of pipe plus fittings, so 11 ft of pipe.

Hfr is the Head Loss of water flowing through 11 ft of 3" pipe at 50 (GPM?). There are many online calculators which will calculate this for you for water, but it's going to be less than 11 ft head loss for sure.

If this is relatively small loop, you seem to be using a ump with a very high head and will need to throttle this quite a lot to maintain flow at 50 GPM. I would strongly suggest a flow meter and proper control valve to avoid breaking your pump by having it operate off the end of its curve. All this energy loss will warm the water up, which you seem to wanting to cool down.

What exactly are you trying to do here?

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

 

[Cheetos]

Thanks LittleInch for reminding me about the worst case. Really appreciate it! I'll also make sure to have a level sensor in the tank to make sure I will always enough water in the tank. The online calculator shows about 0.09 ft of H2O. If I assume 15', I think I have plenty of margin.

I'm trying to test a motor and VFD in the pump application, so I want to throttle the valve to check how the motor+VFD respond based on different pump operating points. I'm planning to have a flowmeter and solenoid valve to regulate the flow.

Since I'm not familiar with the pump industry, I assume 230' and 50 GPM are the end points of the pump curve at 3600rpm? (0 GPM, 230') (50 GPM, 0')?

 

[georgeverghese]

What @artisi says is a note of caution. With VFD, you may be able to keep NPSHr at the operating point reasonable, assuming VFD will be operating on flow control ? Without VFD, if developed backpressure at the pump exit is low, NPSHr will runout to end of the flow curve, and may be even 12-15ft ( ie up to 5m).
Do you need 230ft of head ? What if minimum required pump head is much less ? What is the minimum pressure at the end of this delivery line ? Even if you do have VFD, what will be the operating flow of this pump at the min or max required exit pressure? I suspect it may be much more than 50gpm.

 

[LittleInch]

You calculated 0.1 ft but assume 15???

Even 1 ft is v conservative....

A solenoid valve is on / off, not a control valve.

Pumps tend to be "rated" at close to BEP, best efficiency point. This iften about 70 to 75 % of max flow.

So at each speed you're just finding the pump curve? The manufacturer has already done this.



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

 

[Cheetos]

@LittleInch, I went back and did a hand calculation using Re and Darcy factor and I got about Hz = 2', so my NPSH becomes about 34', which is >> than 15' . Given what you said, I was mistaken about the 50 GPM and 230', so that (50GPM at 230' head) is the ideal operating point suggested by Flowserve.
Getting pump curve has been difficult. The manufacturer finally got back to me today and they are looking into it.

 

[shvet]

soild lines - efficience, dashed ones - NPSHR at BEP