Google NPSH

Jeff
Pipe Stress Analysis Engineer
www.xceed-eng.com

How can I calculate the NPSHa? (NPSHa=Hs+Ha-Hv-Hf)
By referring to the image attached, which Hs should I use? 0?

If a self priming pump is used, is it able to suck the liquid to the inlet of the pump?
Or it is depend on the functionality of the self priming pump, in which I need to refer to its performance curve?
Thank you.

Yes it will be able to suck if its a self priming pump (in principle). You need to make sure that at your flow rate pressure in your high legs is not LESS than vapor pressure of fluid or it will cavitate, loose prime etc.

For your case you have no elevation head and no pressure head at the inlet to your system. Pressure at pump suction will be sub-atmospheric (negative psig, or less than 14.7psia). This is due to line losses in the suction that you need to compute. You need to make sure that the absolute pressure at pump suction is greater than the vapor presssure of the pumped fluid. That difference is your NPSHA. Then check it against your NPSHR from pump curve. NPSHA must be greater than NPSHR

Jeff
Pipe Stress Analysis Engineer
www.xceed-eng.com

Well noted. Thank you for your response and answer.

Is that 1 bara or 1 barg on the inlet?

To fill it you need to pull the fluid over the hump. If you get below vapour pressure it will loose prime.

If you can fill the pipe some how and then keep it full you don't need a self priming pump.

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

Assuming the pump is at sea-level and there are no friction losses, the maximum theoretical suction lift is 10.3 m. At that point, the weight of water in the suction hose equals the force exerted by the atmospheric pressure.

Sorry, my question should be 0.1 barg at the inlet of the pipe.

That would help but isn't enough to fill the pipe if it was empty unless you have a liquid with a density less than 500kg/m3

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

If I fill some process fluid in the self priming pump in the beginning, can it works?(as self priming pump can be functioned when there is liquid and air)

As mentioned by bimr, the maximum theoretical suction lift is 10.3 m, so the maximum magnitude of negative suction pressure (including static head and pressure drop) can a self priming pump function is -10.3m?

Yes, that's what a self priming pump does - can also work with just air in the line

In THEORY, yes, in practice no. The pump needs a certain amount of absolute head ( i.e. below 0 barg) to function. This is often in the range 2-4m. Hence in practice anything over about 5m becomes problematic for low temperature water

Also 10.3m is the maximum at sea level with water about 4 C. Temperature gradually increases the vapour pressure which means your maximum lift is reduced accordingly as the water boils at reduced air pressure.

As you go up from sea level the air pressure reduces and hence there is less force to push the fluid into your pump (pumps don't suck, the air pushes it).

Every liquid is different - what liquid are you talking about?
Is this a real problem or a question?

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

Thank you for your patience and kindly explanation.

From your explanation, I can know that variables: the distance from sea level (atmospheric or air pressure) and the properties of the fluid used must be considered.

The explanation 'pumps don't suck, the air pushes it' make me confuse. Is this means if the air pressure is not enough, self priming pump cannot function?

By referring the link attached below, I think the air pressure wouldn't be enough to push the water to the pump. Then how does the water travel to the pump (pushed by air or sucked by self priming pump)?

http://www.johnbrooks.ca/wp-content/uploads/2013/0...

This is a question. And it might be a problem in my future as the pipe might need to be installed in a certain height to make sure the leakage can be inspected easily (underground pipe is difficult to be inspected visually).

What is depth of the reservoir?

I thought it might confuse a bit but you need to think in absolute pressure. . Air pressure at sea level is 1 bara.So if your pressure at the pump inlet is say 0.5 bars then the air pressure can push the fluid up a 5m lift. At the top of a big mountain let's say the air pressure is 0.4 bara. Then the maximum lift is only 4m.

I fully understand that people understand "suck" because it's something we do to say drink out of a straw. But if you try that worth a closed vessel at vacuum condition you can't "suck" anymore.

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

Assuming the pump can pull the vacuum, consider the length of time it might take for the fluid to reach the pump at start up. Self-priming pumps can overheat if they run dry for too long. You may want to consider check valve at the inlet.

I used to count sand. Now I don't count at all.

You pipe arrangement is not really suitable for long term operation.

You will end up trapping air/gas in the pipe loop and the velocity will be to low to force the air/gas 6 meters to the pump on the downleg.

You will need some type air/gas release valve.

Suggest you revise the pipe diagram.

The proverb is "centrifugal pumps don't suck". Some pumps can be self priming with filling the suction line first. However, once you get to the vapour pressure of the liquid you try to suck - it cant lft anymore. With water at around 15ºC this would just be a around 1700 Pa. So you couldnt lift water more than say 8m oh and as Littleinch mentions - this is only valid at sea level. It you abs. pressure is significantly lower you will have to adjust for this.

0.1 bar is the minimum pressure at the inlet of the pipe (from the pipe header).

Dear LittleInch and MortenA,
https://www.youtube.com/watch?v=enFgN5fOmOA
By referring to the link attached, it seem like the pump is sucking the water up. Am I wrong? Is the air pushing the water to the pump?

Dear SandCounter,
What if I fill some of the process fluid in the self-priming pump in the beginning?

Dear bimr,
May I know, what is the meaning of 'the velocity will be to low to force the air/gas 6 meters to the pump on the downleg'? If the self-priming pump can function, and the operation is continuously, would air trap in the pipe loop also?

Like I said, we humans refer to it as sucking because that's how we understand it from a young age.

however engineers and physicist know that it is actually air pressure pushing on the liquid lifting it up because something ( in this case the pump) is creating a lower pressure location at a higher elevation than the liquid level open to atmosphere.

If the air pressure over the liquid inside say a closed vessel was actually 0.5 barg, 1.5bara, this would allow water to flow out of the top of a pipe 5m higher than liquid level (just). At that point you wouldn't say the air was "sucking" it out of the vessel, you would describe it as being pushed out by the pressure above the air.

Now subtract 0.5 bar from both sides - pressure above the liquid 0 barg, 1.0 bara, pressure at top of the tube -0.5 barg, 0.5 bara - now it suddenly become "sucking"??? No. It's the same process.

All self priming pumps need to be partly filled with liquid to work.

To clear a line of air which has vertical down flow, you need a liquid velocity of >1 m/sec. A bit less might do it for different fluids, but 1m/sec is just about guaranteed. Anything less and the air stays trapped in the highest point.

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

Thank you for your explanation.
My understanding is, self priming pump will creates low pressure inside the pump. As the pressure of the inlet is higher, the fluid tends to flow towards the pump.

I apologize for troubling you again, LittleInch.
From your statement, "To clear a line of air which has vertical down flow, you need a liquid velocity of >1 m/sec. A bit less might do it for different fluids, but 1m/sec is just about guaranteed. Anything less and the air stays trapped in the highest point.", is this means when there is vertical down flow, there will be air traps inside the vertical pipe especially at the highest point if the velocity is not sufficient enough?

Yes

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