13 replies to this topic

[herl_m]

All,
Recently, I've encountered quite infamiliar case. While calculating pump size, the pump suction pressure is slightly below atmospheric. But, NPSHA is positve. The source of suction is a tank operating at atmospheric. My senior does not agree with the negative suction (if converted to barg) and insisted on increasing the elevation of the tank. But, is it true that we cannot have suction below atmospheric? What does it mean? NPSHA is positive, isn't it?

Please help, I'm confused.

[demank]

i think there is misunderstanding between you and your superior.
"My senior does not agree with the negative suction (if converted to barg) and insisted on increasing the elevation of the tank."
Increasing the elevation will give a higher NPSHa result. Maybe your senior not agree of "a negative value of NPSHa", not pressure suction.
Then my advice tell again your senior and be clear between suction pressure and NPSH.

NB: NPSHa must have a positive value (at least 1 m above NPSHr). Below 1 m not recommended.

Edited by demank, 01 September 2009 - 12:09 AM.

[herl_m]

Thanks demank,
but he is fully aware that NPSHA is positive. But there is mixed arguments on the below atmospheric suction pressure.. can the pump still deliver for vacuum pressure at pump suction?

[Zauberberg]

The answer is - yes, and you can see similar cases in pumps serving vacuum distillation tower, or in so called "sump pumps" e.g. having intake from river or basins. The only specific thing with those pumps is that they have to be primed prior to startup. To convince your superior, draw a sketch of the pump taking suction from the basin 5m below the ground level, and pumping water at 20C. You'll see there is plenty of NPSH available for the pump (~4.5m), except that the vertical suction line has to be filled with water prior to pump startup. You can see similar types of pumps in old houses and gardens, I remember my grandfather had one like those.

The concept of NPSH (that is always expressed in units of length/height) is that we should not be interested in pressure but rather in meters or feet of head. That makes things 100% clear.
Best regards,

[Nirav]

Eventhough NPSH term is confusing for many, it's very simple if you think it from basic concept. I would not go in details about how to calculate NPSH-A over here. But I will try to provide input about certain other imp. aspects.

(1) NPSH-R = NPSH required by pump selected which is a characteristic of any specific pump under consideration.
(2) NPSH-A = NPSH available for pump which is a PURE system characteristic rather than a pump.
(3) Always consider reference elevation as "same" in both above terminology. i.e. if NPSH-R@grade level, NPSH-A should also be considered at grade level. Many time, this is one of the confusing points. Because if you do not consider both values at same reference elevation, then your conclusion could be 'wrong'. This is very very important for 'boiling liquid' in particular.
(4) Minimum value of NPSHA-NPSHR should be 1 meter. Sometimes, even 0.6 meter is also acceptable but requires tests by pump manufacturer. So, if possible, kindly stick to value of 1.0 meter as a difference.

Now, in your case, i did not understand what you mean by "calculating pump size".
What I would say is
( A ) Get value of NPSH-R from a rotary engineer.
( B ) Calculate NPSH-A based on your system characteristics.
If difference is not sufficient (i.e. 1 meter), you MUST increase height of suction vessel by the value such that it will give you desired NPSHA.

Edited by Nirav, 01 September 2009 - 12:01 AM.

[shan]

I have no problem with your (1) and (2). However, your (3) has conflict with (1) and (2). NPSH-R is a character of a specific pump. It doesn’t mater where you install the pump. In another word, if the pump NPSH-R is 5 meter, it is 5 meter on the peak of mountain, 5 meter on the bottom of sea, and 5 meter @ grade level. NPSA-A is pressure available @ pump impeller eye (has nothing to do with grade level either). 1 meter of NPSHA-NPSHR is a safety factor that is accepted by many companies). Therefore, 0.6 meter, 0.5 meter, or 0.1 meter makes not difference for the pump operation.

Back to the original question, please clarify suction pressure definition (measured at which point and at reference of what) for discussion

[herrani]

Hi all

I have encountered this situation several times, where I have been involved in the design of tank farms, or similar systems.

Let's say that the following conditions apply:

1) The tank feeding the pumps operates at atmospheric level
2) The minimum operating level in the tank is 2 m
3) The friction losses in the suction line are 5 m
4) The NPSH available is 4 m higher than the required

In this situation, there will be no problem with the NPSH, but still the pressure at the suction nozzle of the pump (2 - 5 = -3 m of liquid) will be below atmospheric. Zauerberg says that this is not a problem, and that many pumps operate under this conditions. I still would appreciate your comments, with respect to the following questions:

a ) Do we need to reconsider the piping class of the piping close to the pump suction nozzle, to make sure it can deal with such vacuum pressures? This could potentially increase costs
b ) Do we need to specify something in particular to the pump vendor in those situations, or is this a normal situation?

I am looking forward to hearing your comments. I have attached a small sketch, in case my description is not clear.

Regards

Ferran

Attached Files

•  Sketch.xls   13.5KB   61 downloads

[shan]

If your liquid vapor pressure is higher that -3 m, your pump impeller will experience gas lock or cavitations caused by gas bulb collapse. I have never seen any pump NPSH-R is negative value. This is why vacuum vessel liquid level is raised high to create positive NPSH-A. You should try as much as you can to eliminate vacuum pocket in the suction line to avoid air intruding.

[herrani]

If your liquid vapor pressure is higher that -3 m, your pump impeller will experience gas lock or cavitations caused by gas bulb collapse. I have never seen any pump NPSH-R is negative value. This is why vacuum vessel liquid level is raised high to create positive NPSH-A. You should try as much as you can to eliminate vacuum pocket in the suction line to avoid air intruding.

Hi Shan

I am not sure if I am missing the point here, but I don't agree with your NPSH calculation.

Just for the sake of the argument, let's assume that the liquid is water. I have attached the NPSH calculations as well as the pressure drop calculations. Since you have to take the vapour pressure of the fluid into account, the NPSH available results in 7 m of water, while the suction pressure is below atmospheric.

If you agree with my calculations, could you kindly give some recommendations to my questions?

Regards

Ferran

Attached Files

•  Sketch + calcs.xls   15KB   107 downloads

Edited by herrani, 02 September 2009 - 03:55 AM.

[fallah]

a ) Do we need to reconsider the piping class of the piping close to the pump suction nozzle, to make sure it can deal with such vacuum pressures? This could potentially increase costs
b ) Do we need to specify something in particular to the pump vendor in those situations, or is this a normal situation?

a )Normally class of suction piping would be considered the same as discharge.Therefore could cover relative vacuum cases.

b )I think this isn't a pump issue and no need to specify anything for vendor in this regard.Reported suction pressure to vendor would be enough.

Edited by fallah, 02 September 2009 - 06:47 AM.

[shan]

If your liquid vapor pressure is higher that -3 m, your pump impeller will experience gas lock or cavitations caused by gas bulb collapse. I have never seen any pump NPSH-R is negative value. This is why vacuum vessel liquid level is raised high to create positive NPSH-A. You should try as much as you can to eliminate vacuum pocket in the suction line to avoid air intruding.

Hi Shan

I am not sure if I am missing the point here, but I don't agree with your NPSH calculation.

Just for the sake of the argument, let's assume that the liquid is water. I have attached the NPSH calculations as well as the pressure drop calculations. Since you have to take the vapour pressure of the fluid into account, the NPSH available results in 7 m of water, while the suction pressure is below atmospheric.

If you agree with my calculations, could you kindly give some recommendations to my questions?

Regards

Ferran

Hi Ferran,

I have no problem with your calculation. My points are listed as the following:
1) NPSH is always positive (it is abbreviated of Net Positive Suction Head to begin with) and always use impeller eye as reference for both NPSHR and NPSHA.
2) There is possibility that pressure value is lower the atmospheric pressure (negative pisg). at some points in the pump suction line. In matter of fact, the whole suction line may be negative pisg for vacuum tower discharge pump or vacuum condenser. You should try to eliminate vacuum pocket as much as you can to avoid air intruding in your example system.
3) Pumps will work perfectly if NPSHA –NPSHR = 0 m. NPSHA-NPSHR = 1 m is added 1 m safety factor to count for possible inaccurate calculations and unpredictable process variations.

Shan

[herrani]

Hi all

Thank you very much for your replies quick replies, and sorry herl_m for "hijacking" your post...

I hope the discussion was relevant.

Regards

Ferran

[herl_m]

Ferran,
I don't mind your "hijacking" at all. In fact, I shud be thanking u for your questions since ur case is similar to mine. I've been busy these last few days and I didn't have time to continue with the discussion.

Yes, thanks for all for joining the discussion.

However, to Shan, what do you mean by vacuum pocket? This is the first time I heard such term.

[shan]

Ferran,
I don't mind your "hijacking" at all. In fact, I shud be thanking u for your questions since ur case is similar to mine. I've been busy these last few days and I didn't have time to continue with the discussion.

Yes, thanks for all for joining the discussion.

However, to Shan, what do you mean by vacuum pocket? This is the first time I heard such term.

Vacuum pocket is a piping system point or section where pressure value is less than atmospheric pressure. It usually happens at top run of piping (when piping with elevation changes) or end of line (pump nozzle). Air may be sucked in from the flanges or other piping elements and vapor may be flashed out due to vacuum situation. This is why your senior does not like negative suction pressure. It is a vacuum pocket.