Does Atmospheric Pressure Count Toward NPSH?

[BronYrAur]

I am trying to pull water out of an open tank, heat it and put it back into the tank. I'm pulling off near the bottom of the tank, going through basket strainers, then into a pump, out of the pump, through a heat exchanger, and finally back to the tank.

My question is with the amount of NPSHa I have. My water level in the tank is only about 2 feet. I thought that meant that my NPSHa was only 2 feet (actually less due to drop across the strainer).

My pump vendor is saying that I have the 2 feet, plus atmospheric pressure contributing to my NPSHA, so it's actually more like 36 feet. I disagree, especially since we are returning the water back to atmosphere. Don't those pressures cancel each other out?

 

[pmover]

BronYrAur,

get a copy of cameron hydraulic data.

npsh is the total suction head in feet of liquid (absolute at the pump centerline or impeller eye) less the absolute vapor pressure (in feet) of hte liquid being pumped.

pump vendor is correct.

good luck!
-pmover

 

[zdas04]

In terms of definitions of terms, you are right and he is wrong. In terms of pump performance, if the pump has a required NPSH of 10 ft, you are pretty sure to cavitate it. You need to look at the height of water, and then subtract any significant pressure drops (certainly the strainer, you need to look at pipe sizes and Reynolds Numbers to figure if friction is significant in this case).

Every NPSH-r I've ever seen published is some number added to atmospheric pressure (I'm not going to argue about whether head and pressure are the same thing or not). So if your pump has 10 ft of head required that is relative to local atmospheric pressure, not to zero psia.

David

 

[BronYrAur]

Thanks. Found a copy of "Hydraulic Handbook" by Fairbanks Morse which explains it very well. So now I just need to get data on the strainers to see pressure drop when they are loaded up.

 

[sailoday28]

There is NPSHR and NPSHA. The NPSH available is referenced to the suction flange and should bel=>than NPSHRequired
NPSHA based on inlet flange conditions
= (P-Pvapor)/rho +V^2/(2g)

or NPSHA based on upstream conditions with referece to flange =
(Patm-Pvapor)/rho + V^2/(2g)+ (change in elevation)-losses
Sinve for stationary inlet water level, V^2/2=0

NPSHA based on upstream conditions at suction flange=
(Patm-Pvapor)/rho + (change in elevation)-losses
Hopefully your NPSHA is = > than the NPSHRequired.

 

[TenPenny]

zdas04, are you sure about that? In my 20 years in this business, every single NPSH calculation is based on absolute, and the NPSHr numbers shown on pump curves are the same. It has to be that way; not every pump is on water, not every suction supply is vented.

 

[BigInch]

zdas, Here's where the difference in pressure and head comes back to haunt you. Head is never referenced as being either an absolute or a gage head. No need. Head is always absolute.



BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

Head is always absolute????
Please check my posted equations. For NPSH you will see there is a difference in pressures. As long as both pressures P and Pvapor are in the same reference gage or absolute, it doesn't matter since one is taking a difference.

Regards

 

[sailoday28]

TenPenny (Mechanical)Perhaps you are stating absolute because chances are when you get Pvapor, it's going to be in absolute units.
Regards

 

[BigInch]

More than chance. P vapor is always absolute. Least I've never seen one that wasn't.

Given what NIST has to say about vapor pressure and boiling points being equal at 1 atmosphere, if VP were ever referenced to 0 psig, we would have to define that as the "Standard Abnormal Boiling Pressure". Never happened.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[zdas04]

TenPenny,
I've looked at a lot of NPSH-r references on pump curves and they always say "ft" not "ft(a)" or some such. I've always taken that to mean that it used height of fluid column plus any applied gauge pressure. Looking at Gould's
excellent web page (

http://www.gouldspumps.com/cpf_0006.html)

I can see that my assumption for lo these many years has been wrong. Thanks for setting me straight.

David

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The harder I work, the luckier I seem

 

[sailoday28]

Biginch
I believe water at 100C boils at 0psig at sea level or 1 atm (abs) 0psig or approximately 14.7 psia.
With regards to head being absolute. Consider elevation, it is always relative, ie from the bottom of a tank, from tlhe center of the earth. The thing is we are dealing with DIFFERENCES in elevation. Similary, the same is with pressures, differences in psia or psig yield difference in psi.

Regards

 

[BigInch]

No argument. Head is in itself a measure of difference, as is pressure and is in fact any measurement, right? Thus a reference point, even if it may be highly localized, is always implied, however a reference point should always be chosen as the most convenient reference point possible. If you want to get technical, even mean sea level varies so much around the world that Mt Everest is not the mountain having the peak that is the farthest distance from the Earth's center, although it is the mountain with the highest elevation above (nearest) mean sea level.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

BigInch (Petroleum) 6 Apr 07 2:19

No argument. Head is in itself a measure of difference, as is pressure and is in fact any measurement, right?

Almost. There are some absolutes. IE, speed of light.
But now, we may be heading way off track from the original question.
Regards

 

[Artisi]

As you haven't been asked the question yet, I will -- is the tank above or below the pump centreline? You post is confusing - you could have 2 ft in a tank positioned below the pump centre or 2 ft in a tank above the pump centre.

 

[25362]

This subject has been discussed in numerous threads.
From the question by BronYrAur I gather it means the pump is below the water level by 2 feet.

When the water in the tank is heated its vapor pressure will go up and the available net positive suction head (=NPSHA) down, but still the atmospheric pressure would be added to it. When estimating NPSHA you better consider the highest temperature attainable on circulation.

As explained by sailoday28, the NPSHA referred to the centerline of the pump equals the pressure, in units of liquid height above, the vapor pressure:

NPSHA = (P[sub]s[/sub]-P[sub]v[/sub])/[ρ]g + u[sup]2[/sup]/2g​

P[sub]s[/sub] = Absolute suction pressure measured at the pump inlet; it includes the pressure over the liquid surface, any static head (positive or negative depending on the location of the pump and the tank) and any friction loss (as negative) in the suction piping
P[sub]v[/sub] = vapor pressure of liquid at the pumped temperature
u = velocity of liquid at the section where measurements are taken
g = acceleration of gravity
[ρ] = density of the liquid

u[sup]2[/sup]/2g is usually small and neglected.

Although a bit out of context, sailoday28, what do you mean by "absolute" when referring to the speed of light ? Non-variant ?

 

[sailoday28]

25362 (Chemical)
My experience in filling out data sheets/specs is to relate conditions to the inlet flange of the pump-Here the v^/2g, elevation and suction pressure/rho/g are known.

With regard to absolute, I was just trying to show that for incompressible flow, the engineering Bernoulli equation is made up of differences (except for the loss term) and therefore absolute and gage, elevation,kinetic energy etc differences are just differences.
With regard to speed of light, it is a fixed invarient number. When we talk of fluid velocity, it is a relative number with regard to a point on the earth.
Regards

 

[lilliput1]

The pump required NPSH is based on standard atmosphere as reference. Howeer to calculate available NPSH you must use the actual absolute ambient pressure acting on the water level at the pump suction. This is important because the available NPSH would be reduced if the pump is at high elevation or if the pump is drawing from a vacuum condenser.

 

[danielthb]

Have you ever seen a pressure indicator in a pump's suction line read below 0 psig and not hear "gravel" running thru the pump? Atmospheric pressure counts.

DB

 

[SeanB]

Atmospheric pressure definitely counts. In your case, your source pressure is 14.7 psia (depending on your elevation with respect to sea level).
Source Pressure: 14.7 psia
- Vapor Pressure: x.x psia
- Line Losses: x psi
- Misc Items: x psi
(suction strainer)
Subtotal: x psia
Subtotal: x feet (subtotal x 2.31/sg)
+ Static Head: 2 feet
NPSHa: x feet