Atmospheric pressure query in NPSHA calcs 1

[rs1600]

Appologies for another NPSH question but i just cant seem to find the confirmation i need on the following query.

pumping water at 30C from an open tank

static head = 0.99m
fluid density = 995.646kg/m3
friction losses = 1.472m
vapour pressure = 0.434m
atmospheric pressue = m


To get vapor pressure figure in metres i do the following:-

Pv = pressure x 0.703 / SG
Pv = 0.6159psi x 0.703 / 0.995646
Pv = 0.434m

To convert atmospheric pressure (assumed at 14.7 psig) into metres, do i use the fluid SG figure also or is there a constant regardless of the fluid SG? is the following correct?

Patm = 14.7psi x 0.703 / 0.995646
Patm = 10.379m

Assuming the above is correct i get:-

NPSHA = 10.379m + 0.99m - 0.434m - 1.472m = 9.463m

Finally, i understand velocity head is generally disregarded due to the minimal affect however an example with real figures would be very helpfull if possible?

if someone would be kind enough to assist then i would be greatfull

Thanks
R

 

[sheiko]

I think you are correct. Just some checks to perform:
- verify that you use the absolute pressure when converting to head
- use vapor pressure at maximum operating temperature
- use minimum liquid level in suction tank
- calculate pressure drop in suction line at max. design flowrate

"We don't believe things because they are true, things are true because we believe them."

 

[micalbrch]

You are correct. The atmospheric pressure has nothing to do with the fluid SG.

Velocity head can be disregarded when fluid velocity is low and pipe length is short. But what are the friction losses in your calculation? They are the losses caused by the fluid velocity. So, they are part of your actual calculation. I suppose your question was more related to acceleration losses. They are usually negligible as long as your pump is not an oscillating PD pump.

 

[rs1600]

Thanks guys, looks like im not too far off then, but still not quite there (been sat on the computer all weekend!)

Micalbrch,
you state 'atmospheric pressure has nothing to do with fluid SG' yet i have used the SG of water at 30C to get my answer?

this is what is confusing me i.e. if the water was at 10C or 60C, or if the fluid was sea water would i need to use that SG figure to deturmine the atmospheric pressure in metres acting on the fluid?

also do i need to adjust the 14.7psi figure to suit the pumping conditions i.e. if it was high up a mountain should i alter this figure to suit?

im not sure if these need to remain constant regardless of fluid / altitude i.e always use Patm = 14.7psi & SG = 1 (this does not seem right) or if the specific conditions outlined need to be allowed for?

hope that makes sence, any feedback to clarify would be appreciated.

regarding velocity head, i think i understand now that this is already included as part of the NPSHA calcultion, so i can ignore, cheers.

Thanks,
R

 

[BigInch]

'atmospheric pressure has nothing to do with fluid SG' yet i have used the SG of water at 30C to get my answer? Water, being relatively incompressible, does not compress significantly until you reach extremely high pressures, 5000 psi and above, therefore pressure effects can be ignored when pressures are low. Water density is much more affected by relatively smaller temperature changes, so if you are doing detailed calculations, you might want to look that up.

this is what is confusing me i.e. if the water was at 10C or 60C, or if the fluid was sea water would i need to use that SG figure to deturmine the atmospheric pressure in metres acting on the fluid? Yes, to be technically correct, you should always use the correct density of your fluid. If pumping sea water, the NPSHa would be less if it was calculated from a pressure, for example, a 20 psig tank pressure would add a head of 20 * 144 / 64 = 45 ft, but it would add 20 * 62.4 / 144 = 46.15 ft of head, if it was fresh water. Pumping from a high reservoir, say from 100 feet above the pump, would add the same 100 feet of head with either sea water or fresh water, but of course the pressure at the pump's inlet would be different, 100 * 64/144 = 44.44 psi for sea water and 100 * 62.4 / 144 = 43.33 psi with fresh water. Just remember that pumps really don't care much about pressure; they only want head and only give head (did I really say that ). Pressure is the result of pump heads and the fluid's density.

The change in vapor pressure with temperature could easily be significant and, if you look up anything, look that one up.

also do i need to adjust the 14.7 psi figure to suit the pumping conditions i.e. if it was high up a mountain should i alter this figure to suit? Yes, if you have a pump station at 3000 m elevation, atmospheric pressure will give you only about 1/2 the head it will give you at sea level. High elevations can have a very significant effect on many things, including engine fuel usage, reduced motor cooling and lower power outputs and low boiling temperatures. Pumping gasoline at high altitues near the equator, where temperatures can also get high, can be difficult to do with the full required NPSHr.

im not sure if these need to remain constant regardless of fluid / altitude i.e always use Patm = 14.7psi & SG = 1 (this does not seem right) or if the specific conditions outlined need to be allowed for? [Technically that is not correct. You should make adjustments for actual conditions when such precision is required by the accuracy you may need in your calculations. SG will of course remain effectively unchanged, but NPSHa to NPSHr comparison should be made using heads calculated form absolute pressures.[/color]

regarding velocity head, i think i understand now that this is already included as part of the NPSHA calcultion, so i can ignore, cheers. Velocity head is often ignored in NPSHa calculations, especially in the USA. Velocity head is a part of NPSHa and it can be added to reach a total NPSHa, if you wish to do so, but you will not have to do that to comply with the NPSHr of pumps manufacturered in the USA.

 

[micalbrch]

" Velocity head is often ignored in NPSHa calculations, especially in the USA." - Interesting point, BigInch. I haven't heard that yet (which doesn't mean anything). Is there a rule (standard) behind that or is it just common practice in the US? What if not velocity head is the 1.472 m friction loss in Diaz' calculation?

 

[ione]

To be precise velocity head should be taken into account when computing NPSHa. Nonetheless in many applications it is such a small amount it is ignored (think of a suction line with a velocity of 6 ft/s that leads only to 0.56 feet of velocity head). Moreover ignoring velocity head is always a conservative approach.

 

[rs1600]

Really appreciate your input guys, i think im there now!

just to make sure - Basically when converting for BOTH my VAPOUR PRESSURE & also my ATMOSPHERIC PRESSURE, from say psi to metres/hd using the follwing equation:-

Head = pressure x 0.703 / Specific Gravity


i now know the pressure figure used should be absolute i.e. to suit the actual altitude (plus additional tank pressure where used). And the Specific Gravity value used should suit the fluid being pumped i.e. water SG will be greater than say petrol SG etc.

i understand i dont nessisarily need to be this accurate for what i am doing but i just want to get the principles straight in my head, i can then consider the level of accuracy required.

If someone can confirm thats correct then im im happy!

off to check out velocity head calculations now, just for my info so any pointers would be great.

Thankyou
R.

 

[BigInch]

That's a switch. I agree and usually I am the one that says you can add velocity head in the calculation to arrive at NPSHa, if you want to. I think it was a thread about one year ago where I actually had to prove to an unbeliever that v^2/2/g was actually a part of available head and that it should be added, not subtracted, as it is here,

http://www.gouldspumps.com/cpf_0007.html

I believe it is mentioned in the Hydaulic Institute's standards, or at the least I can assure you that at the "practical practice" level it is seldom considered, if ever.

Perhaps part of the confusion is generated by the manner in which NPSHr is defined in testing, where a 3% loss of head indicates incepient cavitation. "Head" is measured by a pressure gage at suction inlet and converted to its equivalent head in feet and marked on the chart of the pump NPSHr curve, thus velocity winds up being NOT included in the published NPSH_R values. So, what do we do now when some intelligent engineer realizes that velocity head is a part of NPSHa and can be included. If he starts with a curve developed using the test data above, he might say, "Well, I really only need to supply this much pressure, because I can make up the difference when I add velocity head. The end result of course is that s/he winds up providing a total suction head = NPSHr as required by the chart, even though the chart is only indicating pressure head alone. Well... with that method it's still OK, as the test method provides the most conservative definition. If NPSHr included velocity head, an unconservative mistake could be made. What I was taught is that the US method is more safe, in what is generally a rather unsure understanding of all the processes of cavitation anyway, and where conservatism can often still result in cavitation, so just ignore velocity head everywhere.

I don't remember right now exactly where it is, but when (not if) I find it, I'll post it.

 

[BigInch]

Could it possibly be in API 610? If not, I bet it was in HI.

 

[ciise]

You might wanna use Specific weight when converting Atmospheric pressure to head, !

pressure in kPa= kN/m2

Sp. Weight= kN/M3

Head= Pressure(abs)/Sp. weight

= kN/m2/kN/m3=M!

 

[BigInch]

Converting velocity to head is much easier using metric units.

Don't forget Bernoulli. You have to use some force to increase velocity, so effectively pressure is reduced when it is converted into higher velocity and v/v.

 

[micalbrch]

Thanks BigInch. You answered my question.