Pump NPSH - water or liquid ? 4

[plumberet]

Hope that you can help me here.

In the pump data sheet ( for pumping gasoil ), stated the NPSH is 12 m. Is that 12 m of water or 12 m of gasoil ?

Background to the above question is my attempt to justify lowering the low alarm subject to the NPSH.

Thanks.

 

[katmar]

The short answer - it is 12m of the pumped fluid (gasoil).

For the long answer - see thread378-175054


Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[BigInch]

Shorter answer. "Yes"

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

 

[MortenA]

the message that is coming through is that NPSH = Net Pressure Suction Head is in m of liquid column and that the NPSHa on metres (or ft or whatever) is the same for whatever liquid you are dealing with.

NPSHr may wary due to different properties of the liquid=different dP and different vapour pressure.

Best regards

Morten

 

[BigInch]

Morton. A great and absolutely correct point. It does indeed say "net", which clearly includes vapor pressure. I'll give you a star, because I can't give myself a -star for not thinking it through and just trying to be cute.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

 

[katmar]

Morten, I think you have got your a and r mixed up. The NPSHr (where r = required) is a property of the pump, and is fixed in m (or ft) of the pumped liquid. This is what is shown on a standard pump curve. The NPSHa (where a = available) is also in m (or ft) of the pumped liquid but is affected by the vapor pressure, friction losses, vessel pressure etc.

Harvey

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[MortenA]

yes you are right - i meant NPSHa available NPSH.

Best regards

Morten

 

[javadabdollahi]

NPSHa is acc. to gasoil but NPSHR shall be claculated acc. water.( api 610)

 

[TenPenny]

Brings up my favorite passage (from Centrifugal Pumps Design and Application, Lobanoff and Ross):

"With so much literature available, one might assume that NPSH and its relationship to cavitation is well understood. Nothing could be further from the truth. To this day, NPSH is still misunderstood, misused, and misapplied, resulting in either costly over-design of new systems or unreliable operation of existing pump installations."

 

[d23]

All,

My two cents worth:

Maximum water density occurs at 4 degree C or 39.2 degree F. At 4 degrre C fresh water has a specific gravity of 1.00. In almost all cases NPSHr is plotted in terms of fresh water at a specific gravity of 1.00. You can convert the feet of fresh water to the specific gravity of liquid you have by dividing the catalog NPSHr by the specific gravity. As an example if the gas/liquid mixture has an average specific gravity of 0.85 and the NPSHr is 10 feet the adjusted NPSHr would be:

10 feet / 0.85 = 11.76 NPSHr Required

NPSHa as defined by the Hydraulics Institute is:

NPSHA or the “NPSH available” is Atmospheric pressure + static head + pressure head - the vapor pressure of the pumped product - loss in the piping, valves and fittings. Therefore, NPSHA becomes the amount of pressure available to prevent liquid vaporization or pump cavitation.

In this case the two very real keys to the definition are the vapor pressure and intake piping losses. Crude oil has a very high vapor pressure whereas gas typically has a very low vapor pressure. Long intake piping with a lot of fittings reduces the available intake pressure.

I would suggest a laboratory test for the vapor pressure, limit the amount of friction between the source and pump intake and adjust the NPSHa liquid level accordingly.

The simple thing would be to ask your pump manufacture.

D23

 

[katmar]

d23, I believe you are wrong. Please see the thread referenced in my April 3 post above. The NPSHr is expressed in feet because it is constant in feet irrespective of the fluid pumped.

If your pump curve specifies a requirement of 10 ft of NPSH and you are pumping water you could say that the NPSHr is either 10 ft of water or 4.335 psi. On the other hand, if you are pumping a hydrocarbon fluid with an SG of 0.85 you could say that the NPSHr is either 10 ft of fluid or 3.68 psi. This was generally agreed in the other thread.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[d23]

katmar

I apologize, but I am very short on time right now. I will do a little looking as time allows for support documentation to this post.

I know the published information from HI, Goulds, Flowserve etc. states that NPSHr is in feet or meters so no conversions will be required. I have doubts about that. In general terms NPSHr is the required head in feet or meters that is required to prevent the impeller eye pressures from going below vapor pressure. This value (NPSHr) must account for all intake losses. With liquid other than water the vapor pressure, viscosity etc all tend to change which will affect the NPSHr. I don’t believe there is a blank check for any liquid without knowing the liquid properties. In this specific case there is a gas-oil mixture to be considered. Not knowing the specific application I have to assume the oil is crude at “X” viscosity and the gas is some hydrocarbon base like methane. The gas will have a lower vapor point than water and the oil will have a higher viscosity than water. This means there will be higher intake losses and lower vapor pressures to overcome.

Typically with cases that two liquids are involved they are not mixed very well. In the case of oil and gas the oil is heavier, so it separates from the gas. The oil is on bottom and the gas is on top. During pump operation you will get some mixed flow and some slugs of one liquid. In this type operation I would suggest designing for the worst-case scenario, which would be viscous entrance losses and gas vapor pressures.

The statement that gas bubbles are not the same as “water” cavitation is reasonably accurate. The gas can still compress back into a liquid, which causes implosions. These implosions can still cause damage to the metal they come in contact with.

I will look for some support documentation as time allows however; I believe we will find each application requires review beyond the generic “NPSHr in feet or meters” is the same for all liquids.

D23

 

[BigInch]

There is variation between liquids in NPSHR. The problem is that there are thousands of liquid compounds that would need to be studied to arrive at the correct parameters and, since the variations in NPSHR due to factors other than vapor pressure differ very little from cold water, the study effort required would not be practical, nor likely be conservative. Its just not worth the study effort to obtain an additional 2 to 3 percent of NPSHR, so the reference was established for cold water, not at SG=1, but for cold water at 50ºF, if I remember correctly.

The generally accepted method is to use the NPSHR given by the pump Mfgr for cold water and, when specific problems are encountered for specific products in specific piping configurations for specific applications, ie. gasoline at 44º pumped from a storage tank, is to deal with those based on experience with equivalent systems. In the gasoline case, it is commonly believed NPSHR can be safely reduced, but it still should not be applied as an accross the board solution, due to the risk of increased cavitation damage. Any reduction in NPSHR not sanctioned by the pump mfgr is a very risky business, as many currently believe that the present method of establishing NPSHR based on cold water is not conservative enough for most systems anyway. Recent thought is that the actual beginning point for cavitation may be as high as 20% above that needed for the cold water standard.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

 

[sheiko]

See chapter 3 of ebook:

http://www.lightmypump.com/pump_book.htm

"We don't believe things because they are true, things are true because we believe them."
"Small people talk about others, average people talk about things, smart people talk about ideas and legends never talk."