Effect of dissolved gas on the NPSHa for water injection system

[INHC]

Hi,

The Produced Water (PW) from the oil trains is processed through a treatment system designed as follow:
1st stage: Hydrocyclones
2nd stage: Compact Flotation Units (using fuel gas as back-up to enhance separation)
3rd Stage: Degassing drum

The PW is then injected into the reservoir through a compact Water Injection (WI) pumps (3 stages) fed from the Degassing drum (operating pressure 0,5barg),

The hydraulic shows that the NPSHr for the WI pumps is fully satisfied (considering vapor pressure of pure water at PT)

Actually I'm evaluating the cavitation potential of the WI pumps, due to the dissolved fuel gas in the PW and its effect on the effective NPSHa,
I've been using:
a- Hysys: with Vapor pressure Model (Antoine) and compared with other EoS to estimate the vapor pressure of the produced water
b- C.C Chen application to estimate the effective vapor pressure at the degassing drum.

The results and conclusion:
a- Hysys: depending on the Fuel gas flow rate the vapor pressure is varying, and the NPSHr is not met at maximum fuel gas injection
b- C.C Chen application: the vapor pressure is almost equals to the operating pressure in the degassing drum, and the NPSHr is not met.

I read that some operators added a 4th stage treatment unit to remove the dissolved gas (MPPE: Macro Porous Polymer Extraction), and others inserted a booster pump!!!

I would appreciate your advice on the following:
- Based on operational experience, do you expect any cavitation potential when the pumps are fed with PW from the degassing vessel?
- Are you aware of how this issue have been dealt with wrt. NPSH, vapor pressure?

Thanks

 

[GPRnD]

I've not dealt with your specific application, but we would use a flavor of Chen's paper for calculating the effect of dissolved gas on NPSHa. I will say I think it tends to the conservative side.

The other aspect with using Chen is what f value e.g. f = 0.025 or f =0.03 (both typical values), you choose to use. This in turn depends on the pump and for example whether it has an inducer (which would allow much higher f values to be handled).

I've posted a link for a Pump Symposium paper that covers this topic in case you had not seen it.

http://turbolab.tamu.edu/proc/pumpproc/P15/P1591-98.pdf

The other comment I'd have is that presumably you are assuming that as long as NPSHa > NPSHr, things are ok ? This may or may not be the case depending on the peripheral eye velocity of you 1st stage impeller, the impeller material and how aggressive the water is. Hydraulic Institute has a standard 9.6.1 (2012) that may help in determining an appropriate margin. The webinar linked to below may also help in terms of covering cavitation and NPSH margin (however it doesn't cover dissolved gas).

http://www.gouldspumps.com/Tools-and-Resources/Webinars/Demystifying-NPSH-Webinar-Preview/

 

[Artisi]

Likewise, no direct experience in relating dissolved gas to NPSHa etc. however I would be concerned with the gas handling capability of the pump (although percentage of gas has not be revealed) an excessive amount could put the pump "off prime" or result in erratic pump operation.
You need to consider the fact that dissolved gas will increase in volume in the eye of the impeller (a low pressure area)

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[rotw]

No experience with the specific issue. I know that for boosting pumps materials (e.g. SS) can be chosen to cope with what is called "permanent cavitation". That gives some extra margin on top of the nominal margin you consider NPSHa-NPSHr.

"If you want to acquire a knowledge or skill, read a book and practice the skill".