NPSHA vs NPSHR

[cica58]

Its known that in a positive displacement pump suction line must be take in account the acceleration head (ha)and substract it for the NPSHA. But, what is the value of NPSHA for comparison with NPSHR.
NPSHA = Pa+Hs-Pv-hf-ha?
or NPSHA = Pa+Hs-Pv-hf?
Why?
Thanks

Ciro

 

[25362]

Does it really matter whether ha is deducted from NPSHA or added to NPSHR ? Anyway, Karassik et al., in their Pump Handbook, prefer to include ha in the definition of NPSHA.

 

[onuigbo]

By ha do you mean velocity head? I need the clarification in order to answer your question, the term acceleration head is not acceptable in pump lingo.

We deliver steam as if your life depends on it.

 

[Scipio]

Onuigbo,
Actually Acceleration Head is the correct term, though it's used strictly (in my experience, anyway) in reciprocating positive displacement pump design as a way of describing the start-and-stop nature of suction flow into the pumps, i.e. on a simplex plunger pump the suction fluid is only moving during 180° of the crank rotation. It's additive onto the basic NPSHA calculation, frequently the dominant loss in the suction piping system, and is what usually causes a suction stabilizer to be required.

 

[onuigbo]

Upon further reflection, I believe the term "acceleration head" in PD pumps is the "velocity head" equivalent in centrifugal pumps. The reason (at least in my opinion), why Karassik added velocity head term in calculating NPSHA is because in an existing installation, the gage does not register the velocity head component of the total suction head, the gage reading in this case only records the elevation from pump centerline + suction losses converted to feet of the liquid. Whereas, in new installations, the static head component of the NPSHA already includes the velocity head, hence adding the velocity head in calculating the NPSHA will result in adding the velocity head twice. In other words, by adding the velocity head, Karassik assumes that you're working with an existing system equipped with gages.

We deliver steam as if your life depends on it.

 

[Scipio]

Yes and no, I think you're right as far as why the velocity component exists in one of Karassik's equations, it's required for converting from a suction gauge reading - this is Bernoulli's Equation coming into play. Velocity head, however, is a finite portion of the net energy available in the pump suction region, hence Net Positive Suction Head.

Acceleration head, on the other hand, is actually another loss of energy from that which was available in the suction vessel, over and above the steady-state line losses calculated based on friction and fluid velocity. It's subtracted the same way you subtract line losses. The interesting thing is that it's additive, the head lost as you accelerate the liquid from a stop is recovered as the liquid decelerates, minus some frictional losses, so by installing a suction stabilizer of some sort, you can gain back almost all the energy 'lost', and nearly neutralize the effect entirely.

However, since NPSH calcs rely on the worst case condition, only the losses associated from accelerating the liquid is considered, and deducted from the steady state NPSHA prior to comparison with NPSHR.