Drooping head pump head curve 3

[upendrapatel]

We have testd three high head multi-stage pump in LNG / LPG and obseved that all three pumps (Sp. speed 800, 1000, & 1250) have drooping head curve. Is there any solution or method to change this performance to continously rising curve to shutoff head by reworking on existing impller or diffuser. All pumps have integral vaned diffuser and return vane.

 

[25362]

Have a look at Sam Yedidiah's Centrifugal Pumps User's Guidebook Problems and Solutions (Chapman and Hall).

It reviews the subject and presents some explanations in its chapter on Performance Factors (Item: Recirculation at the impeller outlet).

 

[vanstoja]

Low flow drooping of head-flow curves in centrifugal pumps with vaned diffusers is often caused by flow stall in the diffuser channels that initiates shroud-side reverse flow through the impeller that continues into the suction piping near the pipewall where it is called prerotation because it induces the mainflow near the center of the pipe to rotate in the direction of impeller rotation. This coswirl of the inlet core flow unloads the impeller and reduces its developed head resulting in the head droop. One solution to droop, not always practical, is to eliminate the diffuser vanes and rely on shroud-guided vaneless diffusion or unguided dump diffusion. Depending on how well the vaned diffuser is designed, eliminating the vanes can have serious to minimal effects on performance. My past experience has been with poorly designed diffusers which, when totally eliminated had near-zero effect on pump efficiency but provided a robustly rising head characteristic to shutoff flow. The slope of the head-flow curve changed slightly with little effect on full-range performance. As an alternative to totally scrapping vaned diffusers, you might try to increase diffuser vane inlet diameter to increase the blade-vane cutwater clearance (Gap B in Mackay's nomenclature) thereby reducing the interaction pressure pulses in the gap and the tendency to stalling and flow reversal in the diffuser channels.

 

[svramani5]

I wnat to know feedback on the following queries.
a)Whether for large Circulating water pumps (like in power plant application) variable angle impellers are used with single speed drive so that the pump may be operated for different flow/head due to variation in plant load(output)?
b)If yes, name of such installations?
b)If Yes, manufactureres of such pumps in the industry?
c)Will there be any complication in operation of such pumps?
d)Is it not better to design pumps variable speed drives instead of variable anagle impeller from cost and ease of operation?

 

[TZellers]

Some notes from research that helped reduce or eliminate droop:

Apply a 15 degree or greater bias on the b2 diameter, i.e. trim the impeller so that the front shroud diameter in greater than the rear shroud diameter (equivalent to 15 degrees as viewed from the side. Take into account the average diameter for head.

Impellers with discharge vane angles of 25 degrees or more droop. Also, too many vanes can promote droop.

Scallop the shrouds, but lose 3 to 5 pts of efficiency.

Some things that did NOT help:

add nose cone to impeller
small bias on impeller (4 degree)
increase cutwater
trim only the vanes

 

[vanstoja]

See additional comments of drooping head-flow curves in current thread407-108876 on droop instability.