MCSF calculation? 2

[jet1749]

Anyone know of a method of calculating the Minimum Continuous Stable Flow of an end suction centrifugal pump? A rule-of-thumb or similar would be of help. Not able to find out as the manufacturer Rhodes-Brydon-Yuatt is long since defunct, and the pump manufactured in the 1960's

 

[TD2K]

I have a method (graphical), would need the BEP head, flow and efficiency.

 

[jet1749]

Thanks TD2K, not at my desk until Monday. I have a curve but it may not show the efficiency.

 

[25362]

From old notes: ROT for minimum continuous flows:

1. Minimum continuous stable flow for pumps with discharge 1.5" and larger: 30% of the BEP for the rated diameter. Smaller discharges need only 10% of the BEP for the rated diameter.

2. Minimum continuous thermal flow:

Q = 5.09(HP)/[([Δ]t)(SGxc)]

Where:

Q = gpm
HP = horsepower at or near shutoff.
[Δ]t = allowable temperature rise based on service, ^oF.
SG = specific gravity of fluid.
c = specific heat of fluid (i.e., Water=1.0, HC=0.3-0.6).

 

[jet1749]

Pump details are BEP 250 IGPM (or 300GPM US) @60ft head and 50% efficiency. 1440rpm
thanks
John

 

[25362]

According to Sulzer, for pumps (small) as this one, the minimum continuous thermal flow criterion is considered sufficient.

The estimating formula (Sulzer) for the thermal minimum flow in metric units:

Q_min = (P.3600)/([ρ].c.[Δ]t)​

where

Q_min = m³/h
P = power, kW
[ρ] = density of the fluid, kg/m³
c = specific heat of the fluid, (kJ/kgK)
[Δ]t = permissible heat up, ^oC

 

[vanstoja]

The use of "rule of thumb" estimates for minimum continuous stable operating flows for centrifugal pumps with end suction or side suction inflows is potentially dangerous and needs more substantive analysis for any specific pump design. The 30% and 10% of bep flowrates cited by 25362 are generally too low from my experience and the pump stability papers and articles reported in the open literature. Many cases have been reported of hydraulic instabilities around 60% of bep flowrate and I have seen numbers as high as 80-85% bep flow. Usually low flow instability is triggered by suction recirculation in the impeller channels and may be associated with a particular range of blade inlet incidence angles that engender flow separation from the blades and perhaps cause rotating stall. A secondary instability range may also exist at a range of still lower flows probably associated with discharge recirculation. Low flowrates outside of these incidence angle-related unstable flow ranges may be entirely stable. There is a classic paper by W. Frazer that describes the recirculation instability phenomena and several papers or articles have appeared that use Frazer's data to predict the onset of suction recirculation. A general discussion of "Centrifugal Pump Hydraulic Stability" was published as EPRI CS-1445 in 1980 coauthored by E.Makay, a well-known diagnostician of centrifugal pump operating problems.

 

[jet1749]

Thanks all for the advice, its appreciated.

 

[vanstoja]

Here are some more definitive guidelines and reported hydraulically stable operating ranges for centrifugal pumps indicating that Suction Specific Speed(SSS), impeller blade inlet tip velocity(U_1t) or inlet flow area and fluid specific gravity(SG) are the governing parameters.
Budris,A.R.,1989, "Sorting Out Flow Recirculation Problems", Machine Design, 8/16/89, pp.113-116
Suction Recirc.Factor,SRF=SSSxU_1txSG
LwrLim UprLim MaxSSS MaxU_1txSG
Radial Suction Impeller 550,000 710,000 14,000 80
End-suction Impeller 830,000 950,000 16,000 100
End-suction Inducer 1,400,000 Same --- 120

O'Keefe,W.,1988,"Can State-of-the-Art Research and New Experience Save Your Pumps?", Machine Design, Dec.,pp.39-44
Gam=Hub diameter/Eye diameter MFF=Minimum Flow Fraction
SSS=Suction Specific Speed(US units)

Gam=0 Gam=0.5
SSS MFF MFF
8500 0.35 0.45
16000 0.70 0.80
17500 --- 1.0
19000 1.0 ---

Lobanoff,V.S.&Ross,R.R.,1985,"Centrifugal Pumps-Design and Application, Gulf Publishing Co.
Fig.9-7 Stable Operating Window vs Suction Specific Speed
for 4-Inch(??) pump tested with 8 impellers having same blade profile but different impeller eye geometries. Results may differ for other designs but trend should be similar (following are eyeball estimates for 0.2 FF grid).
SSS Min.FF MaxFF
7000 0.50 1.25
8000 0.57 1.22
9000 0.60 1.19
10000 0.64 1.16
11000 0.68 1.14
13000 0.78 1.11
20000 0.95 1.06

 

[vanstoja]

Tests to determine low flow stability limits of a representative refinery pump were reported by Atkins,RA,Chung,EL&Taylor,HF,1966,"New Monitoring System Warns of Cavitation and Low-Flow Instabilities", Pumps and Systems Magazine, April, pp.12-15. They tested a 6"x8"x11" double suction, overhung process pump for light hydrocarbon duty in the manufacturers test facility using fiber optic Fabry-Perot interferometer(ffpi) pressure sensors good for fluid temperatures in excess of 700F. Test temperature and test fluid are not identified. Given data is N=3600RPM, WHP=125, QD=1688GPM, SSS=11,800(US), NPSH @1000GPM=9.2ft. Perfornamce tests were run at 300,600,900,1200,1600,1800,2000GPM. NPSH suppression tests were run at 300,1000,1600,1800GPM.
Suction recirculation onset was found at 0.62QD with pressure pulsations increasing from 10psi p-p at .62-.90QD to 60psi p-p maximum at 0.40QD. Discharge recirculation onset occurred at 0.51QD rising from 20psi at .51-.80QD to 88psi maximum at 0.3QD. Scaling the preesure-time plots at 1000GPM(0.592QD) and 300GPM(0.178QD) I found p-p pressure spikes of 40-45.5psi,and 80-112psi,respectively. The frequencies of spikes were scaled to be 235-244 Hz which is close to 4X suggesting that they are impeller blade passing peaks from a 4-blade impeller. At 300 GPM, there was also a pressure surge cycle scaled to be 5.37Hz (0.0895X). Unreported parameters derived approximately, assuming water tests at SG=1, 3600RPM and pump designator to represent pipe discharge, pipe inlet and impeller discharge diameters are HD=211ft(91.34psi), SS=1888(US), NPSHD=18.1ft.,inlet pipe velocity=11.03fps, discharge pipe velocity=19.6fps, impeller inlet tip speed=125.66fps, impeller discharge peripheral speed=172.79fps.
The SSSxU1t at SG=1 is 1,482,788 and the 125fps U1t value suggest that an inducer is needed if the pump has end suction based on Budris' criteria cited before. The 0.62QD minimum flow for suction recirculation is somewhat better than the previously cited Lobanoff & Ross Qmin stability limit of roughly 0.73QD for SSS=11,800.

 

[vanstoja]

Since suction recirculation controlled minimum flow fraction is related to the cavatation parameter, Suction Specific Speed(SSS), a useful guide for allowable minimum flow would be relative cavitation erosion damage to the impeller and its effect on lifetime reduction. Curves of SSS vs suction recirculation flow fraction and relative life factor are given in Doolin,J.H.,1986,"Judge Relative Cavitation Peril with Aid of These Eight Factors",Power, Oct.1986,pp.77-80. Four straight line plots are provided for multistage(MS), double-suction(DS), single-suction(SS) and inducer(IN) pumps with endpoints at flow fractions of 0.5 and 1.0 for suction recirculation onset having Relative Life Factors(RLF) of 1.0 and 0.5, respectively. Following are the scaled endpoint SSS values plotted:
Pump Type 0.5 FF 1.0 FF
MS 5536 11542
DS 6580 13420
SS 7623 15229
IN 10823 20841
RLF 1.0 0.5
The 7 other factors or erosion parameters considered besides SSS in impeller erosion pump lifetime estimates are available NPSH, fluid thermodynamics, fluid corrosion, impeller material, speed, flow fraction,and operating hours per year(8000=1.0). An example for a large boiler feed pump results in a lifetime factor of 0.504 (with a factor of 1.0 being approximately 10 years)based on the product of the eight factors.