Pumps, specifically new pumps, or existing pumps in re-rated service, are normally considered in terms of a required flow rate, with an associated required discharge pressure.
Engineers, most often Process Engineers, calculate the differential static pressure between the pump suction centre and the required discharge point. They make allowances for friction and fitting losses, and subsequently arrive at a differential head requirement, to which they specify the required flow rate.
In more complex systems, Engineers might use hydraulic modelling software to simulate the increase in friction over static pressure, for various flow scenarios. Such increase in pressure may then readily be represented as a system curve, with flow rate normally on the x axis and pressure on the y axis.
Such a system curve would further normally have a xero flow origin, at the corresponding pressure representing the static differential. As the fluid’s viscosity, i.e internal resistance to flow, plays greatly into the resultant system curve, such should also be incorporated into the hydraulic model. Pipe diameter, friction co-efficient for the pipe material, i.e roughness factors, fitting losses, all contributing to the accuracy of the system curve.
Theoretically such a system curve will have a definitive limit in terms of flow. A point where one would find only an increase in pressure, with no further increase in flow. This would simply be the point where the friction losses are so high that effectively no further flow would be possible. Basically, exactly the same as if one where to try and force flow into a closed system.
By imposing the pump head/flow curve over the derived system curve, one can subsequently see how the specific pump is going to perform in such a system. Logically, the more precise the system curve represent the dynamic losses over flow, the more accurate the pump performance prediction.
As the fluid’s viscosity, i.e internal resistance to flow, plays greatly into the resultant system curve, such should also be incorporated into the hydraulic model. More importantly maybe, one can see why increasing the pump pressure, proportional to the square of the speed ratio, either by increasing peripheral impeller speed through the use of a larger diameter impeller, or by means of frequency increment, often very little additional flow is achieved.
Furthermore, the intersection point will reveal how far left or right the pump is operating away from its best efficiency point, together with the actual NPSH requirement at the intersection point. Consider NPSH required generally increase with discharge flow rate, and that vibration increase with operation away from B.E.P. either left or right.
Often, I have found that the allowances for friction losses have been greatly over-estimated, resulting in far less system pressure than expected, resulting in the pump running out, while cavitating destructively, not to mention vibration levels of the charts!
Equally important is that pump failures are seldom due to a single event. Damage occurs accumulatively, over time. This becomes extremely important especially when the pump is controlled hydraulically, normally operating behind a control valve.
The shape of the pump curve will directly indicate the risk if pump surging in a manipulated pressure scenario, and could, again accumulatively, lead to seal and bearing failure, while the seal will normally fail first.
Article submitted by: Gideon van Niekerk, The official Hebei Zidong Slurry Pump Agent for Sub Sahara Africa. https://www.linkedin.com/pulse/buying-selling-pump-gideon-van-niekerk/