I am Electrical engineer looking for some hydraulic/mechanical advice on convenience of implementing pressure control by VFD in (3) 40 HP vertical turbine pumps.

 The brand new booster Pumps station has never being in service for 10 years as city customers complain about the high rate/pressure releases a lot of sediment from the century old pipeline system.Other issues arise as potential leaks / ruptures in the pipeline. The older pumping system is still operating at a lower pressure.

The main contractor did not install any water supply pressure control devices such as VFDs, Backpressure valve, Recirculating control valve or reducing impellers diameter.

The motors are controlled through PLC and standard soft-starters/contactors I need to justify technical and economically the high cost of 3 VFDs

My questions are:

1- Considering the affinity laws (flow vs RPM ) and ( flow vs HP) What kind of consideration/limitations exist by implementing pressure control based on VFDs?

2- How to find out the min and max. permissible variation of Pump speed (rpm) from the pump's efficiency curve rated at 1780 rpm. In my research I found out some issues might appear on speed control affecting the pump and/or the motor efficiency, energy cost etc.

From the curve I checked that 2 pumps need to run (in parallel) to meet the target Head and GPM.

///////////////////////////////////////////

The pumps actual performance curves are very important.  Each pump will always follow its curve including the adjustments for shaft speed.  The seemingly minor shaft speed variations with load with a simple induction motor drive actually can be significant enough to contribute to stability with pumps in parallel if the pump curves are less than ideal:

Your minimum RPM2 must still develop the head you need to move Q2 flow at that minimum rpm.   
H2 = H1/1780^2 * RPM2^2
Q2 = Q1/1780 * RPM2

Your maximum rpm must not overpressure the system and still deliver a flowrate at rates that your system can handle, ie must not cause waterhammer if a valve is closed, have a flow velocity too high, draw too much power, must not cavitate the pumps, etc.

If you have a static head, pay close attention to H2 at RPM2 and be sure you still get the flowrate you need.  The pump efficiency may not be the same as what you see on the pump curve.

A common situation would be where two (or more) seemingly identical pumps operate nicely together for a while and then some piping system transient gets one pump to hog load a bit.  Its shaft speed slows a bit, and its head then drops a bit then another pump that had shed some load speeds up a bit and hogs some load, and the party get going!

Now add multiple VFD's trying to impress some sort of control on this circus, and it is most unlikely that stability will improve.

Careful choice of pump curve characteristics is important to having multiple pumps play nicely together.  The old advice to have curves where the head rises continuously to shut-off is valid but can be asking a bit much.  It is most important to make sure that all of the pump curves have suitable characteristics throughout the entire range of operating conditions that can reasonably be expected--including start-up and shut-down.

Pumps operating in parallel with well chosen curves and driven by simple induction motors can be a model of inherent stability.  Add VFD's to these same pumps and motors, and stability can quickly become a very precious commodity.

MORE NEWS