Multi Pumps in Series w/ Single VFD

[DrexelCHE94]

The current scenario is this: 3 pumps installed in series with a bypass on each (w/ check valves). Single VFD to be shared among all 3 pumps so only 1 pump can be on VFD at one time, other two pumps would be across the line. If all pumps are online, does it matter which one is on a VFD thus reduced speed (say 80%)? What would that combined hydraulic pump curve look like with the system curve?

 

[1503-44]

Why would you do that???

You will get the sum of the heads of all pumps. Two will produce head at whatever their curves indicate for the flow going through them.
The VFD controlled pump I imagine will control that flow rate, when it is operating. The lowest flow capacity pump of the other two will probably determine the flow rate of the other two when both are active.

The VFD pump will output its head proportional to the speed squared and flow directly proportional to its speed.

I doubt it makes much difference which one, except ...Make sure the suction components, as well as all pumps can all take the high pressure.inputs and final discharge pressure. You can't stack 3 pumps designed for 1000 psig and expect the last two to survive at 2000 and 3000 psig operating pressures. That may make a difference which one goes first, second and third.

 

[DrexelCHE94]

To control pressure so it doesn't go above MOP, avoiding a control valve to save on energy, and to save money on VFD costs.

 

[electricpete]

You could build the composite curve by adding the dp vs flow curves vertically (with the vfd pump scaled for speed).

If pumps are identical it shouldn't matter which one is on vfd. Unless maybe there is a concern for cavitation on the first pump. If there were concern for cavitation on the first pump then if your vfd is below base speed maybe it makes more sense to put it on the first pump and if the vfd is above base speed then it makes more sense to put it in a 2nd or 3rd position. My logic is that there is not likely concern for cavitation on the 2nd or 3rd pumps because they have a higher pressure at their suction, but maybe there is a concern for cavitation at the first pump depending on system configuration (if not then I'm wasting everyone's time). I'm also assuming that for a given suction condition (pressure and flow rate) a lower speed pump is less likely to cavitate.... I'm not 100% sure that last assumption is right but that's my intuition.



=====================================
(2B)+(2B)' ?

 

[1503-44]

Probably best to use a control valve. 3 pumps in series is not a set up for wide variations in flow rates, which is where vfds usually shine. One vfd for 3 pumps in series suggests that you will have good pressure control between 0.67 to 1 times max final discharge pressure of basically one primary flow rate, but you may need 2 x the 33% pressure to move that same flow in a somewhat normal system curve type system. If you need basically one flow rate at 66% to 100% of final discharge pressure, then OK, maybe it makes sense. But a control valve could very easily do that. In fact, that's what they do best.

With only one pump on vfd, you won't gain much efficiency, which is what vfds do best, so why bother with special motors, all the controls and cables. Just install one valve. Easy solution..

 

[Compositepro]

This seems to be another case of misunderstanding of how control valves work to control flow from centrifugal pumps. A pump running full speed does not always consume full power and the rest is wasted by the control valve. A control valve will reduce flow and waste a small amount of power, but the reduced flow causes the pump to consume significantly less power.

Running pump in series is not to common. Each pump has a seal that will see progressively higher pressures. High pressure pumps will have impellers in series, but only one low-pressure seal.

 

[LittleInch]

The issue is that the other pumps will now be operating not at their BEP as when your VFD pump is at 80% speed, flow will be 80% of the rated flow

Add in the losses of your VFD and you may be using more energy than with a control valve.

Each system is different so do the design work and find out if it will work for you.

But it is quite an odd way to do it.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Valvecrazy]

As was said, "Add in the losses of your VFD and you may be using more energy than with a control valve". Look at your pump curve. Most likely there is very little difference in energy use when restricting with a valve as compared to a VFD. In many cases the parasitic losses of a VFD make it less efficient than a control valve.