VFD vs. Soft Starter (Pump Application)

Well really, that's it in a nutshell. A soft starter cannot change the speed, so if you need to, it will not work. You cannot really compare them in that realm.

VFDs are 5-10 times more expensive than soft starters once you get above about 25A, so you really cannot justify using a VFD in place of a soft starter unless you have some very specific issues:

1. A VFD can allow starting of an AC motor at 100% FLC or less, a soft starter cannot. So if you have a severely limited power source, sometimes a VFD is your only choice. A soft starter cannot start a loaded motor with less than 200% current, usually 300-450% depending on the type of load.
2. A VFD can allow a 3 phase motor to be run from a 1 phase source (with proper selection), a soft starter cannot.
   

Good explanation by JRaef - not only in this thread but in many previous threads as well. One recent thread about VFD and Soft starter I can remember is http://cr4.globalspec.com/thread/50941

In addition, I just would like to add little more:

The intention for using soft starter is primarily to decrease the starting current, not to save energy, not to control speed.

While the intention of using VFD is to decrease starting current, control speed and save energy. However, saving energy by VFD is sometimes misused and misunderstood, VFD can save energy only when it is used for controlling the flow with varying the speed of centrifugal pumps (for liquid) or fans (for air) and the flow varies in wide range and remains at lower end most of the time.

How VFD saves energy?

VFD saves energy based on the affinity law which indicates

Flow ∞ RPM

Head ∞ RPM²

Power ∞ RPM³

Let me explain with this example: In a system, a centrifugal pump delivers 1000 CFM at 1800 RPM and 500 CFM at 900 RMP. The flow of the system varies from 1000 CFM to 500 CFM and most of times the flow remains at 600 CFM.

In this case, if you use a DOL starter, the motor runs at 1800 RPM all the times and the theoretical power consumption of the motor is proportional to the cube of 1800 (as per the affinity law). If a VFD is used here, the theoretical power consumption is mostly proportional to the cube of 600. So, VFD saves power in this case.

Let's consider the same example but most of times the flow remains at 900 CFM (instead of 600 CFM).

In this case, the theoretical power consumption by the VFD is mostly proportional to the cube of 900 and the power saving is not significant as compared to the previous case.

(Note: I mentioned 'theoretical power consumption' because the calculation is not exactly this in practical – there are other things to consider as well)

- MS