I'm jumping into the fray on this a bit late, but here is my 2 cents worth.
Like Dadflap, I too sell VFDs. I cannot speak for the entire industry but there are some general misconceptions throughout the above posts that I feel need clearing. I have read many other posts from d23 and Pumpdesigner and respect your opinions, but it appears to me that you have passionate feelings about this issue, possibly created by inexperienced or unscrupulous vendors. This is most unfortunate but, I admit, all too common in our industry. Regrettably there is no equivalent of an engineering license, journeyman card or operating permit for the sales profession. If there were, there would be a lot less salesmen P-)
1) Most if not all payback calculation programs DO in fact take the normal energy savings of a throttled system into account. True, many vendors who use them do not understand this, but it is done in the algorithms nonetheless. The same holds true for initial cost analysis. In the instructions we are specifically charged with including all aspects of the installed cost, including labor and harmonic mitigation. Do some vendors selectively leave that out in order to make their case more spectacular? Absolutely. Do buyers have a responsibility to protect themselves from said vendors by reviewing the accuracy of the statements? I think this question predates VFDs by a few thousand years. The Romans used the term Caveat Emptor!.
2) While true that as speed goes down, efficiency goes down and heat loss (rate) goes up, this is relative to the fact that energy consumption went down drastically at the same time. To use the early example of 50% speed, energy consumed at that speed went down to 12.5% of that consumed at full speed. So even though throughput efficiency dropped to 95%, that is 95% of the 12.5%, not 95% of the 100%! The same holds true for heating effects as well. The motor is getting more heat PER UNIT OF POWER CONSUMED, but that number just went down drastically so the net efect is that at 50% speed, the motor is seeing less heat than at 100% speed with a throttling valve. Drive losses, same issue.
I think ccfowler summed it up very well in his first response and I voted him a star. VFDs do not work in every application and in fact, I sell quite a few Soft Starts for that very reason. That does not mean that we need to throw the baby out with the bathwater. Energy savings are real and proven, but only when the application alows VFDs to be used at their fullest potential. If you apply them just because a salesman told them to, you get what you desrve.
Final thought: As our friend the Rooster would say, "Just because the fox isn't in the henhouse it doesn't mean the hens are going to be left in peace.... Cock-a-doodle-dooooo!"
Quando Omni Flunkus Moritati