This explanation is going to be a little long so bear with me. The process is a little complicated.
Physics tells us that the efficiency of a pump is highly correlated to the head curve which in turn is correlated to the THP of the motor. It takes more energy and a larger THP motor to produce higher flows and/or higher pressure, so the lower the head curve, the more efficient the pump will be. The actual efficiency of just the motor has an impact on energy use as well but for similar THP pumps, motors tend to have similar efficiencies. Higher THP motors tend to have better electrical efficiency (Watts/BHP) but because it takes more energy/GPM for a larger pump, this will dominate and the larger pump will end up using more energy/GPM even including a more efficient motor.
With two speeds, most of the motors are generally standard efficiency (at least in the CEC list so far) so the efficiency of the pump (wet end) itself is will dominate the total efficiency of the pump and the differences between motors will have a much smaller impact. So again, the head curve drives efficiency more than anything else. So even if you did not have any CEC data, physics will tell you that a lower head curve pump will be more efficient than a higher head curve pump. So I tend to rely upon THP and the head curve to rank order the efficiency of a pump. However, I try not to ignore data either.
Now back to the CEC data and how it is used. First, any pump that does not show a speed is full speed. Next, I don't have a lot of detail on how they do the measurements but there seems to be more measurement error at low speed than high speed but you might expect that depending on how the measurements are down between low and high speeds. One way to tell is that low speed should be within +- 1 GPM of half of the high speed GPM. However, in most cases, this is not true which leads me to believe that the low speed measurements were not necessarily done with a lot of precision. They should have adjusted the plumbing curve for high speed and then just set low speed without any plumbing adjustments. I believe they use a PSI gauge to measure the head loss for both high and low speed and we all know that there is higher error at lower PSI. Plus if they use flow meters to measure the flow rate, those also have significant error at low flow rates. So it isn’t a surprise that they probably have a lot of error at lower speeds.
Another thing that I noticed is that not all of the measurements actually fall on the intersection of the pump head curve and the analytical curve definitions (A,B,C) for both high and low speeds. This means that the plumbing curve measurements may not be the same between pumps and/or speeds.
And finally, the Whisperflo data is all screwed up and they swapped some of the models with others. The uprated versions of a pump should be the same as the equivalent full rated versions but they are not which doesn’t make sense. Also for some plumbing curves, higher THP pumps have better efficiency which again doesn’t make sense so that data must be “corrected†in order to be useful. I notified the CEC of this discrepancy and they are “looking into itâ€. However, I don’t think this is a real priority for them.
So you might ask yourself, then what use is the data. Well it is useful to identify trends one of which is that the efficiency is highly correlated to the THP of the pump. This is why I use the THP and head curves as a proxy for efficiency because theory tells us this plus the general trend of pumps in the CEC data tend to flow this rule.
Also, I use the CEC data in the PumpTools workbook but I do a lot of corrections to data that I believe is not correct. But the PumpTools workbook is primarily a learning tool and to give the user an idea of the energy use that a particular pump might use. For the most part I believe it is fairly accurate but there will be error in the calculations since the data source has a lot of errors.
So on to my list. Given all of the above conditions, I came to the following conclusions:
Hayward SuperPump 1 HP Uprated SP2607X102S
Pentair SuperFlo 3/4 HP Full Rated SF-N2-3/4A
Sta-Rite SuperMax 3/4 HP Full Rated PHK2RAY6D-101L
Are all similar in efficiency simply because their head curves are similar, the THP are close although the SuperPump is lower which is why I give an edge to the SuperPump. But also, the CEC data on Curve-A at high speed shows the SuperPump to have a higher energy factor so that confirms the THP argument. Is it better at half speed? I don’t know for sure but I would assume so given the theory of pumps and motors.
Next, the ¾ HP Whisperflo THP is about the same as the SuperFlo so in theory they should be about the same efficiency and only differ by the head curves. The Superflo head curve is slightly lower than the Whisperflo so it should have an advantage. The CEC data shows high and low speed as having a better energy factor for the SuperFlo SF-N2-1A than the WFDS-24 across most of the data points so that kind of confirms theory. I used the WFDS-24 instead of the 3 because it’s data looks more reasonable and what I would expect from the size pump. But again, they actually should be the same.
So I am not sure I answered all of your question or just generated more but I am happy to clear up anything that may be confusing.
BTW, if there is any doubt that the primary driver for energy use in a pump is flow rate, have a look at the following chart which shows the CEC Curve-A GPM vs Watts. It is pretty clear that the primary driver of energy use is flow rate which is determined by a pump's head curve and the plumbing curve.