First up, I had to test the Vario style control as well as the “PWM” control. So I hooked this up to my XSPC D5 bay reservoir and connected the inlet and outlet ports together thus having nothing else in the loop to provide any flow restriction. I used an Aquacomputer Aquaero 6 XT for control and RPM readout, coupled with the knob on the VPP755 as well. I began by simply changing the knob from Position 5-1 and charting the pump speed with the PWM signal set at 100%. I compared this to a D5 Vario as well:
Please also note that the RPM readout was fluctuating to the point where a standard deviation of 3-4% was common and the average readout is used in the plots, with an error margin only reported here to not further clutter the graphs. We see that the VPP755 has consistently lower pump speed at all positions, beginning with 4511 RPM vs 4830 RPM on the Laing D5 Vario. It got close on Position 4 but then the same trends continued on Positions 3, 2 and 1. The VPP755 had a much more linear change though. In terms of how close the VPP755 came to specs, the pump speed was 4511, 3891, 3192, 2465, 1745 RPM in practice compared to 4500, 3900, 3200, 2500, 1800 RPM respectively. All well within the +/- 10% deviation I allow, which was good to see. So the knob control works well enough. What about PWM control? Knowing that this was not true PWM, and also with Alphacool advertising a patented interplay of the manual knob control with PWM control I set the knob on Position 5, charted PWM signal vs RPM from 100% PWM duty cycle to 1% PWM duty cycle, set the knob to Position 4 and repeated all the way to Position 1. This was done after having bled the simple closed loop of as much air as possible to the point where only microbubbles were present.
There is a lot of information on this chart, so allow me to walk you through it. We know already that there is no true PWM control on the pump, it accepts PWM signal and sets the pump speed based on a variable voltage divider similar to the potentiometer on the knob. As such, there are ranges of PWM signal that correspond to a pump speed followed by a sudden change in pump speed and this trend continues. As before, the RPM readout continued to fluctuate and the average RPM is reported here. At Position 5 on the knob, the pump speed read out was 4511 RPM and this remained the same from 100% PWM to ~88% PWM wherein the pump speed dropped to 3886 RPM (which corresponded to Position 4 and 100% PWM signal, if you recall from the results above) and stayed there till 73% PWM where the next drop in pump speed happened and this (3174 RPM) corresponded to Position 3, 100% PWM. This continued on to where it was fairly obvious what Alphacool meant by an interplay of manual and PWM control of pump speed- both will do the same and offer ~5 average pump speeds ranging from 1740-1800 RPM to 4500-4520 RPM. For all intents and purposes, this is just a Vario style pump and you have the option of having manual or software control of the Vario feature. I think this point should have been made more clear by Alphacool as many, including I, thought otherwise.
The story remains the same at other positions of the knob wherein the PWM-RPM responses simply trigger the drop in RPM corresponding to the lower knob position. This happened almost always at the same position, except at Position 4 where the speed drop corresponding to Position 3 happened at 82% PWM rather than the ~73% it should have been although this is such a minor thing that I can hardly call it a negative point. At Position 1, there is no effect of PWM signal on RPM as now makes sense.
For some more context on PWM control of D5 style pumps, here is the Eispumpe VPP755 (Position 5) compared against two D5 PWM pumps:
The XSPC D5 PWM pump is a rebranded Laing D5 PWM pump with no change done by XSPC aside from re-branding. As such, we note that there is a linear decrease in pump speed as you lower the PWM signal provided and the pump actually shuts down too. The Laing D5 PWM pump just happens to be one of those devices that do not follow Intel PWM spec 1.3 meaning it won’t play well with some fan controllers, although it must be said that Laing never designed the D5 to be used with a PC motherboard for control. Some customers, including EK, used a different motor and reprogrammed the fan curve to play well with PC motherboards and fan controllers and we see here also a linear decrease of pump speed as the PWM signal is lowered, albeit with a higher slope. The Alphacool Eispumpe VPP755 is the outlier here due to the nature of the PWM signal output tied in with the knob control.
Please also note that setting the fan header to voltage/power mode does nothing and the pump speed remains the same from 100-1% and corresponds to that for the knob position. For example, setting 50% power/voltage (6 V) via the fan controller to the pump with the knob in Position 5 doesn’t do anything and the pump speed remains at ~4500 RPM. The pump is not actually getting 6 V here, it continues to get 12 V from the PSU. The only way to do voltage control is to change the input from the PSU side, meaning having a variable controller feeding the full size MOLEX. I do not have the means to do this, although I believe Koolance offers a controller for this very purpose. Voltage control on a Laing D5 was not very good in that the pump shut down at 9 V or less and did not offer any practical RPM response so keep that in mind.
Now that we have a clear picture of how the pump control works, let’s test it out and see how it performs on the next page.