I had a look at these, and from what I can tell they use a single load cell in the middle of the platform.
This makes me think the measured values will vary depending on where the load is on the platform. Is this the case?
Could anyone who owns a set take a look?

 

Any single or dual cell scales are calibrated for the center of the contact path to be centered on the pads. Experience says generally centered, not on the edges.

 

Thanks.
Does experience say that the repeatability from session to session is acceptable with such pads?

 

I have Proform scales and they have been rock solid and the repeatability it very accurate as long as you center the wheel/tire with the center of the pad, the pads are level and the pads are level to the other pads. I've been using mine for 3 years now and now use them to set up cars for others. At some point maybe I'll get into a higher end set, but I've found my pads to be within a couple pounds of the pads use in compliance by NASA and PCA racing organizations I run with.

 

Excellent information.
I'm contemplating making a set using Arduino and some decent load cells. Seems doable, but it's not at the top of my list at the moment.

 

As mikelly said, getting each pad level and all 4 pads level to each other is mandatory to getting all set up data including corner balance. Making your own set of pads sounds like a fun but expensive challenge. The electronics are one thing but maching the pads to the necessary tolerances may be a big hurdle.