Bill,
The best way to run the test is to use simple match. Once you have a velocity increase per grain, or per tenth of a grain, you can weigh the kernels and divide it to see how many kernels are in a grain, then divide the velocity by the number of kernels. You have to keep in mind that the more a case is necked down, the more velocity you get by going up a tenth of a grain. Also the faster the powder the more velocity for a tenth of a grain. An example of a powder and a cartridge that would give a lot of velocity increase per kernel would be IMR4198 in a 22PPC. Using a common powder to BR, N133, the 6PPC gets around 20 fps per tenth of a grain, where the 22PPC closer to 25 fps, and the 30-30 closer to 10 fps.
The main thing to keep in mind is that this is just information that is somewhat useless. Knowing the exact velocity per kernel doesn't help with accuracy, it's just a number. Weighing powder charges closely is what keeps these velocity spreads down, not knowing the velocity per kernel.
There are about four things that determine velocity spread. Amount of powder, primer consistency, bullet seating depth consistency, and neck tension consistency. If I get all of these thing as close as I can, I have seen the 30-30 get down to 2 fps extreme spread for five shots. Since the PPC is necked down more, a person would be doing very well to get velocity spreads down to 5 fps.
Michael