The reason Al doesn't want to supply the data is because it isn't just a simple number, like 4 fps per kernel. It is not possible to give one number that holds true for all cartridges, all bullet weights, and all powders, and all barrel lengths. All of these factors will cause variations, and I didn't even cover all variables like bore and groove dimensions. To get a real number, for a given cartridge, with a given bullet, and a given powder, do this. Load five identical rounds, and then load another five with .5gr more powder. Weigh all powder charges as close as you can. Shoot all ten rounds over the chronograh at the same time, and get the average velocity for the first five, and then for the second five that have .5gr more powder. Determine the velocity increase for .5gr. Let's just say for arguments sake that we gained 80 fps, and you were shooting N133 in a 6PPC. Now weigh out .5gr of N133 on a precision scale. Then count the kernels. Let's just say there were 33 kernels of N133. Now divide 80 fps by 33 kernels, and you'll get 2.4 fps per kernel. Now this is proof that you can chart velocity per kernel, but remember that any change, including atmospheric conditions will have an effect of this data. Making major changes like kernel size, burn rate, case capacity, and capacity to bore ratio will change the velocity per kernel numbers in a huge way. Just because Al could show you some examples, doesn't mean he wants to go to the trouble. A guy sent me a private message wanting me to teach him how to build rifle actions. I asked him where he lived, and he said "Italy". I asked him to give me a call, and I would try to explain it. He then asked me to just teach him with an email. Even my thread about "Why I shoot the 30-30 in benchrest" was a very time consuming post. I think I could spend a week trying to teach someone how to build an action by email, and they still wouldn't understand.
If you follow the simple instructions, you can come up with accurate data for velocity per kernel for one cartridge, bullet, powder, rifle, condition, combination. Now if you are talking about the variation in kernel size in the same can of powder, this formula will still give you the average, because you are weighing a .5gr sample of various kernels from the same can. It all about the average, not knowing how much the larger kernels contributed to the velocity versus the smaller kernels.
Michael