Re: Chargemaster = Expensive thrower
<div class="ubbcode-block"><div class="ubbcode-header">Originally Posted By: rrflyer</div><div class="ubbcode-body">I just received my acculab vic123 and have just done an initial test with my CM vs the acculab.
The chargemaster rest throw was 44.4gns varget.
On the acculab they were consistently showing 44.36 plus or minus 2/100's or .02 of a grain.
20 throws</div></div>
This post caused me to do an experiment. I decided to test my Chargemaster against my VIC-123 against my GD-503. Could it be that my Chargemaster is within a kernel like a Prometheus and I just didn't know it?
I wish I could figure out how to publish my spreadsheet. I calibrated all 3 per instructions, threw a 43.0 gr charge with Varget on my Chargemaster, and weighed on my Acculab VIC-123 and then on my Sartorius GD-503. I weighed the Chargemaster pan on both scales because I need to rezero my VIC-123 before weighing, and this allows me to calculate sample weight. My VIC showed the pan weight to be 142.40 gr while my GD-503 showed the pan weight to be 142.420 gr.
I threw 30 times (actually 31, but one was an overthrow, so I dumped it and redispensed).
What I found was that the standard deviation of the Chargemaster was 0.0415 grains. Multiplying the SD by 2 gives me where I am confident I will be 95% of the time (actually, where my average will be, but no need to get into that). Basically, the Chargemaster is +/- .083 grains. Not bad... that is better than the .1 gr precision that the CM reads to. This translates to +/- roughly 4 kernels of Varget with the CM. That is half of what I was quoting earlier.
The largest mistake was an overthrow of .0922 grains, which happened twice. The largest underthrow was a .0828 grain underthrow. To get the terminology right, I only weighed samples where the CM read 43.0, so the CM was not reading it as an overthrow or an underthrow, it just was an overthrow or an underthrow according to my target weight.
I think the CM performed pretty well. Similarly, the VIC-123 performed pretty well. 8 of the 30 times it varied by about .02 grains, or roughly a kernel. It never varied by more than .025 grains, so it was never as many as 2 kernels off.
I also retested my theory about the weight of a kernel of Varget. I calculated 3 ways... first by weighing a single kernel on the GD-503, then by weighing 5 kernels together and dividing by 5, and finally by weighing 20 kernels.
With wieghing a single kernel, my scale read either .02 gr or .015 gr. The average I came up with was 0.019 grains, but I decided that wasn't a realistic number for the resolution of my scale, so I weighed multiple kernels. Weighing 5 kernels at a time 6 times, I came up with an average weight of .0203 grains. Weighing 20 at a time, I came up with a .022 grain average per kernel. In other words, there is some variation from kernel to kernel, including weights below .02 grains per kernel, but on average, a kernel weighs about .021 grains or so.
Interesting stuff. I learned a little bit. I hope everyone else learned something as well.
