26 November 2009
Veral Smith wrote:
-- May I suggest to anyone who weight sorts. Try shooting everything that falls from your mould and looks good, and compare accuracy to your most carefully weigh sorted lots. You probably will never weigh sort again.
This discussion of the effects of bullet weight variation is fascinating but can also lead down a dead end street. The first question should be “why should there be weight variation” the second “what is the effect and how much?"
Joe Brennan's weight variance of 183.3 to 186.2 is excessive. To be fair, most were within 183.3 to 184.0 but flyers all on the heavy side of the envelope is strange. I normally get no more than 0.20 grains either side. To be fair again, I have never weighed more than 100 bullets in a batch as Joe Brennan has, so it could be argued that by not weighing them all I could have missed the flyers. The counter argument is that if I weigh half the bullets and find no flyers, the chances of all the flyers being in the half I have not weighed is statistically unlikely. Nonetheless, I shall make a point of weighing a lot more than fifty of my next batch.
Up to this point all bullets I have weighed have been within that 0.20 grains each way, no flyers. That tells me that one grain each way is high and three grains should be impossible. If 0.40 grain spread can be routinely achieved there will be no point in weighing because weighing will not find any flyers. But I don't necessarily dismiss weighing for an important match, I'm just making the point that casting can be consistent.
But having said that, lets look at the possible effects of weight variance. It takes a lot of testing under ideal conditions. Joe Brennan's first test showed liitle difference between good bullets and those with 0.20 grains filed off one side, with those with 0.50 grains filed off being slightly worse. In his second test the filed bullets outperformed the good bullets, but that simply showed that shooting in high wind introduces a big variable that renders the test useless.
Bullets stay on track by gyroscopic stability. Perfect gyroscopic stability requires perfect bullet balance around the centre. There is no need to find out whether a bullet with a piece filed off one side will be less accurate. Of course it will. It must be, because it will be out of balance, so the only question is how much less accurate.
Walt Berger told me ( no, I don't know him, I wrote asking a couple of questions ) that eccentricity of a bullet jacket three tenths or less can't be detected at the target from a bench rest rifle, but anything above three tenths can. Filing a nick in the side of a cast bullet simulates that. But if a 0.20 grain nick filed on the outside of the bullet shows a barely measurable difference at the target, bullets that cast within a 0.40 weight spread will be so well balanced that there will be no detectable difference on the target.
It also depends on the rifle and loading technique. Walt Berger also told me that a hunting rifle even if accurised would not group better with bench rest bullets than with hunting bullets because it is not accurate enough to use the superior accuracy of bench rest bullets. Loading technique can be a bigger variable than small differences in bullet weight or balance. Which is no doubt why bullet weight variance is one of the least important factors in accuracy.
Bottom line ? Time is better spent on casting consistent bullets than on weighing them.
Somewhere in this thread somebody said that alloy temp and thus mould temp causes variance in bullet weight. So I tested it by casting two batches of bullets from the same alloy at 610F ( the lowest my pot would go ) and 760F ( set at 750 on the pot but 760 by separate thermometer ).
At 610 the weight of ten bullets was 184.4 min, 184.6 max, 184.5 average. At 760 it was 184.6 min, 184.7 max, 184.66 average. Weight was 0.16 grains more at the higher temp and the spread was half. The reason ? My guess is better fill out at higher temp. As might be expected, at only 0.16 grains heavier, the better fill out was hard to see even with a watchmaker's loupe. About the only place it could be seen was that the line left by the mould interface was slightly more pronounced.
Will I cast hotter in future ? No, because casting hotter = casting slower, and such a tiny difference in weight means nothing in either bullet quality or results at the target. It confirms the claim that weight varies with temp but it is so slight as to be meaningless, and certainly does not explain weight differences of a grain or two.