January 25, 2018 at 12:02 pm #48850
This was posted in a long thread on CBA recently. Many disagree and are of the opinion weight sorting does not appreciably benefit accuracy. There are two ways to weight sort. One technique does not appreciably benefit cast bullet accuracy. The other way definitely does. The quote at the beginning generally describes the arguments why weight sorting does not benefit accuracy. However if you look at the technique I use, the weight sort results and then the proof of the pudding …..on target you’ll get the idea. Granted with just COWW or range scrap alloy shot in old milsurps not much benefit will be realized. However, with an accurate rifle using quality alloy or if shooting past 50 or even 100 yards then weight sorting using this technique is definitely beneficial.
Weight Sort 30 XCB Cast Bullets
“For your theory, that lighter bullets in a batch are less accurate than the heavier ones we must assume: 1. that light bullets are caused by voids, 2. that those voids are not too near the longitudinal axis and, 3. that they are large enough to be significant.”
Ergo is the problem in this discussion. I do not subscribe to any of those 3 assumptions. In fact if you re-read my post with the graph I explain what I’ve found to be the real problem and it is not the suspected or assumed “voids” in the bullets. Yes, that’s what we’ve all been told for probably a hundred years and it is what we’ve based our testing on.
Ten years ago I thought I was casting pretty good bullets, excellent in fact. However. the more I got into shooting cast bullets at HV I found while I was casting good, excellent bullets I too hit the accuracy wall that joeb is alluding to. I also found that when those cast bullets were pushed to really HV (2500 – 3000+ fps) they did not do as well as expected. Back then I was weight sorting as we’ve all been told to. If you line them out by weight you get the so called “bell curve”. In proving insanity I, like you and everyone else, then did the same testing of each .1 gr testing over and over again expecting different results…..we all got the same results; accuracy was not really improved via that method no matter how many times we ran the test. You are asking me now to run the same test and think I will come up with different results? It wouldn’t happen.
Let’s assume we have a mould that will cast perfectly even bullets in all dimensions. Not an assumption but fact is that mould has a finite capacity for any alloy. Thus if we cast with a good alloy giving the best fillout then only those that weigh the heaviest will have filled the mould out completely. Any bullets with less weight are then not dimensionally the same. We may not be able to measure other than weighing that difference but the difference is there in lighter weight bullets none the less. Now that difference in weight (mass) is there but it is not predictable…..we don’t know where in or on the bullet that difference in weight is missing from. The missing weight is what creates the imbalance. I suspect voids in the alloy are not the problem but rather other aspects are which I have previously discussed.
I recently cast 542 NOE 30 XCB bullets of #2 alloy. I have just completed weight sorting them. In the next post I will show the graphed results of the weight sort which should aptly demonstrate what I’m saying. Have to copy, download, etc. so it will be an hour or so.
Here is the results of the weight sort. 542 bullets were cast of Lyman #2 alloy and WQ’d. They were then aged about 12 days before I got around to weight sorting. Here is my set up for weight sorting. I visually inspect each bullet for any defect. If any is found that bullet is rejected to be melted and recast at a later casting session. Those that pass my anal visual inspection then have any remnant of the sprue cut off. That is done on the lead block with a sharp blade on the pocket knife. The bullet is then weighed on the Redding balance beam scale. While waiting for the beam to settle I then visually examine and sprue cut another bullet. With the magnifier in front of the scale I can readily and accurately see what the weighed bullets exact weight is. The bullet is then placed in a bin for that weight.
Of the 542 bullets weighed 22 were rejected for a visual defect or because they weighed less than 186.9 gr which means the weighed ones had passed the visual inspection but still weighed way lite. The remaining 520 XCBs were weight sorted into separate bins of .1 gr increment from 186.9 gr to 188.0 gr……a 1.1 gr spread.
Here is the rough graph of the weight sort. As you can see there is no “bell curve”. The curve rises from 186.9 gr slowly to 187.5 gr and then rises sharply. The “curve” then plateaus out at 187.7, 187.8 and 187.9 gr with 113, 124 and 110 bullets for each weight. The “curve” then falls sharply to just 9 bullets at 188.0 gr. Of those 9 bullets only 2 actually weighed 188.0 gr. The remaining 7 bullets weighed between 187.9 and 188.0 gr. There were no bullets heavier than 188.0 gr.
The weight sorting is showing us the 113 bullets of 187.7 gr, the 124 bullets of 187.8 gr and the 119 bullets (I’ll put the 188.0 gr bullets in with those) of 189.9 gr weight has the highest weight/mass of alloy in them. Since the curve dropped off suddenly we see those weight bullets are the most consistent and the best the mould will produce with that alloy. Those 356 weight selected bullets will be used for best accuracy.
The 187.6 gr bullets will be used as fouler/sighters as I expect they will give very good accuracy also given only a .2 gr +/- difference in weight.
Had we lumped all the visually selected bullets into one group 70% would have been with the excellent bullets, another 15% would have been with the fouler/sighter bullets and the remaining 15% would have been with bullets having a weight/mass difference of 1.1 gr. Now, had I done that I probably would have got nice 1 1/2 moa groups with 7 +/- shots going into moa or less and 2 -3 +/- shots going out of the group in the 1 1/2 moa +/-. How many of you shoot groups like that with bullets only visually sorted?
It is with such weight sorted selected bullets (the 187.7 to 187.9 gr bullets) that I am able to hold moa accuracy to 300 yards and beyond with a 2900+ fps velocity.
That is how I weight sort and why it makes a difference.
January 25, 2018 at 12:42 pm #48851
Weight Sort Test; 30 XCBs of #2 Alloy
In previous posts I posted the graphed results of weight sorting a recently cast batch of 30 XCBs (NOE 310-165-FN 4 cavity aluminum mould). They were cast of Lyman #2 alloy and were WQ’d out of the mould. These were loaded in the 30×60 XCB cartridge with the standard HV load of 53 gr AA4350. That give 2900 fps out of the test rifle; a 31” Broughten Palma contoured barrel with a 16” twist on a BRNO VZ24 Mauser action. The rifle has a Leupold 6.5×20 target scope on it. This rifle was not made for BR shooting. It was made for HV cast bullet shooting to determine how fast a ternary cast bullet could be pushed while maintaining 2 moa accuracy (10 shot groups) linearly to 300+ yards. That goal was achieved. The rifle with weight sorted select 30 XCB bullets will hold 1 to 1 ½ moa accuracy (10 shot groups) to a tested 400 yards (so far).
Let’s keep in mind before we post I shoulda, coulda, woulda tested this many or that many of this or that using different 5 or 10 shot tests of each, etc. that in this batch of 542 cast XCBs there were just only so many of some weights. The number of each weight is on the previously posted graph and listed in the post. The number of some weights, particularly the lower weights, was finite. There were only so many to test….period. Thus I made some judgments on how many each would be a “group”. I strived for 10 shot groups but that was not possible with some weights.
I use the Lyman lino/lead/tin formula to mix #2 alloy. Since I mix my own #2 alloy I’ve found over casting and weight sorting 5,000+ of the 30 XCBs that different batches of #2 alloy usually produce slightly different weights of bullets. In this test I had used a different source of linotype (a pig bar) from the previous batch made with type. The result is that the bullets in this test weighed and average of 1.5 gr less than the previous batch’s 30 XCBs. There was little difference iif any in accuracy or velocity with the slightly lighter weight new XCBs. The previously used weight sorted select XCBs weighed 164 gr fully dressed while the new ones weighed 162.5 gr.
The test was conducted yesterday at a range of 100 yards. Nine test groups were fired plus a sighter target. The barrel was allowed to completely cool and was cleaned between every 3 test groups. One fouler was shot and then the subsequent shots would “go to group” centered on the CBA target. The “foulers were all shot with the heaviest weight sorted select bullets (157.9 gr.) and since it was obviously zeroed no adjustments were made throughout the test to elevation or windage. Each test group was shot with that zero aiming at the center “X” dot.
The fouler target;
January 25, 2018 at 12:43 pm #48852
The first test was with the visually rejected bullets. These bullets were not weighed but had a visual defect (wrinkle, rounded GC shank, cavities in surface, frosted area on the side). The 10 shots group size was 2.280”. Only 2 of those 10 would have “gone to group”. All the other 520 bullets of this batch passed my anal visual inspection.
There were 5 bullets that were visually perfect yet weighed considerably lighter than the rest (155.2 to 156.9 gr). These were tested by themselves. As we see from the dispersion from the group had they not been weight sorted out they could have had a disastrous effect on a CBA score, particularly 2 of them. The 5 shots went into 3.317”.
January 25, 2018 at 12:46 pm #48853
The next test combined 2 weights (156.9 [X]and 157.0 [O]) as there were only 6 of one and 3 of the other. Each weight was tracked on target and is so marked (“X” and “O”). The 9 shots went into 1.620”
Again I combined 2 weights as there was only 3 of the 157.1 gr and 4 of the 157.2 gr XCBs. The seven shots went into 1.515”. That gives a fairly good indication that accuracy is improving as the bullets get heavier. Again the different weights were tracked on target with “X” and “O”.
January 25, 2018 at 12:54 pm #48855
Just when things were going good a guy shows up to test a prototype (they make them here in Lake Havasu) AR in 30-06. It had a 16” barrel with a horrendous muzzle brake. He parked himself on a bench 4 down from me. Every time he shot it felt like a grenade going off next to me. I had to put plugs in my ears plus the muffs over my ears. I tried shooting between his shots but I still managed to flinch off a shot of two as the test progressed. If that happened I called the shot (all went to call) and marked it. Thankfully I had at least 10 shots for the remaining group tests or additional loaded to get a 10 shot group.
The 157.3 weight put 6 of the 10 shots “to group”. There was 1 call in this group. Group size is 1.555” without the call shot.
The next weight, 157.4 gr, grouped into 1.750” for 9 shots. There was one definite called shot (ok, I flinched it off……).
January 25, 2018 at 12:56 pm #48857
As the bullets get heavier and are more completely filled out the groups are beginning to get better and to center back up on target. With the 157/5 gr bullets there was also one call but the rest went into 1.371” with more shots beginning to cluster.
The next test is with a mix of 157.6 gr (3), 157.7 gr (3) and 157.8 gr (4). These are tracked separately on target. Looking back at the graph in the previous post the 157.6 gr were selected to be “foulers/sighters” so I wanted to see how they would hit compared to the “select” weights of 157.7 gr and 157.8 gr. Their impact over lapped the heavier weights. The 10 shots went into 1.267”
January 25, 2018 at 12:58 pm #48858
The last test was with the heaviest 3 weights; 157.7 gr, 157.8 gr and 157/9 gr XCBs which I considered my “select” bullets. There were an equal number of each and they were mixed into one “lot”. Thus I did not track each weight but shot straight away. Again there were 2 for sure calls and maybe the low 6:30 shot but I had sufficient of this load to shoot 3 more shots making the group 10 shots. The 11 shots including the “maybe call” went into 1.304” which is what I recorded. The 10 for sure good shots went into 1.001”……moa. We see here the group has centered back up on target without any flyers.
January 25, 2018 at 1:02 pm #48859
The next test was with the weight sorted XCBs loaded over my match load in the .308W to test in my M70 Match rifle. This is the rifle I use in the “Commercial Rifle” (CBA BR Matches) matches over at Ben Avery Range with the Phoenix CBA bunch. This rifle also has a Leupold 6.5×20 Target scope on it.
This first target is with a mix of the 3 top weights (157.7 gr, 157.8 gr. 157.9 gr) with 3 each of the first 2 and 4 of the last. This group includes the foulers out of the clean barrel as they go very close to group. The first 3 foulers are marked and shots 4 through 10 then went into one small cluster of .584”. The overall group including the foulers is 1.031”
The scope elevation was then adjusted down 3 moa to get close to the aiming point. The next group was with 4 shots each of the 157.8 gr and the 157.9 gr along with the 2 remaining lighter weight sorted out 157.3 gr XCBs. They were mixed when loaded so the test was “blind” so to speak. The group size was 1.692” with 8 shots going into .892”. Wonder which 2 shots were the lighter weight bullets…..?
Well now, one can’t be sure based on just one test so I conducted a second “blind test. Same 4 each of the heavier bullets with 2 157.4 bullets mixed in so I didn’t know which was which when I shot them. The group size was 1.947” with eight shots going into .615”. Again given the almost mirror image of this group with the last one I wonder which 2 shots were the lighter weight bullets…..?
January 25, 2018 at 1:03 pm #48860
Lastly I graphed out the nine 30×60 XCB tests by bullet weight and group size. This gives us a “visual” comparing the group sizes vs. the weight increments. Between the graph and the actual on target performance it is pretty obvious that weight sorting is beneficial to increased accuracy.
Caveat; Many of you think of weight sorting as simply sorting by weight (usually in .1 gr increments) and then shooting those in those weight segregations. The thinking is that accuracy will improve. As most of you know accuracy improves perhaps a little but most often doesn’t using that methodology. The technique I use is to weight sort to eliminate potential flyers by selecting the heaviest bullets of each batch cast. Those are the bullets that are most consistent in fillout and thus the better balanced which gives the best accuracy. These 2 tests demonstrate that.
January 25, 2018 at 2:43 pm #48861GoodsteelKeymaster
- Posts: 196
- Comments: 2365
- Overall: 2561
Solid information Larry. Bravo!!!
I’ve made this a Sticky post.
January 25, 2018 at 3:58 pm #48862
Did you notice, in the 2nd test with the .308W the 23 good shots (with barrel fouled) using the “select” bullets the group average size is .697″! The velocity of that match load is 1980 fps. That says something for the 30 XCB bullet……just not to shabby for 23 shots at 100 yards….! Kind a puts to rest the claim by r5r that the 30 XCB bullet is only good for HV in the tight throat of the 30 XCB chamber………
January 25, 2018 at 4:42 pm #48864GoodsteelKeymaster
- Posts: 196
- Comments: 2365
- Overall: 2561
There are many claims made by many different people, but the fact is, I used more technology in the development of that bullet than has ever been used in history. I was out to create the finest cast bullet that has ever been made, and I believe I came pretty close. The only one that comes close to it (and may actually do better) is the HV bullet made by mountain molds. That guy really knew what he was doing.
January 25, 2018 at 4:50 pm #48865lkydvlParticipant
- Posts: 3
- Comments: 8
- Overall: 11
Nice article. You have me convinced. Have you done a similar test on the effects of weight sorting brass?
January 25, 2018 at 5:34 pm #48866kensParticipant
- Posts: 48
- Comments: 469
- Overall: 517
I recently cast a batch of 35XCB bullets and the majority of them came out heavy and large diameter.
I called NOE to question his mold, and he told me about mold temps. He says that aluminum molds can vary such as I have due to mold temps. I never had thought of it, but aluminum expands more with temperature than iron or brass.
It makes sense to me that a HOT aluminum mold will throw larger bullets, and as such heavier.
When weight sorting as above it makes sense that the heavy bullets are the most filled out, but, here is my next question:
Are you weight sorting by diameter? the larger diameter would be heavier. Now we must ask are you (we) sorting by mold temperature? dont get me wrong here, I am not disputing Larry by any means, but I am only introducing another angle on the weight sorting thing as it may relate to mold temp.
It has all been discussed before about the amount of sizing per particular diameter, to achieve a amount of accuracy, and so on. Personnally, I think mold temp is just as much a factor in all this as the weight sorting myself. I have not tested this, but do believe this to be just as much a factor.
What you others say?
January 25, 2018 at 8:09 pm #48868bjornbModerator
- Posts: 13
- Comments: 166
- Overall: 179
In my testing of the 30 XCB bullet (mostly shot in heavy bench rifles using the 30×57 cartridge), I always weight sorted the bullets identically to what Larry is explaining. I would then ONLY load the “top of the curve” bullets for group testing, the rest would be foulers. When mixing inferior bullets for the purpose of testing, I would always get results similar to Larry’s.
As for the mould temperature issues brought up by kens, I can add that I started out with a 4-cavity brass mould, casting so many bullets with it that it had to go back to Al for adjustments after a few months. I then bought a 5-cavity aluminum mould so I could keep the production going, and I immediately got bullets that differed substantially in weight from the brass mould bullets. We’re talking 2-3 grains weight difference with alloy from the same batch. This didn’t seem to matter much when it came to accuracy; after weight sorting I saw no difference in target results. I still find it worth noting that the 2 moulds produced different weight bullets. A third XCB mould, a 2-cavity alumin, produces identical bullets to the 5-cav.
Thanks Larry for a very well written post.
January 25, 2018 at 11:00 pm #48873ZmanWakeForestParticipant
- Posts: 20
- Comments: 115
- Overall: 135
Very informative post, Larry! Well written indeed!
January 26, 2018 at 12:05 am #48874ArtfulParticipant
- Posts: 155
- Comments: 386
- Overall: 541
Good write up Larry – bjornb, Walt at NEI long ago explained to me the Aluminum and Brass having different expansion properties when heating up will make a difference in the dropped from the same design unless the mold maker takes it into consideration – He was a smart guy – I was ordering two cavity molds of different designs and sometimes he’d say nope don’t do that as the cavity wasn’t big enough to get the bullets to correctly heat the block to match the other side with a larger cavity. I really wish I had been smarter and pumped him for more knowledge back then.
January 26, 2018 at 10:51 am #48875popperParticipant
- Posts: 1
- Comments: 261
- Overall: 262
The technique I use is to weight sort to eliminate potential flyers by selecting the heaviest bullets of each batch cast. Those are the bullets that are most consistent in fillout and thus the better balanced which gives the best accuracy.
Totally agree. Pus it’s almost impossible to do with inexpensive electronics scales. Even a good beam scale needs a shield enclosure over it to prevent draft current errors. Then to take advantage of this method, you need a great set of dies + a very good GC install system.
January 26, 2018 at 12:52 pm #48877slim1836Participant
- Posts: 2
- Comments: 46
- Overall: 48
I weight sort them also, rejecting the low ends and keeping heavier ones that are very close in weight. Re-melt the rejects, load the keepers. I don’t get technical as I only plink at 100 yards or less but I’m in the kill zone for hogs.
You must be logged in to reply to this topic.