Some Thoughts about Pellet Sorting

Jump to latest Follow Reply
G

Guest

Guest
Since the first of the year I've been measuring the head size of just about every pellet I've shot in all my rifles. I keep thinking I have conclusive evidence that this or that rifle "likes" a certain head size and then "poof" my proof goes up in a series of groups which I can't explain. This has happened over and over with different pellets in different rifles. A perfect example was my study of JSB 15.89 grain pellets in my Diana Model 48. The pellet shoots pretty good in the rifle I had a large number of them so I carefully sorted out a couple of hundred each of 5.48mm, 5.49mm, 5.50mm and 5.51mm. Over the last couple of months I've collected 11 five shot groups with each of those head sizes and measured every target using "On Target PC". I decided that the "average to center" value for each target would be the best representation of my individual groups and chose to use MOA as the unit of measurement. This allowed me to discount range in some of the calculations, although in the future I will shoot all groups at the same range to eliminate that variable. For each head size 4 groups were shot at 35 yards and 7 groups were shot at 25 yards. So I had a pretty good sampling of pellets sorted by head size. I then ran the numbers.

Here is what I got:

What that tells me is that there is a very small probability that the size of a group may change very much in that rifle, with that pellet, between 5.48mm and 5.51mm! The actual sizes of the groups are not markedly different statistically! I kept running into this but this case is the one that has convinced me to think about something else.

Almost everyone who tries sorting pellets says that it improves their group size, and it seems to me that a consensus is a consensus, it may be heuristic but if enough people believe it, it is generally a good idea to figure out what leads them to think that way. I have a theory that might explain why there is not a really strong correlation in this data between group size and head size of the pellet.

All those pellets shot an average to center of just about 1 MOA BUT they shot that group to a different POI when I held the zero constant.

At this moment that is just an observation, nothing more than a theory but I was noticing this phenomena happening today as I shot my groups. I have the targets but I have blown the test data because I was adjusting zero between pellet sizes. So back to the shooting table with me. I still have enough of that batch of sorted pellets to test this hypothesis and after I do I'll post the results here.

Food for thought anyway.
 
"Goodtogo"When you get done with your study, do you think you will do one with head size and weight?
Click to expand...
Maybe. Once we isolate what is going on with headsize it should be pretty easy to work out weight with a batch of sorted pellets. This "study" has taken about five months longer than I expected though. Hopefully I can run some more groups late next week. I'm going to swap out the cheapie BSA 2x7 for that SWFA I'm getting and get that zeroed before I continue on this track.
 
Upvote 0
"socaloldman"Does your data track a number or location as in degrees of direction change from the POI relative to the zero constant. That might be an interesting perspective. Hopefully there is no wind involved. Just a thought. Such an interesting study.
Click to expand...
I'd track it on a grid in X and Y and compute the rest of the data from that. The data in the first part is computed in a program that I use. I just take a picture and then upload it to the PC and that program establishes the average center of the group by averaging the coordinates. Then it computes the distance from center for each shot and finally the average distance to the group center. By using the group center and calculating the average to center for the group you eliminate zeroing errors and reduce the influence of the occasional flier. On a five shot group if you only look at extreme spread you are really only looking at a two shot group and that is the group formed by the two shots that are the farthest apart. On the next part I will need to zero one time with a mixed assortment of the pellets and then shoot groups with measured pellets without changing zero. In another forty groups or so I should have some more information.
 
Upvote 0
I had time to finish that study today. So here is what I have:


Above is the original graph of average to center for eleven 5 shot groups at each head diameter. You will recall that I thought I didn't have good evidence of improvement. I thought that it might be possible that group center was moving with head size changes and so I shot the following over the last two days:



Neither graph makes a compelling case for the rifle preferring the 5.51mm head size but both show that it "seems" to favor them a little. A mistake that I am making here is not quantifying the actual difference in improvement between head sizes.

The average over all for this last batch of data was a average to center of 1.186 MOA and a best for the 5.51mm head size of 0.978 MOA. That is a 21 percent reduction in the average to center measurement for the group. My graphs don't make that obvious. I'll take 21 percent over average.

Now here is the interesting graph that I made today:

This graph shows that the extreme spread between the centers of groups for different head sizes is about 1/2 MOA.

My raw targets for the last couple of days of shooting are in that directory; should anyone want to check my numbers. There is also a spread sheet with the data already entered. Microsoft XL
http://www.oldspooks.com/agstuff/pellets

I have not tried to do a serious statistical analysis of this data but I conclude from my graphs that sorting pellets for my Diana 460, with JSB 15.89 grain pellets, can reduce group size by about 21 percent AND reduce extreme variation of group center by about 1/2 MOA. For my test the latter is more significant than the former. A half MOA would reduce the average group size by about 42 percent.

Yes sorting pellets improves group size significantly. On to the ten dimes challenge:
http://www.oldspooks.com/agstuff/Targets/challenges/Ten-Dimes-v2.pdf
 
Upvote 0
A couple years ago I sorted out a couple tins of JSB 18.1 pellets by weight. Then I shot 10 shot groups with my Mrod. I did not see any consistency of group size with weight. Since then I have been shooting straight from the tin. I have not noticed any significant fliers that I could not attribute to me. Of course, I now have a couple of significantly more accurate rifles (BSA R10, FX Wildcat and FX Impact), so I may just retry the test.

Maybe, just maybe group sizes might be due to POI change due to weight, like with your head size? 

In the end, if it comes down to having to sort by head size AND weight to get the smallest groups I am not sure it is worth it. As it is, I am able to consistently hit MOA anywhere from 20 - 60 yards with both my FX guns which is more than adequate for pesting. Beyond 60 yards it becomes a guessing game with holdover and wind so I don't expect that kind of accuracy and don't take those kind of shots.

One of my local gun shops is building a new store with an indoor range that will have 25, 50 and 100 yard lanes. When that is done, then I can get some really accurate data at 100 yards.
 
Upvote 0
"ncrary"
In the end, if it comes down to having to sort by head size AND weight to get the smallest groups I am not sure it is worth it.

Click to expand...
Agree whole heartedly. I spent the afternoon just trying to figure out how sensitive the rifles really were and just how much I was willing to do to get tighter groups.

On my Model 36 I did notice something I had not noticed before. Namely the POI was shifting. I was pretty POd because I had only just hung an SWFA on the rifle and I knew mount wasn't moving around. Then I figured out that there was some sort of play in the system which would move the POI depending upon how I cocked the rifle. I'll have to have a look at that. The rifle was shooting around a minute and a half consistently but groups were moving as much as a couple of mils. Then I figured it out and payed careful attention to how I cocked the rifle and I could literally move the groups around (mostly in elevation). So I might just have to bed that rifle.
 
Upvote 0
"ncrary"
Maybe, just maybe group sizes might be due to POI change due to weight, like with your head size? 

Click to expand...
POI change due to weight is a definite possibility and I don't know of any software which can calculate/simulate that. Chairgun can give you pretty good information about how much your drop will be based upon your muzzle velocity variation.

Toolbox => Vertical stringing.
 
Upvote 0
As it is, I get very good consistency shooting JSBs from the tin. I do notice about a quarter inch shift (at 50 yards) in POI going from 15.89 to 18.1 pellets, all else being the same. But that is to be expected. So, even with a few tenths of a gain shift within a tine, I don't think the POI shift is worth worrying about. When I hunt and shift between 15.89 and 18.1 pellets, I normally don't change my scope or POA and have very few misses. That being said, I have my Wildcat dialed in for 15.89 and my Impact for 18.1, so if I am going to go for a longer range head shot, I will go with the pellet the gun is set up for because I know there is a POI shift. Since accuracy is great with both pellets in both guns, as soon as I finish out my 15.89 last two tins, I will go only with the 18.1 to keep things simple.
 
Upvote 0
You must log in or register to reply here.
Top Bottom
Back
Home
Forums
Resources
Menu