This week over two range sessions I shot 10 five shot COL test groups from my RPR 6.5 CM at 200 yards with a labradar recording every shot. Admittedly I am not a great group shooter but had some interesting results I need help interpreting. Three of the test groups spaced .007, .010, and .015 off the lands returned SDs of between 7.1 and 7.7. the MOAs were 1.3, .88, and .767 respectively. As the jumps widened the SDs went into the teens but never exceeded 15.3. However I had the best MOAs at the widest jumps of .035 (.64 MOA), .045 (.58 MOA) and .050 (.52 MOA) off the lands. I have probably 1500 rounds through this barrel and have shot 140gn Hornady ELDM over 41.5 gn of 4350 almost exclusively. The cartridge base to the lands now measures 2.187. I’ve been shooting COLs of 2.810 since the rifle was new. I’ve hit everything out to 1400 and even once at a mile but now am pursuing FClass groups out to 500. Is there any pepper in the pile of data or is it all fly droppings? Any help would be greatly appreciated.
-
Having trouble using the site?
Contact support
Does small SD indicate anything shooting OCW or COL Testing?
- Thread starter Woodbubba
- Start date
Your lowest SD should net you the most consistency over distance. You said yourself you’re not “a great” group shooter.
In my opinion, anything can group well at 100 but the further out you go you find many things. The biggest thing you find is flaws in your fundamentals.
In my opinion, anything can group well at 100 but the further out you go you find many things. The biggest thing you find is flaws in your fundamentals.
SD's are a statistical number and it tells me how well I'm doing my reloading. When it comes to group size ES is a more significant piece of data in what might going on with the size of a group. And as Darkside-Six says, at 100 yds almost anything can group well (particularly if one is only looking at a 5 shot group), it's at distance that ES is going to be seen as a significant factor.
But, to answer the question regarding SD indicate anything regarding the OCW testing. . . it's really not anything to consider for OCW other than smaller SD's give one more confidence in the analysis between each group for the OCW objective. Sames goes for ES 's as the make it easier to interpret the groups.
But, to answer the question regarding SD indicate anything regarding the OCW testing. . . it's really not anything to consider for OCW other than smaller SD's give one more confidence in the analysis between each group for the OCW objective. Sames goes for ES 's as the make it easier to interpret the groups.
Anymore I don’t even shoot groups.
I mainly look for a velocity flat spot with a hopefully clean extreme spread.
1-2 shots per small change of powder to find a trend then load what looks promising and shoot at distance to confirm.
Then tune accuracy a bit with jump if needed.
I mainly look for a velocity flat spot with a hopefully clean extreme spread.
1-2 shots per small change of powder to find a trend then load what looks promising and shoot at distance to confirm.
Then tune accuracy a bit with jump if needed.
Amen......just looking for that flat spot where if I accidentally drop one more kernel of powder in the brass....it don't friggin matter. Just give me a consistent velocity with a wide enough range of powder weight...and let me worry about jump...then I find that 99.9% of anything else...is just my abilities as a shooter.
Got that, my zero is at 200 not 100 which I believe improves my chances of a smaller spread further out. Basically I’m wondering if the two small SDs are telling me something regarding the internal ballistics performance relative to a barrel node.
Just be careful using SD's with small sample sizes (under 20-30).
I've been doing a lot of testing lately with 30-50 shot sample size and the results are kind of surprising in a way. I haven't done enough to come up with a definitive load development method yet, but I will tell you that 5-shot sample sizes are for sure subject to be a total lie. The more testing I do, the more that statement solidifies.
If I take one of the 50-shot data sets I've collected then have Excel randomly grab 5 of those shots, then grab 5 more, etc.. and build each 5 that it grabs as a data set with an Avg velocity and SD, it will look like the red dots below. Then do the same thing for 10, 15, 20, 30 shots each....
Now what the outer bounds are-- the highest and lowest point-- varies greatly from combo to combo. I've shot rifles that had a 36fps ES, and some with 80fps ES. 0.95" 30-shot groups and 1.6" 30-shot groups. And like I said there's still a lot more variables I want to meddle with...
The point is that every 5-shot group you shoot represents any one of those red dots, and you gotta shoot 6+ of them and average the combined results to get decent data that's worth doing any statistical analysis that you'll use to make confident decisions. 20-shotters for the most part are what seem to be the lowest number of shots to get you a good picture. Same goes for Average velocity and velocity SD. 5-shotters of the same load are subject to bounce around quite a bit. Some are truthful, some are outstanding lies, and some (contain the extreme spreads) are dog shit lies. You never know which until you shoot more.
I've been doing a lot of testing lately with 30-50 shot sample size and the results are kind of surprising in a way. I haven't done enough to come up with a definitive load development method yet, but I will tell you that 5-shot sample sizes are for sure subject to be a total lie. The more testing I do, the more that statement solidifies.
If I take one of the 50-shot data sets I've collected then have Excel randomly grab 5 of those shots, then grab 5 more, etc.. and build each 5 that it grabs as a data set with an Avg velocity and SD, it will look like the red dots below. Then do the same thing for 10, 15, 20, 30 shots each....
Now what the outer bounds are-- the highest and lowest point-- varies greatly from combo to combo. I've shot rifles that had a 36fps ES, and some with 80fps ES. 0.95" 30-shot groups and 1.6" 30-shot groups. And like I said there's still a lot more variables I want to meddle with...
The point is that every 5-shot group you shoot represents any one of those red dots, and you gotta shoot 6+ of them and average the combined results to get decent data that's worth doing any statistical analysis that you'll use to make confident decisions. 20-shotters for the most part are what seem to be the lowest number of shots to get you a good picture. Same goes for Average velocity and velocity SD. 5-shotters of the same load are subject to bounce around quite a bit. Some are truthful, some are outstanding lies, and some (contain the extreme spreads) are dog shit lies. You never know which until you shoot more.
Keep looking for one...heh. Sometimes you have to go slower...eventhough you don't want to...to find that good sweet spot. It will be there though....just got to look for it.
No. I was making a point that not every primer/powder combo will have flat spots in velocity.
My 6gt with cci 450’s and varget will keep climbing on average about 12fps for every .2 charge. From 31gr all the way to 36gr.
Which is why I look at ES to determine nodes.
Here is an example of no flat velocity, but still 3 nodes based on ES:
Attachments
Stop zero’ing at 200. That logic is wrong. Zero range has nothing to do with group size.
That's actually been the most interesting (to me) testing I've done. I have a set of control samples in 2 different barrels (A Wilson and a Bartlein) of 41.00 +/- .02gr (to the kernel, basically) of RL-16 under a 140gr ELD in 6.5 Creedmoor, then repeated the test with a thrower set ~41gr that was +/- about .4gr over the spread of testing.
ES was marginally higher for the thrown charges, SD's were within 1fps for both barrels thrown vs. weighed to the kernel. Wilson shot 1.5 and 1.6" groups for 50 shots, Bartlein shot 1.2 for both. 100yd 5x 10-shot groups with a full cool down between.
That’s great data. It would be interesting, if laborious, to record the position of each shot too. You could then make simulated groups along with your simulated ES.
What I have found is a sliding scale.
The more controlled you are, the smaller sample size you need to be confident in your data. When I say rightly control, I mean neck wall thickness, exactly same shoulder bump, exact same neck tension with a very low ES in seating pressure. And if course powder charge.
Each time you do one less in those steps, the larger your sample size needs to be.
It’s probably true that if you control more variables in the process, you will have less variation. But statistically you need a larger sample size to see smaller differences, not a smaller one.
Yes. But I’m not going to waste ammo if I know there is little reason to think the nodes wouldn’t be the same.
Otherwise you’d burn out a barrel doing load development
Absolutely, if you did enough testing to really know if there was a difference from one load to the next, you would burn out the barrel.
The thing is, I’m not convinced there is any real value in a lot of things people do for load development. I wouldn’t be surprised if most of it turned out to be an empty ritual.
I agree, the more erratic the results, the larger sample needed to get everything to level out and run steady. If the variance is below a certain number at 20-30 shots, I see no reason to mess with it.
I don't get that deep in the weeds with prep, but I have a co-worker that shoots BR and does do all of the OCD stuff. We shot his 6 BRA for 30 shots, too. .95" total group size, 36fps spread, and 8.4fps SD. The best set I've personally seen so far. That said, if we'd have taken his first 5 shots, it would have shown us an SD of like... 2 fps and a group size of .3-.4".
Consider this... You rigidly fix a barreled action into a 2000lb block of steel and fire at a large piece of paper at 500yd in an enclosed tunnel (bear with me...). Fire 1500 rounds, ignoring foulers after cleaning cycles, etc. etc... Just plotting a cone of fire at 500yd for the entire "match grade" life of a given barrel. Looking at this mass of 1500 holes, what are the odds that a random selection of 5 of them is going to contain 2 or more on the outer edges and represent the true precision of the load?
I don't get that deep in the weeds with prep, but I have a co-worker that shoots BR and does do all of the OCD stuff. We shot his 6 BRA for 30 shots, too. .95" total group size, 36fps spread, and 8.4fps SD. The best set I've personally seen so far. That said, if we'd have taken his first 5 shots, it would have shown us an SD of like... 2 fps and a group size of .3-.4".
Consider this... You rigidly fix a barreled action into a 2000lb block of steel and fire at a large piece of paper at 500yd in an enclosed tunnel (bear with me...). Fire 1500 rounds, ignoring foulers after cleaning cycles, etc. etc... Just plotting a cone of fire at 500yd for the entire "match grade" life of a given barrel. Looking at this mass of 1500 holes, what are the odds that a random selection of 5 of them is going to contain 2 or more on the outer edges and represent the true precision of the load?
As far as precision versus SD/ES goes, I don't believe there is any correlation that shows that the groups with the smallest SD/ES also have the smallest groups. In fact, from the reading I've done on the topic from benchrest shooters, including those that have built world record rifles as well as some that hold world records themselves, rarely does the load with the tightest ES/SD have the tightest groups. They don't look at ES/SD, and most couldn't even tell you what their velocity is, let alone what the ES/SD is of their loads.
That said, in our game where we are shooting targets from 100 yards (or less) to a mile +, and everywhere in between, lower ES gives us more ballistic predictability and thus effectiveness, especially the further we go out. We don't focus on one set range, or our priorities and procedures for developing a load would be different.
Looking for velocity flat spots can be misleading, from my experience if you are doing it from a single shot per charge ladder. Ladders, such as those advocated in the "Satterlee method", don't have a large enough sample size to accurately determine any so-called nodes. One shot per charge is not enough to determine any velocity flat spots, simple statistics.
I will use a ladder to find approximate velocity per charge weight, along with signs of pressure. From there, I will load up 5 rounds per each powder charge, well below pressure, and look for the charge weight that produces the best SD/ES. Precision is sorted out with bullet seating tests. For the shooting I do, I look for ~0.30 MOA and an ES of 20 or less (usually end up in the ~15-20 ES range for 10+ shots). Good enough for the shooting I do, but I probably wouldn't beat Alex Wheeler at a 1,000 yard benchrest comp.
That said, in our game where we are shooting targets from 100 yards (or less) to a mile +, and everywhere in between, lower ES gives us more ballistic predictability and thus effectiveness, especially the further we go out. We don't focus on one set range, or our priorities and procedures for developing a load would be different.
Looking for velocity flat spots can be misleading, from my experience if you are doing it from a single shot per charge ladder. Ladders, such as those advocated in the "Satterlee method", don't have a large enough sample size to accurately determine any so-called nodes. One shot per charge is not enough to determine any velocity flat spots, simple statistics.
I will use a ladder to find approximate velocity per charge weight, along with signs of pressure. From there, I will load up 5 rounds per each powder charge, well below pressure, and look for the charge weight that produces the best SD/ES. Precision is sorted out with bullet seating tests. For the shooting I do, I look for ~0.30 MOA and an ES of 20 or less (usually end up in the ~15-20 ES range for 10+ shots). Good enough for the shooting I do, but I probably wouldn't beat Alex Wheeler at a 1,000 yard benchrest comp.
6-7 years ago I couldn’t develop a load like I do now for just the reasons you laid out.
My process in a nutshell. (With very controlled loading process)
3 or 5 shot strings in powder charges .2, .3, or .5 increments depending on the round.
I don’t shoot at paper. I just it as a training exercise. I shoot at about 950 yds with magneto attached. As my charge increases, so does my velocity and wind holds. So I use it as training to spot my impacts and correct.
I then look for adjacent charge increments that have a consistent ES. 2 charges is ok, but 3 is better.
I pick the node I want and then load in the middle. If it’s 32.0 32.2 I load 32.1. If it’s 32.0 32.2 32.4 32.6, I load 32.3.
Then I load that with 3 shot groups with bullet jumped from .005 to .038 in .003 increments.
I use group sizes to pick the seating depth node and will load on long side (for erosion purposes).
Then I will do one more test with that seating depth with groups and chrono with 3 shit groups in .1 powder increments from .5 either size of the powder charge I picked earlier.
Then it’s done and I load
3 or 5 shot strings in powder charges .2, .3, or .5 increments depending on the round.
I don’t shoot at paper. I just it as a training exercise. I shoot at about 950 yds with magneto attached. As my charge increases, so does my velocity and wind holds. So I use it as training to spot my impacts and correct.
I then look for adjacent charge increments that have a consistent ES. 2 charges is ok, but 3 is better.
I pick the node I want and then load in the middle. If it’s 32.0 32.2 I load 32.1. If it’s 32.0 32.2 32.4 32.6, I load 32.3.
Then I load that with 3 shot groups with bullet jumped from .005 to .038 in .003 increments.
I use group sizes to pick the seating depth node and will load on long side (for erosion purposes).
Then I will do one more test with that seating depth with groups and chrono with 3 shit groups in .1 powder increments from .5 either size of the powder charge I picked earlier.
Then it’s done and I load
WHoa whoa whoa....6GT?...I thought we were talking normal rounds here
You go from science to witchcrafty with those. Also, didn't know Varget would be a good powder in those.
I think fairly well. I weight sort the brass and bullets and match bullet bearing length as best I can with Hornady bullets. They aren’t very tight even within the same batch. Talked to their tech support and they allege those tolerances are proprietary so who knows. Neck size for .002 neck tension. Uniform and deburr the primer pocket/hole. Measure powder a shot at a time with an RCBS Chargemaster Lite checking for drift every five rounds or so with a pharmaceutical scale weight set. Seat using a Redding Competition seating die. I measure/calculate OAL with the Hornady Comparator Bushing on a digital caliper not to the variable tip of the ELD composite and finally check concentricity with the Hornady Concentricity checker. I’ve always been skeptical of the 3 and 5 shot sample sizes I read so much about as true statistical analysis calculates a much larger sample size for the total population of shots we are trying to bound. Ledzep is right on for the analysis but Dthomas makes an irrefutable observation concerning barrel burn out. There has to be some middle ground to get to a predictable load. Think I’ll try looking for the flat spot that has been suggested.
Recently I have been using larger samples determining MV/SD/ES when doing load development as well as adjusting LTO etc. As other posters have stated, I have found that with a 5 round sample, SD may be low (under 5). Shoot 10 rounds and it usually climbs to 7-10, 25 rounds... perhaps 10-13 or maybe 15. What I haven't been able to isolate, however, is that , when I find a very consistent load, I may have one round that , for whatever reason, the velocity may be 25-35 fps different than the other 12 or 20. Case prep is identical on all, charges weighed to 0.1 grain. Interesting graph/table.
I've put some effort into finding this one out and haven't yet. I've gone so far as to track pieces of brass to their MV, but even keeping the same pieces of labeled brass over several strings, the velocities that jump around change cases from string to string.
Ya. Point was I don’t pay attention to groups at that stage. Also, when doing groups I only shoot 100yds. Haven’t seen a need to check at longer distances and they always stay we’ll below .5 moa.
Things like powder/primer/brass prep control your sd/es.
Seating depth/neck tension control group size
