.308 load test

Status
Not open for further replies.
~2,000.

I'm joking Bart. It's a chrome lined FN SPR, claimed to be good for 10,000 rounds

Now on the bulk bullets, I'm serious. If I need at least 20 rounds to test each increment in load development, that's going to take a lot of bullets
 
Some loads will rule them selves out without needing that many rounds fired.

I don't know how the F Class and High Power guys do it, maybe they fire 20 round groups, but the Benchrest guys shoot aggregates (Five 5 shot groups for 25 total rounds). If it is obvious the load isn't cutting it before then, there is no need to continue. The goal is to shrink the aggregate. Sure, we try to do that by shooting tiny groups, which is done by not making mistakes, but in the end it is the aggregate that counts. In other words, 25 rounds fired under as is conditions when it is your turn on the firing line with no mulligans or excuses, 5 shots per group for five groups, each group fired separately, one at a time as your turn comes back around.

Bottom line is, results are determined by many rounds fired. So both you and your equipment have to be consistent.

It is difficult to shoot small groups, much more difficult to shoot small aggregates, and even more difficult to shot multiple small aggregates (Grand Aggregate) over a weekend.

It is why the big dogs always seem to be at the top when the dust settles, as they make fewer mistakes than us mere mortals. And all it takes is a handful or less mistakes to take you out of the running.
 
The number of shots each test group has is determined by the shooter's opinions as to what is enough; statistically speaking.

I've shot enough 5-shot strings with the same load to see a 4 to 5 times dimensional spread across all of them. And the composite of all groups fired is larger than any single group. So have many others then the pick the load that shot the smallest group and claim that for its accuracy.

That 10,000 round barrel life claim for chrome lined barrels is popular. Never seen any accuracy standard used to get that number. It may be 2 to 3 inches at 100 yards.
 
I've shot enough 5-shot strings with the same load to see a 4 to 5 times dimensional spread across all of them. And the composite of all groups fired is larger than any single group.
Always is, no puzzle there.
So have many others then the pick the load that shot the smallest group and claim that for its accuracy.
Some people do this of course. Lots actually, as we all like to brag about a small group we shot. That doesn't mean that person is ignorant to the fact that more shots will always add up to a bigger group, or more groups will give them a bigger average. :)
 
Exactly.

I'm not ignorant of statistical analysis. The bigger the population, the bigger the spread. I think you can see that with the results of my test on the original post. It also reinforces what is repeated often here: 3 round groups don't tell you much, 5 round groups tell you more, but may still not be statistically significant.

I do know that me, my rifle and my loads are capable of some pretty impressive accuracy. I'm struggling to find that at the moment and maybe blaming it on the primer isn't the root cause. I won't know until I can get some more. In the mean time it feels as though I'm stumbling around in the dark looking for the light switch
 
I've always had my best accuracy loads with Remington primers. That includes Federal and CCI match primers. This is with over 20 different rifles and chamberings and heaven knows how many different powders and bullets. Sometimes other primers perform well enough for my purposes. But if I'm really wanting bug holes I load regular Remington large rifle primers and small rifle Remington primers. I think 9-1/2 and 7-1/2 respectively. Your mileage may indeed vary, but I always start out with those two primers first! I'll usuall swap to a WW or Federal primer to see what if any difference it makes. Sometimes quite a bit, others, not so much! I've got a couple thousand Fed Match primers I'd trade for Remingtons, or maybe not. Someday I might run into a load where they shoot better than 9-1/2s? Haven't yet, but you never can tell!

BP
 
I treat my load development as step one of many. Typically, I load up 5 rounds at 5 coarse powder steps (starting load will be -10% of published max load and I work up in 2% increments). I'll shoot 5 round groups and measure the groups and the velocity, focusing on the ES. Obviously we look for the smallest groups but I also look at vertical and horizontal stringing. I'll rank the 5 groups into best grouping and best ES. From there, I continue further work looking at how the seating depth / bullet jump to lands affects the group size and ES. From there, I again rank the best smallest group and lowest ES. I choose the best of either two and make up a large batch (maybe 50 rounds each) and continue shooting to see if these rounds are truly consistent or whether it was me just having a good day. As an engineer who dabbles with Design of Experiments one learns that the more repeat tests you perform the more accurate your model. This shouldn't be any different for shooting.
 
I'm a manufacturing engineer by education and I was just thinking about DoE's this morning in relation to load development. Ironic you should mention that
 
WelshShooter comments:
As an engineer who dabbles with Design of Experiments one learns that the more repeat tests you perform the more accurate your model. This shouldn't be any different for shooting.
Even those shooting craps know their dice will average 12 over many rolls and rarely in the 3 or 4 range as well as the 10 or 11 range. The "groups" they roll are exactly like the "groups" we shoot; and the odds of getting 3's and 4's (smallest groups) are exactly the same as 10's and 11's (largest groups). Group shooter's have a range from zero to some value; those shooting craps have a range from 2 to 12. Group shooters have many variables of different bandwidths; crap shooters have only 2.

If one doesn't think group shooting's the same as shooting craps, look up some benchrest results and check out the size of the top 10% competitor's twenty to thirty 5-shot groups over a couple of days shooting the same rifle and ammo. Here's a good place to start:

http://azbrs.com/past-match-results/
 
I'm a manufacturing engineer by education and I was just thinking about DoE's this morning in relation to load development. Ironic you should mention that

I have often thought about it too, but realistically you'll end up with low amount of factors (powder weight, seating depth, bullet weight, primer brand) and high amount of levels (i.e. more than 2) which is kind of the reverse of what you want in a DoE.

Bart, I'm afraid I'm not familiar with the term "aggregates", a quick google doesn't help me much either. Would you mind explaining in layman terms what this means please?
 
If one doesn't think group shooting's the same as shooting craps, look up some benchrest results and check out the size of the top 10% competitor's twenty to thirty 5-shot groups over a couple of days shooting the same rifle and ammo.
Well, craps is pure luck, while group shooting has skill involved. True, it is very hard to shoot the tiniest group possible every time, so group sizes vary from group to group because of changing wind conditions, the fact that even under perfect conditions every single round will not go through the same hole, and of course, plain old human error.

But it's not a crap shoot, any more than High Power or F Class is. All are skilled people with extremely accurate rifles, shooting some fantastic groups, aggregates, and scores. Some 5 shots at a time, some more, but it all adds up to the same thing, great shooters with world class weapons doing amazing things.

I never understood the Benchrest shooters who scoffed at the High Power shooters, or the High Power shooters who scoffed at the Benchrest shooters. Pretty silly crap to me.

I'm not familiar with the term "aggregates"
In Benchrest they shoot "aggregates", which are simple five five shot groups averaged together. One might shoot a .2347 "aggregate, which is good and gives you a chance of placing well, or you might shoot something like a .4513, which is not so good, unless the conditions are terrible and everyone is having trouble.

Five groups:
.247
.351
.183
.246
.283

For a .2620 Aggregate

You groaned after the .381, but thee .183 saved your bacon. :)
 
I have often thought about it too, but realistically you'll end up with low amount of factors (powder weight, seating depth, bullet weight, primer brand) and high amount of levels (i.e. more than 2) which is kind of the reverse of what you want in a DoE.

The biggest problem in conducting a statistically based DoE in load development is creating a control for the "spurious" variable, namely the shooter. Bart often mentions bench rest guys using a free recoiling machine rest. I'm not going that far.

I guess I'll stick to a quasi method of keeping as many variables constant as I can (including me) and continue on with the learning process.

Time to order more supplies........

(p.s. if you guys see any Federal 210M primers show up, send a shout my way)
 
I frequently mention benchrest "free recoil" rifles resting on front and rear supports for group shooting.

Not machine rests with the rifle clamped in its carriage that slides back from recoil.
 
Bag shooting vs Rail guns.

Most people shot the "bag" guns free recoil, but not all.

Rail guns took the gun handling part of human error out of, and turned it into a wind reading contest, assuming each gun/load was capable.

Shooting a bag gun against the rail guns was challenging. :)
 
Ah, thanks Walkalong - makes perfect sense. So it's literally just an average of averages?

The biggest problem in conducting a statistically based DoE in load development is creating a control for the "spurious" variable, namely the shooter.

In that case you'd have to have multiple shooters repeat the same experiment and set them as "blocks" to remove shooter-to-shooter inconsistency :p
 
an average of averages?
An average of five groups. The Grand Aggregate is an average of all groups shot that match (Week or weekend).

An average of groups shot.
 
Some agg's are for eight 10-shot groups at 100 and 200 yards.

Sierra Bullets used to use a 1/4 MOA average at 100 yards (California plant) and 200 yards (current Missouri plant) for their match bullets with 10-shot test groups. They now require all match bullets shoot inside 1/2 MOA at 200 yards. Effectively the same thing.

Hornady match bullets have to shoot 10-shots inside .950 inch at 200 yards.

Both shoot many test groups during production runs.

How credible is one group to represent real accuracy level?

Single group size times max/min multiplier equals approximate size limits of 19 out of 20 groups fired with different numbers of shots per group.

Shots/...... Multipliers
Group ____Max ____ Min
3 _______ 2.45 ____ .40
5 _______ 1.53 ____ .67
10 ______ 1.27 ____ .81
20 ______ 1.12 ____ .89
30 ______ 1.09 ____ .92
50 ______ 1.06 ____ .95

Examples:

If first 5-shot group is 1 inch extreme spread, 19 out of 20 will be from 1.53 inch to .67 inch.

If first 3-shot group is 1 inch, 19 of 20 will be from 2.45 inch and .40 inch.

Here's a good learning tool about groups, aggregates and the statistical reality of shooting groups:

http://azbrs.com/past-match-results/

Pick any event, such as http://azbrs.com/downloads/results/2016-Cactus_Classic-Heavy_Varmint_100yd.pdf then check out the top 20 competitors' group sizes. Then compare them to any other day's Heavy Varmint 100 yard scores and note the group sizes are again, all different.

Had a long talk with a statistician who was also a rifle shooter and he says that all fired groups fall into these categories;

40% of them are about average size.
30% are a little bigger or a little smaller
20% are a lot bigger or smaller
10% are either huge or tiny.

Consider the record benchrest 5-shot group at 100 yards being .0077" shooting a .30-caliber cartridge listed as a 30 PPC,. Its holder has no other records or match wins to his name. It was the tiniest group anyone has ever shot in competition at 100 yards. His others must be too large to talk about.
 
Last edited:
I'm headed to the range with 10 each rounds, CCI 200, WLR and CCI BR2 just to see if I can decern any appreciable difference in grouping or chrono data when isolating for the primer. I'm sure this has been done countless times by many others, but not by me. Hope to learn something. Still looking for Fed 210M's
 
Update:

The frustration has been building with my shooting due to inconsistency in my results. Good loads that don't shoot accurately. Weird flyers. Load tests that I can't duplicate. It's gotten to the point that I'm thinking something is wrong with my rifle and I'm doubting my own skills.

I don't know why I didn't do this sooner. I bought a box of Federal GMM 168g and proceeded to put 10 in a row just inside an inch. Shot the 2nd 10 and got the same results.

So now I believe there's something out of whack in my reloading process. Something common that translates across multiple component combinations. Powder metering variation? Brass prep? Runout? Hell if I know but I'm feeling better about me and my rifle and crappy about my manufacturing process. Any ideas where I might look first would be appreciated.
 
I've heard many say, runout is often an issue. Try to find a smooth flat surface and roll your loads at them. If you can see the tip wobble there's probably too much.
 
Heh, this is the downside to handloading - finding the problem. I did much the same as you this weekend. Before I had a chrono I used to perform load development and settle on the powder weight based on accuracy. For my .308 with 167gr Scenar I settled just over the halfway point between the starting and max load. I bought a chrono last year so I set out to continue the load development. I repeated my "known good load" and continued to work it back up to maximum. The results shocked me. I started with my known good load and have decent extreme spread in velocity (<20fps) and a group size around 0.750" (edge to edge). After that, the new loads were ridiculous; they were all greater than 1.5" at 100m when measuring edge to edge for 5 shot groups. I then proceeded to shoot a 10 round group with known good load (which I made up at the same time as the test loads) and all were within an 0.7" group. This test showed me that my original load was more accurate with a lower velocity spread compared with the continued load development I had carried out. Good news because I don't have to pull the 40 odd rounds I have left over but bad news because I didn't get to increase the velocity of my rounds and made me question what the heck happened. All brass prep were identical, all loads were measured identically and all seating done identically. Same batch of brass, same batch of bullets, same batch of primers, same tub of powder, all loaded at the same time.

Sometimes you just don't get good results. Over the past 5-6 years I've been handloading I've had very good results and not had many bad days. It's worthwhile having bad days so that you can re-assess what you're doing at the bench and to correlate the importance of prep to the output. For example, you might not have trimmed all the brass to the same length and seated your bullets quite long. This would result in a lower bearing surface of the brass gripping the bullet. Could this have an effect on accuracy? Who knows, but you learn to be more uniform in your preparation.
 
Gscotty,

I'm using a Dillon 650. For .308:

Most all of my brass has been once fired by me from new or purchased new, unfired. I have 500 once fired LC brass as the exception but I have not introduced it into the rotation yet.

Brass prep:
Tumble for 45 min
Decap and resize using 650, Dillon lube and Dillion dies to bump shoulder back .003 min from fired dimension, verified with Hornady headspace gauge/calipers. Checking every 10-20 case for varification.
Trim using WFT to 2.005, +/- .002 verified every 10-20 cases using calipers
Chamfer, debur, uniform and clean pocket using Lyman case prep station
Tumble 45 min to clean lube
Case prep complete

Load:
Using Dillion 650 with Dillion dies in all stations, including powder check stations, except Redding micrometer seating die and no crimp die.

If working up a new load with multiple powder charges I will use the Dillion like a single stage, seating primers individually, throwing a starting charge then trickling up to the desired weight with a Frankfort trickle charger and using my Frankfort digital scale.

I'll set up seating die using a bullet in a case that I've cut notches in the neck. Once adjustments are made to achieve the desired OAL determined with calipers I'll seat the first round and check again. Usually this will net an OAL that's .010 longer, which I attribute to the difference in neck tension between the notched set up case and the actual resized case. I verify the desired OAL again and proceed with the rest, checking OAL at 10-20 round intervals

If I'm running a batch using a set load and OAL I'll set up powder throw, checking ~10 throws until I get +/- 0.1 grain consistency before proceeding. Sometimes with powders like 4046 that never happens so I don't use them very often. One I get a consistent charge, I'll proceed checking, checking charge weight every 10-20 rounds, as well as watching the powder check indicator which will give you a rough idea if there's variation in the powder volume (not necessarily an accurate indicator, depending on powder type, as some settle in the case more than others)

Last check is a visual check of every case to determine primer has been seated below flush.

That's it. Too much info? Not enough? Let me know if I'm missing something please.
 
Status
Not open for further replies.
Back
Top