I'm having a tough time with today's shooting. I had loaded up a sample of 308s with 175gr SMK. My standard load. Trickle charged to about 1/32" difference on my beam scale, which is roughly 1 and a half literal kernels of powder. Generally I'm +/- 1 kernel. Historically this results in <5fps SD, the other day multiple strings came to <3fps/SD.

Had some cases that were annealed (my first attempt), which shot excellent. Around <½"@100, basically 4 shots in a .50 hole and one next to it. Both groups noticeably favored vertical spread. The chrono was all over the place. I was getting easily 70fps ES with a SD of 30 on some, best SD of the day was 15fps... absolutely heinous. For 15-20fps SD I might as well just throw charges. The control group I shot was more like ¾-1". Which is fairly large, usually it will do more like ?". Maybe it the nut behind the trigger was loose on that one, but it would be loose on all rounds fired.

The key differences between this lot and previous are that I annealed the case neck/shoulder, and sized using a Redding S. I did load a control test, loaded at the same time, with 5 cases I pulled out of the bin. All the same headstamp and load procedure otherwise. The previously sized cases were around a 2thou bump, but I decided to try a 4thou bump of the shoulder with the type S to ensure easy chambering, as some are a little tight with 2thou on a RCBS die.

Bullet seating was very consistent in both length to ogive, and force used on the handle. Naturally with the control, seating force was rather inconsistent. As noted above, despite the poor velocity numbers, the annealed groups were a massive improvement in group size.

My thought is that the difference in sizing, possibly excessive sizing, coupled with the inconsistent annealing job, took too inconsistent an energy to balloon the case in the chamber. While the annealing was "good enough" to get the job done with regards to getting the bullet in the throat significantly better, it was wild on the ability of the brass to obturate properly.

Maybe I'm just a dipshit and screwed up charging my cases, but I'd like to think that's not the problem. I have my scale zerod to edge-on-point, and have been doing this with excellent results for some time now. As I said, I'm easily charging them to nominal weight +/- 1 kernel, maybe two kernels if they're small. Even if the zero was off, in previous testing I have quite a range I can charge at and be in the zone for good accuracy/harmonics, but the SD/ES would still be good to go.

I'm very flustered at this, but also amazed at the accuracy potential of myself and equipment. Effectively ragged holes at 100yds is nothing to scoff at. One of the groups of the "annealed" bunch literally put 3 shots in the same hole. Or they missed a 10" target... I'm going with the former.

Would love to hear yall's thoughts.

First question, what chronograph are you using, and how far away from the muzzle is it? I've got some pretty wild chrono readings before, turned out it was interference from muzzle blast.

Other than that, it sounds like your method was very good before, getting single digit SD's and all that, so I would probably agree about inconsistent annealing. You said you could feel a considerable difference in bullet seating force between rounds, sounds like maybe inconsistent annealing resulting in large variations in neck tension.

This was on a labradar. It reads close enough to my magnetospeed that I trust them both.

Inconsistent seating force is with unmolested cases I already had sized.

That force being measured with my RightArm mk1 torque-hand. Not the most accurate instrument, but it's been around, ya know...

The force required to seat on the annealed cases was very consistent, at least in comparison. They all went in smooth as silk. While they were "Inconsistent" as far as annealing goes, the process did happen. To throw out some figurative numbers, say it took 20lbs +/- 3lb to seat, whereas with the normal ones, it might be 30lbs +/- 10lbs. Now that I figured out annealing, the next batch might be 20lb +/- 1lb. Look at those numbers relatively, not exactly.

Ya know, hell, I was excited about getting into single-digit SD a while ago, then shooting my last batch with multiple strings (3x 5 shots, of a 50 box, rest shot on paper a different day, with good results) coming in at <5fps SD, two at <3fps, had my juices flowing. It must take more than dumb luck to get those kind of numbers and results on paper.

If I remember correctly, I had shot the rest of that box on paper at 400yds some time ago. I forget the details, but it was certainly sub-moa at 400, discounting the wind. Not bad all things considered... I know I cleaned it with a 200-5x, maybe it was a 7x. Club match, don't really care.

On-target accuracy >> chrono data

OP, the reduction in neck tension is most likely what opened the SD up..try  a touch more neck tension...I went thru similar when I changed my annealing program, accuracy improved but SD and ES jumped up...adding a bit more tension brought them back closer to what they were.....

Guess I'll have to go smaller.

I went with a .335 bush, loaded rounds measure .338 @ neck. Raw brass, no turning. Measured same with factory (175gr fgmm, since I use that brass and bullet). Pretty consistent, seems my digital calipers don't want to pick up much variation when checking different points. They will try to read 5-tenthou, so I'd say being +/- 5tenthou neck OD isn't bad...

I never understood how neck tension can be measured simply with a dimensional measurement. 3thou of soft brass is no where near the hold of 1thou hard brass.

I did think the bullets went in a little "too easy"...

Think setting a light crimp might help? I don't have a FCD, but I could setup the basic-bitch RCBS seater to crimp. I'd just pull the stem out of it after seating on the forster-micrometer. I know a touch more crimp can help get a more consistent burn in straightwalls.

ETA: I'm very excited at the prospect of getting the annealing down, and I think the accuracy improvement will be worth the potential loss of case life from having to size the ~400 cases i have ready to load. Or maybe my cheap ass should just invest in a neck die, but I don't really think it'll be worth it in the long run, as I have no plans to neck-only size being a field rifle. IIRC, you, AKsnowrider, body and neck size separately, correct? Although I believe you are working on a different level of accuracy than I am. But then you wonder... how small can you make a <$1k factory rifle group? When I'm already pushing ?@100 (and previous loads will hold their MOA at 200). Can it do ¼"? Can *I* shoot ¼"? I think realistically I'm not capable of holding the rifle steadier than ¼, at least with a 15x scope. The error in having a target big enough to see at 15x is easily ?". Usually I actually fire when I'm settled so I can't see the target, a cross hair drawn with sharpie.

You also wonder, should I bother picking out that single kernel of powder that fell in extra while manually trickling? Is it worth the effort when I've repeatedly done less than 3fps SD? Is it realistic to even try for better?

I do love the look on the faces of the crotchety old guys and big spender whizbang gun dudes see my targets

Are you trimming your brass? And when you say your node is wide, how wide is it?

I do neck and body separate, but the results are the same, I prefer separate because it allows me to concentrate all my energy on each operation at once, and not have to worry about both at the same time...just makes it easier for me to maintain repeatability easier.....

And yes, a few things can really shrink a group size down, but from there on, it is ever more work for ever smaller improvements, at some point a person has to decide where that line is..Only you can answer if it is worth it..

Yes, brass trimmed to 2.005 on the Giraud. Usually it only grows a few thou, but I throw them all in anyway.

I think I need to stop using the Giraud for ultimate accuracy, it just can't do as well as a lathe. In previous testing it made no difference, but getting down below <½moa, it ought to make a difference. My last batch was still done with the Giraud, although most cases were only 1-2thou long.

Does trim length really matter, as long as it's consistent, and below max?

I've had no measurable change in group size over 39.6-40.3gr. POI will change slightly from the velocity of course. I was loading 39.7 for a while, but went up to a clean 40 for the couple extra fps (20" barrel problems). At the same time I noticed SD went down, I think because of better case fill. Not that ARCOMP has much trouble getting good case fill in .308.

Maybe I need to rework up and test again. It's been around 1krds since I did that testing. About 100rds ago I tested seating out 3thou longer and groups were heinous, so I'd wager throat erosion is still minimal. I can seat shorter, but groups open up. Currently using 2.835 OAL nominal, 2.227 on my ogive tool.

After annealing, size your brass 3-5 times in a row. Just raise then lower and repeat. But make sure it is the same with each piece of brass. This will help bring all the cases back to where they are even again. Then polish the inside neck with  some brass wool. Then be very careful on how you seat the bullet. You want to feel each one the same. If you notice one is tighter or looser, set it to the side to use as a fouler shot.

And your trim length should be as long as your chamber allows you to. The longer neck will always help align the bullet. Just make sure they are all the same length.

Lower enough to get the case clear, but not engage the expander ball?

Good way to check max case length? I figure uniformity is more important than a longer length, and it would seem difficult to get every case to trim+X, where with FC cases I can guarantee I'll be able to trim down to 2.005.

Have you shot for groups past 100 yards?

Usually at 200, but I was time limited last range trip and had other loads needed tested at 100yds.

You need to do a chamber cast. You trim to just under max chamber length of your gun.