Why are my load Standard Deviation numbers bad?

I have a pet load for 2 of my .223 Rem rifles that shoot true .75MOA all day, every day. I'm happy with it but....

The SD numbers are 27.4 over 12 shots, and 26.2 over 20 rounds. 2 different rifles. One rifle a 24" and the other a 20", both 1:9 twist. I'm happy with the load, it seems to work well but WHY are these numbers so bad? What can I do to lower them and get more consistent velocities? I use the same equipment and techniques that produce a SD of 9 in my favorite 270 load.

I'm using a RCBS Chargemaster to drop charges, and have check weights and use them. Scale seems accurate enough to a grain or 2 of powder per .1g.

Different powder? I'm using 24.5 Varget in front of a 69 Sierra. If I understand SD, this means my velocities are not consistent at all.

Any ideas?

have you tried moving a bit up or down to see if you find a load that your gun likes more...

Yes, I originally found this load by doing.3 grain increments with 5 shot groups, selecting for accuracy and group size.

Is the .223 bullet not crimped and the .270 crimped?

Neck tension is probably the biggest factor which ties to consistent brass. All the brass the same? Trimmed the same.
I would prep 20 pieces as close as possible also using a busshing or a Lee collet die can help maintain good neck but not totally nessasary if your expander is smooth.

If it shoots that good I wouldn't care. The paper doesn't lie.

And that is what really matters. I can understand solving this puzzle because it gives you confidence in your next shot. But in the end if you are getting the accuracy you want and expect at the ranges you require then you can relegate the SD spreads to a lesser priority in your load development.

I went down this same rabbit hole back in my old competition days. I would take a perfectly fine shooting load and totally screw it up by constantly changing darned near everything I could to get the SD into low single digits. I finally learned that single digit SD is not necessarily a requirement to accuracy shooting - but trigger time is. It took a lot of lead and powder for me to learn it though. Check out this wikipedia link so you are sure of what you are trying to solve, http://en.wikipedia.org/wiki/Standard_deviation

So then, the list of things to try is long, and in your case some are limiting (such as COAL), but here are a few things you can try to tighten that SD.
1. Cartridge OAL
2. Primer
3. De-burring the primer flash hole
4. Case neck brass thickness uniforming
5. Uniform brass neck hardness (annealing)
6. Indexing brass, but may be somewhat impractical in a semi-auto
7. Weighing brass
8. Only shooting development loads on same temp/density altitude days
9. Changing lots of powders
10. Changing powder
11. Add more to the list that I can't think of right off the top of my head

I didn't go back to my notes, and these were not in any particular order, but you can start to see the level of absurdity I was going towards that never materialized in a better score or more silhouettes hitting the dirt. I reasoned that the proper way of going about this is to completely eliminate every variable YOU can eliminate, and then only modify those things that you cannot. I even went as far as developing loads through one single brass case. And in the end, all I did was get tired of it. I got my SD into single digits, but the accuracy wasn't really any (or much) better than it already was. I chased my tail right into the hole of diminishing returns.

My first suggestion is to put the bullet in front of the powder... that might make a world of difference! Problem solved

Thanks for the comments and knowledge. Solving these kind of puzzles is fun to me, although I do understand the load does work, and works well across several rifles! not just mine.

From the Wiki link: a high standard deviation indicates that the data points are spread out over a large range of values. Yep, I got that in this load. I see total velocity spreads of 96 FPS across 20 rounds. There is a problem, and I think it will shoot more consistently if I can find the cause. And as we all know, consistency is accuracy.

I'm using Lapua brass, trimmed using a Redding trim die, one set of brass had just been trimmed, and the other set due for a trim this time across the bench. I did notice while trimming that this Lapua brass has some variation in neck thickness, but not much. CCI BR4 primer.

I neck size using a Lee Collet die, set where the press just breaks over for consistency in this step. I'll admit, I think this is where I will find my solution. Neck Tension. maybe need to modify the mandrel slightly to increase neck tension. The 270 rounds did offer more resistance during the seating operation. Would consistent light seating tension cause this? The bullets are not loose.

And I seat with a Forster Seater die, no crimp. I'm doing everything I know to do to produce consistent ammo. BTW, these are Rem 700 bolt rifles.

I do understand this is a rabbit hole, and this is my first attempt to fix this. I've got a couple new ideas, thanks to you guys.

problem solved!

I would try the collet die setup the way the instructions show (not camming over) and see if that helps

Annealing if you have not would make a difference as well

Inconsistent neck tension seems most likely.
TC

Rick you are using good equipment, what are the inside of the neck measurements on sized brass? Are they consistent? If not we have a sizing issue.

96 fps is a big swing and a SD of 24 like you and others have seen might cause some accuracy loss at 100 yards it will be worse the further out you get.

Are you just dropping to your charge weight with the chargemaster or drop a few tenths low then trickle in the remaining? The smaller the case volume of a cartridge the more critical this becomes.

I know it can be frustrating but try eliminating one thing at a time.

I think we do have a sizing issue. The mandrel in the Lee Collet die is .220, and the case mouths range from.220 to .224. Hard to measure accurately with a caliper, but I can see that there are inconsistencies in diameter within individual cases by .001, and a few .002.

Maybe I should stop camming it over lightly, as it doesn't seem to be putting enough pressure on the sizing collet. It's tempting to blame Richard Lee, but I won't. I did pop the alum retaining cap out early on putting too much pressure on the handle. Had to replace the cap.
I wish a Lee collet would size without decapping. :-(

I let the Chargemaster drop the load, and then wait for the weight after the count before I touch the scale. I do not tricle the last few grains.

I will modify my neck sizing technique and see if my numbers change. Should I just go to a Redding S type die?

96 FPS is a pretty big swing in 20 shots. I think you're already on the best thing to try first, re-adjusting the collet die like you said earlier.

Just a thought, have you shot across a different chrono or longer ranges by chance to verify that is what you are really getting? I know that large of a swing would also show up in an unknown hit from a called shot break. JBM sez that 96 fps is only .1MOA elevation difference at 100 yds, so that's hardle noticible but like kmon1 wrote it sure would at longer distances.

PS - good equipment.

Rick you could get one of the universal decapping dies and decap before sizing, lots of folks do that so when cleaning the cases prior to sizing, they can get the primer pockets cleaned as well, especially those using SS media and sonic cleaners.

The collet dies do a good job but remember it takes 25lbs of force applied so with the press camming over you might not be getting the consistency. Try not camming it over on some cases and see if that doesn't get those measurements more consistent.

Thanks guys! I think I have a plan. I appreciate all the help! I will report on progress. I can't be the first to need this info.

Consistent necks after sizing is what I would look at first. The second is primers.

For a swing that large sizing is where I would look the hardest, I would bet you have something going on with your necks. Size 10-20 and then start measuring for consistency. Also, in my experience lessen neck tension to help. I rarely use more than .002 neck tension...not saying there aren't loads that will shoot tighter tensions I just normally don't search there.

I use primers to fine tune things...just as an example, I recently increased the speed of one of my loads and it appears to be shooting pretty good, yesterday I ran it on a chrono Fed match versus Win and the difference was SD14 and SD18 respectively.

For those that never shoot over 3-400 yards SD and ES isn't important but if you stretch one out further you will see a difference and the load you have right now no matter how good it shoots at 100 will be crap at 600+ yards. Good luck...fwiw, I'm not overly happy with a SD of 14. I like to have the SD under 10...the awesome bench gun that I have SD is 2-3.

With a sample size that large can you post a histogram of the data? If you can can you post the raw number?

I normally shoot at 100 yards, and occasionally shoot longer distances, so it really doesn't matter, except that it does, if you want to create premium ammunition. I already loaded a few to take to the range this week using the Collet die, and consistent pressure on the handle. I can't measure that force, which is why I was trying to cam the die. I think that is the problem. I can see how my method could very well cause the problem.

I know the chrono will drive you crazy chasing unseen problems, but I'll take it over guessing velocity from a loading manual. Cleric, I will PM you the data, it makes an ugly post.

Well fellow THFrs, modifying my method on the collet die didn't change a thing. How disappointing! I'm gonna check my chargemaster against a beam scale next. I tried to figure out the Redding bushing die sizing info and it just gave me a headache. Maybe I should micro adjust my FL sizing die and let the expander ball take care of it. ??? Or maybe dump Varget. CRAP. LOL

Series 5 Shots: 20
Min 2768 Max 2858
Avg 2803 S-D 21.5
ES 90

Beautiful range day here in DFW. Took a Korean era Garand with me, so it's been a fun day!

Redding bushing dies sizing is easy at the step your at, when you don't have compenents is when it is hard. Take on of your loaded shells and measure outside of the neck...subtract .001 or .002 and that is the neck bushing you need.

Your charge consistency could definitely be an issue but I really doubt it would cause that big of a swing. I'm still leaning toward necks.

Something I just thought of....are you crunching powder with those loads? I ask because if you are crunching powder it could be pushing the bullets out over nite and causing you to have significant swings in OAL which could cause it too. Just a thought...

I'd try another primer, but you'll basically be back to square one with load development.

Thanks for the suggestions gentlemen! I developed this load using CCI-41 primers. Switched to CCI-BR and the groups got noticeably smaller. I did not do another development phase when I switched. It shot good, and then great! No Chrono at that time. Seemed to help, got tighter groups by prob 15%.

I think I'll look for a new neck sizing die. I was noticing lots of variation in my outside case neck dimensions during this analysis. The Lee collet fingers (4 of them), could cause this, Lee even suggests you size twice and turn the case 90 degrees in between press strokes. I tried that suggestion with the last 20.

I'm going to go away for awhile and work on this. If I find it, I will let you all know. Thanks for your time and knowledge! She still shoots great!

Not to change the subject on tension, but have you checked concentricity it of your reloads? You may not be entering the lands squarely on all your loads. I try to stay under .003

Redding S neck die made a marginally better difference. Did not solve the velocity spread. I'll keep looking, the groups were better with the Redding die. :-)

It is possible to have a load that yields inconsistent velocity yet groups well, but what you may find is that if you extend the range you're testing these loads at then more vertical spread will show. It could be that Varget just isn't going to yield consistent velocity and you might try something like Re15, H4895, or something else.

What can happen is that you hit a node where the bullet is exiting as the muzzle is approaching the top of its excursion. The theory is that if the bullet is moving just a tad slow it will exit a little later with the muzzle pointed just a tiny bit further upward which makes up for the increase in drop, due to increased flight time, due in turn to the lower velocity. This phenomenon can work for you to make up for velocity variations, but it will only work well within a limited range of distances.

Testing at 200 or 300 yards might dissuade you from thinking the accuracy really IS good, which in turn can make the decision to switch powders a little easier.

Get a balance beam and check you weights, The 270 is a larger case and slight variations in powder do not have the impact on charge like the smaller cased 223. Do your modified ladder test again at 200/300 on a calm day. Possibly change powder. The 80 grain smk's like R-15. How many rounds fired, think anneal.

Have you verified your chronograph is giving you accurate data.

Thanks for the responses. It may be the chronograph, and admittedly the 270 small SD number was a 5 shot sample. I had a SD=8 for the first 5 rounds of my last 223 string.

I think a new powder is next. and I'll dispense and balance beam weigh each load. I do let the Chargemaster warm up, out of drafts...etc. If the elementary math I did is on track, I would need a several tenths swing in charge weights to cause the velocity spread I'm seeing. I have the tools to check and I will.

It would seem to me that during testing, velocities should be consistent for each load, no matter downrange results. So if true, it shouldn't take many rounds to find a more consistent powder. Any experience otherwise? I do have some RL-15, 4064, 4198, BLC(2) and TAC on the shelf.

I didn't read quite all of this, but most. Something I didn't see anyone mention was clean barrel or fouled. I would assume that could cause a large range in the first few shots on a clean barel as opposed to a fouled one.

Could be the chonograph. May want to shoot through a friend's to see how it reads.
Could shoot a known good sd batch through your's, say Fed GMM 168's or something.

Ooops not 308, but get some match ammo and check the sd with your chrono.

I was using 25.4 gr of varget over SGK 65's in 223 and getting good sd, not spectacular,
+- less than 15fps so about 30.