Load Development and other topics- part 1

[Harrier]

Load Development to match M193 and wandering down paths along the way.

About three years ago, with the cost and supply of milsurp ammo uncertain, I began testing variations of 223 reloads with the goal of duplicating M193 performance for practice loads from my rifles.

As a velocity baseline, I used surplus XM193 and IMI M193, various lots of Lake City, South African DNL and Singapore HB milsurp. I shot them at 100 and 200 yds, from a 20" milsurp barrel, then my other rifles to get a performance baseline for each.

The TM-43-0001-27 Army Ammunition Data Sheets (1994) spec for 5.56 indicates (on page 10-5) a velocity of 3250@ 15' (not 78' as in M855 and other sources) Since I normally setup my chrono at 15' I didn't have to reverse calculate what that would be for my test. None of the surplus ammo tested that fast out of my 20" EMC 1:9 barrel. The averages were XM193=3194, IMI=3225, DNL=3189, LC=3214 and Singapore HB=3208 for an average of 3206 fps. I rounded that to 3200 fps as a velocity for my 20" to match and additional testing showed 3000 fps for 16" barrels. Since 5.56 is already at max pressure levels, I decided to not try and makeup the extra 50 fps.

I'm not going to mention the exact loads I used because they are worked up in my guns and many of them exhibited excessive pressure along the way in some cases, so to avoid potential misunderstandings or accidents, we'll leave the exact load research to the individual user.

When selecting 55gr FMJ projectiles to use, I only had a couple hundred IMI on hand but I did have a couple thousand recently acquired BulkBullets product on hand, so that's what I used. Looking at the TM the overall length of the M193 bullet is 0.76". Measured samples of the BulkBullet were 0.74" ...close enough.

My goals were to match Milsurp velocity and improve accuracy if possible, with cheaper reloaded components for practice. Most M193 Milsurp I have tested only delivers about 4"MOA so any improvement over that is welcome. Initial baseline tests were done several times over different days with the various milsurp rounds to get at least 20 samples each and verify performance in each rifle.

After each day's initial baseline with a rifle, the rest of each session was testing various loads of powder, primer and case combos. I started with 4 powders- BLC2, H-335, AA-2230 and AA-2460.
I quickly selected AA2230 because none of the others would get the velocity needed with published load data and also supply issues at the time- the fact I had 8# of 2230 also helped.

Load testing started in July 2015 and I thought I would be done in 2 or 3 range sessions... the most recent session was April 2018. This was much longer than I anticipated but it was/is fun to compare different formulas and I think I improved my shooting during that time too.

When I replenished the 8# of AA-2230 powder a year later, I did a lot comparison test. This definitely showed me why you need to rework loads with a new lot of powder and why I should buy 8# jugs or all same lot of my go-to powders in the future.

Here is my range setup. I'm very lucky the owner allows me to shoot here. There are only 3 or 4 others given permission to use his property. I usually spend all day out there, shooting and checking targets at my own pace... I couldn't do what I do at a public range.



All test loads were hand weighed and matched as far as primers, powder and case headstamps, however different sets of components were used over various range sessions. Some tests were comparing only one component change to see how that item affected results.

Loads are color coded with a sharpie . I've developed a system of dots, lines and colors that will handle a large number of loads without duplicating symbols. A few of each load are also marked on the side with the load specifics, in case the box tips over (which has happened on a couple occasions) or I get confused which is which (which has never happened... :think: :? ). It also helps later during primer and case review.

View attachment 2

During testing I kept detailed records in Excel and made adhoc comparisons of various component variables, such as a primer change, as I went. When at the bench I have a range sheet I record shot and chrono data for later correlation with my spreadsheet.



In addition to velocity, performance at 200 and 300 yards (max available) was also looked at. During testing I'm not as concerned with POI on the target (as long as I hit it) as I am with group size and vertical stringing. When i settle on a load, then I can dial it in.
We frequently have 5-10mph cross winds out here, so horizontal drift is normal and I look at it as more practice. My standard testing range is 200 yards (better info than 100 and easier to walk than 300) but several promising loads were also shot at 300. The typical thing I run into over and over is a great 3 or 4-shot group then one or two flyers to open up shot string. I admit I pull my share of em but some are just out there on their own... I'll just blame bullet quality or runout or sump'in.



As I narrowed the load to 2 or 3 charges. The highest charge (A) would consistently make the expected velocity, but sometimes went inexplicably higher and a noticeable number of pressure signs were observed in various guns... (cratered primers, head swipes, hard bolt lift, top-hats and even a few blown primers). These problems were mainly in guns with 223 chambers, but enough were noticed in guns with 556 chambers to suspect this load was on the edge of safe pressures and over pressure when certain conditions occurred. What were those conditions? primer used? seating depth? cases? Temperature? and... could I correct for that while maintaining desired velocity and this powder? Those were the paths I needed to explore before settling on a standard load for all rifles.

My #2 load (B) was .2gr less powder. It would make the desired velocity about 1/3 of the time out of the 20" and 85% of the time from a 16"... but not always in either... how come? was the question.

The 3rd charge in line (C) was another .2 lower and it would meet minimum velocity about 45% of the time in 16" but only 1 time out of 8 from the 20"- about 50fps under my spec... what is going on here? Probably barrel length robbing umph... so not much hope for this one

Additional analysis again showed the velocity of some groups much higher than others with the same basic load. There also seemed to be a lot of Hi/Lo velocity overlap in loads that should have had a distinct gap. Since the SD and ES was often similar between these anomalies, perhaps it was within normal statistical distributions. I ran various non- standard deviation tests and compared... +/-95% limits, RSD, MAD, ESD... Nothing stood out to explain the variances.

My standard prep procedures are fairly extensive and comprise about 14 steps. Test loads are all done by hand, so I'm fairly confident it's not something during load assembly. In spite of this, I considered that I may have marked some of the loads wrong so I pulled a few extras and they were of the correct weight.

My thoughts then centered on component differences ... primers, powder lots, cases, etc. I started looking through my records & comparing those.

I mainly use velocity and % delta as the primary factors in determining component differences.
First I sorted for only the 20" and 16" barrels, that left 104 shot strings to look at. I then separated the two powder lots and noticed that the older lot produced significantly higher velocities for the same powder charge- so I eliminated them from the analysis leaving 47 recent 5-shot groups, the majority in 16" barrels.


analysis phase continued in the thread... part 2

 

[Harrier]

Part 2

Next I looked at primers, since a lot of velocity difference can result from just a primer change.




I counted I had tested 7 different primers ... CCI 400 & 450, WLR, S&B, FC, R-P and Wolf.cu.




Some were mainly for data collection during a primer comparison test. For use in the AR I have settled on CCI-450, RP 7.5 and S&B as they may be available-




Overall I found less than 0.5% velocity difference between primers, where differences within a given charge was as high as 3.5%, so that doesn't quite explain it...

I looked at cases next. Although I tried to use the same set of HB cases throughout most testing, over the 3 years I had ended up using a large variety of headstamps. I counted 14 different cases tested overall, most of them during a couple of "brand A vs brand B" or "heavy vs light" cases tests.

At least for .223/5.56 there doesn't seem to be any weight distinction between military and commercial branding but there is a noticeable difference between heavy cases and normal to light weight cases... (I'll explain later why even that difference is not what it seems to be).

After trimming to 1.750", weighing and averaging 10-20 cases of each headstamp, I divided them into 3 general groups by using a +/- 2% spread from the median weight of 92 gr (decapped & trimmed) ...
The majority of cases fell in the 'Normal' middle-weight category. These ranged from 90 to 94 grains trimmed & unprimed (or 93 to 97 gr primed).

The next group were heavy cases. These were generally over 100gr but some were as light as 97 gr unprimed. Most of these were foreign brands, with both commercial and military headstamps.. DNL, PMP, IVI, Aguilla, FNM, RG, L2A2, TAA-10.

The last group were a few brands of light cases- under 90gr. These included HB, Hornady Match and LC from the 70's (LC got significantly heavier in the mid 80's onward).

Sorting the loads by velocity I noticed for each charge group the lighter cases tended to be at the lowest velocities and the heaviest cases were at the higher velocities. This would be an expected distribution due to case volume for the powder charge. Normal weight cases comprised the middle with a few intermixing with the other velocity groups at both ends of their range, probably due to something else like primer, seating depth or individual barrel shot in - now something was making some sense.

I did an additional round of testing with the three selected charge weights and a variety of weight selected heavy vs normal cases.

My current thought is to make up two batches and use a lighter charge in heavier cases and a slightly heavier charge in normal/lighter cases, hopefully keeping all velocities within the milspec range and given barrel length.

Following are the results, however there were a couple unexpected surprises in the data.
I'll construct a couple tables to help visualize the averages -
below, we are comparing charges A (heaviest), B (medium) and C (Lightest) with case weight Heavy-Normal-Light

16" barrel velocities
Cases > .... H ......... N ......... L
A .......... 3125 ..... [highlight=yellow]3102[/highlight] ..... 3087
B .......... [highlight=yellow]3119 [/highlight]..... 3036 ..... 3006
C .......... 3087 ..... 2982 ..... 3069
...........................3003 (w/o S&B)
20" barrel velocities
Cases > .... H ......... N ......... L
A .......... 3260 ..... 3227 ..... 3196
B .......... 3248 ..... 3175 ..... 3169
C .......... 3176 ..... 3165 ..... 3134

Analyzing charge A, it is one that always produces milsurp velocities from both 20" and 16" but also exhibits over pressure in some guns under certain conditions.
This would make a good practice load as velocities and POI should be the same as milsurp however when shot in a heavy case there are pressure signs and accuracy is terrible, averaging 3" MOA at 200 yds with several flyers.

When charge A is shot from a normal-weight case the groups improve to 2.4). I got several groups under 1"MOA and two from the 20" under 0.36"MOA at 100yd "... and had 2 groups under .75MOA at 200yds with this load. During the 100 rounds shot, there were a couple instances of Hi Pressure noted with a bolt gun and one set of cratered primers on an 16" AR, so this one might be eliminated for safety reasons after additional testing... time for some primer/seating analysis....

Finally when charge A is loaded in light cases the velocity barely makes my minimum spec from the 20" barrel (I'll give in on few fps) but easily exceeds the 3000 fps minimum from a 16", in addition all of the groups are right at 1" MOA and no pressure signs were noted, so this is a viable option when used with light cases only.

Moving on to B -the middle charge... From the above tables, the 20" tube will it meet velocity specs only when shot with heavy cases . Most groups were in the 1.4"MOA or less range which is more acceptable than milsurp and great for practice fodder.

In the 16" barrels, all case weights, load B has acceptable velocity and no pressure. In limited testing, even my Savage bolt 223 was able to shoot this without pressure signs (it normally shows pressure early). Precision varied from 1.3" to over 3" MOA so there is still some work to be done here.
However, in the 16" chart, Load B with heavy cases has a higher velocity than the mid-weight of charge A. These appear so similar they might be considered interchangeable, so for now this one should be put on the shelf for additional testing to make sure pressure signs are investigated further.

When considering charge C, none made the grade with the 20" barrel, falling short of minimum velocity by 25fps... its 1.1" groups could be a balancing consideration if needed.
The 16" length barrel still performs above minimum with the Heavy and Light cases but something happened with the middle group as velocity is almost 20 fps under the bar. I would have to eliminate half the readings to get it over 3000 fps so clearly something is affecting it. Case-Primer combos were LC-CCI450, LC-S&B, LC-WSR, RP-WSR, RP-S&B -hmmmm... the lower velocity ones (avg 50 fps lower) all had S&B primers. When sticking with CCI or WSR primers, the middle group can be included in load options.

With further primer analysis, these numbers could be refined more, but due to future supply unknowns, using a primer variety needs to be an option.
When making up bulk loads for storage or use over long time, Loads should always be marked & ID'd for a variety of reasons, especially when used in multiple guns.

The easiest way I have found is to ID each load by head-stamp, (which will be separately classified as Heavy-Normal-Light) and by a 2 or 3 color code in the rim groove using a sharpie (it survives a lot of abuse and can tell you at a glance what you have once you know your codes). Then keep a detailed load list in Excel or other log book.

In summary
⦁ Charge A, often produces hi pressure signs with heavy cases and sometimes with normal weight cases, so these should not be used. When loaded in very light cases it appears to work fine out of both 16" and 20". It also just happens to match the velocity of charge C when loaded in heavy cases giving an option for using light/heavy cases.. Charge A with Light cases is the only one within velocity spec For the 20"... so using light cases (weighed) appears to be the only viable option for charge A at this time.

⦁ Charge B appears to be usable in all case weights for the 16" barrel but only with heavy cases for the 20"

⦁ Charge C is only usable in 16" so will be eliminated from the standardization unless required for bolt guns.

So from the above I have come up with a plan to have 2 basic loads for heavy and normal/light cases.
Currently the load for the majority of normal weight cases is still up in the air because a few showed excessive pressure in 16" barrel, so for that... My current thoughts on my next step for this load ...

1. I could ignore it and hope it was an anomaly (until the next one appears- then what do i do with 500 of that same load.
2. I could drop Load A 0.1 grain and raise Load C 0.1 grain and see how those work.
3. I could look closer at primer selection and play with seating depth to see if mods have a positive affect.

and now the kicker to all this that I mentioned in the beginning....

One of the things I do fairly regularly is measure brands of bullets and cases I use. I keep excel spreadsheets of all the data for later reference. I note scale weight after the case has been de-primed and trimmed to length and assign it to a H-N-L group. Along with the case info I usually include water capacity averages from 10-20 cases specially prepped the same way but before de-primed (I also assign them to a H-M-L H2O capacity group based on 3% +/- mean variance arbitrarily picked with no thought whatsoever- it just sounded good... [actually 5% was too much and 2% didn't work with the actual range]).

Now everyone knows from reading the internet that heavier cases have less capacity than light cases and that's the basis I used for the above analysis. Somewhere along the way I happened to sort the H-N-L weight group and noticed that with some cases, their sister column for water capacity showed them in a different group than their weight dictated... further sorting showed...

All of the cases marked Lowest capacity in the water classification were also marked as Heavy in the weight group, so that was good (but, not all heavy cases were marked as low capacity).

Interestingly all of the cases marked as having the highest 3% of water capacity (should have been lightest cases), were listed in the Normal weight group, with some of them on the lower edge of the highest weight group! Further searching showed the lightest weight cases were not classified yet, due to various reasons, making them oranges... but WTF is going on with all those normal cases?

Also, in a reverse twist, 6 of the heaviest cases weight-wise fell well within the normal water capacity range. All of them, DNL,FNM, IVI, RG and Aguilla have consistently shown higher velocity per charge, so something was clearly amiss here. Investigating this further, the one uniform difference I can see in their prep code (as opposed to normal ones) is that they were neck sized while the others were full length sized... so that has to be the factor - a little more body volume, throwing them into the normal water capacity range... I'll fix that! ... or maybe that's a solution for some applications?

Until now, I thought most of the development work was finally done and I was close to picking my standardized 55gr practice load but I still have a few questions to resolve before I can start setting up the Dillon.... work to be continued...

This may all just be an exercise, as I am moving toward heavier bullets and looking at the MK 318 options and what to use to load a similar round....

Below are a couple of the better targets during testing. These are from the 20" milsurp barrel from the 80's that has had an unknown number of rounds thru it when I got it in 2003.

 

[Jack Dupp]

I stopped reading after the first post (too impatient), but I appreciate the effort that goes into making threads like these. That said, I tried to accomplish the same thing about 10 years ago. I also settled on AA2230 powder, although I'll use TAC, H335, or surplus WCC844 in a pinch.

I found there's really no economical FMJ that can touch the Hornady 55gr FMJBT bullet. I'm pretty sure my load is 25gr with a OAL of 2.250", but it's been a while since I loaded up a batch. All my ARs get 1.5MOA with this load. Obviously some ARs are more precise-shooting than others and I get the occasional sub-MOA 5 shot group at 100-300 yards.

 

[Harrier]

No Problem,
I've been known for being wordy and giving the Gettysburg Address when a simple answer would suffice.
I agree that the Hornady bullet would probably be a better choice if buying fresh, but I got what I got.

I try to give enough info to answer questions before they arise. Many of you already know or have done a lot of this stuff, maybe a newbie can learn a tip or two from my ramblings...

besides, I included pictures to keep ya awake .

 

[Suck My Glock]

My first guess (and that's all it is) as to why some of the cases displayed water volume capacities that did not seem to match their case weight,...I know composition of materials used to make brass varies from foundry to foundry. Wikipedia explains this is often for reasons of producing different mechanical and electrical properties. I expect also that differing recipes for brass alloy lend themselves to easier/better drawing and forming. So could it be that various case producers use differing brass compositions in their manufacturing, producing slightly differing weight of brass per cubic inch lets say from manufacturer to manufacturer?

 

[Harrier]

Brass composition would certainly be a factor in overall weight, as would other aspects such as web and case wall thickness and being of foreign manufacture would support the different alloy too... but overall IMO one would expect the heavier cases to also have less water capacity which they didn't. Especially in light of the pressure signs and chrono readings.

My reasoning is that the fired cases expanded to fill the chamber and since they were only neck sized, retained the larger volume that would have been squeezed down during full length sizing.
I will round up some fired cases in the near future and do a water capacity test both ways to see if my theory holds water (so to speak) .

 

[backcountry]

Harrier,

Awesome post ! I love your detail. It reminds me very much of myself. Marking the many recipes with the sharpie is a great tip. I think I will color code them to match my spreadsheet color. I'm recently into powder coating (PC) and the colors are real purdy.

I have developed an Excel spreadsheet myself through several versions over the years.
It has proved invaluable in analyzing and keeping my data accessible.

I would really appreciate getting a copy of your spreadsheet for comparison.

thanks for the post,
b