From: Lloyd D Reid <reidld@ecf.toronto.edu>
Subject: Velocity testing .308 Palma loads

I did some more testing at Jim Bullock's yesterday.  I thought that
some actual test data may be of interest to the techno-masochists
here.  I'd like to hear any comment/criticisms/questions on my
methods or results.

On the handful of occasions that I have chronographed loads, I have
always been disappointed at the uniformity of my loads.  Even when
I employed the most careful loading techniques, I always seemed to
get velocity spreads of 100-200 fps, in a typical 5-shot string.
Hearing about sub-10 fps standard deviations here, made me cry.
I finally decided to get serious about it, and wrestle my velocity
deviations down to "rec.guns boasting territory" ;-)

My primary interest is long range target rifle shooting.  We are told
that it is desirable to have as small a deviation in muzzle velocity as
possible.  The reasoning behind this goal, is to reduce vertical stringing
at long range (800 yards - 1000 yards) due to shot-to-shot time-of-flight
differences.  Even though a rifle may shoot a tight, round group at 100
yards, if different shots take more or less time to reach the target,
they will drop more (or less), leading to a vertically strung group.
(This is ignoring the complications of "compensation," and barrel
vibrations.) At these ranges, it takes over one second for a .308 bullet
to reach the target.

These time-of-flight differences make much, much less of a difference
at shorter ranges.  At 300 yards and less, only a benchrest shooter
(if anybody) would notice the difference.

Bart Bobbitt has said that, in his experience, RWS primers have
given the best results.  Federal match primers (as used in Jensen's
92 Palma load) are 2nd best, followed by Winchester, with CCI
"a distant last".  Bart, please correct me if I've misquoted you.

His argument is that RWS provides the mildest, and most consistent
flame, which is (he argues) better.  Since Bart is (much!) more
experienced than I, can shoot a helluva lot better, and seems
to make pretty rational arguments, I thought I'd give it a try.

In a previous article I mentioned that my results with RWS primers were
less than stellar.  I was not at all impressed, and was getting ready
to declare to all the net that Bart was wrong, and I had proof!  (Thought
I'd relish the rare opportunity ;-).

Bart replied that this was due to my loads being below max (I used 43.0,
44.0, and 44.5 grains of IMR 4895).  He pointed out that my figures showed
a drop in SD as my charge was raised from 43.0 to 44.5.  Well, I wasn't
convinced, but I thought I'd at least give it a whirl...

So, I loaded up 15 each of 44.5, 45.0, 45.4, 45.8, and 46.2 grains of
IMR 4895, using my standard "best" .308 loading techniques.  Powder
charges were trickled to the nearest tenth of a grain (kinda time
consuming ;-).  Note that these charges are quite hot, and are probably
not at all safe to try in any .308, without careful testing.

When the dust settled, there was good news, and bad news.  The good
news is that I got far better uniformity in velocity than I ever
have before.  The bad news is that Bart was right, and I'll have to
wait 'til another day before I can call him a liar in public ;-).

Remember that the object here is _not_ to gain the highest possible
muzzle velocity; rather, it is to get the most consistent load that
is safe, in _my_ rifle, under _controlled conditions_.  I don't want
"excessive" pressures, because it's pretty discouraging to have the
primer pockets on your lovely hand-prepped match brass go loose,
and refuse to hold a primer for that wonder-load you just worked up.
I have yet to deprime these cases and see what happened to the tightness
of my primer pockets.  The "success" of my experiment will of course
be contingent on how long my brass lasts; these are hotter loads than
I used last season.

OK, here's the data:

Rifle:  My Winchester Model 70-based Palma rifle, caliber .308 Win.
Barrel is 30" long, .307" bore X .298" groove diameter, 1 in 14" twist
MacLennan barrel.  Barrel had 2602 rounds fired through it since new.
Chamber is .308 Obermayer.

Chronograph:  An older Chrony F-1, approx. 10' from muzzle.  Sky
condition: overcast.  Chrono seemed to pick up all shots fired.

All loads: '92 Palma brass, neck turned to 0.122", F/L sized with
Redding die to be 0.002" shorter than chamber headspace.  The standard
0.3068" expander ball used, giving a measured 0.003" of neck tension.
(This is more neck tension than desired - I'm working on fixing this).
Bullet is Sierra 155 Palma, powder is IMR 4895, lot E93JA08 L7875.
Powder charges trickled to the nearest tenth of a grain on beam
balance, verified on electronic powder scale, before charging (i.e,
powder charges are +- 0.05 grains).  RWS #5341 large rifle primers used.


First, some data from my earlier 18 Dec 93 tests.... these are what
Bart commented on.

Weather:  mild, no wind, estimate temperature +5C/40F

43.0 IMR 4895, Winchester WLR primers (only test case to use
Winchester primers)
xbar= 2802.29 SD= 24.2501 Nsamples= 14
2822	2785	2830	2828	2822	2758	2783	2791	2809
2820	2830	2775	2774	2805

'xbar' is the average, "SD" is "sigma sub (N - 1)".

43.0 IMR 4895, RWS primers for this and all the rest:
xbar= 2812.42 SD= 51.1174 Nsamples= 12
2797	2818	2772	2772	2828	2809	2774	2755	2949
2805	2850	2820

Note that, within the SD, there is no difference in muzzle velocity
between RWS and WLR primers.  And, the RWS-primed ones really suck.

44.0 IMR 4895: xbar= 2877.53 SD= 29.215 Nsamples= 17
2861	2869	2921	2936	2881	2863	2856	2865	2861
2877	2945	2881	2879	2854	2848	2848	2873

44.5 IMR 4895: xbar= 2908.12 SD= 19.1405 Nsamples= 17
2919	2912	2879	2906	2887	2925	2904	2896	2915
2894	2915	2896	2898	2883	2929	2927	2953


Yesterday's (12 Feb 94) testing:

Temp: (est'd) -10C/14F

Ammo temp: unknown.  Since we were firing in a semi-heated enclosed space,
it was proabably "somewhat" warmer than outside temp.  Have reason to
believe that ammo temp was fairly constant, for the duration of testing

Before testing, fired 3 'barrel warmers' - military surplus .308.
Chronograph survived this, so the day was off to a good start...

44.5 IMR 4895: xbar= 2864.13 SD= 15.2356 Nsamples= 15
2898	2859	2881	2887	2865	2867	2861	2850	2854
2859	2863	2844	2869	2842	2863

45.0 IMR 4895: xbar= 2872.4 SD= 10.1686 Nsamples= 15
2885	2887	2873	2879	2867	2869	2873	2859	2877
2850	2879	2885	2871	2865	2867

45.4 IMR 4895 xbar= 2912.33 SD= 7.0778 Nsamples= 15
2919	2915	2923	2908	2917	2915	2908	2915	2912
2925	2906	2904	2900	2906	2912

45.8 IMR 4895: xbar= 2950.8 SD= 17.1556 Nsamples= 15
2986	2945	2977	2940	2966	2951	2958	2962	2929
2951	2940	2929	2932	2938	2958

46.2 IMR 4895: xbar= 2960.21 SD= 14.0449 Nsamples= 14
2977	2968	2966	2950	2929	2968	2945	2973	2960
2960	2960	2947	2958	2982


Tabulated, we have:

Charge		Velocity	SD	Speed per addtn'l grain of powder
44.5/RWS	2864		15
45.0/RWS	2872		10	16
45.4/RWS	2912		7	100
45.8/RWS	2950		17	95
46.2/RWS	2960		14	25

18 Dec 93 testing:
43.0/WLR	2802		24
43.0/RWS	2812		51
44.0/RWS	2877		29	65
44.5/RWS	2908		19	62


It doesn't look like the "speed increment per grain" is a particularly stable
number.  Perhaps it is noisy because my sample size of 15 is too small?
Anyone out there play a statistician in Real Life?  (That's your cue,
Henry....)

Well, it looks like what Bart said was true.  Fancy that ;-)  I will
run some more tests, centered around 45.4 grains.  I'd like to get
SDs for a number of different runs, and see if I've really found
a 'valley' in velocity spread, or if the statistical noise is just
trying to enchant me.....   Also, I will test ammo with slightly
reduced charges at elevated controlled temperatures, to get a better
idea of what will work at the range in summer, at 75F-100F.  Hey,
I'm not _that_ wet behind the ears.. ;-)


That's it for the serious testing, here's two other tidbits:

Special test run on the side:  5 rounds of 44.5 IMR 4895 were cold-soaked,
to see what effect (if any) the cold primer/cold powder would have:
xbar= 2844.6 SD= 22.1991 Nsamples= 5
2846	2840	2822	2881	2834

compare this to the ambient (unknown ;-) temperature, probably about
5 or 10C warmer:
44.5 IMR 4895: xbar= 2864.13 SD= 15.2356 Nsamples= 15

It seems like the cold-soaked ammo was slower, and more erratic.  I
dunno is it is statistically meaningful, though.  Next time, I think
I will measure the ammo temperature, and see what I get from it.
I'll also run a larger batch of cold-soaked ammo.

Another thing tested was a quick'n'dirty groundhog load that I threw
together at the last moment:  berdan primed military brass, a fairly
mild 44.5 grains IMR 4895 ('cause that's what my powder measure was
set up for ;-), and Nosler 125 grain Ballistic Tip bullets.  This shot
~3/4" 5-shot groups, both from my Model 70 and from the ORA Mauser.
It looks like it has real potential, especially if the jerks behind
the trigger were a bit smoother ;-).  This accuracy level is damn
near the best I can shoot from a bench (I'm no screaming hell of a
stool shooter, sigh).

I've only ever heard good things about Ballistic Tips.  It seems like
they perform much better than one should expect from a sexied-up hunting
bullet.  Boy, am I going to be hell on those groundhogs this summer....


- Daniel