Given the scarcity of smokeless rifle cartridge propellant, I think a lot of us have been motivated to try different things. Over the course of the last two years I managed to get a few pounds of Shooters World Precision Rifle powder. Having sat on this powder for some time I recently had a chance to try it with something I hadn’t planned on: the .224 Valkyrie. I have a Palma match coming up at the end of May and decided to give SWP a try in order to preserve my other propellants and to see how it would do.
Quickload predicted excellent muzzle velocity and safe pressures in my Starline cases, that averaged 32 grains of H2O capacity after being fired but not resized. A load of 24 grains under a 90 grain Sierra MatchKing bullet was predicted to give 2700 fps from a 26″ barrel with the rounds loaded at 2.36″ OAL for single round feeding during the Palma match, and 24.5 grains should give 2750 fps at 96% fill and almost 100% propellant burn. Not bad!
All loaded up and ready to go, I benched the 26″ rifle that I built. It has no gas system and I run it as a straight-pull bolt action using a side charge upper receiver and bolt carrier. A 3D printed handle helps work the action.
Right out of the gate, two things were apparent: This ammo has excellent accuracy, but velocity is well below prediction. The 24 grain load ran 2609 fps average and the 24.5 grain load ran at 2650. However this ammo was shooting 1/3 minute groups without trying very hard. I turned to Gordons Reloading Tool because it has a feature that will tune a powder based on real world results. In this case GRT predicted that 24.9 grains of SWP would result in 2694 fps at the same ambient temperature as the previous shooting session. So with the new ammo loaded up, off to the range.
Match rifle with 26″ 224 Valkyrie barrel for upcoming 1000 yard Palma match. Bench target on left was shot at 200 yards. Reduced MR Target was shot prone with a sling at 300 yards. Shoot-n-See is about the width of the 10 ring and the SNS 9 ring is about the width of the X ring of the reduced MR target.
In this case the ambient temperature was about 10 degrees warmer than the last shooting session, and GRT provides a field to account for this difference. The updated prediction was 2700 fps, for a difference of 6 fps. Shooting the ammo on a Lab Radar gave 2715 fps which is close enough in my books! Best of all, the accuracy remained excellent.
I shot on MR targets that were reduced to shoot from 600 yards down to 300 yards, which is the maximum distance at my rifle club. I shot many more X’s than 10’s and I think I’m ready for the Palma match. While the velocity is nothing to write home about for a 1000 yard match, I think the accuracy will play a more important roll, and the velocity is sufficient that all I need to do is make good wind calls in order to shoot well. I’ll report with range results after the match at the end of May.
In preparation for an upcoming 1300 aggregate Palma match to be fired at 800, 900, and 1000 yards over two days, I have measured the volume of 200 Winchester 223 cases that are otherwise ready to shoot. These cases have a mean as well as median capacity of 30.4 in grains of H2O (from here forward I will refer to grains of H2O as simply “grains”). The standard deviation was 0.12 grains, with a minimum of 30.13 and maximum of 30.88. With the expectation of 20 to 35 fps muzzle velocity per grain variation in case capacity, I expected the velocity variation to be between roughly 15 and 25 fps between the cases near the min and those near the max.
Ordering the cases from low to high volume gives the following distribution:
The majority of cases are between 30.2 and 30.55 grains with tails beyond those thresholds. I removed all these cases in the tails from the batch that I will be shooting at the upcoming match.
These rounds will be loaded with Hornady 90 grain A-Tip bullets to an OAL of approximately 2.45 inches. I expect to reach approximately 2600 fps with this ammunition at the muzzle.
For this study, I took the high and low six cases at each end of the distribution to measure velocity for comparison. My plan is two-fold, two see if these outliers (using this term loosely) can significantly effect my scores at 600 to 1000 yards, and if so, by how much. This is a small sample size as I want to keep the cases that are not at the tail ends of the distribution for the upcoming competition.
After shooting the 12 rounds I found that the muzzle velocity of the high-volume cases was 2581 fps with SD of 9.7 fps and the muzzle velocity of the low-volume cases was 2606 fps with SD of 14.9 fps. This gives a total spread of 25 fps, at the high end of the expected range.
Notice how the high capacity cases have LOWER velocity than the low capacity cases. This reflects the higher pressure generated by equal powder charge weights in a smaller volume, and is in agreement qualitatively with Quickload as well as general experience.
Running this data through a custom ballistics calculator that uses the mean velocity to get a fixed launch angle for all rounds gives the following vertical variation at the target for distances between 200 and 1000 yards:
Will this humble 223 load be capable of 1000 yard performance? Here’s how the bullet speed drops with distance for this load using Hornady’s G1 BC of 0.585.
Taking the rule of thumb for the transonic limit of 1.2 Mach, in which we select a Mach 1 value of 1125 fps corresponding to air at 68 degrees F which gives a transonic limit of 1350 fps. Our ammunition comes out right at the limit, and its performance has been proven in competition as well.
In my experience a bullet need not be going too fast at the target to perform with good accuracy. Provided the bullet is properly stabilized it will continue to exhibit good accuracy to Mach 1.1 and potentially lower. It is running out of gas for sure at Mach 1.2 though so that is a good speed to aim for at the target.
So what does this all really mean for a competitive target shooter or anyone else who wants to place rounds accurately at long range? Let’s start at the low end of long range, 600 yards, usually referred to as mid-range by high power competitors. Here is where the rubber hits the road.
Suppose a sling shooter and his weapon are able, all things being equal, to hold the 10 ring most of the time at 600 yards and 1000 yards. We can simulate such a shooter, who would have a mean radius right around 0.9 (see ballistipedia.com for a proper explanation of mean radius). We can also neglect wind for this study but add in the vertical variation due to variation in muzzle velocity. At 600 yards, how many points and X’s would this shooter give up due to high or low capacity cases? Let’s have a look at the extreme example in which the shooter only shoots the cases at the tails of the distribution. Simulating 10 different groups at 600 yards we would have the perfect cases on the left and the tail-capacity cases on the right.
Note: the inner circle is the F-Class X ring. Clearly these are not F-Class groups being simulated and I will address the effect of muzzle velocity in competitive F-Class in an upcoming post.
We see our shooter dropping a couple points and X’s due to case capacity variation most of the time. But we know that most of our cases are going to be in the good portion of the distribution, not at the tail. So an outlier could cost us a point, but we see why top shooters who do not measure and account for case volume rarely drop points and keep on winning anyway. Thank goodness that shooting, especially from a sling, still comes down to shooting ability, especially the general fundamentals and wind reading. But still, a competitive match could still come down to one or two rounds with relatively high or low muzzle velocity.
At 1000 yards the situation is a little bit more dramatic, but still we’re talking a few points or X’s and this for all cases at extreme ends of the distribution:
I find it amusing that sometimes inaccuracy in the weapon and ammunition can work to our favor. If we throw a shot high that would otherwise have been low due to other random variation in the weapon, wind, and ammunition, then we get lucky and save a point or two. But in the aggregate we cannot achieve winning scores consistently in this way.
So we see that muzzle velocity variation due to case capacity variation is another knob we can turn in our pursuit of perfection, but we still have to be good at shooting to reach our full potential and win matches.
All long range competitive shooters agree that ammunition must be accurate at short range in order to be accurate at long range. They also know that this is a necessary condition, but not a sufficient one. In addition, the ammunition should have very low variation in muzzle velocity and the bullets should have a good ballistic coefficient.
Many factors influence muzzle velocity, and variation in these factors will lead commensurate variation in muzzle velocity. These include:
Powder charge
Case neck tension
Case mouth uniformity
Flash hole uniformity
Primer consistency
Bullet weight
Bullet seating depth
Case volume
Recoil technique (strange but true – a topic for another post)
Each of these factors has a limit to which we can minimize variation, and at some point the effort to decrease variation leads to diminishing returns. In this article I will consider ammunition that is between reasonably and superbly controlled for most of the factors in the list and both how and how much controlling additionally for case volume can result in improved variation in muzzle velocity.
Suppose Case Volume was the Only Factor
The simplest place to start is with ammunition that is perfect in every way except for variation in case volume. Quickload and experiment have both shown that for many typical cartridges, powders, bullets, etc., muzzle velocity varies with case volume at a rate of approximately 20 fps to 30 fps per grain of H2O as a measure of case capacity. From here I will use “grains” in place of “grains H2O”, the “H2O” being implied.
Consider ammunition from a simulated population of 500 284 Win cases in which the volume of the cases is normally distributed about a mean of 69.1 grains with a standard deviation of 0.175 grains. I got the mean and SD used here from real world measurement of 100 cases. Assuming that a case with 69.1 grains capacity produces 2800 fps muzzle velocity for a 180 grain Berger Hybrid, and a 20 fps per grain volume variation, a randomly generated population is shown in the following graph:
Most of the muzzle velocities are centered around 2800 fps as expected and we see a high an extreme spread of about 20 fps, which is not surprising since the population of cases has an extreme spread of case volumes that is about 1 grain.
So what does this mean in practice? How will this otherwise perfect lot of ammunition shoot, all else being perfect? Using a typical G1 drag model for the Berger 180 Hybrid we get the following vertical distribution on paper from 600 to 1000 yards down range:
And again with vertical dispersion measured in minutes of angle instead of inches:
I prefer looking at the plot that shows POI vertical variation in minutes because it is easier to relate to score in high power rifle competition. At 200 yards the vertical variation is very small at just over +/- 0.1 minutes. I don’t know many people outside short range bench rest who would lose too much sleep over that. At 600 yards we’re approaching +/- 0.5 minutes, now that’s something. In Service rifle that can cost you an X or a point, and in F-Class we are definitely talking points to lose for a shot that would otherwise be near the inside edge of scoring a 10, and at 600 yards these days, X-count often separates winners and losers.
For those of us who shoot long range matches, which are typically from 800 to 1000 yards and shot on an LR or LR-F target, you can see that we’re talking serious points to lose with the vertical spread approaching +/- 1 moa.
Adding Measurement Error
We cannot measure case capacity perfectly. How much does measurement error influence our results so far? With the Bison Armory Case Capacity Gauge, I have performed some experiments to determine measurement error and a normal distribution with zero mean and standard deviation of 0.025 grains fits the data. Adding random measurement error to the muzzle velocity vs case capacity shown earlier, in which we keep the muzzle velocity where it was for the perfect case and only vary the measurement due to error we get the following:
I think it is clear from this figure that case capacity measurement error is not a significant factor. If case volume was the only contributor to muzzle velocity variation, and we can measure case capacity as accurately as indicated in the figure, then it would be a simple thing to produce a batch of ammunition for a match that had minimal velocity spread.
Of course it’s not so simple. In the next post I’ll add muzzle velocity variation from all other sources and we’ll see how that complicates matters, and also how to make the best use of case capacity measurements to decrease muzzle velocity dispersion.
I conducted a brief study of case capacity and its effect on muzzle velocity this weekend. Such studies are easy because spending time at the rifle range is fun. They are also difficult because time is limited and it takes a lot of trigger time to get statistically significant results.
This study is not rigorous in that insufficient data was collected to prove any correlation between muzzle velocity and case capacity for a given brand of case, but enough data was collected to show a link over several brands of cases. The difficulty here is that there is more to muzzle velocity variation than case volume, but if the variation in capacity is great enough, we see the effects clearly.
Starting with the case capacity in grains of H2O between a selection of new and once fired 308 Win brass from Lapua (once fired), Federal (once fired), SSA (new), Winchester (new), and Hornady (once fired).
SSA has the lowest capacity while Hornady and Winchester were about the same at the highest capacity. Approximately 2 grains of H2O capacity separate the lowest from the highest. We expect that all else being equal (i.e. the same powder charge and bullet weight etc.), the cases with the lowest capacity will exhibit the highest muzzle velocity and vice-versa. Here’s the results from the range session shooting off-hand with an M14 (shot pretty well, one 10-round string was 96-2x)
In this figure clear correlation between case volume and muzzle velocity is apparent. Obviously other factors influence muzzle velocity besides case volume as there is significant variation in muzzle velocity that does not correlate with case volume. For example, the SSA brass (grey dots) has lower muzzle velocity than the Federal brass (orange dots) even though it clearly has lower case volume, which generally correlates with higher muzzle velocity.
Given that the powder charges were thrown by an Autotrickler to 41.2 +/- 0.02 grains of H4895, the powder charge is the most consistent thing besides bullet weight at 168 grains for the Sierra Match King bullets used in this test. Notice also that for each 10-shot group except for the group shot with Hornady brass, the variation among the group does not correlate much at all with case volume. This is to be expected with sample sizes this small. Even so, the correlation among the data in general agrees with the prediction made by Quickload between 20 and 30 fps per grain of case volume, all else being equal.
The Hornady brass did show good correlation between case volume and muzzle velocity so let us consider it more closely.
This correlates with the prediction given by Quickload but is still too small a data sample to be taken as strong evidence. And there lies the problem as always with load development and accuracy: the difficulty with which we obtain meaningful results due to the constraints involved in gathering statistically significant data. Barrels heat up, fatigue sets in, Lab Radars fail to register a shot, and so on.
Ideally I would turn necks and be very careful about neck tension, flash holes, and the rest, and then shoot 50 to 100 rounds of each brand case. I’ve also found that correlation is stronger if the volume of the fired case is measured before resizing and compared with the muzzle velocity from the previously fired shot.
So take the data as it is, a point from which we can move forward, no more, no less, and an indication that what we expect is true, so now we have to be more careful to prove it.
In an upcoming article I will discuss strategies for using case volume measurements to inform load development for match shooting at 600 yards and beyond.
