Pressure Testing the 33XC
Load Developing for ELR cartridges comes with special considerations, both for accuracy, but especially for cost. Every time you pull the trigger a lot of money is heading down range. The projectiles are significantly more expensive, the powder charges are large enough that powder costs skyrocket, and last but not least, the cost of barrel life is now impossible to ignore.
The is especially true for ELR wildcats. Fireforming ELR cartridges adds yet another cost and step before heading to the "capital L", Long range. Fireforming using the old COW method or hydro-forming is too intimidating for most sane folks and many just use a cheap bullet and a reasonable load for the unimproved cartridge. Most ELR shooters fire their brass once anyway as SD/ES and POI change between 0x and 1x.
Fortunately, pressure testing provides us with a way to save ourselves and our customers time and money. How, you ask, does pressure testing save us and our customers money? Easy! By answering questions with real world data. This data has historically been hard to come by for numerous reasons. It's expensive to produce and very dependent on the exact components involved. What is true for x bullet with y powder in z barrel won't be true if any factors or components are changed. This can even be true for identical barrels chambered with identical reamers!
Factory ammunition goes through extensive pressure testing for many obvious and some not-so-obvious reasons. The primary concern for factory ammunition is SAFETY. The companies didn't make it this far by collecting lawsuits. After safety, comes velocity and then finally accuracy. The factories don't release pressure data, nor do they tell you what powders they're utilizing. Also, they're not pressure testing in your barrel or your weather. And last time I asked, Hornady has no plans to release 375 Cheytac Improved and the pressure data for such wildcats will not be coming from them.
At the last ELR Southeast Match, we fielded our FNH Ballista with its new custom chamber in David Tubb's new cartridge, the 33XC. (For those of you unfamiliar with this cartridge click HERE to read about it from the horse's mouth or head over to SnipersHide where you'll find a large thread on folks working with the cartridge.) We had very little time to load develop this rifle before the match but we quickly found a 300gr Berger OTM load that seemed to work well enough by the typical standards of 1/2 MOA and single digit SDs. Unfortunately, we did so with H50BMG which, unbeknownst to us, is going bye-bye.
Even with this satisfactory load, we struggled during the match with numerous mechanical issues including, but not limited to, FN's trigger falling apart, our muzzle brake loosening, and what we thought were light primer strikes. We also saw flyers throughout the practice day that were cause for concern. Fortunately, we rounded out the podium in the light division in second and third.
As soon as we returned to the coast it was time to re-load develop the Ballista. The first question was which powder to select to replace Hodgdon's venerable 50BMG powder? A quick look at the burn rate chart gave us several candidates.
Lovex D100
Vihtavuori N170
Vihtavuori N570
Vihtavuori 24N41
Vihtavuori 20N29
Alliant RL 50
Alliant RL33
Since we knew from David Tubb's write-up, that RL 33 and 50BMG were safe powders to work from in this cartridege, we wanted to find out where we were pressure-wise with these powders. We had a decent load with H50BMG so a vishay gage was installed on the barrel and we used the remaining rounds from the match for verification loads. Below is the pressure readings from those rounds.
Now with a standard to work from there was a little work to be done with the not-so-trusty QuickLoad to find relative/predicted pressures to come up with safe starting loads with the possible powders to put behind the 300gr Berger OTMs. Here is where the pressure system delivers in spades. Even though a strain gage system only gives relative pressures, by matching with the parameters of predicted behaviors from QuickLoad a reasonable assumption of correlation can be gained. Below is what QL had for the same load.
*From the above two images a slight correction factor of a few kpsi, or splitting the difference between predicted pressure and velocity gives a pretty good match to the chronograph data from our verification loads. Alternatively, the presence of a looser than nominal bore could cause the decrease in measured pressure. Predicted muzzle exit times correspond to the pressure software's calculation.
By looking at the two curves it is obvious what is so attractive about H50BMG in the 33XC. The shape of the curve produced is calm so to speak, not very spiky as to indicate burn behavior that may be prone to pressure spikes. Also, there was reasonably consistent ignition. (More on this later) Also, the pressure was at the lower end of the 15% window below max pressure, and gave approximately 100% fill of usable case capacity, and produced acceptable velocity.
*DISCLAIMER: There is a lot going on in QuickLoad and you should understand and take heed of the disclaimers within that software and follow established safe reloading practices. ELR Cartridges carry a large volume of powder. The consequences for straying from appropriate powders can be dire. (This is a good reason to have a group like ours do your pressure testing for you!)
Now that a representative profile is built within QuickLoad it was time to see the predicted pressures of the other powders appropriate for the 33XC. At first glance it looked as though Vihtavuori N570 would produce great speeds in this cartridge. This was correct. By far the fastest loads we tested came from N570. This velocity came with a cost. While the bolt did not lock, the lift was heavy. The pressure system showed a relative pressure increase from the verification load of around 14,000psi! See pressure readout below:
This is a good illustration of the the limitations of QuickLoad. With all of the same parameters as the verification load, the software predicted the data below:
This load was not a starting load but if QL's inputs for N570 were correct, the readings from the pressure system should have been much lower and the load wouldn't have been so close to the upper limits of the margin of safety. As is apparent from the pressure readout, the attractive curve profile from 50BMG was not produced by N570. This curve much more resembled the spike handloaders wish to avoid in creating stable loads. The pressure and velocity were significantly higher than predicted by QuickLoad. Also, the fill ratio was very poor; around 91%, which could lead to all kind of issues.
Here would be a good time to talk about the concept of margins of safety. Tubb states in his write-up on the 33XC that the Peterson brass' primer pockets begin to loosen around 87,000psi. This is an extreme amount of pressure and just because the brass can handle this kind of pressure doesn't mean your action or your bolt lugs could handle these forces once or repeatedly. In this rifle, the N570 load produced the highest reading seen during testing. The pocket of the piece of brass fired above was indeed a little loose after this single firing. Could the reading from the pressure system be off? This is a possibility as it read approximately 6% low with the verification loads, theoretically.
Almost locking up the bolt with the N570 load was a cheap lesson. The shooter suffered no harm and neither did the weapon. The primary reason for this good fortune is these overlapping margins of safety built into the system. The action is built to withstand higher than normal pressure spikes, the brass is made to handle over-pressure events, and the barrel is certainly built to withstand extreme pressures. Had this not been a wildcat, loading practices for SAAMI and CIP cartridges would have had established practices and a large amount of load data that would allow handloaders to approach high pressures incrementally to avoid a pressure spike. Had the test of N570 continued or started with similar charges as those powders nearby in the burn rate chart, dangerous pressures would have likely been observed.
The performance of Vihtavuori 20n29 and 24n41 were such that they are not appropriate for for the 33XC. Even loads above 100% fill did not produce adequate velocities to warrant further testing
The next powder tested was the Lovex powder D100. This powder was designed for 50 BMG and similar cartridges and sits very close to Hodgdon's H50BMG. As with N570 an appropriate starting charge was chosen and loads were tested until acceptable velocity was reached. Below are the predicted and measured pressure curves for D100:
D100 produced very well behaved curves and pressure-to-velocity ratios were quite good. The rise and efficiency of the D100 loads tested were also very good. However, fill ratio was well below optimal and the rifle did not group well with these loads. There also was another interesting aspect to D100. It's a ball powder. Right off the bat, precision and ELR shooters are biased against ball powder. From the curves produced there maybe some evidence to suggest the Federal 215M primers were not consistently igniting this spherical powder. However, it would take a tremendous amount of research to isolate all variables to prove this. More testing needs to be done anyway with this excellent powder to see which cartridges it will be best for. Hopefully a hotter primer or modified flash hole could make this powder more suitable for the 33XC in ELR competitions.
Next, the Alliant powders were tested. RL33 and RL 50 are both at the slower extremes of the burn rate chart. Both are suitable for overbore bottleneck cartridges firing heavy-for-caliber projectiles. RL 50 was tested first and edged out RL 33 for a few reasons and only in minor ways. RL50 produced slightly higher velocities with slightly lower pressures. RL 50 and RL33 would both be better powders if they were a bit bulkier. In the end, the decision to go with RL 50 came from the very consistent standard deviations, more consistent ignition and good grouping. In the case of both Alliant powders, QuickLoad under-estimated the velocities that were seen across the MagnetoSpeed.
The bulk and burn characteristics of Hodgdon's H50BMG made it the perfect powder for the 33XC and hopefully the powder companies will find enough reasons to develop more varieties for this area of the burn rate chart. See below for the pressure curves for RL33:
...and for RL50:
As mentioned above, throughout the first day of testing there were what could have been interpreted as light primer strikes. In the case of this FNH Ballista, it would have been next to impossible to accurately diagnose the "primer" "issues" without the pressure system. The following lags or flat spots were observed in many of the traces.
Several variables were tested but in the end the main contributing factor to these "ignition problems" was greatly reduced by polishing the fire control group. The inside of the bolt body, spring, shroud, firing pin, and sear were polished. An additional spacer was added to the spring assembly as well. Once again, these ignition delays were not apparent to the shooter. Groups in earlier testing were unsatisfactory and mimicked a stressed receiver. Once polished, all loads were tested again. Also, other primers were tested and as would be expected, Federal 215 match proved the most reliable. RL 50 remained the top performer of the powders. D100 still had some ignition issues, ie. flat spots, but more testing of this powder will be done to see if those can be overcame.
