Wolfe Publishing Group

    Primer Flash Hole Diameter & Black-Powder Cartridge Accuracy

    Hoke tool for the reduction of the flash hole diameter.
    Hoke tool for the reduction of the flash hole diameter.
    I was very careful to keep the drill bit aligned with the axis of the existing hole. I did not want to end up with oval, oblong, or egg-shaped holes while enlarging the hole. Running the drill at a relatively slow speed kept the bit from grabbing and resulted in a nice, clean hole. After completing seven holes, I held the pepper shaker lid up to the shop light to look for burrs. The new holes were uniform and round and would allow the black pepper to pass through with less resistance resulting in a bit spicier meal.

    Based on the positive pepper-shaker experience, I felt I should put my drilling skills to good use. Being in my reloading room, I glanced at a container of brass cartridges and wondered if modifying the primer flash hole diameter would have any measurable impact on black-powder cartridge accuracy. Although anecdotal, multiple shooters had shared their significant positive results from primer flash hole modification. It was decided to “drill down” and see if there were any accuracy gains to be had. Does a particular primer flash hole diameter result in more uniform powder ignition and perhaps better accuracy?

    Primer Flash Hole Diameters

    The plan was to test three flash hole diameters along with one variation on the factory case flash hole diameter. The base case would be the standard diameter as punched by the brass manufacturer. In this test, 38-50 Remington Hepburn brass is based upon the parent case, Winchester Western 30-40 KRAG brass. Not having gauge pins small enough to measure the primer flash hole diameter, I utilized drill bits as a substitute and came up with 5⁄64 inches for the diameter of the factory flash hole. The unmodified factory brass was considered the “base case” for this test.

    Being a fan of the book by J.S. and Pat Wolf, Loading Cartridges for The Original .45-70 Springfield Rifle and Carbine on loading for the 45-70 Trapdoor Springfield, the second diameter selected was 3⁄32 inches as recommended.1 Wolf recommends a .096-inch flash hole using a No. 41 drill bit. Although, a 3⁄32-inch bit resulting in a .094-inch hole was stated as being equally as effective as the .096-inch flash hole. This was as simple as using the appropriate size drill and carefully enlarging the factory flash hole.

    A very accomplished cast bullet shooter that happens to be a friend of mine, shared that he had very good luck with reducing the flash hole, followed by redrilling to 1⁄16-inch. Being a machinist, he made his own primer flash hole swaging tool to affect the flash hole reduction. Granted, this was for smokeless powder, but I felt it was worth including within this black-powder test. I used a Hoke tool to reduce the flash hole before drilling, which worked very well. If desired, the Hoke tool includes a pin that can be used to set the diameter of the flash hole during the swaging process. I used the tool without the pin, swaged the hole slightly smaller than desired, and drilled it out to the final diameter.

    The 3⁄32-inch flash hole has approximately 44 percent more area than the factory flash hole, while the 1⁄16-inch flash hole has 35 percent less area.

    Since this test was going to be a significant effort, I wondered if there wasn’t another flash hole diameter test that could be included as part of the program. I decided upon using a paper wad on the inside of an unaltered case, to cover the flash hole. This is commonly used by some shooters to increase accuracy. The theory is the paper keeps the powder grains from passing through the hole and resting directly on the primer. If you have ever charged an unprimed case, you have found that some of the powder passes through the flash hole until the powder bridges the hole. So, the paper wad does indeed keep the powder from passing through the flash hole and resting on the primer.

    I do not believe the paper wad dampens the primer’s force of the flash. I say this because I once tried aluminum from a pop can to keep the powder away from the primer. I placed strips of aluminum between the primer and the cartridge case before seating the primer. The primer seating process neatly cut an aluminum wad. I only loaded a few rounds because I thought the primer would not ignite the powder. Well, it did, and punched a nice round hole in the center of the aluminum wad. Further testing indicated no benefit. I would not duplicate this method.

    Test Plan

    I have previously reported on a completed neck tension test. I have used a similar methodology for this testing program, including a modified mean radius method (to represent group size), vertical dispersion, and group sizes (measured between the two widest shots) to present the results. You can find this test in The Black Powder Cartridge News, No. 118, (Summer 2022). The following is a summary of the modified mean radius method.

    “To obtain the mean radius of a five-shot test group, measure the heights of all shots above a horizontal line drawn through the lowest shot. Averaging these measurements will give the center of the group above the horizontal line. In the same manner, measure the horizontal distance from a vertical line drawn through the furthest left shot. Averaging these measurements will locate the center of the group from the vertical line. Where the vertical and horizontal averages cross is the group center. The distance from the group center is measured for each shot. These measurements are normally averaged to arrive at the Mean Radius.”

    The distance from the group center to each shot was measured but I did not average the five shots, as would be done using the mean radius method. I used the distance for every shot from the group center in the analysis. Using this methodology allows for the inclusion of the variation from each shot in the analysis which results in 35 data points for each flash hole diameter, rather than seven data points (i.e., seven, five-shot groups). From a statistical view, any sample size less than thirty is considered a small sample. This is the rationale for 35 shots for each flash hole diameter. As part of the testing program, I planned to complete an analysis to determine if there was any statistically significant accuracy difference between the four tests.

    A summary of the test process:

    • 35 shots were completed for each test, for a total of 140 rounds plus fouling shots

    • Shooting was completed during similar range conditions for all tests. The range conditions were very good during all of the tests.

    • The shooting order was varied for each test

    • Each test was completed as a five-shot group, for a total of seven, five-shot groups for each test.

    • A modified mean radius method was used so that the dispersion of every shot from the center of the group was included. resulting in 35 data points for each test.

    • 15 cartridge cases were used for each test. The cartridge cases which were drilled oversize, 3⁄32 inch, and reduced 1⁄16 inch, were marked and kept separate from the standard factory cases. It was necessary to reload these particular cartridge cases multiple times to complete the test. I was careful to mark the cases to ensure they were shot upon the appropriate target.

    Each of the four tests was color-coded to avoid confusion at the range during testing.
    Each of the four tests was color-coded to avoid confusion at the range during testing.
    A CPA Stevens 44½ chambered in 38-50 Remington Hepburn was used for the test. This is an excellent rifle with proven accuracy. The rifle has a Green Mountain barrel with a 12-inch twist and a DZ Arms 6x scope. All shooting was from the prone position over cross sticks with wiping between shots to remove fouling. Fifty-seven grains of 2Fg Swiss ignited with CCI Large Pistol primers, HDPE wad, and a Paul Jones 367-grain Money bullet cast of 20:1 alloy. I used an inside cartridge neck diameter of .376 inches for the .376 diameter bullet. After firing, the case neck would accept a .379-gauge pin that would indicate a three-thousandths expansion, or a bit more if we consider the brass spring-back. An ultra-tight chamber neck is not necessary for good accuracy and is probably not desirable.

    With an overall length of 3.340 inches, the completed cartridge can be fully seated with thumb pressure well into the lands, due to the tapered bullet, but does not require any tools or levers to seat.

    All tests were fired at 200 yards from the prone position using SPG 18-ring Schuetzen targets. These targets are a nice size (approximately 12 inches square), and large enough to keep the shots on paper. They can be easily retained for later reference.

    Two wet patches were used to clean the bore followed by a chamber mop to remove any excess moisture. After completion of four, five-shot groups, the rifle was cleaned and allowed to cool. Before the next 20 shots, two fouling shots were fired.

    Modified Mean Radius Method

    The four tests; are 1) the factory unaltered case with a 5⁄64-inch flash hole diameter, 2) the factory case 5⁄64 inch with a paper primer wad, 3) a 3⁄32-inch flash hole diameter, and 4) 1⁄16-inch flash hole diameter. There was no statistically significant difference in accuracy between the four tests when analyzed using the modified mean radius method and testing for statistical significance using a one-way ANOVA (Analysis of Variance.) This means there is not any true accuracy difference between the four flash hole tests. Notice the reported variance for each test is very similar. Results are presented below.

    Vertical Variance

    The results presented in Table II are the largest and average vertical measurement plus a one-way ANOVA analysis based on the variance from the horizontal center of the groups measured in the vertical direction. To explain in another manner, we are measuring how tall the groups are. In this analysis, only the vertical dispersion was evaluated. This was done in the interest of determining if there was one primer flash hole size that reduced high and low shots. Conversely, was there one primer flash hole size that resulted in more vertical dispersion? We can adjust for wind; we cannot adjust for random high-low shots.

    This analysis was undertaken as I noticed some of the test groups were greater in the vertical dimension than in the horizontal direction. See the sample groups for the 1⁄16-inch flash hole and the paper primer wad.

    There was not a statistically significant difference in vertical variance between the four tests at a 95 percent confidence level. However, there was an interesting result.

    Reducing the flash hole diameter to 1⁄16 inch resulted in round groups.
    Reducing the flash hole diameter to 1⁄16 inch resulted in round groups.
    As background, Ronald A. Fisher (1890-1962), along with Karl Pearson (1857-1936), were both founders of modern statistics. Their primary focus was on random variability versus statistically significant variability between tests. One thing that Fisher established, under the influence of Pearson, was the guideline that for a test to be considered statistically significant it should meet criteria of 95 percent confidence or greater. Without too much detail, this requires a P-value of .05 or less, (decimal equivalent of 5 percent i.e., 1-.95). Great debates were undertaken over the merits of being 90 percent confident or perhaps 99 percent confident. It was deemed correct by Fisher that 90 percent was not a high enough confidence and 99 percent was too high. He settled on 95 percent. In reviewing the statistical analysis for the vertical variance for the flash hole tests it was found that the P-value was .075. This means that we are only 92.5 percent confident there is a difference between the tests. Meaning one or more of the tests are not the same as the others. I will let the reader decide the validity of the test related to vertical dispersion.

    Notably, the variance from the center of the groups was very similar for all four tests, but the vertical dimension was higher for the paper wad test. An ad hoc test was run, which confirmed the paper wad test was the outlier. To the sharp-eyed reader, it was probably noticed that the paper wad case had the highest average vertical as well as double the variance of the other groups; note the variance 2.05 (factory case with paper wad) vs .96, .92 and .75. Adding a paper primer wad during the reloading process may increase vertical stringing.

    Group Size

    The vertical stringing in these groups was typical.
    The vertical stringing in these groups was typical.
    For each test, there were seven, five-shot groups. This would not be considered a large sample but is of interest. Each group was measured for the distance between the widest shots. The statistical analysis of the test was not found to be statistically significant at a 95 percent confidence level. With a P-value of .069, we can state that there is a difference in the tests at 93.1 percent confidence. The paper wad test was the outlier and produced the largest groups. Again, I will let the reader determine the validity of the test.

    I would feel much better if I had 30, five-shot groups for each test but that involves more bullet-casting effort, reloading components and time than I have available.


    At a 95 percent confidence level, the test results demonstrated there was no statistical difference in the accuracy of the three primer flash hole diameters using the modified mean radius method to measure the group sizes. The use of a paper primer wad did result in increased vertical stringing at a 92.5 percent confidence level. The paper primer wad also increased the five-shot group sizes, when compared to the three primer flash hole sizes, at a 93.1 percent confidence level. The paper primer wad does not appear to be of any accuracy benefit. Ultimately, the reader should decide if the paper primer wad is worthwhile or if the time would be better spent on other aspects of reloading.

    1. J.S. and Pat Wolf, Loading Cartridges for The Original .45-70 Springfield Rifle and Carbine, 2nd Edition. Published by Wolf’s Western Traders, Sheridan, Wyoming 1996.

    Wolfe Publishing Group