Annexe C — Distribution de poids des douilles
Appendix C — Case Weight Distribution
Le tri par poids est un proxy pour mesurer la consistance du volume interne. En pesant un lot important de douilles du même fabricant et lot, la distribution suit typiquement une courbe en cloche (loi normale).
Données de distribution
Le tableau ci-dessous montre à quoi ressemble un tel lot : 500 douilles .223 Remington réparties en classes de 0,2 grain, chacune étiquetée par la première et la dernière lecture à 0,1 grain qu'elle contient.
Ce tableau est un modèle, pas une mesure. C'est une loi normale de moyenne 92,07 gr et d'écart-type 0,35 gr, tabulée et arrondie au nombre entier de douilles ; il illustre la forme à laquelle s'attendre. Un lot réel ne tombe jamais aussi juste : un test du χ² d'ajustement sur ces effectifs rend χ² = 0,41 à 8 degrés de liberté, un ajustement qu'aucun échantillon de 500 n'atteint — sur 200 000 lots simulés tirés de cette loi exacte, pas un seul n'en a approché. Servez-vous du tableau pour reconnaître le motif, et de votre balance pour les nombres.
| Plage de poids (grains) | Nombre de douilles | Pourcentage |
|---|---|---|
| 91,0 – 91,1 | 2 | 0,4 % |
| 91,2 – 91,3 | 8 | 1,6 % |
| 91,4 – 91,5 | 24 | 4,8 % |
| 91,6 – 91,7 | 58 | 11,6 % |
| 91,8 – 91,9 | 92 | 18,4 % |
| 92,0 – 92,1 | 114 | 22,8 % |
| 92,2 – 92,3 | 98 | 19,6 % |
| 92,4 – 92,5 | 62 | 12,4 % |
| 92,6 – 92,7 | 31 | 6,2 % |
| 92,8 – 92,9 | 9 | 1,8 % |
| 93,0 – 93,1 | 2 | 0,4 % |
| Total | 500 | 100 % |
Bénéfices du tri
Le bénéfice principal est la corrélation entre poids et volume interne. Des douilles de poids identique du même lot ont très probablement un volume interne identique, menant à une pression de chambre plus consistante et donc une vitesse initiale plus uniforme. Ceci est particulièrement important pour les cartouches à faible capacité (.223 Rem, 6 mm BR), où un grain de variation de poids représente un pourcentage significatif du volume interne.
Le tri fournit aussi une fonction de sécurité :
- Douilles exceptionnellement légères : peuvent indiquer des parois minces ou un culot faible, plus vulnérables à haute pression.
- Douilles exceptionnellement lourdes : volume interne réduit, générant des pressions plus élevées à charge égale.
Pour le plus haut niveau de compétition, triez en lots de 50 ou 100 dans une fenêtre de 0,3 à 0,5 grain. Les douilles hors fenêtre restent parfaitement utilisables pour l'entraînement.
Sorting brass by weight is a proxy for measuring internal volume consistency. When weighing a large batch of brass from the same manufacturer and lot, the distribution typically follows a normal "bell curve."
The table below shows what such a batch looks like: 500 pieces of .223 Remington brass sorted into classes 0.2 grain wide, each labelled by the first and last 0.1-grain reading it contains. By identifying the most frequent weight range (the peak of the curve), shooters can select a highly uniform subset for match ammunition while setting aside the outliers for practice.
This table is a model, not a measurement. It is a normal curve of mean 92.07 gr and standard deviation 0.35 gr, tabulated and rounded to whole cases, and it is offered as an illustration of the shape to expect. A real batch never falls this neatly: a chi-squared goodness-of-fit test on these counts returns \(\chi^2 = 0.41\) on 8 degrees of freedom, a fit no sample of 500 achieves—in 200,000 simulated batches drawn from exactly this distribution, not one came so close. Use the table to recognise the pattern, and your own scale for the numbers.
| Weight Range (Grains) | Number of Cases | Percentage |
|---|---|---|
| 91.0 – 91.1 | 2 | 0.4% |
| 91.2 – 91.3 | 8 | 1.6% |
| 91.4 – 91.5 | 24 | 4.8% |
| 91.6 – 91.7 | 58 | 11.6% |
| 91.8 – 91.9 | 92 | 18.4% |
| 92.0 – 92.1 | 114 | 22.8% |
| 92.2 – 92.3 | 98 | 19.6% |
| 92.4 – 92.5 | 62 | 12.4% |
| 92.6 – 92.7 | 31 | 6.2% |
| 92.8 – 92.9 | 9 | 1.8% |
| 93.0 – 93.1 | 2 | 0.4% |
| Total | 500 | 100% |
Sorting Benefits
Weight sorting is useful because case weight and internal volume are related—but the relationship is looser than it is often presented, and it is worth being precise about its limits.
Within a single manufacturer and lot, heavier cases do tend to have less internal volume, which in turn tends to raise chamber pressure and shift muzzle velocity. The effect is most pronounced in small-capacity cartridges such as the .223 Remington or the 6mm BR, where a one-grain difference in brass represents a larger fraction of the powder space. Across different manufacturers, however, the correlation largely breaks down: head and web geometry vary far more between brands than wall thickness does, and two cases of identical mass can enclose noticeably different volumes.
The reason is geometric. Most of the mass variation in a batch of brass sits in the head and web—the thickest part of the case, and the part that contributes least to the powder space—rather than in the neck and body walls that actually bound the charge. Weight therefore ranks cases usefully, but it does not measure the quantity that governs pressure. When the actual number matters—when entering a case capacity into an internal-ballistics estimator, or when chasing the last few feet per second of consistency—measure water capacity directly, as described in Chapter [ch:brass]. Weight sorting is the screening step; volume measurement is the arbiter.
Weight sorting also provides a genuine safety function, and this benefit does not depend on the volume correlation at all. Exceptionally light cases in a batch may indicate thin spots in the walls or a weak web, making them more prone to failure at high pressures. These outliers should be set aside for practice ammunition or discarded entirely. At the other extreme, unusually heavy cases will tend toward reduced internal volume and higher pressures at the same powder charge—a tendency worth accounting for when working near maximum loads.
For the highest levels of competition, sorting into lots of 50 or 100 rounds—all within a window of 0.3 to 0.5 grains—ensures that every shot in a match is fired from a near-identical case. The remaining cases that fall outside this window are perfectly serviceable for practice or training, where absolute consistency is less critical.