Anecdotes like these are hardly evidence. Anything could have happened. They couldn't figure out what happened yet they came to a conclusion. Not because of any evidence but because a faulty process of elimination. The load isn't listed, the powder density to case volume isn't defined. Why is it almost always the .38 Special? What is "very light"? What powder density to case volume causes this?
For years they were blaming 2.8 gr of Bullseye for blowing up all those Model 14's. Is that a "very light" load?
As the first guy said, multiple ballistic labs at manufacturers say it can't happen. But we're supposed to believe this anecdote over them? There's lots of reasons not to use very "light loads" but "detonation" isn't one of them.
I think it's more likely he made a mistake because he was unfamiliar with the Star. Those are more complex than your average beast.
I agree the theory about the powder laying in the bottom of the case and being ignited at both ends with resulting pressures waves starting at each end and meeting in the middle is suspect. In fact I’ll go one farther and say it’s total hogwash. The powder would not start burning at both ends, it would start burning along the entire exposed surface.
Generally speaking one of the concerns with extremely low load densities is that way too much powder is exposed to the primer flash, and or is scattered by the primer blast so that it is much more exposed to flame and ignites faster and creates a much larger initial pressure spike.
Why is the .38 special often implicated? Probably because it is a low pressure cartridge (17,500 psi) that started life in 1898 as a black powder cartridge. As such it has a very large internal volume by smokeless powders standards (which it started being loaded with as early as 1899).
The .357 Magnum was made 1/8” longer just to prevent it from being chambered in the .38 Special chamber. It’s also over sized for what it needs, but the larger charges used to generate the higher pressures (up to 35,000 psi) usually make that no more of an issue than it is in the .38 Special.
Where it all comes off the rails load density wise is when a hand loader wants a very light .38 Special target load but also wants to use it in a .357 Magnum case. Minimum charges for .357 Magnum are usually about 1 to 2 grains higher than for the .38 Special.
Looking at old (2nd edition) Hornady data, 2.5 grains of Bullseye is the minimum listed charge for the 148 gr LSWC and significantly there is a blank space to the left of it on the table (I.e. if you want a lower velocity use a different powder). The .357 data doesn’t show the 148 gr bullet but shows a minimum charge of 4.2 grains of Bullseye compared to 3.3 grains for the .38 Special.
The 11th edition shows a minimum charge of 2.1 grains of Bullseye for the same 148 gr LSWC bullet in the .38 special (and doesn’t show Bullseye data at all). Why would the significant 16%) difference in minimum charges exist? Two possibilities.
1) You’ll often see some minor differences in min or max charges due to lot to lot variation in a powder. For example look at H110 and Win 296, or HP38 and Win 231 data in most manuals. They are respectively the same powders wig different labels. Yet you will almost always see minor differences in min and max loads. That reflects lot to lot variation in the powder.
2) Despite having the same name, powder formulation for a given powder changes over time.
If you look at the gun powder data base for the National Center for Forensic Science…
National Center for Forensic Science
…you’ll find distinctly four different formulations for Bullseye dating from its production by both Hercules and Alliant:
1) Nitroglycerin and Ethyl centralite
2) Nitroglycerin, Ethyl centralite, and Dibutyl phthalate
3) Diphenylamine, Nitroglycerin and Ethyl centralite
4) Nitroglycerin, 2,4-Dinitrotoluene, Diethyl phthalate, Diphenylamine, Ethyl centralite, Dibutyl phthalate and 2-nitrodiphenylamine
I’m not a chemist but my understanding is that changes in established powders over time are the result of changes in production methods and costs as well as adding features like increases flash suppression, increased powder stability, reduced temperature sensitivity, etc.
In other words today’s Bullseye isn’t necessarily the same as yesterday’s Bullseye.
—-
How does that all relate to a light charge of Bullseye in a Model 14 and numerous case studies of one that have occurred over the years that claim a light charge of Bullseye in a .38 Special caused detonation and a kaboom?
Your argument is that ballistics labs claim it can’t happen. I am sure that is ***probably the case - but only within the parameters and experimental accuracy of their research. But that means a there are a lot of “ifs” involved:
- What formulation of Bullseye was used? What if your formulation of Bullseye is different?
- What did they consider a light charge? Did they keep reducing to 0 or did they stop at a minimum recommended charge weight? What if your “light charge” is different?
- What powder lot was used and where was it in respect to the range of powder burn traits for Bullseye? What if your powder is on the other end of the range?
- What was the water capacity of the brass used? What if your brass has slightly larger water capacity?
- What firearm or test barrel was used? What if your firearm isn’t as strong, has different chamber, leade or forcing cone dimensions?
All research has confidence intervals and qualifiers that have to be considered that effect the extent to which any conclusions can be generalized to other, let alone all other, cases. Not everyone summarizing that research mentions those qualifies and caveats that impact the limitations of the data.
*** I’ll stick with “probably” here as some of those labs may well have conducted the research for very specific legal purposes in response to law suits, or to proactively address concerns about future liability, and/or along very narrow constraints that may augment a legal defense but not necessarily capture all of the possible real world occurrences.
——
In short I am not going to discount a large number of reported (and admittedly potentially suspect) cases based solely on lab results that by definition have constraints that limit their application to the entire population of possible low low density kaboom events.
It’s much like the argument between the data junkies and the ballistic gel junkies. Ballistic gel gives repeatable (reliable) results, but those results lack any real meaning until you can compare a bullet’s lab results in ballistic gel with street results in a significant number of real world shoots.
At best you, all you can do is establish a range of performance in ballistic gel that is associated with an acceptable range of performance in real world shoots. Yet the gel junkies forget that last part of the sentence or the role that data on actual shoots plays in validating the gel data.
You need both.