There used to be a lot more intelligent examination of it in the reloading press. It seems that, as Rocky points out, it has been explained to the satisfaction of those who understand such things, but remains a mystery to those of us in the great unwashed. What I find truly curious is that there exists an editor of a well-known publication who maintains that there is no such thing and that the problem is solely due to those stupid handloaders who overcharge cases.

That's mostly where you hear about it but apparently it can happen with "moderate" charges under certain circumstances. In the .243 example Kikkonen wrote "then new" Norma MRP. I'm guessing a range of loading data was being worked up when the test barrel exploded much to everyone's surprise. So surprised that the test was repeated to verify and one wouldn't risk such expensive equipment lightly. It's anecdotal and apparently an unusual case, but makes for a BIG caution sign. Note that about everybody would consider MRP too slow to be useful for reduced-loading a .243 WCF.

I don't think there's a consensus on what causes S.E.E. It seems to happen mostly with light charges of slow for volume, relatively hard to ignite powder in big cases leaving a lot of air space. Jacketed bullets which would offer more resistance in entering the bore seem to exacerbate the problem.

I know only what I've read. It's not something that I want to personally experience and exercise an abundance of caution to avoid. Particularly since nobody can define the boundary at which S.E.E becomes a possibility.

If you look at the decades-old data from Dr. Brownell, the evidence for instability in light loads of slow burning powder is clearly present.

He shows a graph of pressure vs. load. In the lower part of the curve, greatly increased pressure variability is clearly present. Undoubtedly, he noticed that effect. But, as far as I have found, he did not comment on it, or what may have caused it.

I have a hard time accepting the notion of pressure waves standing in the cavity, and hold that explanation as tentative. There is just too much you have to assume in order to accept that idea, as far as I can tell. However, SOMETHING clearly happens. Danged if I know what. So I can't criticize that theory, because I don't have anything that is as good, let alone better.

Denton, alternate theory from Kikkonen, slightly embellished, no idea if it's right.

Slow powders with a heavy deterrent coating are hard to ignite. Given a large air space in the cartridge and the powder not necessarily near the flash hole the energy of the primer would to a greater extent dissipate rather than couple into the powder. Further any gas from burning would have more volume in which to expand, slowing the rate of pressure (heat) increase. The powder does not ignite cleanly and "smolders" which produces explosive products of nitrocellulose decomposition. Heat from compression, burning through the deterrent coating, or something else causes chamber temperature to rise and ignite the explosive gas.

In favor of this theory Kikkonen noted that there was something less than a 100 millisecond delay between the primer firing and the explosion in the .243 example. He also wrote that people experiencing what was likely S.E.E noted a slight hang fire.

Again, I don't know.

--- edit

Also in working with very light loads (with very fast powder) I noted a point at which powder residue reduced markedly. This correlated with a small jump in velocity, below and beyond which velocity increased more or less in proportion to powder charge. Not that this necessarily means anything, particularly since it wasn't under controlled conditions.

He's got a couple fingers that are an inch shorter than normal.
You go real close to his shop on your way to/from work.

The whole thing may never be truly understood - unless and until we find a way to observe what actually happens inside a cartridge. Right now, we cannot. What we "observe" now are merely secondary or even tertiary results of what happened. We guess at pressures based on what effect that pressure has on something else, for example. We observe long-after-the-fact effects and infer a cause, but we cannot directly observe the process at all, much less in anything like real time.

I read long ago that some people who know about such things said it was a detonation instead of an explosion.

A detonation would seem more logical, because considering the chemical construction of powder, it doesn't seem there would be enough stored energy in a small amount of powder to cause that much destruction.

Clue:

Solids like wood, for example, don't burn as solids until initial heat converts 'em to gas.

I used to demonstrate this — back in the days when coffee cans had steel lids — by filling a coffee can half full of sawdust, punching a nail hole in the lid, heating the closed can on a stove, igniting the hot gas that spewed from the hole, blowing the flame out, letting the can cool-off, and finally opening the can to show that the sawdust wasn't even browned or charred — certainly not burnt.

All this takes place with wood over a much longer time than it does with powder.

When you really, thoroughly grasp this principle and assimilate it into your thinking, several ways that the "SEE" can occur begin to become very clear to you. (This isn't the only phenomenon that's part of it, of course.)

123...,

It's not how much energy is in the powder so much as how it's released. Detonation, explosion, deflagration, whichever it is, it happens very quickly, at least faster than the bullet can move (inertia) and make enough space to control the evolving gas.

nighthawk and Ken...

At first blush, that sounds like a more plausible explanation. It certainly requires fewer assumptions.

With all the extra space, the powder would burn more as it does in open air at first. Also, a small amount of oxygen would be available for the first little while.

In Brownell's writings, you can clearly see numerous very fast spikes riding on the normal pressure wave when using small charges. They appear to be fairly regular in frequency, but very irregular in amplitude.