azdukhuntr said:
BT, I'm not trying to light firestorm here or pop a cap on anybody but I'm extremely curious why there is a 3000+psi difference between what CIP feels is acceptable loading pressure and SAAMI does not. Purely speculation on my part but would it be possible that SAAMI is taking such a conservative approach to pressure maximums due to the fact that we exist in such a litigious society.In other words is SAAMI just covering their collective ass because stupid people do stupid things. I read that blog and then did my own research on CIP and SAAMI. CIP also puts a maximums momentum on their ammo although not sure what that has to do initial chamber pressures. I find it hard to believe that the Europeans produce such superior strength steel, over everyone else, that they can justify safely loading to chamber pressures in excess of 15,000psi where here SAAMI believes that anything over 12,000psi in a 20ga load is cause for grave concern. I'm not a metallurgist nor a mechanical engineer but I would like someone to logically explain why SAAMI believes that loading to CIP pressure would be considered so dangerous.
azdukhuntr,
It would take some graphs and paper, and an understanding of statistics to explain the answer to this. Essentially BT is correct up above where he says it's really about what they are measuring and how. I'll give it a quick try. I've got a degree in probability and statistics, at one time I could have whipped off the gaussian distributions and formula for you, but that was many years ago.
First, you have to realize that in reality there IS NO ONE NUMBER. That is, NO phenomena ever precisely replicates itself--maybe things happen close enough that you and I looking at it say it was the "same", but it really wasn't. For example, the commercial shotgun shells (and from your experience loading even) may be SUPPOSED to have 30 grains of powder in them... but they REALLY have between 20 and 40 grains, say, distributed along a "bell" or Gaussian curve with a very high spike around 30, dropping off on either side steeply. We have ways and terms to measure how strong that spike is, how steep the drop off is, etc, and we usually place (arbitrarily but because it's measurable) emphasis at the spots on the curve away from the center at one, two, and three "standard deviations" away from the center spike. We TEND to say in day to day talk that the chances of a powder drop > 3 standard devs from the center is zero, and generalize it like "They drop between 29.8 and 30.2 grains", but that's just hopeful thinking that the universe will follow it's normal statistics.
Second is the idea that when setting RISK AVOIDANCE NUMBERS, what we REALLY must set is a number whose "curve" is offset enough from the CURVE of the "bad things happen" number that they don't intersect in any appreciable chance amount. If we were SURE that we could load a shotgun shell whose pressure was ALWAYS 18,000psi or less, then we could of course load right up against the proof tested limit for the guns... but we can't be sure of a "pressure number", we can only predict it's gaussian curve based on enough samples of those shells.
We THEN actually have to guess at and predict those curves by a random sampling of shells, because we can't really fire them all, so we calculate the "probable error" in our curve predictions based on the sample size and how wide the curve is from -3 to +3 (sometimes 2) standard devs, and THEN we realize that calculation has inherent error guesswork too, and we calculate how sure we are of IT! We call those error guesses "alpha" and "beta", and they shape our thinking for every gaussian distribution which represents some "repeatable physical occurrence" for it's repeatability.
Anyways, here's what combines to dictate that there is a LARGE SPACE between SAAMI pressure limits and the Proof-test limit for our guns:
1. Proof loads are "random" shells too! We only HOPE they tested the gun out around 20,000psi, they MAY have only tested it to 16,000psi!
2. Shells are random, some may spike high
3. Pressure inside shells is not a linear phenomena, it follows PV=nrT BUT the P is changing by a chemical process which varies (how much volume of gas is produced by what is in the powder as it reacts) AND the T isn't linear either... So it's not a fault-tolerant system! Any high-end error may create MUCH higher pressure spikes than expected from merely a linear distribution.
4. The pressure at which the guns explode is gaussian, some may break early
5. The quality of the metals and how they take repeated stress is gaussian, some may fail to deform elastically earlier than others
6. Pressures inside all those shells rely on the interactions of numerous gaussian distributions: hull size, primer hotness, crimp depth, powder components, powder drop, powder burn rate. None of which can be directly controlled by the testing guys.
Anyways, it all dictates that if we set 11,500 for 12ga ammo, and we know how inexact our proof loads are, we need to proof "at 19-20k ish" to be able to HOPE/GUESS/PREDICT that we'll be ACTUALLY proofing sufficiently to say the gun is safe. SAAMIs math is good and correct for their spread between the two--and this is why SAAMI says that for the CIP hi-pressure loads, they would actually require a higher proofing than what CIP does for it.
Dave