A few months ago, we started an investigation into whether the PRE (actually, pressure ring diameter) and the CHE (case head expansion) methods provide adequate information to guide a reloader.



We have our first batch of data, and the analysis shows that neither method performs well. More data should follow in due course, and perhaps our understanding will be better when that happens.



Two different tests were performed. One only requires repeated micrometer measurements on matched pairs of cartridges. This test checks how well the measurements repeat, given the same input conditions. The other test checks the correlation of PRE and CHE to a strain gauge system, which has previously been shown to be very consistent.



PRE is the better of the two systems, so I'll use that to illustrate, assuming that the GIF attachment thing works.



This is a graph of KPSI, measured by the strain gauge, and the diameter of the pressure ring, measured by a very nice micrometer. Practically all the random error is found to be in the PRE system.



The blue lines contain 95% of the data. If you pick a dimension, such as .4890", and run a line up to the first blue line, and then across to the vetical scale, you get the lowest pressure that .4890" could represent in my rifle. If you do that again, for the upper line, you get the highest pressure that .4890" could represent. Those numbers are 44,000 PSI and 73,000 PSI. If you take that measurement, you don't really know where in that band of possible values you are.



http://www.huntchat.com/attachment.php?s=&postid=241974



Then you have the problem of getting an acceptable standard to measure against. If you want to use fired commercial cartridges as a basis, you'll have to average 140 of them to get the random error down far enough to be comparable with the obsolete CUP system.



In short, the PRE and CHE methods are extremely coarse, giving only the very roughest of indications of pressure. You can easily be 15,000 PSI above where you should be, and not know it. If you have used these methods for 30 years without blinding yourself or blowing your fingers off, you have proved that you have a conservatively designed firearm, or that you're lucky.

Thanks for the effort Denton. I've suspected as much for several years and have not tried to use such info to correlate pressure. Sometimes I'm smarter than I think.

Dan

You mean this graph?







What pre-processing methods have you employed? Or are these the raw data?

Nitroman...

Thank you!! For some reason, the silly thing did not work for me. It's much nicer to have the picture on display.

It's just raw data... simple linear regression, with 95% prediction intervals. The dimension data are from a Mitutoyo electronic micrometer, with resolution to .00005", strapped to a granite inspection plate for stability and repeatability.

We also did another test, which I did not include in the original post. It uses only micrometer data.

As you can see from the regression plot, the test included cartridges from about 35,000 PSI to 65,000 PSI. Actually, there were two identical cartridges for each load. By running those pairs, it is possible to see if the same pressure will reproduce the same case dimension. The tool is called Interclass Correlation Coefficient. An ICC of .70 to .90 is considered marginal, and an ICC of .90 or greater is considered good. 1.00 is ideal. PRE yielded an ICC of .359, so it didn't even halfway make it to marginal, and CHE is even worse.

You are right, neither method is worth a hoot! Higher pressures sometimes show more expansion in cases from the same lot, with identical powder charges, same bullets, same number of reloadings (or even new cases), etc., but higher pressures sometimes show LESS EXPANSION THAN LOWER PRESSURES!People who use this method to determine what is or is not safe are kidding themselves!! But they do usually end up with a load which is below the cartridge's potential, VERY SAFE, BUT UNDERLOADED!!



This is particularly true for people who use "expansion of factory loaded ammo" as their maximum expansion limit!

Thanks for running the tests Denton et al. Many reloaders have been persuaded to use the techniques such as PRE and CHE under the guise that it was safe. Clearly it is not when you can have that much variation in pressures for a given load.

Kind of puts the comments that both Ken Howell and OKShooter have had on the topic into a whole new light. Apparently they knew what they were talking about!


Turok

I suspect that Ken Howell and OKShooter knew exactly what they were talking about.



Ken Waters made very modest claims for the PRE system, which he correctly believed to be superior to CHE. Unfortunately, he was a bit optimistic for even the very modest claims that he made. It would be very convenient if we could actually get pressure measurements from our brass.



Ken Waters, Sept '82, Handloader Magazine, page 6a:

Quote:

---

Ballistics laboratories utilize special pressure barrels and copper or lead crusher gauges�or, alternatively, electronic transducer strain gauges�to measure chamber pressure.



I have agreed to explain in detail the system that I�ve been using for some thirty years in developing handloads totaling countless thousands of rounds for a variety of both standard and wildcat cartridges. It has yet to let me down.



Two terms that I use are somewhat different from those in wider use concerning the same concepts�developing rather than working-up loads, since it is frequently necessary to work down from early loads and unsafe to work up; and judging rather than reading pressures since any such method [he's talking about PRE here] is far too inexact to qualify for any term such as reading that implies a greater precision than is actually the case.



Now of course, no such system of judging pressures can reveal the actual pressure in pounds per square inch or copper units of pressure. It must be understood that this is only a means of determining comparative pressures, with nothing more to be expected of it. With the data provided, the pressures of my handloads can be classified as moderate, normal, near maximum, maximum, or excessive�which is all that is necessary to ensure the safety of the shooter.

---

Ken Waters knew that the results he was getting were only comparative, and that they were very rough. He contrasts his system to the copper crusher and strain gauge systems, which he acknowledges as giving real measurements. All that is bang on.



The only place Waters is contradicted by the data is that PRE "ensures the safety of the shooter". It is actually quite dangerous.

HotCore- Do you think that PRE better than CHE for most reloaders to use to compare pressures.?Most

reloaders don't have super accurate mikes, and being the pressure ring expands more than the case head, is it a better indicator?And easier to measure with less accurate

mikes?Ed.

I check pressure ring also in my wildcatting, as wel las use my formula and a chrono. But PRE really is only a comparative process, not able by itself to give you a pressure range.That is where my formula and a chrono comes
in the way I do it, OR where a pressure guage comes in.I say this for the following reasons.
a- Measuring pressure ring is measuring case expansion
above the base after firing.In includes the vagaries of
how hard the brass is in that transition area of the case,
and it isn't uniform from case to case(which affects springback),
and the differences in base size to chamber size, difference in chamber finish, and headspace (as case stretching can throw Pressure ring expansion off.
b- measuring the expansion in the middle of the case, with
pressure equipment while firing, seems to me more accurate for getting pressure.The brass there is more consistant,
it is thinner and doesn't spring back(flexes with the chamber sides more closely) and with formulas based on the elasticity of brass and steel, with the measurements of the barrel and brass thickness will be
pretty accurate.
c- the jump around in pressures in Denton's tests didn't cause a similiar change in velocity as vel is a function of
the room under the pressure curve.BUT IN ALL LOADINGS THE PEAK PRESSURE TIME FROM IGNITION CAN CHANGE WHEN IT OCCURS IN THE FIRING
SEQUENCE, THUS CHANGING ITS PRESSURE READING, AND THOSE CHANGES CAN MAKE THE CASE PRESSURE RING EXPAND IN WAYS THAT DON'T REFLECT THE ACTUAL PRESSURE.Acts funny so to speak.
D-So I think there is a place for pressure testing, and
it is just another tool to use.Ed.

Denton,

Cool post, I also suapected that these measurements were not as accurate as we might have been led to believe.

One which I have found to be of some limited use is shoulder measurement, how far it moves upon firing.
Crude but effective ( I think... )
More = more
Less = less
Different for each gun and brass, but seems somewhat consistant for the combination.

Seems much more likely that this works better than eyeballing the primer, especially if you are not very familiar with the specific primer you are using.

Input appreciated, I have not come to any specific conclusions as to a method that really works.

Travis F.

Measuring length is an interesting idea. Might give you a longer base to work from.

Just working from "book" loads is a good, safe route, too.

Most pressure signs don't kick in until it is too late, except that my primers flatten around 50,000 PSI, which is too low to do as much good as you'd like.

A strain gage isn't that expensive, and it's pretty precise. SigmaE is the standard deviation of the random error in the measurement system. Smaller is better. In terms of PSI, I have measured or calculated from published data the following SigmaE numbers for a single cartridge test. This provides the simplest way of comparing several systems side by side.

CHE method, 7,500
PRE method, 6,800
Copper crusher method, 1,827
Commercial piezoelectric, 1,366
Well designed strain gage, 667

If you're careful to run all your tests at one barrel temperature, you can actually produce more repeatable results at home than the figures you find in some loading books.

Thanks Denton. I figured HotCore was full of $hit!

Seriously, I knew the accuracy and precision could not be good, just too many facts point to its weaknesses. Just playing with it for some time now (along with my strain gage setups) I found about the same thing as you. I get about where I feel I'm in the normal loading range somewhere but don't know exactly where I'm at with pressure. It helps some, I give it that, but I simply find a concensous of three loading manuals much more usefull as to where I'm likely at with regaurd to max pressure/MV.

Working up a load to max pressure takes just a few shots with a strain gage. Watching the consistancy while switching components, varying OAL or what have is insight not learned any other way, MV is the closest indicator. One thing I've seen just too many times is pressure does not always show itself until it's up way too high. It's absolutely no wonder why people regularly post MV 100-200 fps faster than manuals state, and feel they are their safe max load.

I feel I'm a litle bit of a pressure pusher too, but 70,000 psi for a regular load is about all I want to play around with and subject the rifle to over any length of time. I generally get somewhere around 50-70 fps more than book loads if that at those higher pressures though, and often just what the book says or less... at the higher pressure.

When someone says "I get 3400 fps with a 180 in my 300 Ultra", I already know where they're operating at, and probably how long their brass lasts too. The question remains, how many cycles will a good action hold up on at 75-80k psi though, cause I know brass "will" live at those pressures, just not long.

http://www.reloadingpro.com/Web_Expansion_Measurements.htm

It is apparent that PRE and CHE measurement are a religion. Keep the faith baby!