AIU...

Well, you could start a thread on what constitutes overbore!

Well there is 5 things to choose from about
what is going on-

1-All Sensors, machines, operators are malfunctioning,
but only on certains types of loads, 0 chance.

2-Ionized gases, making magnetic fields on
only certain loads. 0 chance.

3-Overlapping pressure waves triggering sensors
only on certain loads, giving false readings.
Wouldn't all loads have overlapping
relected waves. Some chance.

4-Reflected waves causing slowed powder burn to accelerate the powder burning with bullet slowed, causing a spike that is a true reading on sensor, on certain loads. Better chance.

5-A slowed bullet and powder burn that catches
on fire again,in reaction to slowed bullet raising pressure back up, and with slowed bullet causes spike giving a true reading, on certain loads.
Better chance.

Some may think this research and discussion
is not worth the effort, but I bet an ass kick,
that there is enough guys with equipment, enough of us quarterbacks watching, that with a
little co-operation we will find the cause and cure here on AR........Ed.

PS-And one cure has been practiced for years,
IE, lower velocities with lead bullets, though
for other reasons.And most guys doing high
velocity with heavy lead bullets knew something
was out of wack.I think we can find out why.

I shoot 400 gr real hard lead bullets in my 458
wildcat at fast speeds(29-3200).Acts just like
jacketed.Spikes may not occur in big bores.
We got to test to find out some time.But I know high velocities are doable with hard lead, but we got to find right way.Ed

Ed, I'll bet you're right. If we keep picking at it, we will get an answer.

I also think we can eliminate #5. Here's the argument: A full charge of powder contains just so much chemical energy. If the bullet is more than a few inches down the barrel, even a full charge of powder does not have enough energy to re-pressurize the entire barrel to 60 KPSI. I think that it has to follow that no remaining part of the load has enough energy to do that, especially if the bullet is near the muzzle. The result of all that is that the secondary spikes are not true pressure at the chamber, but must be a real physical event nearer the muzzle. There isn't enough energy available for it to be real pressure.

I do have an oscilloscope that would give four channels, but I would have to build little differential amplifiers for each strain gage, and stick 4 gages on a rifle in order to do the experiment suggested for watching the wave propagate up and down the barrel... lots of work, and lots of competing experiments to do.

BTW, according to a source at White, the Oehler Piezo system has a feature to suppress the secondary spikes. Once the circuit detects a sustained downtrend in pressure, it quits taking data. So it is not correct to say that we know that piezo systems would not detect the spikes, if allowed to run for the full time.

Denton, you make a very good point about the over bore thread (LOL), and what's worse is that this and that thread are the most popular on the forum! Maybe I should start a thread on quantum weirdness and its implications on the nature of secondary spikes - that is, the mere act of measuring the phenomenon changes the phenomenon.

I'm just curious, I've been following this topic for quite sometime now.Just how many blowups would you say have been caused by this. And if the phenonemon causes any real damage to a rifle, how is it that yours is ready for testing after each and every firing???? Methihks this begins to look more and more like a shell game.

Denton--Mtngun muzzle sensor picked up spike, but Ithink spike starts about 50-60% the way down, according to his chamber sensor.We need to test with sensors at 60-50-40% of the way.

The theory is---
If spike starts that early there may be enough
unburnt gases to ignite added on to the
balance of regular powder burning in the catchup phase to make
spike.A little gase unburnt I think can cause
a condition approaching detonation,IE has more
power than normal for that portion that was
gaseous, and didn't burn exactly at the time
it was released from solid state.IT might be able to raise pressur in half the barrel
under those conditions.You say you made spikes in a 7.62 russian......Is this an expensive heirloom, because
I would kindly ask if you'd cut it off a couple inches at a time to see if spike goes away, using spiky 123gr bullet load that you had..
Or if not possible we can stick sensors on
another so we can get another 7.62.That might give enough info to differentiate
whether the event and/or readings are from wave action or one of our
ideas.

JP-I think he has only done it on the 7.62
and he stopped...The fella on other forum is
still doing testing and is doing it as easy
as he can, but I am sure he wouldn't be surprised if barrel muzzle let go. He acquired an extra barrel.Ed.

quote:

I do have an oscilloscope that would give four channels, but I would have to build little differential amplifiers for each strain gage

If you have a 4 channel scope, it already has the amplifiers and offseting circuits built in. You just need 3 plain resistors per bridge circuit to get the voltage offset down to where the 'scope can see it. A single 1.5V battery can provide excitation for all circuits. These circuits will not be calibrated, but they will show the time offsets correctly. Still, it is a lot of work getting all the gauges on, etc.

Just a thought, and I might be off base, as I know next to nothing about these pressure measurement systems. You guys are looking at this data in the time domain. Is it possible to get the raw data in amplitude and sample time and do an FFT/DFT on it. You could then look at the phase and magnitude of the various bins. A lot of time things jump up at you in the frequency domain.

Just a thought.

-Steve

Some answers to some questions:

Does the spike go away when the barrel is shortened? According to Charlie Sisk, yes. After blowing the muzzle off one of his barrels, he recrowned it and continued shooting it, without spikes. I plan to shorten my barrel within the next month, so we'll have another data point.

Why don't don't more guns blow up from spikes? Sisk's blow ups took place with deliberate overloads -- he won't say exactly what kind of load but apparently it was up there. He was intentionally pushing the limits to see what would happen. There have been a few ancedotal reports of ported muzzles damaged by hot handloads. Remington barrels are reputed to have a 130,000 psi yield strength, so there is a significant safety factor. Instead of worrying if our barrel is going to blow up, it's more realistic to ask if the spike has an effect on accuracy.

I have an experiment or two planned to test Oehler's ion alibi. I think their theory is pretty lame, and they are just desperately inventing theories to cover their ass, but it should be easy to test.

Hawkins, all of the strain gages on my barrel see the mega-spikes with cast bullets. There may be 15 - 25 ksi discrepancies, but that doesn't explain a 100+ ksi spike. The gage near the muzzle does not see the full pressure curve, except as low level noise.

The gage at the muzzle sees only a single significant pressure event, which tells you something very important if you think about it. If the spike were due to re-ignition or wave harmonics or some other event unrelated to what is happening at the base of the bullet, there would be two pressure events recorded by the muzzle gage: 1) normal muzzle pressure as the bullet passes by the gage and 2) the second pressure event caused by reflections, re-ignition, or what have you. The fact that there is only the single pressure event measured near the muzzle tells us that there is no second, seperate event. Whatever is happening, is happening at the base of the bullet.

Someday the data points will come together. What confuses me if it is a real pressure at the muzzle; Why does a gage at the breech see it, and why does a gage on the bottom read lower than a gage on top?. (recoil moment at the breech would give the same reading). Why does hanging a weight on the barrel change the "spike pressure . (it would change the barrel vibrations).
Finally, Why has this been discovered only when more people are measuring pressure with strain gages?.
Good Luck!

We need a test with a sensor 60% down
the barrel length from breech.

The muzzle sensor is only picking up
what pressure is left on the
high spike when bullet goes by it, but the
spike already started back up the barrel.
It didn't start exactly where muzzle sensor is.
That may be why muzzle sensor reads spike lower
as spike dropping off by the time bullet gets
to muzzle sensor.

A sensor in the middle of barrel might pick up
the full height of the spike,IE read like what
the breech sensor showed about the spike.

popemann(mtngun) put wt on barrel I
think, and it didn't make a great deal of change.Still had spike.Ed.

Hubel, A 15 pound weight dropped the pressure fron 106 kpsi to 86 kpsi. Why would the weight change the pressure?. It would change barrel vibrations though.
Good luck!

Ed, the spike appears to be happening about 3 - 4 inches from the muzzle, while my muzzle gage is located 2 inches from the muzzle. We should learn more when I start hacksawing the barrel one inch at a time.

Hawkins, the large barrel weight may have bent the barrel enough to affect bullet friction. A better test would be to apply tension to the barrel with a barrel sleeve & nut at the muzzle like a Dan Wesson revolver or else just encase the barrel in concrete. Either of those things would radically change the barrel vibrations without bending the barrel. In retrospect, my barrel weight was one of those experiments that intended to isolate a variable but instead may have introduced a new variable.

Regarding the small discrepancies seen in the magnitude of the spikes as measured by different gages on the same gun, that's a valid question. Assuming the discrepancies are real, and not just shot-to-shot variation, it may have something to do with the frequency response of the different gages.

Hawkins, regarding bending moments, at your suggestion I read Vaughn's book and the bending moments he claimed to measure were on the receiver, not the barrel, and were quite small in magnitude, about 400 in-pounds. That is the equivalent of 18 pounds of force on the end of a 22" barrel.

Why wasn't the spike seen until recently? According to RSI, Norma developed the "catch-up" theory to explain the spikes 20 years ago. Americans rely primarily on Oehler pressure measuring equipment which, as Denton pointed out, may be deliberately filtering out the spike.

I'm trying to keep an open mind as I have no axe to grind here and would love to return to shooting high velocity cast bullets without worrying about damaging my gun. However, there is enough evidence supporting the spikes to at least warrant caution and further investigation, which is what I am doing.

Popenmann--I'm sure spikes are starting
back toward breech rather than last 4 inches of
barrel.
Here the theory----
On your timeline on the traces,the spikes start
at about 2/3 the way on the time trace.
And due to bullet accelerating from rest the
aproximate time it takes bullet to get a third the way down barrel is same as time to move the last 2/3, as the accelerating rate is increasing....So at the halfway on the time scale bullet is a third the way down barrel,
we would say 33% the way...At the 2/3 mark on the timeline the bullet would be about
50-65% down the barrel, 12-15 inches on a 24
inch barrel.Or a reverse way of figuring acceleration it will take about 2/3 the time to
get halfway and the other third to get the
last half and out the barrel.And traces show spike starting about 2/3 on the timeline.
So that is why I think
that spike is starting about halfway
of the bullet travel...Ed.

Just a brainstorm here, but Denton's comment about the fact that there is not enough energy in the barrel to go back to 60 KPSI got me thinking.

If this event IS a true detonation, then the whole barrel would not HAVE to be pressurized! Detonation would mean the pressure could go up fast enough locally to create that level of pressure, without containment! It would then disperse through a shockwave throughout the barrel. FWIW, Dutch.

quote:

Originally posted by Dutch:
Just a brainstorm here, but Denton's comment about the fact that there is not enough energy in the barrel to go back to 60 KPSI got me thinking.

If this event IS a true detonation, then the whole barrel would not HAVE to be pressurized! Detonation would mean the pressure could go up fast enough locally to create that level of pressure, without containment! It would then disperse through a shockwave throughout the barrel. FWIW, Dutch.

Imagine a slow powder that hasn't yet turned to gas. The bullet slows because of increased friction. The "slowing" causes a higher pressure wave to reflect back to the still burning powder. Most smokeless powder burn rates increase with pressure and the pressure just suddenly raised.

The trigger for this event would be a sudden increase in the friction coefficient of the bore to lead bullet interface. It is what would happen when you exceed the shear rate of a lubricating oil. Remember how you can snap a Big Hunk candy bar?

Now this scenario would result in a pressure spike at the breach, but it wouldn't blow the end of the muzzle off.

Dutch--You've got a point as the spike
on the traces look
just like what you might think you'd get from
a shockwave, TO THE BREECH SENSOR..
But on sensors down the barrel they will look
like aprimary spike, as to what sensors down the barrel see, it is "their" primary spike.
And it may be just a mini detonation with the
powder and gases not burning,re-igniting and getting it on so to speak, with the rest that was still burning.

dla- It might blow the muzzle if you kept adding
powder to a spiky load like Sisk did, as that would make spike(which muzzle sensors are picking up the tailend of I think) bigger and longer to overcome the strength of the thinner
muzzle on the average rifle.

Something is occurring and
happening and it is being registered on breech sensor either because barrel is pressurized that high or a pressure wave is refected
back to it.

Another thought--

In the good old days they showed
pressure curves based more on bullet position not time, mostly military work or related to military supplied by private ammo co.Ok they made their tests at various positions along barrel and plotted results.

I am sure that if a load was put together that
showed an out of place spike at some bullet position down the barrel, loads were changed and the records burnt so to speak.It was asked above why no blowups in billions of military stuff shot over the years, well they didn't
use combinations that had problems.Ed.

Having a while to think about it, on long drive today, I think one issue can fairly simply be resolved: Is it a "ring" wave, propagating down the barrel? The answer is derived from one suggested here earlier.

All it takes is two strain gages, and two diff amps. One is the conventional over-the-chamber gage, and the other is located about halfway down the barrel, someplace well behind where the bullet is when the event occurs.

You then have to hook both gages up to a dual trace oscilloscope, so you can see both traces in true time relationship.

If the secondary spike is a ring wave, you will see it at the mid-=barrel gage before it appears at the chamber... simple as that. If the speed of the pulse is about 10,000 fps in steel, a one foot distance between the gages should yield a 100 microsecond difference in time, and that is easily resolved.

You can't do that with the PT, because the PT is a single channel device, and it triggers with respect to the signal it sees. If you put one on the muzzle, you'll see a little blip, at the far left of the screen.

I have a good scope, and I have one 7A22 diff amp plug in, which is perfect. Those are now cheap as dirt on the used market, about $35. All I have to do is find one on eBay that looks good, yank my standard dual channel plug-in, and slap in a second 7A22. From there it is duck soup... regulated supply off a little battery, three resistors and a TrimPot to make up the bridges, and a second gage slapped on the barrel.... RV battery with an inverter to power the scope, and something to power the camera. Hope they still make 107 Polaroid film.

If all goes well, I should be able to get this done before full summertime.

Ed, the spikes in question typically begin at 1.1 ms compared to the exit time of 1.25 - 1.3 ms. The distance the bullet has traveled at 1.1 ms can be estimated with Quickload. The answer is about 18 inches for the loads in question. A hacksaw should provide a more certain answer.

Denton, do you really think that sound waves traveling through steel have enough energy to cause massive strain and stress?

There is a lot of misinformation floating around in the shooting community confusing the different kinds of waves. Transverse waves do not travel at the speed of sound.

I thought that exit was about 1.5 m-sec.
With exit at say 1.25, the 18 inches is about right. 6 inches back from muzzle.

There may be energy waves of some kind in
the steel, and there may be energy waves in the
gaseous burning powder medium.
The one idea above is that the spike the breech sensor sees is a wave in the gase back toward
breech from the explosion causing the spike
reading on the breech trace.
Other idea is a traveling wave in the steel,
from the explosion,
causing an expansion the sensor sees.Maybe not a sound wave but radial expansion of the metal?

I think we will find out with folks doing
the testing and research.Ed.

quote:

There is a lot of misinformation floating around in the shooting community confusing the different kinds of waves. Transverse waves do not travel at the speed of sound.

That's right. I had forgotten that for the moment. How fast do they travel in steel?

I don't know the answer to your other question... just a mental experiment: If there is an obstruction in the bore, not big enough to cause the barrel to burst, how would the barrel respond? I think with a ring wave of considerable amplitude. Don't know.

quote:

that is, the mere act of measuring the phenomenon changes the phenomenon.

Well, we could bring up Schroedinger's cat... that is in the same vein, and would surely get lots of discussion!!

There are some pretty sharp minds on this board... no telling where that would go!

quote:

Denton; The common mode rejection may be 90 db at 60 hertz, but it reduced 20 db per decade of frequency. Rapidly changing waveforms have greatly reduced rejections. In what way is the gage shielded, either electrically of magfnetically.

Shielding: The leads to the gage are a shielded twisted pair, except for the last 1.5" or so, which are just a tightly twisted pair. On the circuit board, the PT uses four layer board, with one layer dedicated just to ground, and one dedicated to power, which is also signal ground. That's the really good, low noise way to do it. The Fabrique doesn't do that, and is noisier.

CMRR: It's true that CMRR does fall off with frequency, but I don't think it is quite so quickly as 20 dB per octave. I pulled out the manual for my Tek 7A13 (70's technology), and CMRR is flat to 100 KhZ. I also looked at the specs for the instrumentation amplifier I use (90's technology--similar to the one in the PT, but not the same). It just specifies 120 dB or better, at the gains that are likely to be used (100X, 1000X). I think CMRR is at least that over the bandwidth, which is just 200 Khz. So a 1 volt signal induced in both inputs would appear as 1 microvolt, which is about 1000X smaller than the gage signal...negligible, and I can't imagine inducing a signal as large as 1 volt, even if I tried.

quote:

Originally posted by denton:
That's right. I had forgotten that for the moment. How fast do they travel in steel?

.

Transverse or shear waves travel at ~8us/inch. Longitudinal waves travel at ~4.2us/inch.

Since the pressure in the chamber acts radially upon the walls of the chamber to compress the wall, I believe the wave launched is longitudinal, not transverse. It is the same as if the pressure were applied to the end of the end of the barrel.

No, Denton, let's not go there - although some of the comments on this forum have remind me of sundry discussions carried on regarding quantum paradoxes.

That aside, when the chamber expands following detonation, subsequent barrel expansion will follow the bullet down the barrel creating an accelerating expansion wave. Right behind that expansion wave will be a rebound contraction wave as the chamber and barrel return to their normal diameters. But, some energy from the burst of high pressure will be absorbed by (or transferred to) and stored within the elasticity of the barrel steel – now potential energy. This potential energy will cause rebound contraction of the chamber and barrel to a diameter slightly smaller than resting state. Next, like a bouncing ball, the chamber and barrel will show some rebound expansion, although diminished, until all the energy is dissipated as heat, sound, etc. Moreover, this effect may be interacting with a pressure wave(s) reflecting back from the distal barrel to the proximal barrel and/or from the tenon of the action, which will also be the recipient of some pressure-derived energy that is stored as potential energy in the elasticity of the steel.

Clearly, this is a very complex set of harmonic events that are likely specific to each gun and cartridge fired. Thus, it is reasonable to posit that under some gun/cartridge conditions the harmonic events synergize to create enough secondary chamber expansion to create a measurable secondary spike. Under other (more common conditions) the harmonic effects essentially cancel, and no secondary spike is observed.

Maybe this is all bullshit, but it sure sounds good (LOL) - like discussions of quantum paradoxes. Regards and safe/happy shooting, AIU

quote:

Originally posted by Ackley Improved User:
According to Oehler technicians, the reason for the peculiar secondary pressure spikes with the strain-gauge PSI systems is from the "electrical effects" of the ionized gases from the muzzle blast interacting with nearby connecting cables. They believe that they can reproduce the effect by strategically placing the cables such that this effect occurs, and they can eliminate the secondary spikes by simply moving the cables away from the muzzle blast. In other works, its an artifact created by the "electrical effects" of the ionized gases of the muzzle blast. Interesting idea - any thoughts?

If true, it would certainly cause one to question the validity/accuracy of pressure data obtained from such a system!

Got this from RSI site about research and
testing to get best accuracy.---

""In short, Chris believes when each shot is fired the chamber swells and produces an "annular wave" or "pressure pulse" which is then bounced back and forth along the barrel between the muzzle and receiver. This wave, or pulse, is like a "doughnut" traveling up and down the barrel and opens the bore diameter slightly so if the bullet exits the barrel coincidentally with the wave at the muzzle, the barrel will behave as if it has a bad crown. Others have espoused a similar theory but Long is the first to develop a way to predict when the wave(s) are at the end of the barrel so the theory can be applied. For more information about Chris' epiphany, click here.""

You can go to site and read more.The reason I posted this is that I also think that this pressure pulse or donut effect traveling in
the barrel structure exists. These folks figuring it out so as to get around the lossof accuracy that can occur if bullet exits atwrong time.My reason for posting is that these pulses
in the barrel were not of enough strength to trigger muzzle sensors on Mtngun's test.It was only triggered by the much greater expansion of the barrel
from pressure behind the bullet when bullet got there.So this to me is one thing pointing away from pulses or harmonics causing false
readings.Pressure waves in the gases may
be the culprit, and be a different thing if the spike is like a mini detonation, because the speed of the explosion is much faster than the normal propellent burning speed.And it may be traveling back through gase if that is possible
to breech sensor as outlined above.
The following is also from RSI site--

""In every instance when secondary pressures are detected they can be eliminated by using a faster power, heavier bullet or a bullet with more bore contact area. Normal "tweaking" of loads may change the peak of secondary pressures but will not eliminate the problem. Below is the list of factors we now know can cause secondary pressures.

Powder burn rate too slow for the bullet.
Bullet weight too light for the powder's burn rate.
Bullet bore contact area less then normal for the bullet weight
Barrel longer then normal
Bore severely worn or incorrectly lapped (loose/worn toward the muzzle)
Moly in bore or moly coated bullets that reduce bore friction
We are often asked when secondary pressures are too high. Obviously secondary pressures more then 25,000 or 30,000 PSI at the weakest part of a barrel represent a safety issue. On a more practical note, loads that exhibit secondary pressures often show significant variation in barrel timing (when the bullet exits at the muzzle). Even if the timing does not vary shot to shot, it certainly will when the temperature changes so these loads rarely shoot well. Our advice is simply to avoid all loads that produce secondary pressures and keep peak pressures where they are supposed to be, in the chamber. If you shoot factory ammo, try a different brand. If you reload, use a slightly faster powder or heavier bullet.

I recently had the occasion to work with a friend's .223 Douglas barrel that had shot perhaps 6,000 rounds of an off brand ball powder sold as "Data 2200". My understanding is the powder was actually reject 2230, a powder we know produces secondary pressures in small calibers. The owner shot around 2,000 rounds of the load through the barrel each season, then re chamber his good ol' shooter as the throat washed out. When I looked down the barrel with a bore scope I could see rings just down from the muzzle spaced exactly the amount that was removed from the chamber end each time the barrel was re chambered. This convinced me secondary pressures will eventually damage a good barrel.

No doubt there are shooters who will debate our conclusions. We welcome anyone to pose other possible explanations. The "catch up" theory is the only one we have found both fits the evidence and can be used to eliminate these problems. To date, only shooters who do not have access to pressure testing equipment argue adamantly against the "catch up" theory. Professional ballisticians we have talked to whose job it is to formulate powder seem to be in full agreement. Even if the theory is not the "entire" explanation, it is certainly useful."" unquote.
So this writer has seen
evidence of the little bit of bulging from
the phenomenon in a 223 Douglas barrel.
at about the same point fromm breech as
the barrel was set back due to throat wear.
A physical manifestion can't occur from a
ghost event. It is real, however the actual
event is happening...Ed.

I think we can rule out any kind of longitudinal wave. Strain gages are sensitive only in one direction. Cross axis, sensitivity is about .3%.

Also, we refer to them as spikes, which is probably OK, but they are not truly spikes. I think this illusion is created by the way the PT system responds. To meet its design function, there is no reason why the PT needs to be able to show the strain gage in compression, and it cannot.

The following was observed once, and I need to be able to reproduce it a few times, in different rifles, to make sure I'm not seeing just a special case: As I increased powder, first I noticed a little flat spot in the decay of the pressure. As I continued up, I saw a single cycle sine wave in the decay. With just a little more powder, the amplitude of the sine wave grew, until the negative half hit zero, where the PT circuitry limited it. The positive half then continued to grow rapidly with increasing powder, until it limited out at the upper limit of the PT. If that's correct, we aren't seeing a pulse at all. We are seeing the positive half of a superimposed sine wave, and the PT isn't letting us see the other half... not within its intended design parameters.

With the two gage experiment, I'll use dual supplies for the gage bridge, and put the quiescent condition at about zero. That will let me see the full superimposed wave.

As I said, there are a lot of good heads out there working on this. AIU's argument is an interesting one. It leads to the reverse question: Why is something like this not always present?? Maybe it is, and just not within the display window.

I'll make it a point to take a little care in placing the two gages at exactly some chosen distance. If we get a ring wave, we will be able to see how fast it propagates.

quote:

To date, only shooters who do not have access to pressure testing equipment argue adamantly against the "catch up" theory.

Ummmm, well.... I just can't swallow the "catch up" theory. And I have pressure testing equipment. First, I can't get a mental picture of how it could actually happen. And, if it did, I can't see how it can produce anything but longitudinal waves, which a strain gage would ignore.

Don't have enough info to really argue against it, but, so far, have not accepted it or any other theory.

Denton-You may be right.But there is the possibility that if the "catch up" buring phase
is more like an explosion or mini detonation
it might have the power to send
a wave back through the gase to breech sensor,
or repressurize the combustion space so the
breech sensor picks it up.They stated on RSI site that the spike trace sensed and printed out
is actually longer on the graph than what it is in the real, due to nature of the machine, so perhaps it takes very little of a real sharp
mini detonation to make machine read and print out a larger event than the real.And that goes along with what you said about there not being enough enegy left in powder to re-pressurize whole barrel.Under regular burning conditions. But a real short sharp mini detonation might be the thing.And in a way that may inadvertantly be ok for shooters, for if
a detonation started with say a 1/4 of the energy left
there would be more than blown muzzles.
The whole gun would unglue.

If a propellent burns say 95% away under regular propellent burning rates, and the last 5% is a gase and it detonates, there is more power in that 5% than the rest, especially
if applied wrong.And denonation being so fast and erratic doesn't do anything for extra for bullet speed or accuracy.It may bind the bullet, as mtngun's spiky loads sometimes were slower than regular.

If your setup is put at the breech and picks up a conventional sine wave, with positive and negative aspects, that may confirm the wave pressure in the gases, as caused by small amount of powder left in gaseous form, detonating causing the wave.Maybe that is how a wave would affect sensors set up to pick up pos and neg parts of a wave.I understand high explosive exhibit pos and neg pressures
in the waves they send out from the "point of the explosion". So with certain loads 16-18
inches down the barrel something is creating spike and sensors are picking it up.The how has to be determined for sure.To protect shooters and guns.There is a poster over on Campfire
who had spikes with his 338Win.No one
even responde. And many others seem to try to
ignore the problem.Reason I mention this is
if this problem is a catchup of the combustion process that has slowed and it is a mini
detonation, it isn't going to catch up or increase bullet acceleration, THEN IT CAN BE
A LITTLE BOMB, if too much powder/gase combination goes off.Like Charley did on purpose.Problem should be solved.Ed.

quote:

Shielding: The leads to the gage are a shielded twisted pair, except for the last 1.5" or so, which are just a tightly twisted pair

2 POints here to reinforce the 'can't be electrical interference' issue... First, there is a reson we EE types use twisted pairs...so that any noise that gets injected into the wiring gets injected equally into both leads, so that the differential (critical word...more in a sec) amplifier rejects it equally.

Second, the differential amplifier only amplifies the difference in the 2 leads...so noise is automatically reduced/eliminated. A really good circut doesn't even need the shielding (and doesn't actually have a true ground), but it (shielding) does help some, especially with large noise pulses that could induce enough voltage to damage the relatively sensitive input circuitry.

Denton, when you set up your experiment, the timing is likely to be tight enough that things like the length of the leads may be critical. I forget the exact propogation speed of electrical signals in copper, but it is a variable that is easy to control. Matching the gauge factors would be good too. That would make switching the gauge leads between gauges easy...to validate your results...insuring that circuit delay isn't skewing the results.

I still believe the answer lies in some sort of 'hoop stress wave' and some harmonics hitting just the right frequency, but hey, I am an EE, not a ME! If you can get a digitizing o-scope, a snapshot of the display is easy and would make great study material for a us nerdy types!

Have fun!

quote:

I can't see how it can produce anything but longitudinal waves, which a strain gage would ignore

A true detonation would create a gas pressure wave that would expand the barrel in exactly the same manner as the normal combustion...just with much higher magnitude for the same amount of energy released.

quote:

Originally posted by denton:
I think we can rule out any kind of longitudinal wave. Strain gages are sensitive only in one direction. Cross axis, sensitivity is about .3%.
.

I respectfully disagree. I believe that the wave is longitudinal, not transverse. In fact, there isn' a method of creating a transverse wave by a pressure acting within a cylinder - unless the cylinder ruptures Also, I can't see how you could mount a strain guage to ignore a longitudinal wave - the strain gauge is going to react to the barrel expanding (thickening).

But I think we need to be open to the idea that there may be two effects at play here. The two are separate, but may be viewed similarly by the strain gauge measurement system.

The two are:

1. Bullet friction variation. This is where the bullet to bore friction changes abruptly, most likely increasing. The bullet slows suddenly, creating an incident higher pressure wave that propagates through the super-heated gas at about 20us/inch (@1100*C) and influences the still burning powder (4350/4381, etc). The burn rate increases (perhaps detonation?) due to the high pressue wave, sending a higher pressure wave propagation down the barrel and catching up with the bullet.Evidence in favor:Barrel ringing.
   
2. A longitudinal wave that gains in amplitude from the reflection from the muzzle, and loses amplitude from reflecting from the receiver. This wave may cause permanent thinning of the barrel if the pressure exceeds the yield point. Otherwise this wave will simply distort Strain guage measurements. Evidence in favor: Piezo tranducers do not "see" the secondary pressure peak. The piezo transducer is not influenced by the barrel deforming, as is the strain guage
   

Also I'm wondering about the bandwidth of Piezo transducers. Obviously the Piezo system can record intial combustion, where the rate change (min to max) is about 300 us. But the rate of change for bullet "stiction" could be much faster.

Thanks for the thoughts guys. This problem is intriguing.

dla wrote--

""Bullet friction variation. This is where the bullet to bore friction changes abruptly, most likely increasing. The bullet slows suddenly, creating an incident higher pressure wave that propagates through the super-heated gas at about 20us/inch (@1100*C) and influences the still burning powder (4350/4381, etc). The burn rate increases (perhaps detonation?) due to the high pressue wave, sending a higher pressure wave propagation down the barrel and catching up with the bullet.Evidence in favor:Barrel ringing.""

This sounds like a resonable idea.I would add
to the beginning of that description,--A light or slippery bullet jumps out fast--letting pressure off of powder a little- combustion slows--then bullet slows--raising pressure--
combustion picks up---like a mini detonation-- and so on.

In description of idea 2 you state that piezo
transducers don't pick spikes up.But I know
that in the barrel sensors
would if they were placed at the point of the
bullet travel where spike occurs.I am
sure in old days when testing was done at
all points on the barrel and pressure curve was plotted based on bullet position, that they found spikes on loads.They discarded same as
unsafe and unusable as well as no public info
about it...Ed.

quote:

Strain gages are sensitive only in one direction. Cross axis, sensitivity is about .3%.

Denton, are you sure about that 0.3% figure? I thought it was far higher, due to Poisson's ratio. Three gauge rosettes are sold to allow one to sort out all the cross axis effects.

Also, earlier you mentioned someone at "White" reports some piezo systems filter out secondary pulses. I take it that is H.P. White labs? If so, that ups the odds this is all truly created by high gas pressures.

Poissons ratio is 0.3 not 0.3%.
The "differential" amplifier responds to the difference in the gage leads and the rest of the resistance bridge in the equipment.
Any noise pickup in the gage leads will not be rejected.
If you doubt if a barrel has a "doughnut" wave when fired, why will placing the bare barrel on a hard surface cause the gun to shoot high?.
One of these days we may understand this stuff, if we do perhaps it will result in higher velocities.
Good Luck!

.3% is the strain gage manufacturer's spec for cross-axis sensitivity. They just don't sense much except in one direction.

Now if the steel is somehow actually translating longitudinal vibration to transverse, then there is really something there to sense.

Yes. "White" refers to H. P. White Labs.

The multi-gage assemblies that you can get are really slick. For one thing, all the gages are stuck to the same object, so you get phenomonal temperature compensation. They even have some very slick models that sense torsion.

Denton: you are confusing terms. Cross axis sensitivty is the gage responding to a major stress axis perpendicular to it. Poissons ratio is a physical property. A cross axis stress will have a 0.3 effect on the material itself.
Good luck!

Well here goes.I may get hollered at for this.

Ok- slippery, light, easy to move bullet jumps out fast--outruns regular propellent burn--
burn slows--bullet slows--causing pressure to
reignite powder and gases--they have a mini
detonation as the burning catches up--a spike occurs that breech sensor sees-- spike pressure on the rear of bullet
2/3 the way down the barrel, causes the rings seen by some(or if high enough in thin barrel,
splits thinner muzzle)..

So to me it is a MINI SEE with bullet 2/3
the way down the barrel, as opposed to
regular SEE occuring behind a parked bullet
at begining of the cycle.

Secondary Explosive Effect with bullet moving
and down the barrel.............

This Mini SEE could occur in all parts of the
combustion chamber made up of the barrel from breech to where bullet is at the time..Like a regular SEE flashing over the whole length.
Or the Mini SEE occuring in front half of the
burning length and sending wave back to breech sensor. Denton's sine wave detection set up may help figure which way.And Mtngun cutting off barrel in his experiments, may make event not happen as bullet might be out of barrel, at the time powder
slowed a little so it couldn't repressurize
burning powder mixture.

Is this too simple--Ed.

quote:

Originally posted by hubel458:
Secondary Explosive Effect with bullet moving
and down the barrel.............
.

I think that is a good definition.

I think it is important to consider that the powder doesn't stay all bunched up inside the cartridge. It spreads down the barrel as the bullet moves.

To create a reflected wave, the bullet must slow very quickly - a step function. We know that slow powders are prone to detonation under the right circumstances, and I think that the incident wave caused by the abrupt slowing of the bullet could cause a detonation anywhere.

Also, most folks apply a good crimp to facilitate a uniform ignition with slow powders (e.g. W296/H110 in large pistols). Has the effect of crimp been measured? It is pretty easy to buy a $13 Lee Factory Crimp Die and vary the crimp pressure from "not much" to "outrageous". The FCD will create more than 2X the holding force of a roll-crimp. This might be a good way to check the bullet jump issue.