Secondary pressure spikes in 223

[603Country]

a recent thread about using BL-C(2) powder in the 223 brought up the issue of potential secondary pressure spikes. Naturally, since I have some of that powder, that concerns me. So I did some research on the issue. Understand that I am not speaking from the position of having great knowledge on this, but rather from a position of wanting more info.

What I found was that BL-C(2) is just one of the powders that can, in certain circumstances, have secondary pressure spikes. If a person has the wrong combination of barrel characteristics, several powders can give problems. AA2230 is also one of them, and is a powder I use also.

As I understand it, those 'wrong' barrel characteristics would be: over 20 inches long; worn bore; rough bore, and shooting light bullets in that barrel.

With that in mind, as soon as the gunsmith will give me my rifle back, it'll have a 20 inch expensive custom barrel on it. And I am wondering if perhaps now I can use the BL-C(2) and AA2230 (and H335) and my favorite little bullets (Nosler 40 gr BT's) and not worry about the secondary pressure spikes.

I know that Unclenick is knowledgable on this issue. Any info or opinions he, or any of you, can share would be appreciated.

[bluetopper]

Personally, I'll rank this concern right below global warming.

[F. Guffey]

a recent thread about using BL-C(2) powder in the 223 brought up the issue of potential secondary pressure spikes.

I have read a few threads about secondary spikes, my opinion, those posting could not read right and up and did not know where the bullet was when the spike occurred.

I am the fan of the running start, I want my bullets to have 'the jump'.

F. Guffey

[Unclenick]

Bluetoopper,

Some folks use "Secondary Spike" to refer to SEE (secondary explosion effect) or detonation, which is, indeed, rare. That's not what it refers to here, though. Here they are radial expansions that occur down barrel when the bullet has moved out faster than the powder can make gas to keep up with expansion. As it burns, though, it builds enough pressure behind its own mass to propel that mass to catch with and slam into the base of the bullet. The resulting radial expansion of the metal sets up a transverse wave that shows up in strain gauge readings at the chamber looking like a chamber pressure spike, even though there's actually no excess pressure at that location. What is seen is actually a reflection of the collision that momentarily stressed the bore radially further down. Instruments that read only chamber pressure, like copper crushers and piezo transducers, don't see the reflected wave, as their mounts tend to move with it. You have to use strain gauge instrumentation to pick it up.

Texas gunsmith Charlie Sisk had some photos up on, I think, THR or 24 Hr. Campfire at one point, showing some .338's barrels he had blown the muzzles off of with about ten rounds of a load that created a particularly bad secondary spike of this type. Jim Ristow of RSI, maker of the Pressure Trace instrument has an article that discusses this starting about half way down. He describes borescoping a ringed 223 barrel that had been beat on by enough rounds that they caused that effect, and seeing separate rings corresponding to how far the barrel had been set back for rechambering.

Some folks are in denial that the phenomenon is real, instead thinking it must be some instrumentation phenomenon. I was in that camp until Sisk put his blown up muzzles photos up. That was enough proof in the pudding for me.


603Country,

The secondary spikes seem to care mainly about powder burn rate and bullet mass. Barrel condition will affect friction, but friction is down around 3-6% of the resistance against pushing a bullet down the tube (the rest is bullet inertia), which I don't expect to be enough difference to help you out much. I'd be looking at a powder like Reloader 10X or even 4198 for a 40 grain bullet in the .223. Unfortunately, buying a Pressure Trace and putting a gauge on the barrel is about the only way to be sure of what's happening in your particular gun.

[603Country]

From what I've read, the spike occurs pretty much only in long barrels, and rarely occurs in barrels of 20 inches or less. My barrel is now 20 inches long.

In one particular recorded instance of the spike, it happened at 21.6 inches of the barrel (in two cases). I can't imagine that it would be 21.6 inches in every case, because it depends on too many factors. Further, because of the variables involved, such as a worn barrel or really rough bore, I wouldn't think that the secondary pressure spike could be predicted in any barrel or at any point in the barrel. The variables or factors I refer to are:

- worn (loose) bore
- rough bore, either pitted or rusty
- long barrel, though what qualified as long wasn't stated, but was over 20"
- light bullet for the burn rate of the powder

I may have left a couple of those factors out, but these I remember. The article also suggested that the secondary pressure spikes would not normally allow for great accuracy, since it would affect the barrel residence time of the bullet and the bullet exit would vary with the inconsistent barrel harmonics. Therefore, it suggested that if you were getting great accuracy you are not having issues with that secondary pressure spike.

So...I don't fear for my life, but I do worry a bit about my new expensive barrel. Since I mostly use H335 and now have a new supply, this whole thing becomes moot. But, I have a good bit of BL-C(2) and AA2230 and intend to use it with my 40 gr Noslers, which is the bullet I launch most often in my 223, though I suppose I could just use those powders behind the 64 gr Nosler BSB or 65 gr Sierra GK. Minor issue, but I probably won't live long enough to use up those powders behind those bullets. I'll have to explain all this to my grandson, so he won't have to read all these posts after I'm gone.

[bluetopper]

Due to this thread I'm going to take a hacksaw to my rifle barrels and cut them down.

[F. Guffey]

From what I've read, the spike occurs pretty much only in long barrels, and rarely occurs in barrels of 20 inches or less. My barrel is now 20 inches long.

If you want to see a secondary spike chamber an 8mm57 in a 30/06 chamber then pull the trigger. There is a chance the receiver will be rendered scrap, there is also a chance the case head will fail.

F. Guffey

[308Loader]

I have a can of blc2 for sale.

[Jim243]

Boy am I glad my barrels are only 16 & 18 inches long and use 55 grain or heavier bullets. I have at times thought about using light weight bullets for my 243, but I think I will just stick to the 85, 95 and 105 grain bullets for that one.

Thanks 603 that is good info to know.

Jim

[F. Guffey]

Then there was B-Square, they decided to blow up a rifle, it was an M1917, they almost gave up when they switched to pistol powder. They did manage to split the barrel, when finished the barrel took on a flower design.

I do not have problems with long barrels, the longer the barrel the slower the powder.

F. Guffey

[603Country]

Jim243, don't take what I said as gospel. You know where I got the info...the internet. Unclenick would be the person most likely to have the best info, but I'm trying to understand as much as I can on the issue. Being an engineer, I always want to understand things rather than just hear it and believe. Doesn't always work that way though. My plan going forward is to keep using the H335 as my primary powder behind the 40 gr Noslers and the BL-C(2) behind the 64 gr Nosler BSB bullets. I've used a lot of AA2230, but always behind the 65 gr Sierra GK.

My gunsmith just told me that my rifle should be ready late this coming week. I'm excited.

[Longshot4]

I have been using BLC2 for about 30 years in a Rem. 700 in 222 Rem. BLC2 is my #1 choice for the 222 for accuracy. Although my choice primer to go with it is Rem. 7 1/2 do to it is what gives best accuracy. That might be what the spikes are about. How did you find pressure spikes?

[Unclenick]

Longshot,

They show up in strain gauge measurements of pressure (see images below). BL-C(2) is popular, and its fine with heavier bullet weights for sure, but it is 7.62 powder. The version the military likes for 5.56 is sold as H335. Same powder, but faster burn rate due to lower surface deterrent concentration.

Whether it has caused your guns any issue or not depends, in part, on bullet weight. Bore scope examination is the only way I know to look for a problem positively. You can ring a barrel a little and never notice it on the target for the same reason some pitted barrels can still shoot pretty well.


Bluetopper,

Set down the hack saw. Pop the cap off a bottle of brew. Relax. Don't worry. There's no need for anything as drastic as a muzzlectomy.

These spikes reliably disappear if you either increase bullet weight or change to a powder with a faster burn rate for the cartridge and bullet weight.

"First contemplation of the problems of Interior Ballistics gives the impression that they should yield rather easily to relatively simple methods of analysis. Further study shows the subject to be of almost unbelievable complexity."

Homer Powley

"Perhaps there is no more persistent crank than the Rifle Crank. He has more theories to the square inch than there are hairs on a dogs back."

Theodore Roosevelt III

603Country,

Well, you've gone and opened a can of worms. With kind permission to hot link from Jim Ristow at shootingsoftware.com, take a look at the two misbehaving .223 loads below. One has a muzzle velocity of about 3100 fps, and the other about 3500 fps, but notice the secondary spikes start and peak in about the same number of milliseconds. You can bet the two different velocity bullets are not the same distance down the barrel when these peaks occur. As near as I can tell by tweaking QuickLOAD for pressure match, it works out the 40 grain bullet's pressure peak occurs right about the time it would exit a 20” barrel, but the 55 grain bullet is only about 17.5” down the barrel. Assuming it took some time for the barrel distortion to travel to the gauge at the chamber, the actual bullet positions would be closer to the breech. The peak is additive to the chamber pressure curve, further confusing matters.





And then, just to add insult to injury, a load that I measured at a snail's 2792 fps @ 15 ft:



The tiny secondary spike starts late, but peaks at about the same time as the others. The bullet would have been about 12.5 inches down the bore at the time of the peak. Indeed, if you look through RSI's stuff, only their 7 mm STW load has different timing; a peak at around 1.3 ms.



So, by now you should be questioning everything you thought you knew affected where the bullet was when these secondary “pressure” readings occur. That coincidental peak timing is exactly why I thought it was some kind of instrumentation anomaly when I first saw the effect. It doesn't seem to matter what the chambering is, or what final velocity the bullet is, much less any differences in barrel condition or friction. That first plot is for a moly-coated bullet, demonstrating that reduced friction doesn’t affect the peak timing.

But the temptation to think it’s just the instrumentation is roundly knocked down by the fact you can make the phenomenon stop with heavier bullets or faster powder. Plus, there is the 800 lb gorilla in the room, Charlie Sisk's blown-off muzzles. These both say, no, unless you believe you are living in the Matrix, it's not just a computer illusion.

It dawned on me eventually that what the two .223's and the '03 might have in common was they were all guns with 24” barrels. The 7 mm STW probably had a 26” barrel, as a lot of longer case magnum sporting rifles do. So what the similar timing proves, I think, is a barrel disturbance or wave that strains the chamber area enough to look like a high pressure. Indeed, you can't strain steel enough to look like a high pressure without some serious stress, though whether that is all due to pressure directly or not, I can't say without more data than we have. Characterizing the strain better looks like it would require putting gauges around the barrel to catch different axes, and several hundred dollars worth of strain gauges up and down the length of the barrel, also on different axes, then a data logger fast enough to capture it all and show the relative timing. And, of course, we'd want to know where it looks like the bullet actually is each step of the way, which multiple gauges would reveal.

The spike does not appear to be a resonance, as the curve decays on the back side, same as a pressure curve. Varmint Al, a former Lawrence Livermore National Laboratories engineer, did a lot of modeling of firearm events on their fancy FEA software that you can see on his site. He concluded:

Originally Posted by Vamint Al

Maybe the "consensus" was that a rifle barrel vibrated in one or more of the mode shapes when fired. That was because the mode shapes and frequencies were easy to calculate and they did seem to answer some of the questions. From these FEA dynamic pressure calculations, it appears that the recoil and forced deformations are much more important than the natural vibration modes in determining where a barrel is pointing when the bullet exits the muzzle.

So, what are we seeing exactly? I don’t know. My current working theory is that when the powder mass catches up with the bullet, the bullet is upset outward during the attempt at sudden momentum transfer. If that stretches the bore radially even a little it will greatly increase the apparent friction. At that point pressure will abruptly straighten and push forward on the bullet base and bore. There are a couple of kinds of elastic event you can imagine occurring from that. One would be like Al’s Mode 6 on the harmonics page linked to above. Looking at how a long barrel deflects in his rifle animation, you can imagine that something pulling the barrel straight would crack a deflecting muzzle like a whip, which would be the other form. The problem is, I would expect an elastic event like that to ring, and not just stop and be damped out, the way the spike is. Though, of course, the stock bedding may do some of that damping, as might the receiver mass. Still, I’d expect some amount of reflection and ringing.

That’s about as far as I’ve got with it. Expensive measurements appear to be needed to get further in a definitive way.

[603Country]

Thanks for the update Unclenick. All of this is very interesting to me. As for what to do from now on, I'll stick to H335 behind the 40 grain Noslers. I'll use the BL-C(2) for the 64+ grain bullets. With my fancy new barrel, it ought to shoot all my bullet types pretty well, so I'll just go ahead and expect good accuracy.

Sorry about getting everyone all riled up over this. That wasn't my intention.

[Unclenick]

It's a controversial thing because you can tolerate some of it, but not too much.

Just to add to the mess, a number of experimenters have shown spherical propellants have more tendency to cause the spikes than similar burn rate stick powders. It's probably because spherical propellant progressivity depends entirely on deterrents, making it easy for a drop in pressure to slow them down, while stick progressivity is more geometry dependent, and so they have less deterrent to impede a burn once it gets underway. This makes them a bit less sensitive to the rapid expansion and less prone to squibbing out (though I have seen examples of it having happened, so this is a matter of degrees rather than all or nothing).

Personally, for light bullets in .223, I usually use IMR 4198. It seems to have OK indifference to small charge variations, so the fact it doesn't meter as well as a spherical powder doesn't bother me. In Hodgdon's published data its maximum load of 22.2 grains fills the case about 100% under the bullet and gets within 25 fps of the fastest powder loads they publish. That maximum load of 22.2 grains beat 28 grains of H335 and 28.5 grains of BL-C(2) by about 75 and 40 fps, respectively, so it saves you about 25% on powder weight, which should be close to the cost savings.

Note, by the way, that Hodgdon lists that bullet seated to 2.280" COL. That's over the SAAMI max. Alliant uses 2.155" COL. I'm thinking Hodgdon typo'd 2.180", but I'll check with them on that.

One thing I haven't tried with light bullets are the Ramshot spherical powders. They may be less spike prone. BL-C(2) and H335 and H380 and H414/760 and 748 are all older St. Mark's Western Canon (WC) powders. The deterrent coatings on these powders are notoriously more difficult to light than stick powders and are the reason CCI changed their magnum primer formulation in 1989 just for lighting these powders up, and it is why magnum primers, in general, get better results from them. The Ramshot line is newer, however, and their tech told me they use a more modern deterrent chemistry that ignites more reliably so you don't need magnum primers with them. That same easier ignition should also make them less prone to slowing burn or squibbing out, so I'm speculating they may be less prone to making the spikes. But without testing it, I won't know for sure.

Significantly increased performance comes from Alliant's Power Pro 1200-R, which claims to get to get the same Nosler 40 grain BT Hodgdon tested to a fairly astonishing 3800 fps in a 24" tube with a 26.8 grain maximum charge. It is spherical, but they fire it with a Federal 205M primer, not a magnum primer, so this is likely another modern deterrent coating powder. It may, in fact, be the same or about the same as Ramshot Exterminator, which claims to get the same bullet to about 2750 fps with a 27.0 maximum charge. It wouldn't surprise me if the turned out to be the same powder from the same manufacturer, with the performance differences due to the variations in test gear and case and primer choices and conditions.

[F. Guffey]

So, by now you should be questioning everything you thought you knew affected where the bullet was when these secondary “pressure” readings occur.

Not me, I am the fan of the running start, I do not want my bullet to hesitate, I give my bullets the 'running start'. When reading time and pressure the person reading the graph/scope should be able to tell 'when it happened' and how much pressure was created.

F. Guffey

[603Country]

Unclenick, it appears that we are on the same page in a few areas. One is the stick powder versus the spherical, with the spherical (some of them anyway) being harder to light off. So, since I have a good bit of IMR3031, I'll give it a try as soon as the gunsmith will let me pay the ransom on my 223 and get it back. I'd go get some 4198, assuming that Cabelas still has some on the shelves, but the truth is that I just have too much darn powder now, so I need to shoot up some of what I have before I add to the pile. I'll move the BL-C(2) to the grandson's 308. It won't matter to him.

As for primers, all I use these days are CCI BR primers. Maybe you can tell me if they are anything other than standard strength. I've never bothered to find out, though that's info I should have taken the time to get.

And, something of minor interest perhaps is that when I graduated from college in Chemical Engineering, one of the job offers I looked at was the powder plant in St Marks, Florida. I don't remember who owned it at that time. Didn't matter, since I took a higher paying job in New Orleans, which looked like a fun place to live.

[F. Guffey]

Then there is seating the bullet into the lands and the claim seating the bullet into the lands will increase pressure and always following that is the saying a reloader should reduce powder by 'X' % when seating bullets into the lands and then etc..

If there is/was any truth to any of that a graph should show a pressure spike showing time and pressure. Absent in all graphs is a hint of a spike in pressure indicating the bullet hit the lands.

As suggested the last time graphs were posted chamber an 8mm57 Mauser round in a 30/06 chamber. When fired everything looks normal then! the ship hits the sand. the case head will crush/upset, the distance from the top of the cup above the web to the case head will shorten. The flash hole/primer pocket will increase in diameter, then there is that part about catastrophic coming into play. If the case head crushes to the point the case head is no longer supported there will be a separation between the case head and case body.

If someone wants to understand what happens and when I suggest loading the 8mm57 round down to start with.

F. Guffey

[F. Guffey]

match the graph to the 'Old saying': The firing pin drives the case, bullet and powder forward to the shoulder of the chamber and then the primers is crushed and fires the round. So I wonder (not really), if the graph shows time and pressure how much time does it take for the firing pin to reach the primer and traveling very slowly, how much time does it take for the case to go from the rear of the chamber to the front of the chamber?

I have said I have fired cases that never made it to the front of the chamber, and I have suggested reloaders should be able to determine if the case formed or stretched. Again, on a graph the graph records events in time and pressure.

F. Guffey

[Bongo Boy]

I'm not quite following the pressure plots--are these measured or modeled, and what are t1, t2...t4?

[F. Guffey]

what are t1, t2...t4?

T1. T2 and T4 are events happening in a part of the barrel I do not have. I do not have barrels that can tolerate those pressures, the only parts I have on rifles that can tolerate those pressures are located in the chamber.

There is a bullet maker in the very far north with a barrel hanging on the wall, the muzzle is opened up about a foot from the exit hole. The barrel is identical to a barrel pictured on the Internet. A shooter near Houston forgot to remove a bore scope, then there was the M1917 that hung on the wall in Ft. Worth, TX. It was named the 'Buck Horn Rifle' because of the curled metal at the muzzle.

F. Guffey

[Unclenick]

Bongo Boy,

These are measurements made with a strain gauge datalogging instrument. The resolution is only 8 bits, so with the 50,000 psi scale, the resolution jumps in 195 psi increments.

Strain is how much something bends or deflects in response to the stress of an applied force. The strain gauge used here is the most common type that is a zig-zag pattern of an electrically resistive foil sandwiched between two thin insulating film layers. You glue them to things you want to measure strain on, and wires run from the ends of the foil to form an arm of a resistance measuring bridge. After gluing these gauges to something, when that something stretches or compresses, the gauge stretches or compresses with it, causing its resistance to go up or down. By measuring that resistance change, and knowing the bending properties of the material it is attached to, you learn how much the something was strained (deformed) by the stress force applied to it.

On metal, strain gauge measurements are extremely linear, out to many decimal places, just so long as the metal is strained within its elastic limits. This is how laboratory scale load cells are made; by gluing strain gauges to small metal beams that the scale pan sits on. In effect, the metal beams are a spring and the gauge changes resistance in proportion to how much that spring is bent by the weight on it. On the outside of a rifle chamber, the gauge changes resistance in proportion to how much the steel expands underneath it due to the pressure inside. This is a quick event, but the traces in the images are showing that resistance change and what pressure would produce the amount of stress that results in the amount of strain that the changing gauge resistance shows.

T1 is test one, T2 is test two, etc. Each of the tests is a different shot. Across the bottom of the scale is the time starting from the trigger pressure set on the test instrument. The vertical left scale is the corresponding pressure based on the thickness of the steel wall of the chamber under the gauge. Each peak pressure labeled T(n) corresponds to the round that produced the pressure plot that is the same color as the lettering.

In the plot I made of the .30-06 in my dad's '03, T1 had the cable connector come loose during the measurement, so the measurement just showed a lot contact make and break garbage, so I removed it. That's why that graph starts with T2.


Mr. Guffey,

Below may be a plot that will interest you. It shows three pressure traces from three rounds fired touching the lands, T1, T2, and T3, and then four pressure traces of the same load with the bullet seated 0.030" deeper, T3, T4, T5, and T6. You can see the pressure peaks are about 20% lower for the last four rounds. In this instance, the measurement timer starts when the strain corresponds to 10,000 psi. At just past the 1.0 millisecond point beyond that, the bullet has cleared the muzzle. It then takes just a fraction of a millisecond longer for the drop in pressure to reach the chamber and the reading to drop zero. The fact the reading is slightly above zero, as many are, is some sort of direct current offset or drift, caused either by a capacitor taking a charge, or by glue recovery hysteresis in the gauge interface surface. I don't know which.



In 1965, Dr. Lloyd Brownell ran the same experiment, also using strain gauges. The plot below is of how the peak pressure changed as he went from touching the lands to seating very deeply. It's interesting for a couple of reasons. One is that there is a pressure minimum, below which seating even deeper raises pressure again. The other is that the drop in pressure is continuous as you come off the lands, and not a sudden jump. Brownell's explanation was not the usual static vs. kinetic coefficient of friction one you hear which is based on the assumption there is a more sudden change in pressure from touching the lands. Instead he suggests that after the neck expands to release the bullet and before the bullet can reach the throat, there is gas bypassing it, and that stops the pressure from building as much before the bullet is running away from the chamber, expanding the burning space. That running start then prevents the remaining powder from reach as high a peak. This plot of peak pressure vs. seating depth is for a round nose bullet in .30-06. A round nose has to move further than a Spitzer nose does to make as big a gas bypass opening around the bullet. That's why the pressure change in the 6 mm plot above is more sudden for a shorter seating depth difference.



Nick

[griz]

I occurs to me that most of these powders, and barrels longer than 20 inches, were in use for decades before the ability to plot pressure versus time. I find it hard to believe that a load published 50 years ago will suddenly start blowing the end off of barrels because we discovered that it has been spiking the pressure all along.

[Unclenick]

That seems improbable, but you have to be aware that powders distributed to handloaders often wind up differently-sourced over time and that 50 years ago they didn't control burn rates of canister grade powders the way they do now. Also, as Allan Jones points out, primers change more often than people realize, and without public notice, and none of the ones we have today exactly match what there was 50 years ago. The two facts combined means many loads from 50 years ago are never replicated exactly today, even though it may appear from pressure signs and velocities that they are.

IMR 4895, for example, has come through at least three different companies and plants and two different processes in that 50 year period, with good powder from DuPont, followed by some pretty terrible production out of Radford for a few years, then taken over by GD and moved to the Valleyfield plant in Canada, and more recently going through a process change as Valleyfield announced the old IMR process was no longer economically feasible to maintain. That process change is one reason you are seeing some new IMR powders being introduced.

We all know that when you go through a load manual there are loads that are more accurate than others. It is quite possible the secondary spike is the reason at least some of those loads never tune in well.

Internal barrel ringing may be invisible without a borescope, but the .223 barrel Jim Ristow described had three rings, spaced exactly by the amount and number of previous setbacks and rechamberings done to the barrel. Is it possible the ring is contributing to accuracy problems when it gets large enough, and that is the real reason the barrel had to be set back to start shooting again? I don't know, but it's worth exploring.

At any rate, even when they do no damage, those pressure events aren't helping anything, either. Until I know more about them I'm going to avoid them. One of the problems with all this stuff is that only the military has the funding to do really detailed laboratory research on it. I know I can't afford to run millions of rounds through thousands of guns to look for a statistical effect. I just hope the military sees it and studies it so the rest of us can get a better idea what is going on.

Jim Ristow says the spike doesn't happen in 20" and shorter barrels, but since the bullet often isn't that far along when they occur, and I have no idea what sample size and variety of barrel contours his examination is based on, I take it with a grain of salt. Without a military study, anecdotal evidence figures big in this and other problems. As Jack Belk points out, just because you've flown hundreds of times and never been in a plane crash doesn't mean you shouldn't believe they can happen. But that's how a lot of folks interpolate their personal experience anyway.

Time passes; measuring improves; we learn more. At least, we should expect to.