Replicating Federal GMM 308 ammo

[Rimfire5]

For what it is worth, I started on this quest 2 years ago when I started hand loading for my .308.

The Federal GMM (maroon and gold box) with SMK 168 grain bullets was rated at a muzzle velocity of 2650 fps.
The older Federal GMM (buff and blue box) with SMK 168 grain bullets was rated at a muzzle velocity of 2600 fps.

I tried Reloader 15 and got better results using 43.0 grains of RL-15 at 2.805 OAL and 42.9 grains at 2.805 OAL.

I was happy but not satisfied so I tried H4895.

I got even better results using 41.8 grains of H4895 with multiple depths from 2.800 to 2.815.

I changed bullets to Nosler Custom Competition 168s and got even better results with 41.5 grains of H4895 at 2.830 OAL. I also tried Berger 168s and got good results with H4895 with 41.5 grains and 2.820 OAL.

Then, using SMKs and Nosler CCs I tried Vhita Vuori N140 and got even better results using 41.9 grains with SMKs at 2.805 OAL and with Nosler CCs at 2825 OAL.

After I got slightly better accuracy with my Reloader hand loads than with factory ammo, I have given up trying to replicate factory ammo and have concenterated on getting the absolute best accuracy using whatever powder makes my .308 Savage Law Enforcement model 10 FP shoot as well as it can.
So far, N140 and Nosler CC bullets outshoot anything else I have tried in my rifle.
The SMKs shoot great, Hornady Match 168s shoot about the same, but the Noslers shoot just a bit better.

[Unclenick]

Originally Posted by jimro

"Free Space Loss" of a traveling wave comes into play (the authors epiphany of guy wires transmitting a wave is stupid, there is no fixed point at the muzzle to reflect energy back towards the action) . . .

Jimro,

The place the guy wire epiphany fails by analogy is not the anchoring of its opposite ends, but rather that the observable reflection is a transverse wave, like those in Varmint Al's FEA work, rather than a compression wave as the OBT concept describes. But still, I can see how the visual would encourage someone to contemplate all modes of wave travel through steel.

I assume you get this, but for anyone not following I’ll add that you don't need an anchored end to get compressive wave reflection. All that is required for a pressure wave to make a return trip is for the stress packed into the dead end of the travel to relieve itself by expanding back in the direction from whence it came. Think of the muzzle steel compressing locally as that stress arrives, then, because the steel has good elasticity, it snaps back to shape, sending the compression zone back the other way.

The elasticity of steel isn't perfect, and you do lose a little of the energy as sound and heat with each cycle, but steel is pretty efficient in this regard. Consider that a steel tuning fork rings audibly for many thousands of cycles before gradually fading to silence. Obviously that efficiency is affected by the hardness of the steel, which is spring hard in a tuning fork, and barrel steel can't be that brittle. So a barrel will have more loss per cycle. But even so, if you're talking hundreds of cycles in barrel steel rather than thousands as in the tuning fork, you're still sustaining barrel distortion that is little changed during the few cycles it takes a bullet to clear the muzzle.


Bart,

Dan Newberry can probably defend himself, but I don't see how his OCW system is in any way dependent on the OBT theory. Newberry and Long know each other, but I think that's a case of the theory (OBT) being embraced after the OCW system concept originated because it seemed to predict observed results, and not the other way around. Correlation is not causation, however, so I expect Mr. Newberry would entertain alternative theoretical explanations for the observations, too.

If you read Newberry's pages, you find he began by observing the GMM .308 168 grain SMK load worked very well in a large variety of rifles with various barrel lengths. Maybe it wasn't the very best load in all those rifles, but was consistently pretty darn good. He calls it a "Chocolate Ice Cream load"; maybe not everybody's favorite flavor, but liked at least pretty well by almost everybody. He reasoned that there would likely be other combinations of components that would be similarly pretty darn good for a large range of rifles and set about figuring out how to identify these. He settled on the best indicator of a candidate OCW load being a generous span of powder charge maintaining the same point of impact at moderate ranges (up to 300 yards). That candidate then needs to be verified by a number of people testing it in different guns before being accepted as an OCW load.

That's what the OCW system does that I can see. No theory is required to make it work; just empirical testing.

Even if you don’t buy into Chris Long’s idea about the pressure wave and stick strictly to Varmint Al’s transverse barrel deflections, you still want to hit an optimum bullet exit time, as otherwise Audette ladders would never show a predictive result, even with thin, whippy barrels that exaggerate them. Nor would barrel tuners work to synchronize muzzle deflection with bullet exit. I would distinguish between a calculated optimum barrel time (theory dependent) and the empirically determined barrel time of any apparent sweet spot load, OCW or otherwise. I would expect any such established best barrel time to work with different component combinations. Muzzle velocity alone does not determine barrel, which is why your observation that copying a velocity does not necessarily get you to a sweet spot is right.


Dan Newberry,

Welcome to the forum.

There are two reasons I recommended the Remington and Lapua cases to the OP rather than Winchester. One goes to Bart's point about velocity not being a reliable stand-alone performance predictor, and I had recommended a chronograph as a load finder in this particular instance. The Remington and Lapua cases are closer in capacity to that of the Federal cases than Winchester cases are. For that reason the same powder charge will tend to produce similar velocities and barrel times in his particular rifle. If he were to use the Winchester case, as you say, it would take more powder to get to the same velocity. Because that larger charge weight makes more total gas, it will achieve a matching velocity with a lower ratio of peak pressure to average pressure. That means a smaller portion of the total acceleration happens in early bore travel, so barrel time is longer. That could move him off a sweet spot’s barrel time and force him to go for a slightly higher muzzle velocity to get back to an optimum charge weight.

The other reason is that I've noted a number of complaints about Winchester brass recently, particularly on the CMP forum; primer pockets getting loose in just a few reloads and unusual numbers of split necks, in particular. All the kinds of complaints you more often see about Federal brass. I have to wonder if the move of Winchester’s center-fire ammunition manufacturing from Alton, IL to Oxford, MS, announced in 2010, is having an adverse impact. You lose skilled labor when you do something like that, and I don’t doubt there’s a learning curve for new equipment operators.


For anyone interested in temperature sensitivity:

Board member Denton Bramwell published some experiments in which he found the correlation between pressure and barrel temperature to be much more significant than powder temperature. The temperature stability of the powder burn rate alone would be more important for first shot consistency in a cold barrel, but once the barrel warms up, that apparently ceases to be the dominant term. The only theoretical hypothesis that comes to mind is that the coefficient of friction is increasing between the bullet and the bore as the barrel temperature goes up. It’s normal for friction coefficients between pairs of solid materials to increase with temperature, though I don’t have numbers for steel and gilding metal specifically.

So, to minimize sensitivity to the barrel being hot, I think what’s needed most is a powder whose burn rate changes least as pressure increases. Burn rate charts just show how different powder burn rates compare under a standard set of test conditions. As soon as you start changing chamber pressures, their burn rates change and which is faster or slower can change places.

In the Precision Shooting Reloading Guide, Dave Milosovich describes loading a .308 to a series of fixed velocities using IMR 4895 and IMR 4064 under a 180 grain bullet. At 2200 fps it took a heavier charge of 4895 than 4064. At about 2400 fps they were about the same charge weight. At 2500 fps and higher a heavier charge of 4064 was. As pressure and velocity increased, 4064 had its burn rate increase less. Because it tended to burn more consistently under changing pressure, it should be less affected not only by temperature, but also by barrel temperature, charge dispensing errors, or any other factor that changes pressure. The results amount to 4895 producing 67 fps/grain and 4064 producing just 47 fps/grain.



Unfortunately, there does not seem to be universal corroboration of the above. Hornady’s data (for their 178 and 180 grain bullets) shows no significant difference between the two powders in fps/grain over their span of listed velocities. Hodgdon gives IMR 4064 under the 175 grain SMK an advantage, but only about half as much as Milosovich reported. But then, we know nothing about the barrel temperatures they allowed during testing or how much was actually tested and how much was calculated. About the only thing all three sources seem to agree on is that Varget has the lowest fps/grain in .308 loads with bullets in that weight range, ranging from 36 fps/grain (Hodgdon) to 45 fps/grain (Hornady). IMR 4064 and Varget both have earned outstanding reputations as match load powders.

The disagreement suggests you need to run a test for yourself under your typical shooting conditions at your or maybe even under your range of shooting conditions. You don’t need to go through all of Milosovich’s steps. Just average a good load’s velocity (not too near maximum pressure) and knock the load down maybe 5% then measure enough rounds at that level to get a good average. Divide the difference in the average velocities by the number of grains the charge difference was, and there’s your comparative number. Unless your upper load is causing a problem like forcing uneven bolt lug contact to iron out or is starting to exceed maximums and is stretching steel, that should be all you need to predict velocities from other charge weights.

[Powderman]

A lot of guys I know are issued Fed GMM for their duty ammunition, and want to duplicate its ballistics for training beyond what they are authorized by their department.

Exactly.

My department issues GMM2--the .308 with the 175 gr. MatchKing. My practice load is 41.5 of RL15 under the 175 grain MatchKing; cases trimmed to 2.000; COAL @ 2.820, with CCI #34 primers. Excellent practice load.

[Jimro]

Unclenick,

I understand about compression waves, but I think that the softness of barrel steel isn't going to retain energy very well beyond a few cycles. I wish I had enough engineering skill to calculate the energy retention of steel through a vibration cycle to prove/disprove that thought.

To get back to the tuning fork analogy, the frequency doesn't change with time but the amplitude does. No matter how you ring it the "natural frequency" will come out, but the volume (amplitude) will depend on the initial input of energy. But you can see how the quickly amplitude can decrease.




Jimro

[Unclenick]

Jimro,

You are right in principle but off in magnitude. When I referred to cycles I did not mean frequency or pitch (a vibration rate), but rather to the total number of cycles of vibration without regard to how long that took. That's what I was interested in, and the Wiki Commons plots you borrowed show how you work this out.

In electronics engineering we usually use Q in resonant circuits to predict this. Q is the amount of energy at the start of the cycle divided by the portion of that energy that is lost by end of a cycle. Your equations use a different measure of the loss factor called zeta (ζ). It is the ratio of the loss that you actually have per cycle to the loss per cycle that produces critical damping, the minimum level of loss that prevents ringing.

What I did was replicate the Wiki Commons plot in Excel so I could change zeta to different values. The first plot is the lower ζ=0.3 plot which is the lower plot in Wiki Commons illustration, but with the time scale widened. The next is ζ=0.1, same as the upper Wiki Commons plot. I then went on to give you ζ=0.03 and ζ=0.01 so you can see that as the loss factor gets smaller the ringing cycles take longer and longer to decay.

Take a listen to the tuning forks in this You Tube video. You can hear they sustain for a number of seconds before they lose much volume. From that example, if I suppose a tuning fork tuned to A above middle C (440 cps) takes 5 seconds to lose half its magnitude, then it has a zeta value of about ζ=0.00005. My last two plots show what this looks like on both the same scale as the first plots and then on a much wider scale of cycles.

As we both mentioned previously, the barrel steel is softer and won't ring as long, but keep in mind that a brass bell is softer than barrel steel and it still rings pretty well. Maybe only a fifth as long as the tuning forks. My expectation is the barrel steel won't be any worse than that, though I'd have to check on it. The bottom line is I expect loss in magnitude to be only a few percent by the time the bullet exits, and not major losses in magnitude.

[Jimro]

Unclenick,

Totally following you. I have enough credits to almost qualify for a minor in "music theory" and while my electronics background is sorely lacking I still have to be able to calculate free space loss between antennas for my current job.

I looked up the modulus of elasticity for steel (in various alloys and hardness) but couldn't find a way to translate that data into a form useable to answer the question, "how quickly will barrel steel lose energy?"

Intuitively the answer is, "longer than it takes to get the bullet out the bore." But my experience with mild steel in a machine shop is that it is a very poor conductor of compression, so I am interested in how much muzzle deformation actually occurs by the the time the bullet exits.

Jimro

[Bart B.]

Unclenick, I agree that the charge weight that shoots the bullet out at the optimum time produces best accuracy at some range. But that accuracy is only "best" in a portion of the range from muzzle to a far distant limit. For example, from muzzle to 600 yards, the smallest MOA value may well be at 400 yards and the MOA value at other ranges a bit bigger. A different charge weight will send the bullet out at a different place in the muzzle's vertical swing. And that'll move the most accurate range to another one and make the worse accuracy number get bigger. This is what happens with muzzle velocity spread vs muzzle exit angle compensate for bullet drop. There's one area on the muzzle axis vertical spread that's best for accuracy up to some range. The Brits knew this well with their arsenal .303 ammo shot in their SMLE's; it shot more accurate at 600 yards and beyond than the Mauser 98 action rifles did with the same ammo. But the Mausers (with the same barrel dimensions) shot more accurate at ranges shorter than 600 yards compared to the SMLE's. Each rifle had its own barrel whip issues that differed because of how recoil forces moved each barreled action around in the stocks before the bullet left the muzzle.

Regarding the "shock wave" the OCW believers have traveling in a barrel, where does it start at? I think that if it's any place other that at the case head, there'll be two of them when it starts; one going backwards toward the bolt face and the other forward towards the muzzle. That doubles the number of shock waves in a barrel.

Jimro, videos I've seen of rifle barrels whipping like a fishing rod does showed the vibrations damped out after about 6 or 7 cycles. The major ones have the most effect on muzzle axis angle have resonant and harmonic frequencies up to about 400 Hz. After a 2 to 4 millisecond barrel time (depending on centerfire rifle muzzle velocities), the bullet's left the barrel and nothing the barrel can do after that will effect its path. So there's only about one plus major whip cycle while the bullet's going through the barrel.

[Jimro]

Bart,

I don't disagree with the transverse wave (fishing rod type bending). What I was looking for was the horizontal compression waves through steel, and how quickly that will disperse. The transverse waves will vibrate with the "natural frequency" of the firearm, while the compression wave will move with the speed of the transmission medium (steel).

That "fishing rod" action is a result of the the exact same force that makes a fishing rod whip, the inertia of the rod resisting change at a different rate because of the elasticity of the rod. This is caused by the internal recoil and gravity. The barrels have this transverse wave in the vertical plane, as there is no forces acting in the horizontal plane. In theory if you could get a rifle to recoil directly to the rear, you could minimize this barrel whip (such as comparing the barrel whip on an M16 with that of an AK).

The "horizontal compression" wave is moving through the barrel, from chamber to muzzle. As the wave moves the effect is seen at 90 degrees to the direction of wave travel, which is supposedly measurable as a diameter increase and decrease.

The frequency of sound doesn't change the speed of sound. The frequency of the transverse wave (fishing rod whip) will be somewhere along the realm of a few hundred Hertz (long tube of steel means a relatively low harmonic frequency), but the horizontal compression wave supposedly travels much faster (the wave that causes the increase/decrease of muzzle diameter).

In air, we have "free space loss" so that things further away get quieter, or dimmer. In the barrel the number of cycles of the muzzle increase/decrease is determined by the speed of the wave and the length of the barrel (multiply the number of cycles and the difference in magnitude of the first to final and you can calculate the signal loss over distance, unfortunately numbers that I do not have for a compression wave in steel). If we go for the low estimate of around 12,000 fps for this wave propagation, then there will be 2 or 3 cycles minimum before the bullets leaves the bore, and depending on barrel length possibly more.

Personally I'm not totally sold on the compression wave idea as the main evidence cited was muzzle velocity SD and group dispersion. It makes sense, but do do other explanations.

Jimro

[Unclenick]

And that's sort of where I am. I don't entirely buy the OBT theory, as I've posted in other threads, but the idea is interesting to investigate and seems to predict results fairly accurately in many instances.

The issue with estimating sustained ringing is about the losses. If you whip a barrel on any axis in a stocked rifle that becomes a tail wagging a dog, and there are lots of opportunities to transfer energy from the barrel to the stock. That makes it hard to say a steel property necessarily had a lot to do with dominating how fast the ringing damped out. Even a metal that's not hard (brass bell) can ring for a while, provided its elastic limits aren't exceeded and it isn't transferring energy out of the system to anything other than air. How much energy is lost as heat due to the flexing the metal itself, I just don't have information on. I’ll try to look for some. I’ve got about half a dozen things backed up now that I need to go the university reference library to find. The web doesn’t have everything yet. I could also make some measurements.

A source for a compression wave is pretty easy to come by. If you look at hoop stress equations you find that any time you apply radial stress from the inside to expand a tube, the expansion pulls on the tube longitudinally. I've gone through the hoop stress equations to estimate muzzle strain from that. It didn't appear to directly affect the bore diameter more than a couple of ten thousandths of an inch in most instances. But that's a static analysis and not a dynamic one. It needs to be born in mind that the OBT theory is geared toward finding only the neutral dead spots in the compression wave.

As Bart suggests, hoop stress due to pressure would create two such waves, with one going forward down the tube and the other making the short trip to the breech. From the breech, the rearward traveling wave reflects back forward to chase the other going forward. The OBT calculator does, indeed, create differently spaced null times resulting in pairs of barrel times that aren’t too far apart, followed by a longer period during which no null spot appears.

As an aside, in an email correspondence with Chris Long several years ago, he said he had one rifle that defied his calculator completely. It was in a custom receiver heavy barrel .308 bench gun. He said he machined the threads as class 3 super tight threads, and they were so snug he couldn’t thread the barrel by hand. The whole length needed a barrel vice and wrench. He concluded that the tight thread had tied the receiver strongly enough to the barrel that it became an extension of the barrel, greatly spacing the vibration period out.

Possibly apropos of that last paragraph, Harold Vaughn, in his book, Rifle Accuracy Facts, had some illustrations of normal thread loading, showing 36% of the load is on the first turn of a barrel thread coming off the shoulder. This is because the further away from the barrel shoulder a thread is, the greater the span of material that can stretch. The end result, he showed, is the rearmost threads can actually slip radially, allowing more barrel swing. In effect the barrel is pivoting around the first turn, then. This indicates the tie to the receiver, from the stand point of transmitting a pressure wave, is not normally great.

Vaughn also shows there are a lot of recoil moments in a real gun. Even just the asymmetry of having a gas vent hole drilled on one side of the receiver and not the other produces an asymmetry that causes a moment he could measure. So it's sensitive and lots of opportunities exist for the system to introduce odd deflection angles and opportunities to dump energy.

As to the OCW, I’ll repeat myself: it doesn't depend on the OBT theory being true to work. The OBT theory is just one possible explanation, but the OCW system itself is dependent only on observation. While everything Bart said about sensitivities of group sizes at different ranges is true, I’m not seeing how that affects the OCW concept. The OCW concept is just to develop loads that mimic the accuracy and broad applicability observed to exist in good quality factory match ammo. It’s not trying to be the equal of hand-tuned loads for a particular gun at a particular range. The OCW development system eschews all benchrest techniques, like case prep, and uses fully resized factory brass and factory COL’s, then sees how far you can get with that toward making a load that works well in a range of rifles. It makes ½ moa groups possible at shorter ranges in many instances, but it’s not geared toward bugholes.

I think you can use a target round robin like Newberry’s to help find bughole loads for a particular rifle at a particular range. In that regard it may be used like a latter day Audette ladder, but having higher resolution and every dispersion axis taken into account. But if you’re going to use this round robin method with prepped brass and to check the effect of changing seating depths on a load, then you’ll want to be firing your tests at the range you intend to shoot the load at, and accept that the final load may only be really good in your particular rifle at that particular range. If that’s the case, by definition, it’s not an OCW load, but a gun-specific load.

In this thread the OP asked how to mimic a factory match load, and that’s what makes the OCW concept so applicable to it.

[Jimro]

Unclenick,

We sure do travel down some rabbit holes, don't we?

I'm not complaining, have more math to play with now :-)

Jimro