by Tom Gaylord, a.k.a. B.B. Pelletier
Part 1
Part 2
Part 3
Part 4
Part 5
Part 6
Part 7
Part 8
Part 9
Part 10
Part 11
Part 12
This is the summary report in this series. I’ll give you my thoughts on how this test went, and I expect you to comment, as well.
The barrels
Three barrels were used in this test. One was the factory barrel that comes with the .22-caliber AirForce Talon SS. It’s a 12-inch Lothar Walther barrel that has a choke of about a half-thousandth inch reduction in the bore diameter for the final 2 inches of length. That makes all the pellets of uniform size as they leave the muzzle, and it may potentially stop any in-bore wobble. This barrel has the standard airgun twist rate of 1-turn-in-16-inches of bore travel, written as 1:16″.
The other 2 barrels were made by Dennis Quackenbush. Neither barrel is choked. One is a 1:12″ twist; the other is a 1:22″ twist. They’re also about 12-inches long and are held in the gun by AirForce Talon SS barrel bushings. Several comments have suggested that because these barrels are different than the Lothar Walther barrel, this test is somehow not fair. But the results of all the shooting prove otherwise. Sure, there are variations from barrel to barrel, depending on the power used and which pellet was shot. But the results are so close between all 3 barrels that whatever differences there might be are overridden by the similarities. In other words, I’m suggesting that if Lothar Walther had made all 3 barrels, there would be similar differences.
The 3 barrels used in the test. Factory barrel in the middle.
I believe the twist rates are what drive the results. We weren’t searching for the most accurate barrel in this test. We were looking for behavior changes as conditions were changed. And we got that.
Velocity
The first thing that was tested was velocity. Both pellets — the 14.3-grain Crosman Premiers and the 15.9-grain JSB Exact Jumbo were shot from all 3 barrels at each of 3 predetermined power settings. These settings were marked on the gun so they were kept constant throughout the test.
The power settings were the power indicator screw all the way to the left (the lowest setting), and the power screw centered on each mark (settings 6 and 10).
I reviewed the velocity for you in Part 8. Here’s a summary of that report.
In all cases, the velocity increased the most between power setting zero and setting 6. The velocity increase from setting 6 to setting 10 was always smaller than the increase from setting zero to setting 6, and that’s irrespective of the twist rate or which pellet was shot.
What you’re seeing here is the slowing down of the rate of velocity increase as the air flow increases. That’ll become clear in a moment when I discuss the rifle’s maximum velocity potential.
As the twist rate slowed (1:22″ is slower than 1:12″), the velocity increased at most power settings with most pellets. There was one instance with the 1:22″ barrel when the JSB Exact pellet actually went 2 f.p.s. slower at setting 10 than at setting 6; but with all other barrels and pellets, there was always a velocity increase as the power setting went higher.
Focusing on the 1:22″ barrel for a moment, we see that the velocity increases between setting 6 and setting 10 were not as great as they were in either the factory (1:16″) barrel or the 1:12″ barrel. This suggests what we have suspected all along — that the twist rate of the barrel does slow down the pellet as it gets tighter. And we can see from this test that the phenomenon is most apparent at the lower power settings. At the higher power settings, the differences seem to shrink, indicating that the influence of the power setting is overriding the influence of the twist rate. I believe this is an important finding, and it sets up the next observation, which is that the top velocity of the gun was fairly close for all 3 barrels, regardless of the twist rate. The type of pellet made more difference to the top velocity than the barrel twist rate did.
It should be obvious from these results that the Talon SS powerplant has upper limits that cannot be exceeded by forcing more compressed air through the barrel. This illustrates the relationship between barrel length and velocity in a pneumatic airgun.
A second thing I found interesting is that power setting 6 is very close in performance to power setting 10. In the case of the 1:22″ twist barrel, it’s remarkably close…but it’s close for all three barrels. A prudent airgunner might consider this when setting the power wheel adjustment on his Talon SS, knowing that a lower setting uses less air, yet gives velocity that isn’t that much slower.
A third thing is that the velocity performance of the 1:22″ barrel is so good at power setting 6 that it makes power setting 10 useless. Take that thought just a little farther, and you’ll see that all power settings above setting 10 are pretty much a waste of air in a Talon SS with a 12-inch barrel, regardless of which pellet you use.
Accuracy testing
Next, I tested all 3 barrels with both pellets shot at all 3 power levels at 10 meters (11 yards) and 25 yards. Following that, I tested all 3 barrels and both pellets, again, at 50 yards, only I didn’t use the zero power setting. This was where my eyes were opened regarding the effects of twist rate.
I analyzed the accuracy in 2 different reports. One (Part 9) was the 10-meter and 25-yard accuracy and the other (Part 12) was the 50-yard accuracy, alone. Now, with the table above we can combine these results and analyze all the accuracy data together.
The first observation I’ll make is that at 10 meters, I got 10-shot groups that ranged from as small as 0.092 inches to as large as 0.578 inches. The factory barrel gave the best results with the JSB pellet; but with the Premier, it was the 1:22-inch twist that did the best. Curiously, that pellet and twist rate didn’t seem to change that much as the power was increased (at 10 meters). With all other barrels and pellets, the group size did change a lot as the power changed.
It’s too simple to say the factory barrel with the 1:16-inch twist rate is the best; but of the 3 twist rates in this test, it certainly is the most flexible across the board. However, you’ll notice that the 1:12-inch twist barrel did shoot the best single group (with JSB Exact pellets) at 50 yards. That group is so close to the Crosman Premier group shot by the 1:16-inch barrel that I can’t call a clear winner — BUT — here is what I CAN say. The Quackenbush 1:12-inch twist barrel is certainly capable of shooting 50-yard groups at least as tight as those shot by the Lother Walther barrel; and in my mind, that puts the barrel-equivalency question to rest.
Another observation is that the 1:22-inch twist barrel was just as good at 10 meters as the other 2 barrels, in general, but look at how the groups opened at 50 yards! That says something very strong about the relationship of the twist rate to accuracy. And it also brings up a second observation.
Premier pellets and JSB pellets performed differently throughout this test. Just look at the 50-yard results for Premiers and JSBs with the 1:12-inch twist barrel, and you’ll see what I mean. This is one more bit of evidence that barrels have preferences for certain pellets.
Final observation
This will be my final remark in this series of reports, and it does not come from the data collected in this test but from the 5-part test of the Diana model 25 smoothbore. In that test, we saw that the smoothbore was able to place 10 JSB Exact RS pellets into a group measuring 0.337 inches at 10 meters. However, at 25 yards, the same pellet loaded the same way made a group that measured 3.168 inches. That difference tells us clearly that spin and not aerodynamic drag is the main key to pellet accuracy. I think we now see that twist rates do matter a lot, and the standard rate is the best all-around rate for now.
At 10 meters, 10 JSB Exact RS pellets made this 0.337-inch group.
At 25 yards, the same JWB-Exact-RS pellets, seated to the same depth, made this 3.168-inch group. They are clearly not accurate after 10 meters.
Like the charts. Easy to see why the 1:16 is so versatile.
Sounds to me like things from this test could be taken into consideration if a person was trying to find the correct diameter and weight when trying to swage a bullet for a air gun.
How much pressure is behind the bullet, how well the bullet seals to the rifling (fit and twist rate), and weight. Sounds to me like it is getting things balanced accordingly.
And I believe that the factory barrel having choke helped it some. Maybe that would of helped the other 2 barrels some also ?
This test seems to verify that the factories had the correct choice all along in terms of general usability with their 1:16 barrels. GF1 made a good point abbot swaging bullets though. If someone was going to decide to make our use their own “different” than standard pellet, and they knew the velocity range they wanted, then this data is helpful in choosing barrel specs for that purpose. I don’t see many people wanting our able to do that though.
Now, can you shoot long heavy pellets at long range out of the fast twist barrel, please? I’m really curious to see if they’ll stabilize better like they would in a regular firearm.
/Dave
/Dave,
I presume you meant to say “a regular airgun”?
B.B.
BB,
No, actually I meant firearm in that context, but using your airgun. I didn’t explain myself very well, as usual….. I think longer, heavier projectiles will follow the same pattern in your airgun as they would in something like an AR15 with its faster twist barrels used for heavier, longer bullets. So, I guess at the same time, if like to see long range groups from both the fast and the slow twist barrels as kind of an addendum or sidebar to what you’ve completed already. We pretty much already have the data for this, but I’d like to see the super heavy weight pellets added. The test would go as follows…
Your choice of 2 types of pellets- one short and light, one long and heavy (like a Eun Jin), plus a .22 round ball. The two types of pellets will be shot on their “best power/accuracy” setting from both barrels, and then out of curiosity, the round ball at least shot from the slow twist barrel (if not both also). Just group size being reported, no velocity needed.
Thanks,
/Dave
/Dave,
Okay. I don’t think the round ball will be accurate at 50 yards, but the Eun Jin just might be. I do want to know the velocity, though, because that will be important. Since velocity determines rotational speed with a given twist rate.
B.B.
This topic is not exhausted but I’m thankful for these experiments.
How does one measure the rotational speed of the pellet?
Chris,
Easier than you might think. Take the muzzle velocity and factor in the twist rate per foot. That gives the revolutions per second, and it doesn’t slow down very much as the pellet travels downrange.
For example, if the twist rate is 1:12″ and the velocity is 1,000 f.p.s. at the muzzle, the pellet leaves the bore turning at 1,000 RPS, which is 60,000 RPM.
B.B.
I figured it was simple enough. But isn’t it a large assumption that the pellet “doesn’t slow down very much as the pellet travels downrange?”
Chris,
Not at all. It has been proven by ballisticians since the 19th century. The rotation speed simply does not taper off as fast as the directional speed.
B.B.
BB,
well done on what I consider a very involved and probably exhaustive but very interesting test. Like everything else in this world, twist and power settings in the airgun world (and probably in the powder world) are a compromise based on what you want w/r/t speed, distance and accuracy.
Fred DPRoNJ
BB,
To my mind, this is an iconoclastic study. There is little doubt from the results that pellets appear to be spin stabilized to an extent that I would have never expected. I hope that pellet designers (or those who take that task on themselves) will look long and hard at these results, perhaps do their own experiments to confirm/extend, and incorporate the findings into better pellet designs. Given these results, I don’t think there should be much thought given to drag stabilization for long range designs; in fact, I suspect drag should be eliminated as much as possible for best accuracy at longer ranges, just as in firearms.
Can you compare the pellets (i.e. length and diameter) and twist rates used in your test to the Greenhill formula? It would be interesting to see how well it predicts these results.
Even though it confounds conventional thinking, this test will likely result in better pellet designs, more informed pellet decisions, and less mysterious pellet-pickiness suffered by knowledgeable users!
BG_Farmer,
Thank you for that praise. As for the comparison using the Greenhill formula, I will leave that to someone else who may be able to explain it better. The combination of drag stabilizing and spin stabilizing is something that merits a study of its own. I am not the one to do it. I am still puzzled why .177 through .25 caliber pellets are all stabilized by the one twist rate and at various velocities.
B.B.
BB,
You’ve worked hard on this, so the praise is deserved!
RE: the Greenhill formula, I just meant measure the pellets and plug them into the formula and see what twist rate it predicts. I can do the math (if I go slow), but I don’t have those pellets (or any pellets) in .22, nor do I know where to find the dimensions online. Obviously, the diameter is ~5.5mm, but the length is just as important in the formula (which is only an approximation/simplification itself — air rifles might require a different setup in some places).
I agree there is still a lot to figure out. .22 cal. may be the most suited to the “standard” twist rate, since it was likely taken from a .22 cal. rimfire barrel. .177 might actually be slightly different, not to mention all the others. What is important is that your results indicate that spin is doing a lot more than the “authorities” led us to believe! Whereas before I thought drag did all the work and spin was just to even out inconsistencies in a pellet, it appears that the opposite is quite likely the case.
BG_Farmer,
What I meant was these pellets are both hollow and the Greenhill formula is for bullets. Don’t you think the drag and lack of weight will upset the results?
B.B.
BB,
I suspect drag and low weight (relative to bullet) has a significant effect, just not sure what for either, or if they cancel each other out to some degree :)! The commonly used Greenhill formula has a term built into it that cancels out for solid lead and a more complex variant can be used for less dense (or maybe hollow lead) projectiles. I also just don’t know what drag does to the formula in this case — it definitely slows the pellet down quickly. Here also, there is a constant related to the muzzle velocity that is typically only changed above or below 2000fps, but I suspect could be jiggered both for lower pellet velocities and the effect of increased drag.
I would run the formula first in its standard form:
twist_rate = (150*projectile_diameter^2)/projectile_length
just to see what it predicts is best for each of these two pellets.
Then the longer form:
twist_rate = ((C*projectile_diameter^2)/projectile_length) * sqrt(SG/10.9)
where C could be played with (it is typically 150 for v<2800fps and 180 for faster projectiles),
and SG is 10.9 for solid lead and could be changed to more accurately reflect the density of an actual (hollowish) pellet.
Oh dear… Should we add sectional density and ballistic coefficient to the parameters of the experiment? Say similar pellet designs with two or three SDs, or BCs… <G>
The 2 calibers that stands out to me all the time is the .20 cal. and .22 cal. pellets. Not by preference to shoot but by design.
They are both the closest together in diameter out of all 4 calibers. But the weight of the pellets differ along with how long they are if you compare similar brands and designs in the 2 different calibers.
And if I remember right the fps ends up being pretty close most of the time. But sometimes the .20 cal. is slower than the .22 cal. My first thought was that the .20 cal. would shoot a faster fps. because it was a smaller diameter.
That makes me think maybe the .20 cal. would improve with a different twist rate barrel.
GF1,
Actually, .20 caliber Premiers are longer than .22 caliber Premiers of the same weight and are more difficult to stabilize. They also go slower in PCPs because less air gets behind them.
B.N.
Along with possibly more friction from that longer body…
Is that maybe why there isn’t that much of a pellet choice for the .20 cal.
So maybe there needs to be a design change for the .20 cal. pellets or barrels for them to shoot better.
I have always wondered why they don’t seem to be as popular. I have never owned anything in that size so I really don’t know the characteristics of that pellet.
But I have thought about getting one of the 12″ barrels for my Talon SS and giving them a try.
And I think that would give more availability of something to start with if I wanted to swage a bullet for that cal.
Something like using the .22 cal. round balls. Do you think that could be a good starting point for a blank to swage to the .20 cal. size?
B.B.
Congratulations, you’ve made a tremendous job! The data collected and your findings are worthy of a book! I wish I could be able to make a reserch on such a scale one day.
duskwight
duskwight,
Thanks! I wish I could design and build a recoilless air rifle!
\B.B.
B.B.
I am a happy man.
I am very glad that the standard twist rate proves to be the most reliably accurate at various distances. As picky as I am about accuracy if one of the “custom” twist rates were more reliable I may have been very tempted to get new barrels for all of my guns. An expense and task I would not have relished.
I wonder if your results had conclusively shown that one of the other twist rates were the most accurate if the various manufacturers would have been compelled to do their own tests and perhaps change the barrels of their guns to the more accurate twist rate?
G & G
G&G,
You have to understand, the airgun pond is small, and I am a very small fish within it. If there had been a world-beating discovery here we might have seen some change in several years, but it would have been small at first. It would take decades before any large changes would have come about.
B.B.
B.B.
I really appreciate your humility; however, even though you may be a small fish I believe you are a very visible one. How many other blog authors (in addition to your other accomplishments) have managed to get their own gun design into production?
In any event, I would not have expected any overnight changes. I am sure it would have taken years for large scale change to occur. On the other hand, I believe that the first to get a new product on line with increased accuracy from a new barrel design would have greatly benefited from it.
Again, your work continues to impress.I look forward to many more years of it.
G & G
Pardon?
I think that’s supposed to 0.092 inches
Well, with the final set of numbers available I’ve updated the R-system data file (I had to put in dummy “power 0” entries to keep the color/marker sequencing working correctly — that explains the “group size 0” at power 0/velocity 0 intersection).
The Updated (and Final?) Pairs Plot is in the same place:
http://wlfraed.home.netcom.com/TwistTest.png
I’m curious about the best rifling twist for .20 and .25 pellet rifles, However with long pellets in .177 and .22 wouldn’t a slower rifling work better ie barracuda or kodiak pellets. Also I notice that having more lands ie Lothar Walther barrel works excellently for pistols as my 2300Custom outshoots my 1377B which has 8 lands and is choked towards the end.
Patrick,
Yes, I think a .22-caliber round ball would be a good starting blank for a .20 cal swaged pellet.
B.B.
BB
I was asking above about the .22 cal. round ball to swage a .20 cal pellet. And trying a .20 cal barrel in my Talon SS.
I was going to make the tool this week end and give it a try. Already got some .22 cal. round balls. Just got to get the barrel at some point in time.
The Greenhill formula is based on a specific bullet shape and density, and an artillery projectile was the one used. Pistol bullets tend to be very short for their weight since the diameter is large compared to many rifles (though not all). As the bullet gets farther away from the original one used to establish the formula, adjustments must be made. The physical length of the barrel has no affect on twist rate. A one turn in 12 inch barrel will have tow complete revolutions in a 24 inch rifle barrel, but only half a revolution in a 6 inch pistol barrel. The twist rate is still the same.
The Greenhill formula is based on a specific bullet shape and density, and an artillery projectile was the one used. Pistol bullets tend to be very short for their weight since the diameter is large compared to many rifles (though not all). As the bullet gets farther away from the original one used to establish the formula, adjustments must be made. The physical length of the barrel has no affect on twist rate. A one turn in 12 inch barrel will have tow complete revolutions in a 24 inch rifle barrel, but only half a revolution in a 6 inch pistol barrel. The twist rate is still the same.
Very very interesting results. This has me curious about the effect that the twist rates would have on my Crosman AirGuns (they group realy tight already). I am glad that you did this article, as I have not been able to find any info on how much of a difference different rifling makes for pellet rifles, other than the normal smooth bore has poor accuracy.
By the way as loose of groups as your getting with the factory barrel I would complain to AirForce about that, with a factory barrel even on a cheap MSP you should be doing better than 0.75 inch groups at 50 yards with any pellet that the gun likes, especialy with a .22 (I could see failing to compensate for the wind with a light .177 pellet).
I think it is time that I made a few barrels. So I guess that I am going to make 3 with each of the following riflings (one each for each gun used to test) :
1:14, 1:15, 1:16, 1:17, 1:18, 1:19, 1:20, 1:21
As well as dual riffling with a main lead of 1:18 and a 3 inch exit going to each of the above, and the same with a lead of 1:20, and 1:16. For a total of 24 barrels per gun to make.
And then I will test each and every one for accuracy at every range from 10 yards to 150 yards, as well as get a chrony and test the MV of each, and do all testing with every pellet that these guns seem to like. And in a few years when I finish all of the testing post my findings here, and stick with the best barrel on each gun out of the bunch.
In the world of firearms velocity vs. twist rate has a large impact. Given that pellets have significantly more drag that typical Spitzer bullet, I would expect that the velocity change over distance is dramatic. Are the group sizes reflecting the spin RPM of the pellet at each distance? Has anyone done the calculations based on the B.C of these two pellets and constructed a spreadsheet or table the shows the velocity and spin RPM at 10m, 25 yds and 50 yds?
CindiB,
Welcome to the blog.
Nothing has been done that you ask. All my testing was empirical.
B.B.