Barrel length, velocity, and vertical, oh [**]

[DownhillFromHere]

I posted this response in another thread, but the topic(s) seems relevant to a wider audience. Main takeaway for me is that my 50-year-old college statistics knowledge lead me to conclude that lower .22LR match round velocities introduce higher potential vertical dispersion (2nd bullet point). If my support as shown for the hypothesis is in error, I'm open to learning.
-----------
So there's theory, there's observed data, there's statistics...

So, two theories:

• "Longer" barrels "should" give better velocity ES/SD than "shorter" barrels with the same decent-quality ammo.
• Lower velocity doesn't matter, just add elevation and done.

Longer vs. shorter barrel
There's theory that a "longer" barrels should give improved velocity ES/SD values over a "shorter" barrel. We'll leave the definition of "longer" at ~26 inches and "shorter" at ~16 inches; it really doesn't matter for this example. Anyway, my longtime gunsmith/friend had the same opinion. So when he cut my RimX barrel (28" Shilen taper-lapped blank), I specified 26 inches.

Over a sample size of a few hundred rounds of lot-tested Center-X in this new RimX and over four thousand in my Vudoo, the 26" Shilen barrel gave me a 50fps lower average velocity than the 19" ACE barrel (~1035 vs ~1085 at ~72 degrees F.). The standard deviation was very close to identical for both barrels at 8-10fps.

I recounted the saga of getting the RimX to shoot accurately here. The RimX barrel was "bobbed" to 22 inches which gave me back 30+ fps velocity. Again, ES/SD on the lot-tested Center-X remained very close to 30fps ES / 8-10 SD at both 22" and 26" lengths.

Would this repeat over many barrels of differing lengths? I don't know - I'd like to, but no one has posted hard data that I've seen - but my specific example is the one important to me.

In any event, slower velocity for good-quality .22LR ammo does have an impact, as shown next.
-----------

Lower velocity doesn't matter
EDIT 04/19/2023: I think the table below demonstrates how it can matter. My original table had an error - see the strikethrough value in corrected table below; I had entered a wrong elevation. While there is a measurable increase in vertical dispersion, it isn't significant. END EDIT
Dropping velocity 50fps and adjusting elevation accordingly isn't the end of it - the lower velocity also increases potential vertical dispersion. It's been 50 years since college statistics. If my conclusion and supporting data are erroneous, I'm open to correction.

I mentioned Hornady centerfire test data earlier [in the thread from which this is copied]. It was correctly pointed out that centerfire and rimfire are two very different worlds. The commentary from Hornady's Jayden Quinlan I had in mind was this: the longer a bullet takes to get to its target, the more time external forces, including gravity, have to act on it.

So I grabbed a recent LabRadar series of 20 rounds of a new lot of SK Long Range Match (I haven't kept any large series of my Center-X, which was tested/purchased three years ago and of which one case remains). I replaced the static values from the CSV files with fomulae and started playing. I think I've shown that lower velocity, combined with standard deviation and distance, introduces wider potential vertical dispersion, and this potential dispersion increases as distance and SD increase.

So, I held distance and SD as constants and looked at vertical dispersion only:
- Shots: 20
- Distance: 300 yards
- Velocity: 1130 | 1080 fps (actual velocity | actual velocity-50fps; please also see note after the table)
- ES/SD of 30 and 8.7 respectively for both velocities (calculated, not assumed)
- Elevation values from StrelokPro for my Vudoo using Lapua drag curve
[Correction: the red text labels below should read "Potential vertical dispersion increase at lower velocity (mils)"

So - same bullet, same everything except velocity difference of 50fps shows a potential 0.5 mil increase in dispersion at 300 yards (1.2 mils vs. 0.7 mils) at 95% confidence level (two standard deviations). That's about five inches - and the potential dispersion increases as velocity drops (going back to Quinlan's comment about gravity over time). Comparing 1080 to 1030 fps - which is close to what I actually observed but did not record with my Center-X in Vudoo vs RimX - shows a .9 mil potential difference(2.1 mils vs. 1.2 mils), or about 9 inches more potential dispersion at a 95% confidence level.

Is that enough to matter to you? [Shrug] Taking the confidence to 99.7%, or three standard deviations, ramps the potential vertical spread even higher.

I think this shows how lower velocity for a .22LR, within the realm of "normal" .22LR match ammo velocities, does indeed translate to measurable increased POTENTIAL vertical dispersion. If my logic is wrong, show me how. The bullets don't lie, but statistics might, or at least confuse (Example: A 95% confidence level may be stated as (mean +/- (SD*2)) - but in my example, with hard data, that translates to (1130 +/- (2*8.7)) = (1130 +/- 17.4) = (1147 highest, 1113 lowest, difference of 34 - which exceeds the ES).

 

[justin amateur]

I followed y'er thinking and y'er math, then let it ferment a few hours.

I think y'er sample size is skewing y'er conclusion.
I think the results are being affected by the number of shots fired from each barrel.
It's rimfire, and I can't obtain similar chrony numbers
from consecutive boxes from the same brick.
The ES and SD varies from the same rifle under the same conditions on the same day.

I think the sample size is insufficient to provide a reliable conclusion.
Pointing the finger at barrel length or velocity range,
when the actual difference in numbers is caused by
variations in components, chemistry and cartridge assembly.

The same problem occurs during factory lot grading.
That's why we get CenterX that produces better results than Midas+ or X-Act.

 

[obx22]

Y’all have lost me on theorizing that lower velocity equals higher dispersion.
Gravity is a constant.
That leaves time of flight as the variable, and any resistant forces from the atmospherics.
By that rationale a series of rounds 50fps faster or slower should still fall within the same dispersion if in a tunnel or other controlled conditions.
Outside, a longer time of flight allows more chances of variables of outside forces acting on the slug as it makes its way to the target. So I guess the answer is, depends?

 

[obx22]

Y’all have lost me on theorizing that lower velocity equals higher dispersion.
Gravity is a constant.
That leaves time of flight as the variable, and any resistant forces from the atmospherics.
By that rationale a series of rounds 50fps faster or slower should still fall within the same dispersion if in a tunnel or other controlled conditions.
Outside, a longer time of flight allows more chances of variables of outside forces acting on the slug as it makes its way to the target. So I guess the answer is, depends?

As for barrel length, and it’s mitigation of velocity variables, I think the true test would be to have a long barrel with sensors located all along the tube which detect when the slug passes. That should show any areas where velocity inconsistencies narrow or expand?

 

[acudaowner]

 

[littlesister]

I posted this response in another thread, but the topic(s) seems relevant to a wider audience. Main takeaway for me is that my 50-year-old college statistics knowledge lead me to conclude that lower .22LR match round velocities introduce higher potential vertical dispersion (2nd bullet point). If my support as shown for the hypothesis is in error, I'm open to learning.
-----------
So there's theory, there's observed data, there's statistics...

So, two theories:

• "Longer" barrels "should" give better velocity ES/SD than "shorter" barrels with the same decent-quality ammo.
• Lower velocity doesn't matter, just add elevation and done.

Longer vs. shorter barrel
There's theory that a "longer" barrels should give improved velocity ES/SD values over a "shorter" barrel. We'll leave the definition of "longer" at ~26 inches and "shorter" at ~16 inches; it really doesn't matter for this example. Anyway, my longtime gunsmith/friend had the same opinion. So when he cut my RimX barrel (28" Shilen taper-lapped blank), I specified 26 inches.

Over a sample size of a few hundred rounds of lot-tested Center-X in this new RimX and over four thousand in my Vudoo, the 26" Shilen barrel gave me a 50fps lower average velocity than the 19" ACE barrel (~1035 vs ~1085 at ~72 degrees F.). The standard deviation was very close to identical for both barrels at 8-10fps.

I recounted the saga of getting the RimX to shoot accurately here. The RimX barrel was "bobbed" to 22 inches which gave me back 30+ fps velocity. Again, ES/SD on the lot-tested Center-X remained very close to 30fps ES / 8-10 SD at both 22" and 26" lengths.

Would this repeat over many barrels of differing lengths? I don't know - I'd like to, but no one has posted hard data that I've seen - but my specific example is the one important to me.

In any event, slower velocity for good-quality .22LR ammo does have an impact, as shown next.
-----------

Lower velocity doesn't matter
I think the table below demonstrates how it can matter. Dropping velocity 50fps and adjusting elevation accordingly isn't the end of it - the lower velocity also increases potential vertical dispersion. It's been 50 years since college statistics. If my conclusion and supporting data are erroneous, I'm open to correction.

I mentioned Hornady centerfire test data earlier [in the thread from which this is copied]. It was correctly pointed out that centerfire and rimfire are two very different worlds. The commentary from Hornady's Jayden Quinlan I had in mind was this: the longer a bullet takes to get to its target, the more time external forces, including gravity, have to act on it.

So I grabbed a recent LabRadar series of 20 rounds of a new lot of SK Long Range Match (I haven't kept any large series of my Center-X, which was tested/purchased three years ago and of which one case remains). I replaced the static values from the CSV files with fomulae and started playing. I think I've shown that lower velocity, combined with standard deviation and distance, introduces wider potential vertical dispersion, and this potential dispersion increases as distance and SD increase.

So, I held distance and SD as constants and looked at vertical dispersion only:
- Shots: 20
- Distance: 300 yards
- Velocity: 1130 | 1080 fps (actual velocity | actual velocity-50fps; please also see note after the table)
- ES/SD of 30 and 8.7 respectively for both velocities (calculated, not assumed)
- Elevation values from StrelokPro for my Vudoo using Lapua drag curve
[Correction: the red text labels below should read "Potential vertical dispersion increase at lower velocity (mils)"

So - same bullet, same everything except velocity difference of 50fps shows a potential 0.5 mil increase in dispersion at 300 yards (1.2 mils vs. 0.7 mils) at 95% confidence level (two standard deviations). That's about five inches - and the potential dispersion increases as velocity drops (going back to Quinlan's comment about gravity over time). Comparing 1080 to 1030 fps - which is close to what I actually observed but did not record with my Center-X in Vudoo vs RimX - shows a .9 mil potential difference(2.1 mils vs. 1.2 mils), or about 9 inches more potential dispersion at a 95% confidence level.

Is that enough to matter to you? [Shrug] Taking the confidence to 99.7%, or three standard deviations, ramps the potential vertical spread even higher.

I think this shows how lower velocity for a .22LR, within the realm of "normal" .22LR match ammo velocities, does indeed translate to measurable increased POTENTIAL vertical dispersion. If my logic is wrong, show me how. The bullets don't lie, but statistics might, or at least confuse (Example: A 95% confidence level may be stated as (mean +/- (SD*2)) - but in my example, with hard data, that translates to (1130 +/- (2*8.7)) = (1130 +/- 17.4) = (1147 highest, 1113 lowest, difference of 34 - which exceeds the ES).

This reply is a good example of why I don't chronograph or log data on 22lr ammo. All you'll ever learn is how inconsistent 22 ammo is. Inconsistency goes down with the top grades of match ammo but even this not 100% reliable! As to why longer 22 barrels generally shoot a few fps slower is simple. 22lr propellent burns out in 16 to18 inches depending on weather conditions. After 18 inches the bullet is own its own and friction comes into play and scrubs off some velocity. The longer the barrel is past 18in the more velocity will be shaved off. We're only talking a few fps here and that really makes no difference as to accuracy. A lot of shooters like longer barrels believing it gives the bullet time to settle down after the propellent blast. I also buy into this theory especially at longer ranges. As to velocity. A few feet per second up or down within the advertised range of top grade match ammo makes no difference as to accuracy. Only vertical scope adj. I have found good lots of slower and faster ammo shot in the same gun. It's all about the quality of the particular lot not the speed.

 

[10ring1]

Comparing Cf length to RF length is apples to oranges

 

[DownhillFromHere]

I think y'er sample size is skewing y'er conclusion.
I think the results are being affected by the number of shots fired from each barrel.
It's rimfire, and I can't obtain similar chrony numbers
from consecutive boxes from the same brick.
The ES and SD varies from the same rifle under the same conditions on the same day.

Original post edited 04/19/2023 after @littlepod found an error - hypothesis busted.

I think the issues you raise confirm my hypothesis more than contradict it. If I read your words correctly, you're saying my ES of 30 and SD of ~9 are low and you expect higher spreads/SDs as sample size increases.

I agree. But higher ES/SD values add support my hypothesis, which, simply stated, is that a given bullet is statistically more likely to show increased vertical dispersion at a lower mean velocity than a higher mean velocity within the realm of "expected" .22LR match ammo velocities (~1025-1125fps). That likelihood increases with higher SD.

That statement is based on this methodology:

• Find mean velocity and SD, the larger the sample size the better.
• Feed the statistically-derived (via SD) highest and lowest velocities into a ballistic calculator and note their elevations.
• Run the numbers again with a lower/higher average velocity.
• The lower velocity shows a measurable increase in probable vertical dispersion.

That probability bit me last year. The rimfire matches I most enjoy feature ~1.5MOA targets out to 150 yards and 4-6MOA targets as far out as 400 yards. Last season at a little monthy match, I had cleaned a 100-yard target set - 4 rows of 5 steel targets, smallest was 3" IDPA (1.8" wide), largest was 5" IDPA (3.0" wide) - except for one last 5" IDPA. No wind. Good trigger pull. And I missed. I would have bet $100 in a heartbeat that I would hit that target with that shot - and I missed. I shot that match six times last year and missed 2 targets out of 124 shots (clean score gets a chance at a 2" IDPA bonus; I hit all four of those). There is no doubt the miss was attributable to the rimfire ammo inconsistency you have so ably demonstrated to this community.

Bottom line is, I think I've shown that Center-X or other decent ammo at 1085fps mean velocity has a statistically lower vertical dispersion than the same round at 1035fps, and your main concern - inconsistent rimfire velocity - actually supports my case.

Thanks for commenting, and for all the testing you've done.

 

[justin amateur]

If I follow y'er thinking correctly, any decrease in average muzzle velocity
dropping by more than about 0.5% of accepted standard velocity for match 22lr
pushes the chamber pressures into an area where minor cartridge differences
have greater effect on the burn rates/pressure peaks and causes more variations in exit velocity.
I can relate that to ladder testing with centerfire reloads when looking for the velocity plateau.
Is the 22lr indicating that same type of plateau, a more even pressure curve in the 1080 fps range?

 

[littlepod]

I think your table has some bad data. Your drops are wrong for 1097

I plugged in some data... to say if the SD was slightly better on the longer barrel. Comes out to be the same spread. So it'd be a 24 ES, vs a 32 ES.

	Average	2SD	-2SD	SD
Velocity	1110	1126	1094	8
Drop		12.4	13	0.6
Velocity	1080	1092	1068	6
		13	13.6	0.6

 

[DownhillFromHere]

I think your table has some bad data. Your drops are wrong for 1097

I plugged in some data... to say if the SD was slightly better on the longer barrel. Comes out to be the same spread. So it'd be a 24 ES, vs a 32 ES.

	Average	2SD	-2SD	SD
Velocity	1110	1126	1094	8
Drop		12.4	13	0.6
Velocity	1080	1092	1068	6
		13	13.6	0.6

You are correct! I appreciate your taking the time to run some data and catch the error. Peer review was a significant part of my daily work, especially early in my career. It's a little embarrassing to submit something with an oversight error like that, but I'm correcting it. The whole thing remains a mildly interesting academic exercise (well, if you're geeky enough).

So, with a corrected value of 12.7 instead of 12.3 (from my solver), the increase in potential vertical dispersion is measurable but pretty much falls into the realm of noise. Hypothesis busted.

Thanks again.

 

[littlepod]

You are correct! I appreciate your taking the time to run some data and catch the error. Peer review was a significant part of my daily work, especially early in my career. It's a little embarrassing to submit something with an oversight error like that, but I'm correcting it. The whole thing remains a mildly interesting academic exercise (well, if you're geeky enough).

So, with a corrected value of 12.7 instead of 12.3 (from my solver), the increase in potential vertical dispersion is measurable but pretty much falls into the realm of noise. Hypothesis busted.

Thanks again.

Faster is better, not as drastic of a change but it is. The only reason why longer is better is if the SDs/ESs are lower, otherwise, the percent of ES / Velocity is the factor. So 32/1110 or 24/1080. For me longer = better weight balance for PRS, and slightly lower SDs. Maybe not be drastic enough to offset the velocity, but then I can also find hotter ammo like Lapua Long Range will shoot 1100 out of my 25"

 

[grauhanen]

The only reason why longer is better is if the SDs/ESs are lower

Are you aware of any verifiable testing that's been published in print or online to confirm that longer barrels produce better SD and ES numbers?

 

[littlepod]

Are you aware of any verifiable testing that's been published in print or online to confirm that longer barrels produce better SD and ES numbers?

Nothing with a large enough sample size to be scientific.

 

[RTH1800]

My 28” barrel is stunning quiet with Eley Tenex. Less noise than a suppressed 16” barrel.
Huge advantage squirrel hunting.

 

[HPIguy]

My 28” barrel is stunning quiet with Eley Tenex. Less noise than a suppressed 16” barrel.
Huge advantage squirrel hunting.

What does that have to do with what's being discussed in this thread?

 

[RTH1800]

Advantages of a long barrel. It’s a hell of an advantage.

 

[12Bravo20]

I've seen a few videos discussing this and found them interesting. Some videos show a difference in vertical dispersion between different lengths of barrels while others show no measurable difference.

Now if my 21" 1-12 twist Bore Buddy barrel will settle in and stop having flyers then I can shoot it and my 16" 1-16 CMMG barrel and my Savage MkII BTVLSS with it's 21" 1-16 twist barrel to compare all three. I know the 1-12 twist will be different than the two 1-16 twist barrels. And I now a bolt action versus a semi auto will come into play too.

 

[Milo 2.5]

Advantages of a long barrel. It’s a hell of an advantage.

Entirely too much restraint! Hell, I'll listen to you just based off your rifle pics. lmao

 

[RTH1800]

Since it shoots like this at 100 yards I guess the 28” barrel qualifies as fairly accurate as well.