Fascinating Phenomenon in Cylinder Gap Data

During the past few days I have been absorbing the data collected during the BBTI Project’s cylinder gap tests. I noticed a strange anomaly. For each of the cylinder gaps tested, as the barrel length increases so does effect on velocity increases (as compared to having no cylinder gap at all), up until a certain point, then the effect on velocity starts to decrease.

I would have expected that as the barrel length increases, so the gap would have an increasing effect on velocity (because the bullet stays in the barrel longer so the gap will bleed off more gas).

The only explanation I can come up with is because they were using handgun, not rifle, cartridges (.38 Special & .357 Magnum), once the pressure inside the barrel decreased to a certain level the friction from the barrel begins to have a much greater impact on the velocity than the bleeding of gas, thereby decreasing the relative impact of the cylinder gap. Can anyone else come up with any other explanations?

Steve Johnson

Founder and Dictator-In-Chief of TFB. A passionate gun owner, a shooting enthusiast and totally tacti-uncool. Favorite first date location: any gun range. Steve can be contacted here.


  • Matt Fulghum

    I expect that flow “chokes” at the cylinder gap, meaning that it cannot flow out of the area of the gap faster than the speed of sound in the burnt propellant gas. Increasing the gap width will increase the area out of which gas will flow, so your total mass flow of gas out of the gap will increase proportionally. With less gas propelling your bullet, you have less ability to transfer momentum, leading to reduced velocity.

    With a shorter barrel the bullet is being accelerated for less time before it leaves the barrel, and since the speed of sound in the propellant gas is almost solely dependent on temperature (which doesn’t care about barrel length) you will see less mass lost from the gap in the time the bullet is in the barrel, and therefore less useful momentum is wasted… which explains the reduced effect at shorter barrel lengths.

    With a longer barrel the mass loss rate from the gap is still about the same as with the shorter barrel, but more momentum is transferred to the bullet from the gas in the barrel before it leaves, so the gap should have a proportionally smaller effect.

    So with those arguments, I would expect the situation to be worst case when the barrel length is of medium length.

    Even still, I wonder what the error bars are on these velocity comparisons…

    • Aurelien

      I would agree. Physics would probably show that the amount of gas expelled through the gap between cylinder and barrel comes to a maximum after a certain barrel length (of course depending on powder and wideness of gap).

      From then on, the longer barrel would not lose any compression anymore, and therefore the velocity would once again go up.

      Another thing to factor in would be the carbon deposit which would gradually fill the gap (it’s the reason the gap is there in the first place). The more you shoot, the more the gap fills.

      The interesting fact is, those studies would allow to chose a near perfect barrel length to ammo ratio for any revolver on the market.

      • Matt Fulghum

        I’m not sure what you mean by the longer barrel not losing compression anymore… As long as you have pressure in the barrel that is higher than the surrounding atmosphere, you will lose gas from the cylinder gap, and the pressure in the barrel remains significant long after the bullet has left the muzzle. It’s just that with a longer barrel the loss that occurs will be less meaningful overall in the gas->bullet momentum transfer.

        Also, the cylinder gap comes down to manufacturing tolerances. It would be vastly expensive to hand fit every cylinder, and a cylinder that is too long will not fit, so manufacturers have to compromise by allowing a certain amount of gap in the design to maximize productivity. I would not be terribly surprised if performance center revolvers used cylinders that were cut too long in the general assembly process, but were otherwise dimensionally good.

      • Aurelien

        Well let me explain my theory then :

        When you empty a water bottle, only a certain amount of water can leave the bottle at a given time, dependding on the opening.

        Same thing with gasses. At some point the gas lost in the gap between cylinder and barrel will come to a maximum, at which point you will still lose gas, but the compression inside the barrel will remain the same until the bullet leaves the barrel.

        That’s all i’m saying.

        Also, the cylinder gap allows the revolver to work properly even if you shoot like a maniac. Never got a cylinder stuck because of heat expansin and/or carbon deposit ? Happened to me on tight-machined competition revolvers.

  • John

    I remmembered something along the lines of an auto exhaust, backpressure or something…

  • noob

    Interesting. I wonder if this will affect the development of the LSAT which uses a tilting chamber to contain it’s CT or Caseless ammo and may have a cylinder gap.

    • Dave C


      Checkout the NIDA “Joint Armaments Conference, Exhibition & Firing Demonstration” website: http://www.dtic.mil/ndia/2009infantrysmallarms/wednesdaysessioniv8536.pdf (sorry, I don’t know how to hyperlink on this site). Powerpoint slide 6 explains it nicely. They use gas pressure from the propellant combustion to expand a sleeve in the chamber against the barrel face and breech. Elegant solution indeed.

      BTW, one may pleasantly wile away time reading the various presentations on the NIDA presentations. Lots of cool info there.

      • Matt Fulghum

        I like it. I’m sure someone’s feeling (rightly) smug about that design somewhere.

      • noob

        Thanks! that’s pretty awesome.

        I hope that there’s a latch to pull out the axel (if it’s called that) so you can remove the chamber halves entirely to clean them, otherwise it’s time to go in with a dental pick and angle brush to get carbon out from that tightly toleranced interface between the two halves.

  • Lance

    Don’t care revolver make up for there low capacity and some models loses ballistics is made up by reliability and accuracy.

    • Yoda

      Just stop for God’s sake.

    • armed_partisan

      It looks to me that some people have such a strong bias against revolvers they refuse to see any advantages to them, which can only mean that they have limited or no experience with them. If you don’t think that the average revolver is more accurate than the average semi-auto, I have news for you: you are a lousy shot.

      • SpudGun

        @armed partisan – I think you are missing the point entirely. The topic relates to pressure loss with pistol barrels over a certain length versus the pressure loss of the cylinder gap, it has nothing to do with ‘glocks rulez, revolvers sux’.

        Lance has received so many thumbs down because he posted a comment that has nothing to do with the topic at hand. He might as well have written ‘Dallas Cowboys are the bestest!’ in terms of the relevance it possessed.

    • armed_partisan

      If that was true, Spudgun, then why is my unrelated comment above, which has to do with the data and the numbers presented exclusively, also receiving thumbs down? Nobody has responded to it, as there are no counter arguments. These are the facts. I presented the formula I used, discussed my method of evaluation, and the accurate results that I found. The amount of velocity lost through a cylinder gap is marginal to irrelevant. So many other factors effect velocity to a greater extent, like twist rate, rifling type, powder selection, ambient temperature, etc, that the only reason anyone could take any issue with my original comment was is their bias prevented them from acknowledging that their favored gun might not possess as great a benefit as they imagine.

      • JMD


  • Sian

    That would be my guess. faster-burning pistol powders would be burnt completely at the 9-10 inch mark, and with longer barrels you’re now fighting the barrel friction with quickly declining pressure. With no cylinder gap, even though your powder is burnt up, it has nowhere to go but down the barrel, keeping velocities fairly even past the optimal point.

  • armed_partisan

    I spent a long time on this site when you first posted the link. I did a number of calculations on several different loads they had posted to calculate muzzle energy. That’s done as follows:
    (Bullet Weight in Grains)x(Muzzle Velocity in FPS)^2/450395=Muzzle Energy in ft/lbs

    I compared the differences in velocity between the “no gap” and the .006″ gap for each given load at their largest and narrowest points (usually zero) on the chart. If, for example, a given load showed that it had the biggest gap in velocities at the 14″ barrel mark, and another showed it’s biggest gap at the 17″ mark, then I compared it at those points. Usually, your talking less than 100fps (many times much less) difference in velocity between the largest gap (.006″) and no gap. If you calculate the muzzle energy difference of the given bullet weight, you’ll find that the difference in energy is NEGLIGIBLE.

    Not only less significant than I would personally find acceptable, but insignificant enough to be almost statistically irrelevant. The largest gap in muzzle energy I found for a given load of the ones I calculated was LESS than 2.5 foot-pounds of muzzle energy difference. That’s less than a pellet guns worth of muzzle energy, and I’m not talking a high-end hunting air rifle, I’m talking a cheapo POS airgun. There are airsoft guns that produce more muzzle energy than that, and that was the LARGEST variation I found.

    Apparently, the major factor in velocity of say, a 9mm fired out of a 4.49″ barreled Glock 17, and that same load fired out of a 6.5″ Ruger Blackhawk Convertible is not the fact that the Blackhawk has a cylinder gap, but the fact that it has a significantly longer barrel, and will therefore out perform the Glock in terms of velocity and muzzle energy. A Blackhawk with a 4.62″ barrel will theoretically gain more velocity over the 4.49″ Glock 17 Barrel than it will lose due to cylinder gap.

  • Spiff

    Just for the sake of experimenting, try a Glock and a revolver of comparable barrel length and compare real velocities – a polygon barrel tends to seal more gases giving a modest increase in velocity over a button rifled barrel using simular ammunition specs.

    • Cymond

      Remember, pistol barrel lengths are measured from the rear of the chamber while revolver barrel lengths indicate the length of barrel in front of the cylinder.

      For the sake of easy math, let’s assume we have a 1″ long chamber. A 4″ pistol has 3″ of barrel in front of the chamber, while a 4″ revolver has 4″ of barrel in front of the cylinder.

      • Spiff

        I was simply trying to find a commonality between polygon rifling and button rifling in hand guns. Some cartridges head space on the rim others on the case length (.38super/auto) – I think a polygon rifled barrel in either a rifle or pistol will produce a slight bit more velocity, and tends to to leave less “debris” in the barrel due to the hot gases consuming the powder…This is one reason why Glocks can continue to function after 6,000 + rounds without cleaning, and seem to achieve a slight increase in velocity…My Glock, an AF series, has had about 12-14,000 rounds through it without any stopages, although I must admit I clean it every now and again…

  • ThomasD

    These are smokeless powders. If the DUT had used black powder the results might more closely track your expectations. But smokeless powders behave quite differently depending on the pressure under which they burn.

    So total losses of gasses at the gap may not be the entire problem so much as the delay in time to peak (or ideal) combustion pressure.

    If so this phenomenon may not be readily apparent in ultra short barrels – as they never have the opportunity to ‘reach peak’ anyway – the bullet already having departed the muzzle.

    But as barrel length grows the phenomenon can becomes apparent. It may also diminish in overall effect as barrel length continues to increase beyond some point.

    While a longer barrel represents a larger volume pressure vessel it also means more time spent at higher than atmospheric pressure (these pressures being what accelerate all bullets.) The greater time spent under pressure (even though slightly reduced from absolute maximum by the gap) allows for greater opportunity for total acceleration of the projectile and velocities begin to rise.

    As indicated in the data there is an inflection point where one performance aspect outweighs the other.

  • DaveR

    I think the original speculation sounds the most-plausible.

    Namely that there is a maximum velocity for any given load, and that velocity is determined by the equilibrium reached between the gas pressure acting to accelerate the bullet (which is itself a function of the powder AND the velocity of the bullet at any point in time) and the bullet/barrel friction acting to slow it down. This is reflected in the fact that if you look at the raw data, the No-gap barrels plateau off at mid-length and then never rise.

    Gapped barrels take longer to reach this plateau because there is a pressure bleed behind the bullet. However, that bleed is never great enough to allow the pressure to drop below the level where friction with the barrel causes an equillibrium to be reached.

    My speculation would be that, if longer barrel lengths were used, the velocities from the gapped barrels should plateau off as well, albeit at a slightly lower maximum velocity that the ungapped barrels.

    • Peter H

      I looked at the velocity graphs for all 13 loads tested in the cylinder gap test.

      1. At only 10 rounds per variable set, the curves were not very smooth, and no statistically valid conclusions about minuscule velocity differences can be drawn.

      2. At 16″, there were numerous anomalies, probably testing artifacts–not real phenomena.

      3. Nevertheless, the *overall trend* for all ammo tested showed an increase in velocity with increase in barrel length.

      4. As would be expected, velocities were lower with increased gap size.

      5. All the ammo appears to be carbine compatible.

      6. Some loads appeared to have much more consistent velocities than others–better quality control or luck of the draw?

      7. Nothing more to see here. Move along.

      • SpudGun

        @Peter H –

        6. Some loads appeared to have much more consistent velocities than others–better quality control or luck of the draw?

        If Steve’s assertion that friction is coming into play is correct, then the relative size of each manufacturers bullet diameters could also be a factor.

        Not that I wish to make this process even more complicated (Lord knows I’m having trouble figuring it out with my idiot brain), but if a more generous bullet is creating a better gas seal in the barrel, will this increase the gas vented at the cylinder gap or will it push the bullet faster down the barrel? Or both?

        If a less generous bullet was travelling along the barrel, would the gas pressure drop as the bullet / barrel seal decreased? Would the less generous bullet reduce back pressure towards the cylinder?

        Then there is also the question of forcing cone design and the size of the respective cylinder throats. So you could have a really small cylinder gap but still be losing pressure as the gas wastefully expands the cartridge in the cylinder throat / forcing cone.

        My brain hurts, I give up.

  • mikee

    Just looking at the graphs the periodic nature of the velocity vs. barrel length jumps out at me. Does that mean anything?

  • I’m not a fluid dynamics engineer nor do I play one on TV. There is a time element here too. As Matt Fulghum pointed out, a short barrel vents the pressure faster than a long barrel and this reduces the effect of the flash gap pressure loss. In a longer barrel, as the bullet propels forward, it leaves behind an expanding gas chamber. This also means the path of least resistance for the gas changes from the flash gap to the barrel as the bullet accelerates through the bore. Initially as the bullet passes the flash gap, the building pressure against the base is vented sideways. But as it moves up the bore, the gas flows straight forward, compressing the gas in the barrel – only excess pressure is vented sideways. This pressure venting at the gap should decrease rapidly as the bullet moves forward.

    Not being a physicist, would not the friction drag become a reducing percentage of the bullet’s velocity over time? That is, as the bullet’s momentum builds, the friction matters less because it takes less work energy to move the bullet a given length down the barrel. I would think this especially true once the length of the bullet has been grooved by the rifling.

  • It’s worth cross-referencing to Ballistics By The Inch here.

    Those guys have done some excellent work. Note that .357Mag has finished accelerating after around 15″ to 16″ of barrel length, depending upon load. After that, additional barrel length actually slows the bullet down.

    In a revolver, the gap would become inconsequential at that point, hence the trend data we see here.

    In other news, I’m continuously baffled when I see manufacturers putting out lever rifles and carbines chambered in .357Mag with barrels longer than 16″. Surely they must know, right?