Operating Systems 201: Bang vs. Gas Trap

The 1927 model Bang rifle, paradoxically uses the simpler and more advanced gas trap operating system!

Now, based on the four previous articles on gun operating mechanisms, some of my readers may be thinking “jeeze, Nathaniel, we already know all this stuff! Why are you telling us this?” Don’t worry! I haven’t forgotten about you, and that’s why I am sprinkling in more advanced topics as we talk about the more basic ones. We just spoke about the principle of gas operation, so let’s jump up a grade to 201 level, and talk about two of the most conceptually obvious incarnations of gas operation. I’m writing, of course, about the Bang, and gas-trap methods of tapping propellant gases.

  • Bang-type gas operation: Where gasses are tapped against a forward-moving muzzle cap, which is connected via a seesaw linkage to an operating piston.
  • Gas trap operation: Where gasses are tapped into a stationary muzzle cap, and then bled to the rear to act against a gas piston.

Because of their similarities, these two systems of operation are often confused; both use muzzle caps that harness the gas pressure without needing ports drilled in the weapon’s barrel, but they differ in how that gas is utilized. The Bang system is an earlier system, which heavily relies on mechanical devices. Gas is assumed to drive forward from the muzzle, so the muzzle cap moves forward. That motion must then be converted to rearward force to operate the mechanism, so a seesaw linkage is incorporated to provide that. This is a very natural progression of design, but results in a very complex and cumbersome design.

Pure gas trap weapons, on the other hand, utilize the less intuitive principle of gas flow, where gas can be tapped against a stationary muzzle cap and produce rearward force directly, operating a purely rearward-moving piston, no linkage required.

To illustrate how these two mechanisms are different, YouTuber Bloke on the Range took to the blackboard:

The Bang-type system was not widely used; only the earlier designs from the Dane Søren Bang utilized the principle, as well as a handful of other experimentals like the obscure Chinese Liu rifle. The gas trap system with fixed muzzle cap, however, got further: The early models of the US M1 Garand used the mechanism, as did the German G.41(W) and G.41(M) selfloading rifles.

Readers should note that so far as I know, the simpler gas trap mechanism is a direct development of Bang’s forward moving gas cap mechanism. In fact, the earliest rifle I know of that used the gas trap system was Bang’s own 1920 model rifle!

Nathaniel F

Nathaniel is a history enthusiast and firearms hobbyist whose primary interest lies in military small arms technological developments beginning with the smokeless powder era. In addition to contributing to The Firearm Blog, he runs 196,800 Revolutions Per Minute, a blog devoted to modern small arms design and theory. He is also the author of the original web serial Heartblood, which is being updated and edited regularly. He can be reached via email at nathaniel.f@staff.thefirearmblog.com.


  • pbla4024

    Now, is M1895 potato digger bang operated when muzzle moves kind of down? 🙂

    • err, no. The muzzle doesn’t move anywhere, there’s no sleeve. There’s a conventional transverse gas port but with the piston on a swinging arm.

      • ostiariusalpha

        It’s technically not a piston on the arm, but more like a proto/quasi-piston. It has an impingement face that transfers mechanical energy, but it lacks a cylinder, which would be requisite to make a piston assembly. Because the impingement face moves in an arc, the indentation that it seats into can’t act as a cylinder.

        • RocketScientist

          Not to argue with ya (got a lot of respect for you Ost), and at this point we are having the most pedantic of debates (which are my favorite kind), but I’d say you CAN call it a piston, looking at schematics/cross sections of the action. In the image below it seems clearly illustrated to me. It shows the lever in the fully closed (top) and fully opened/rearward (bottom) positions. We can clearly see a narrow gas port that is drilled through the wall of the barrel. This opens up in a longer-than-it-is-wide opening. We see that a peg-like protrusion of the action lever (don’t know the actual terms so I’m making it up as I go along… the part labeled “l”) seats in this opening/recess. I think it would be entirely accurate to label that recess as a cylinder and the protrusion on the action lever as a piston (albeit one with a fairly short length/time of action, but thats not entirely different from many short-stroke/action piston guns out there). I’m coming at this from a mechanical engineering POV (as thats my training/profession) and it certainly meets the definitions for those parts used in my experience.

          Further, as an interesting aside, the arrangement between the action lever (labelled “l”) and the connecting rod (labelled “c”) and the slider in the action it connects to are classic examples of a crank-slider mechanism. So you’ve got pistons and cylinders turning rapid gas expansion into an angular (rotational) motion and then into a reciprocating motion. In a broad sense, that describes steam engines and pumps almost exactly. Not a real big surprise that browning went with a few well-understood mechanisms of the era (glory days of steam and all that) to make a revolutionary firearms.

          Anyways. Hope you found this discussion at least interesting.

          Image for reference:

          • ostiariusalpha

            Ha ha, it is just a pedantic, but not irrelevant, distinction. I don’t mind at all that you’d make a counter-argument here, I admit that I’m wrong about stuff plenty of the time. And that is good work digging up the diagram. But if I may defend my assertion a bit, that it isn’t a true piston, by pointing out that a piston operates by building up a pressure well inside the cylinder as it continually pushes on the piston (or the piston pushes on it) until it either comes to a mechanical stop or the piston reaches a point that a relief valve in the cylinder channel lets off the pressure (often by the piston simply exiting the cylinder altogether). In contrast, the M1895, while it does get a little boost from the space that the peg seats into, does not create any kind seal between the “piston” circumference and “cylinder” walls, but rather gains almost all of its energy from the initial impulse, with the pressurized gasses from the barrel escaping around the peg even before it has fully exited its seating space. I tried to be careful to not say that the peg is a pseudo-piston, but that it acts as a quasi-piston; doing very similar work to what a true piston does, but lacking the necessary qualification of sealing the pressure in the cylinder. This is actually a good feature on a black powder autoloader, as a true piston would quickly jam from fouling, whereas the large clearance around the impingement peg would keep the machine gun going for quite a while longer.

          • Most (all?) pistons are very much not sealed in any way to their cylinder…

          • ostiariusalpha

            Hmm, you might want to look up how a piston and cylinder work together, Bloke. I had to look it up myself to be sure, and found the technical engineering term: the peg is actually called a “displacer” when acting as a quasi-piston.

          • then almost no “piston” in a firearm is a piston, since there’s almost always play and no seal. Sometimes a lot of play.

          • ostiariusalpha

            Nope, even an AKM piston seals pretty well in its cylinder. At least, mine does, and every single one I’ve ever handled is the same. If your piston has a lot of play, then it is worn and you should look at replacing it.

          • My brand new Spaghetti M1 has play. Every SA80 I was ever issued has play. My SVT had play. the piston AR15 upper I had from new had play. My L1A1 had play. These things are not at all gas tight!

          • ostiariusalpha

            The only play they should have is on the op rod. This articulation is actually necessary because of the fit between the piston and cylinder, since it requires the reciprocating part to adjust to the stationary part for proper alignment. In fact, Century Arms had a lot of issues with AK rifles that they sold, because their piston didn’t wobble on the op rod and would not self align for seating into the cylinder, ending up with damage to the piston ring.

          • Os, do me a favor and disassemble a gun with a piston (AR-15s count) sometime, after it’s been shot. Take photos of where the gas residue is.

            It will disprove your theory.

          • ostiariusalpha

            I don’t think you understand what a gas ring seal is, Nate. It doesn’t mean that absolutely nothing gets past the gas rings, that would be mechanically impossible for a low-friction clearance fit component like a piston (a high-friction component would obviously be something such as a bullet in a barrel, but even those get gas seepage on conventional rifling). Not to mention undesirable, as the high pressure gas acts as a lubricant in the miniscule clearance between the piston rings and cylinder; it is actually because the clearance is so small that you need the gas to perform as a lubricant at all. Having residue behind the piston is not proof that piston rings don’t form a seal, the fact that you’d even bring it up as some kind of proof rather more indicates that you aren’t getting why gas rings are necessary for a piston in the first place.

          • Lay off the salt, bro.

            The word “seal” has a very clear definition. It doesn’t seem like you’re using it correctly. You are saying the piston seals the cylinder, then saying they don’t seal the cylinder.

            Maybe you understand what you’re saying, but how about you take a step back and think about how you’re saying it, instead of attacking other people as know-nothings, hmmm?

          • ostiariusalpha

            LOL! If you thought that was salty, then get a taste of this. Yes, I did assert that the piston seals against the cylinder, but I stand in relatively good company in doing so: David Williams (Patent #2090656), John Garand (Patent #2369669), Eugene Stoner (Patent #2951424), Jim Sullivan (Patent #20120180354), and every single engineering text that deals with pistons. Like all mechanical seals, piston rings are an imperfect seal and blow-by is an understood fact that has to be engineered around (that’s why the crankcase in a car engine needs a relief valve to keep the engine block from rupturing a gasket, despite that the typical automobile piston has three separate piston rings to minimize blow-by). I am entirely curious as to what authority you are appealing to in claiming that the piston rings are not seals, since as far as I know, none exist. Perhaps if you and Bloke had taken less umbrage at being being corrected, and simply Googled it, this whole diversion would have been easily avoided.

          • So far as I know, neither of us are angry. I certainly don’t think either of use were offended by what you said.

            You say you’re “in good company” by saying it seals against the cylinder, listing off some firearms designers and their patents. However, as Bloke knows, those designers didn’t write the text of their patents, a patent attorney did. So you’re in the company of their patent attorneys, I suppose.

            Like I said before, it seems to me like you’re attacking others because of a misunderstanding in your own wording. Look at the context for how you use the word “seal”, and see that it is extremely confusing. You say the 1895 does not have a piston because it doesn’t “seal”, except that it seals to the same extent as any other design (that is to say, hardly at all):


            Now, I could see the argument for the original Browning “muzzle cup” prototype gas operated rifle not having a piston, as it has something a bit more like a trap that moves forward, but the 1895 has a piston just as much as, say, an M1 does. In fact, since we’re talking about patents, Browning’s attourney describes the gas mechanism eventually used in the 1895 thusly:

            “The construction of the lever F with the recess surrounding the nipple prevents to a very considerable extent the escape of gases between the aperture and lever until after the lever shall have commenced its opening movement and received its initial force, for were there no such protecting recess between the aperture and the lever the gases would escape laterally between the aperture and the lever, and thereby a considerable portion of the initial force would be lost, and, further, the gases so escaping would be in the form of a flame which would be objectionable, if not dangerous; but by protecting the apparatus by the recess between the aperture and the lever the gases are confined to their initial operation, so that such an escape is prevented, and after the lever starts the gases will then follow the lever and without the lateral spread of the gases which would otherwise occur.”

            The fact is that no gas piston can totally seal its cylinder. To an extent they do seal, but the miscommunication occurred when you suggested that the 1895 machine gun somehow didn’t seal its piston. Does that make sense?

          • ostiariusalpha

            Ha ha! This is why I like you.

            First of all though, I do find it somewhat bad form for you to suggest that Garand, Stoner, and Sullivan were idiot savants that didn’t understand what technical terms their patent attorneys were using, or that they wouldn’t care to inform the attorney if he had used a term incorrectly (Mr. Williams also had a sophisticated understanding of firearms engineering, as I gather it). That’s not going to be a very fruitful avenue to defend your contradiction of their patents. Mr. Sullivan, at least, is quite alive and can still defend his involvement in the wording of his patent.

            As for my “attack,” if it seems to me that the person I’m talking to doesn’t understand (or has a defective understanding of) the principles involved in the discussion, I will clearly tell them so, with the hope that they have knowledge, understanding, and motivation to adequately clarify, defend, or modify their position. This is very much separate from those times when I have not too infrequently actually attacked people’s intelligence, character, and ancestry with straightforwardly ornery insults; to be clear, my questioning of your understanding is not an attack on your intelligence or character.

            Now, to the meat and potatoes of our disagreement. I have no indisputable proof that John Moses Browning had a deeply detailed understanding of what exactly makes a piston what it is, but I consider it far more likely that he (or at least his patent attorney) avoided using that common engineering term for a very good reason. From the language they use about the recess and the nipple (or as I have referred to it, the displacer) it is clear that they are using the recess tube to direct the gasses toward the lever, not as a pressure well such as is found in a cylinder. Because the nipple/displacer does not form a mechanical seal with the recess, it cannot create such a well either; in fact, it’s designed to avoid contacting the channel walls of the recess altogether (unlike the Garand and its .525″-.526″ piston tolerance, plus Go/No-Go gauges for its cylinder). Rather, the nipple is (in their understanding) performing primarily as a valve head extension on the lever, allowing it to receive directly from the aperture hole the maximum impinging energy of the pressurized gasses; a simple impingement surface. Though as the nipple travels out of the recess, they do something quite interesting and useful together despite not creating a mechanical seal… they become a flash suppressor. The pressurized gasses flow past the peg, but it creates turbulence in the gas and allows the recess channel to cool it enough to disrupt the flash. Frankly, the nipple on the M1895’s lever works more than fine without being a piston.

          • Second paragraph: Putting words in my mouth. I said no such thing, I just said that the patents were written by attorneys, which is true.

            Third paragraph: Still missing the point that your argument was confusing to anyone with a sound understanding of the terminology and mechanics, of which Bloke and myself are two. Don’t believe me? Ask Bloke what he does for a living.

            Fourth paragraph: I suspect his patent attorney didn’t use that term because he was a different attorney than the guys Garand, Williams, Stoner, and Sullivan had. Also, he was working during a completely different era than any of the others, and their conventions for wording may have been completely different then.

            It seems to me from what you wrote in the rest of the fourth paragraph that you didn’t closely read Browning’s patent. Go ahead and give it another go:

            “The construction of the lever F with the recess surrounding the nipple prevents to a very considerable extent the escape of gases between the aperture and lever until after the lever shall have commenced its opening movement and received its initial force, for were there no such protecting recess between the aperture and the lever the gases would escape laterally between the aperture and the lever, and thereby a considerable portion of the initial force would be lost, and, further, the gases so escaping would be in the form of a flame which would be objectionable, if not dangerous; but by protecting the apparatus by the recess between the aperture and the lever the gases are confined to their initial operation, so that such an escape is prevented, and after the lever starts the gases will then follow the lever and without the lateral spread of the gases which would otherwise occur.”

  • RocketScientist

    I remember reading that one of the things that hindered the development of really effective self-loading gas-operated rifles was the very strong doctrinal aversion most of the major nations war departments (or equivalent organizations) had against gas ports drilled into the barrels of firearms. There were (in many cases valid) concerns about fouling of small gas ports and erosion/wear of the barrell. In an era where clean-burning smokeless powder was a relatively new development and materials science was advancing rapdily, its easy to see why a bunch of grey-haired generals would be slow to understand the implications, or reluctant to move forward with them, and in many cases for good reason. However their almost religious opposition to holes drilled through a barrel is said to have seriously hindered small arms development in the interwar and WWII periods, especially on the german side, even after the problems with gas ports had been mostly solved or ameliorated. Arms developers were forced in many cases to use an older, more problematic gas-trap mechanism for no reason other than the army required it. Good thing we’ve got the whole small-arms-development process fixed up and figured out these days…. oh wait…

  • Dave

    The Søren Bang rifle was a seven-shot, charger-loaded piece, yes? 1911 and 1920s were very similar, or so I thought. Nathaniel F. seems to indicate or imply that the 1920s version used a gas trap instead of the “true” Bang system, whereby a slender op-rod pulled forward actuated a highly modified lever-action system to perform the operations of extraction, ejection, and reloading…

    As J. Hatcher wrote in his notebooks, few army pun-sters knew how close they’d come to being issued with “U.S. rifle, caliber .30/.276 M1 Bang!” The story about the Bang includes the down-side that the rifle’s barrel was necessarily light and therefore prone to overheating and vertical stringing, a failing common to the Johnson as well.

    • The manual – assuming I am reading it correctly – indicates that the 1927 Bang was a gas trap design. Hatcher – I think it was Hatcher – also states that the 1920 model was has trap as well.

      • Dave

        Ah yes, thanks! There it is on page 10. “The trapped gases under the pressure thus built up in the expansion chamber pass through the port to the gas cylinder where they act upon the piston and push it to the rear. This movement of the piston is transmitted to the cover through the piston rod.”

  • Alexandru Ianu

    So, the Heinemann toggle lock rifle is a Bang system without the seesaw since the action moving forward is intended.

    • Nope, as the Heinemann isn’t gas operated, it’s gas unlocked delayed blowback, a system I will get into later.

  • Organgrinder

    If memory serves, Browning used what could be considered a prototype Bang mechanism to create the world’s first self loading rifle. He mounted a hinged plate in front of the muzzle of a lever action with a link connected to the lever, and a return spring. Later he used this setup to create the world’s first self-powered machinegun.

    • Yep, the mechanism used on Bowning’s earliest selfloading prototypes was very similar, although he took it in a different direction with later prototypes

  • MPWS

    I used to ask myself why some weapons had gas tap at very front (and even at muzzle as article explains) and some quite happily somewhere in middle of barrel’s length and in both cases it worked to satisfaction.
    One answer to this is burning properties of gas – speed vs. pressure. Other and even more important is due to (not mentioned here yet) is travel-pressure curve. Due to continuous expansion during burning process the gas level progressively drops along the length of barrel.
    Here comes the crucial point: you can make small port along the length of barrel and expect high pressure (and harsher extraction) and short duration OR you can have lower pressure exit and longer duration (with better working system). Which one would you pick?