“Flapper Locking” Makes A Comeback in AA .338 Lapua

Now this is something that I did not expect to see make a technical comeback with the focus of modern firearms on rotating bolts. The new Alexander Arms .338 Lapua Magnum rifle (the “Ulfberht”) utilizes a flapper-locking bolt design. The base design was first patented in the late 1800’s (1870) by a Swede named Friberg

Similar in concept to the roller-lock designs that typified mid-century H&K’s, the flapper locking mechanism uses flaps attached to the bolt to retard the bolt’s rearward movement giving time for the powder to burn and safely unlock the firearm.

For those looking to understand how a flapper-locked firearm works, our own Ian from Forgotten Weapons has a great video showing the field strip of a flapper-locked German rifle.

Nathan S

One of TFB’s resident Jarheads, Nathan now works within the firearms industry. A consecutive Marine rifle and pistol expert, he enjoys local 3-gun, NFA, gunsmithing, MSR’s, & high-speed gear. Nathan has traveled to over 30 countries working with US DoD & foreign MoDs.

The above post is my opinion and does not reflect the views of any company or organization.


  • Max Popenker

    Well, mixing up HK roller-DELAYED action, MG-42 roller-LOCKED action and flap-locked action such as this of AA or Degtyarov DP / RPD / DShK is hardly correct.
    Of cause, there were attempts to create delayed (retarded) blowback actions with flaps instead of rollers, but I believe that AA system is not one of those 😉

    • Giolli Joker

      Ok, this clarifies a lot… I watched Ian’s video thinking that we were talking about a delayed blowback as implied by the article, but I couldn’t get how the momentum was imparted to the firing pin extension to allow unlocking.
      So basically we have a gas operated firearm where the piston unlocks the flappers instead of a rotating bolt head?

      • Ben


      • J-

        It boggles my mind that this would be pulled out of the history books, especially with the .338 Lapua.

        It was common for repeating rifles of the 1800’s and early 1900’s to have long bolts that locked at the rear in a machined receiver. The Winchester lever guns all locked at the rear of the bolt. So did the British SMLE. The M1918 BAR uses a long tilting block that locks against the top of the receiver, pretty much an upside down SKS.

        As an engineering design, this is bad. A long bolt has more room to flex and so needs to be bigger and heavier to handle the stress. Same with the receiver. The M1918 uses a huge, machined steel receiver. It is a beast. The SMLE and Winchester rifles used low pressure ammo (.30-30 and .303 which were early smokeless loads) and can’t handle higher pressure ammo. When the No 4 Mk 1 Jungle Carbine was made from the SMLE by lightening the receiver, the receiver would stretch under firing and deform leading to poor bolt head alignment and the notorious wandering zero.

        The BAR and M1 Garand both fire the same cartridge, the big difference is the Garand locks up near the breech which allows for a much lighter receiver. The AR10 used a barrel extension and took all the firing stress off the receiver, which is what allowed it to be made from aluminum.

        The flapper design requires two long, stiff flappers and a heavy receiver. To use that in conjunction with a super magnum cartridge seems so dumb to me.

  • MPWS

    When looking at something like this, one has to ask: where is innovation? The hefty cartridge application actually begs for something like spring supported barrel/lockup assembly within outer envelope. Or, momentum transfer using accelerator. This instead is just traditional hard lockup transferring recoil kick right into shoulder.
    As to “flap lockup”, it does not bear any particular advantage part of straining larger volume of material, maybe. Well yeah, maybe compression in flaps is better than shear in lock-nut. Six or half a dozen.

    • Timothy G. Yan

      Hmmm….the use of the Degtyarov flap lock is what make the AA Ulfberht so shootable w/o using a very annoying muzzle brake. It recoils like an AR-10.

      • MPWS

        Oh really! So as it says, there is no substitute for personal experience.
        in any case, as I see it, it is traditional hard lock. Perhaps it is a tad on heavy side. No need for brake is definitely a plus.

        • iksnilol

          Then you add a suppressor and you have something downright enjoyable to shoot (or as enjoyable as a .338 Lapua gets).

      • MPWS

        …hold it for second. This needs closer examination; If the lock surfaces are just a small amount over square that might add to gradual giving just before puff comes. The gas assist might be needed to give recoiling parts speed. This is just a speculation.

        • Timothy G. Yan

          The flaps transfer the recoil impulse to the steel receiver. It’s the compression of the steel that handles the recoil. The flap size doesn’t matter. The longer the receiver, the more steel it has to work with. The Dushka HMG holding up just fine with some that have been in service for more than 50 years.

          • J-

            Steel has a high modulus of elasticity. There is not enough receiver flex to absorb that much recoil. If it did, you are well into range of causing fatigue failure of the receiver. The Dushka fires an old cartridge, from 1938, best I could find was 50,000 psi. The 338 Lapua is rated at over 60,000 psi. A higher pressure cartridge is going to wear down on that system a lot faster. Especially in a shoulder fire-able rifle vs a crew served weapon.

          • Timothy G. Yan

            I could only tell you what the designer told me and how it shoot.

          • Zapp Brannigan

            I think you are misunderstanding how things work. Steel is the best material for the receiver in this rifle. Steel is very strong and it is unique in having an infinite fatigue life (if stresses are kept under roughly half of yield). Want less stress? Add more material in the appropriate areas.

          • tts

            It doesn’t actually have infinite fatigue life and its only for stresses that are moderate to small percentage of the steel strength that steel has ‘infinite fatigue life’ depending on the alloy used.

            Also adding more material is a real expensive and heavy way to add strength on guns firing powerful high pressure cartridges like .338 Lapua. There is a reason why flapper actions aren’t used much and were quickly abandoned by gun designers early on.

          • Timothy G. Yan

            No, flapper design is still being made and in use by major military around the world in their HMG and auto-canons. Rotating bolt is not the end-all-be-all. I would suggesting TFB readers to read up on I.Hogg and A. Williams’ books on auto action.

            On a technical note, the Friberg-Kjellman-Degtyarov system is extreme strong and it prevents out of battery detonation by design.

          • tts

            They’re being made and used because the tooling still exists to crank them out and sell them to poorer countries. That and they’re way over built (assuming you’re talking about the Dshk).

            There is nothing inherently impressive or strong about a flapper system and any of the others can be overbuilt similarly too if necessary. Don’t buy into or repeat the BS you’ve been told by some sales or company rep. They’re just trying to sell their product.

          • Timothy G. Yan

            No, those are new designs with new tooling and I was not talking about the Dushka. So, I am sure you know more about gun designs than Bill Alexander and on why the system work for this weapon. Plus you know more about how it suppose to shoot (have you actually shot one?). Good for you! I’m sure you know more about weapon action that Hogg too. Double good for you!

          • tts

            What new designs? And you were conflating this guns durability with the Dshk which was the only other flapper design you’d brought up before. One is a back breakingly massive HMG and the other is a semi auto sporting rifle, you can’t compare them just because they both use a flapper action.

            And I never claimed to know more than any gun designer or how it shoots but I was essentially repeating the common knowledge of the flapper action and continuous recoil actions’ effects on recoil, none of which you’ve factually disputed nor are they controversial.

            Fact is they got some very expensive guns to sell and it wouldn’t be the first time a smallish manufacturer or a salesmen embellished things a bit to make some sales.

          • Timothy G. Yan

            You have no clue and I was even nice enough to tell you to look up Hogg’s and Williams’ books on gun action. Bye.

          • J-

            I’m pretty sure that I understand how things work. You are right that steel is strong and that it does have an endurance limit.

            But the point that I was making is to Yan was that flexion of the receiver is not responsible for soaking up recoil. More importantly, this design, to be strong, requires a massive, heavy, bulky receiver. Which is that happens when you add material. If the point is to produce a man portable weapon, you don’t want it to weigh a ton. A long bolt that locks at the rear against the receiver is an inefficient use of materials. There are better designs that are stronger that use less materials because they don’t stress the receiver like this.

          • tts

            They’re fine cuz’ its a massively overbuilt 75lbs machine gun/light auto cannon with a moderate rate of fire.

            The Ulfberht on the other hand weighs about 20lbs and while semi auto only is firing a cartridge with a higher pressure.

            I doubt the flapper action or the flexing of the receiver has much of anything to do with reducing the recoil in of itself. That probably has more to do with the continuous recoil design of the bolt carrier more than anything else.

            They’re known for reducing recoil on different actions at the cost of rate of fire and a longer receiver. For a semi auto weapon priced firmly as a rich people’s range toy rate of fire doesn’t matter too much though so its a good design choice especially for such a powerful cartridge.

        • bull

          it probably has more to do with the amount of force needed to unlock the action… and if the bolt slams into the reciever.

      • Dan

        From the videos I seen of people shooting the Ulberht it doesn’t appear to beat them up.

  • anon

    good god, that’s quite a piston

    • notalima

      Ran out of ammo? You can use your piston/carrier assembly as a club!

  • Zapp Brannigan

    If the goal is to make an accurate semi-auto rifle, especially if that rifle isn’t going to be shooting a lot of rounds at a sitting, the best design to emulate is the AR with its direct impingement. It has the added bonus of being a design that is familiar to many.

    • Vitor Roma

      There is this crazy myth that DI is more accurate. As long the barrel is high quality, free floating and the bolt locks well, the gun will be accurate be it shorts stroke, long stroke, DI or delayed blowback.

      • Zapp Brannigan

        One can make any semi-auto rifle into a tackdriver, just that it’s a bit easier (aka cheaper) to do that with the AR’s direct impingement than other designs.

        • Giolli Joker

          With roller delayed blowback you get full free floating barrel…

    • Bill Alexander

      Actually the AR makes a very poor starting point if one wishes to create a semi automatic weapon in 338 Lapua caliber.

      We initially considered this approach as it is an obvious one and even went as far as testing several designs based upon this concept. The results clearly indicated that the configuration was less than optimum. Familiarity in this instance is a disadvantage

      With the 338 one has to contain the pressure, support the cantilever of the barrel, constrain the recoil of the cartridge as it pertains to leverage on the barrel and match the weapon timing to unlock the breach. To be effective as a semi automatic weapon the system requires stability such that the shooter can effectively observe the fall of shot and at minimum not loose target designation. The heat management of such a system with the requirement to shoot multiple rounds is the last requirement.

      So a solid barrel foundation is an absolute given. With a reasonably heavy profile 27″ barrel to deal with heat build up, it is problematic to mate with a light alloy receiver and provide contact stresses that are low enough for a minimum 5000 round duty cycle. This is not to say impossible but the size of the trunnion or barrel extension becomes excessive, or the system runs the risk of shooting loose. Working pressures also drive trunnion size. So the expedient route is to fit the barrel directly to a steel receiver suitable to absorb the recoil, heat, and contact stresses. It then makes more sense to lock the breach to the receiver directly and not incorporate a superfluous component.

      To time the weapon correctly the extraction cycle must be set to allow the chamber pressure to fall to very low levels before unlock. This dictates that the gas is drawn further down the barrel and then acts upon a large area piston as there is less pressure and duration to provide the necessary energy for cycling. Benefits are that the system is extremely accurate as the mechanism movement has minimal effect on the firing event and the support for reciprocation is now 22″ long. Clearances between parts may be relaxed without incurring detrimental angular deviations. A cyclic rate of 450/500 rpm is typical.

      Removing the rotating bolt head with locking lugs to disengage the barrel extension saves nearly 1″ of overall length and 2″ of movement that must be in addition to the length of the cartridge. The savings translate to weight savings and allow the movement to be substantial behind the magazine. The system can now be set as a constant recoil as a variation in the bolt position at the rearmost part of any cycle is well behind the next cartridge from the magazine.

      Admittedly we were nervous of such a mechanical deviation from the norm and invested considerably before making these steps. Even so the potential still existed for the rifles to shoot badly or for the variation between weapons to be unacceptable in a production environment. This required that the trial fleet be a larger sample set than normal and that the initial production be closely monitored. We expended somewhere in the order of $175k in ammunition alone before the design was proven.

      The rifle has shown a very good tolerance to a production route without introducing stringent tolerance requirements, or unit to unit variations. The design targets for accuracy and durability are exceeded. The typical production unit will report 1/2 moa or better with Black Hills ammunition.

      PIP formed part of the weapon design process as did life cycle analysis and FMEA. We have ongoing projects to examine both weight and accuracy of the system as well as operating enhancements to reduce any cold bore effects.

      • Vitor Roma

        Thanks for the amazing post. A bit late but hope people will catch all this fascinating info. The piston/rod is very long, where in the barrel is the gas port located? 14″, 15″ mark?

        • Bill Alexander

          19″ from the breach . The piston is essentially the carrier with a 22″ bearing surface. What one may consider the carrier typically, is about the size of a Snickers bar and simply provides a location for the cam tracks. and the second bearing surface for the mechanism.

          I wonder if people will review the information or simply dismiss it as a sales pitch.

      • Giolli Joker

        This comment deserves much more visibility!!!