Israeli Dror LMG: Shooting & Disassembly (Video)

The Dror was an Israeli copy of the Johnson LMG, which was made domestically in Israel as part of Israel’s war of independence in the late 1940s. The design was developed through a combination of reverse engineering a Johnson LMG (clandestinely, in a New York hotel room) and technical data leaked to the Haganah by sympathetic employees of the Johnson company. The first version was chambered in .303 British, but this was quickly replaced by an 8mm Mauser version (ammo Israel had significant stocks of at the time) using modified BAR magazines.

Unfortunately, the Dror turned out to be rather unsuccessful in testing. Hopes were very high for it, and it was put into official production before field testing was completed – only to find that it failed many tests pretty badly. Ultimately it was used for training, but saw little or no actual combat use – the Bren and MG-34 were much more successful guns.

This particular Dror is a semiauto rebuild provided by my friend Mark (thanks, Mark!).


  1. “the Dror turned out to be rather unsuccessful in testing”
    Now, I am wondering which was flawed: reverse-engineering XOR conversion-to-other-cartridge XOR manufacturing process?
    I know some example, even when legal conversion of fire-arms to other cartridge failed miserable: French Chauchat variant for .30-06 cartridge and Hotchkiss machine variant for 7.92×57 Mauser used by Polish Army (as Hotchkiss wz. 25)

  2. Hey Ian,

    As a bona-fide Israeli and a gun-nut I have to give you a lot of credit for getting the terminology and Hebrew down, especially when it comes to history.

    Kol Ha-kavode (kudos) to you and Merry Christmas from your friend in Prescott Valley!

    Steve M.

  3. From all I can see, the original Stoner Armalite rifle design essentially was a Dror with the addition of the direct gas-impingement system of the Ljungman AG-42 rifle. Plus of course uniting the magazine well with the lower receiver and permanently anchoring the barrel in the upper.

    I suspect one of the reliability issues was the magazine interface, very like that of the earlier FG-42 or later AK-47 and M-14. I was watching the catch “bounce” and the BAR magazine “wobble” forward and back in recoil. That would be a likely cause of feeding problems.

    Retaining the original BAR magazine catch, etc., would have been a wiser course. Note that the Armalite does essentially that. The StG-44 type catch would also have been a bit “steadier”.

    The thing is, if you left off the full-auto setting and the iron sights, and put a scope on it, it seems to me that the Dror would have made a good Designated Marksman Rifle (DMR). The barrel is essentially free-floated, it fires from a closed-and-locked bolt, and all movement of barrel and bolt is linear, no tilting. All of which would tend to help consistency in shot grouping, which is what a DMR needs.

    Having a semi-auto DMR in addition to a bolt-action scoped rifle for the sharpshooter team is a modern concept. As Plaster points out, the semi-auto DMR gives the team a useful anti-materiel/anti-squad capability on top of the one-shot, one-kill capability of the accurized bolt-action.

    In the kind of counter-terrorist warfare the Haganah was fighting both before and after the War of Independence, that would have been a very useful thing for the “night squads” to have.

    BTW, Leonard Slater’s book The Pledge devotes an entire chapter to the Dror story;

    No, nobody’s getting my copy.



    • “From all I can see, the original Stoner Armalite rifle design essentially was a Dror with the addition of the direct gas-impingement system of the Ljungman AG-42 rifle”

      If the gas key was a blind hole, you’d be right. But Stoner’s gas handling was a truly innovative improvement on the simple “gas tube acts as a fixed piston; gas cylinder is integral with the bolt carrier” of the French rifles and the Ljungmann, and eliminates all off-bore-axis forces from the system. The original patent (US2951424) is quite informative on the thinking that went into it.

      “it seems to me that the Dror would have made a good Designated Marksman Rifle (DMR). The barrel is essentially free-floated”

      Except for the fact that the barrel is actually floating rather than being rigidly mounted, and interacts with the barrel sleeve at the front bearing. Fixed barrels which are free-floated forward of the mounting thread give good accuracy, moving barrels typically do not (and particularly with useful tolerances).

      • Actually, the Walther WA 2000 sharpshooter’s rifle used fundamentally the same barrel support setup as the Dror, except for being gas-operated rather than recoil. And there ere no complaints about its accuracy- just how much it cost to build, which is why only 176 were ever made;

        Given a bit of adjustment to the tolerances, I’d say the Dror could probably shoot side-by-side with the 2000 out to 500 meters or so.

        I’d love to see what such an “accurized” one in .30-06 could do with a 10X scope and match ammunition.

        Hm. Now there’s a semi-auto repro project for you…



        • You could easily name M82 as an example of sniper weapon which has fronted moving barrel.

          People (quite a few) appear to be of the fixation that rifles with moving barrel(which this is as well) are “less accurate”. I advise them to consider that barrel being front supported has only very short portion which is free to vibrate, therefore angular deviation is lot less than with “free floating” barrel.

          In reality though, bullet is long gone when barrel gets into its usual act. Two-three milliseconds of bullet barrel time is not enough to get it vibrate to its full amplitude. To have full benefit of floating barrel it has to be thick beyond practical means.

          • It only has a moving barrel though because that’s how it’s mechanism operates, not for the purposes of achieving accuracy.

          • I am not saying it is negative to accuracy, I don’t know… Are you saying it’s potentially positive for accuracy Eon/Denny?

        • “….fronted moving barrel’.
          Was meant “front supported moving barrel”.

        • The Walthers barrel doesn’t move though, as a consequence of its gas system. I am not particularly enamored by the Dror personally, I think it’s barrel latch is ok at a push in relation to it facilitating how the barrel is required to function. Mind you, didn’t really like the Johnson concept either.

        • The WA2000 barrel is fixed at both ends, and the bar for the front support is like a structural girder.

          Frankly, with a Dror-type setup you’ve got looseness from the tolerances of the front and rear bearings, plus the flex of the barrel jacket to deal with. Not a setup lending itself to accuracy, since there’s no guarantee that the barrel will start in exactly the same position with every shot.

          And the Barrett M82 is not anywhere near as accurate as Barrett’s bolt-actions…

      • When the bolt head is in the locked position, gas via the tube hits the back of the bolt and bounces off it against the carrier pushing it back which rotates the the head via a stud that passes through it being engaged by a cammed surface in the carrier. Is that right? Although a gas piston, can push the carrier back in the tappet configuration achieving the same. So why couldn’t a gas tube that was configured like a Mas49, avoiding putting carbon around the bolt.

          • Not being, may have implications for it’s use in an Sa80. In regards it returning to battery, wonky offset’ness as oppose a tubular, tube like fit… We often have to forward assist, sometimes at most inopportune moments such as an Afghan Police officer deciding to shoot you. The direct impingement in regards gas going in behind the bolt may ensure the carrier gets a decent wack to rotate the bolt out of the cavern were the smoking ejected rounds pass through and grit gets stuck inside etc. A gas tappet seems to unlock the Sa80 fine though, mind you it has two settings. The Mas49 didn’t have a rotary bolt, never mind one of the seven lugged variety. The AR doesn’t have an adjustable gas system, perhaps because it’s really always on full which is good in regards the above but bad in regards the fouling inside the bolt sometimes causes AR’s problems given it must turn inside the carrier. The Tavor bolt head has moved away from the usual layout, then there’s the Ak’s of course- Rifles famed, for reliability.

    • The British Sa80 Lsw could be a cracking designated marksman rifle and a good LMG at once, if it was in .308 open/closed bolt arrangement for semi/full I reckon.

  4. Thank you Ian.
    Beautiful gun, pleasing to the eye.
    Seems that is drawbacks can be corrected if anyone wanted to reproduce it.
    Although, I’d take mine with selective fire and pay the tax.
    That is if the Federalistos opened it up.

  5. Sorry to be a nudnik, but I just thought of something after my posting.

    How very appropriate that you should do a video on the Dror in the week following Hanukah.

    For everyone who doesn’t know; “dror” is Hebrew for “freedom”. Hanukah celebrates the world’s first known war fought exclusively for the purpose of religious freedom, about as noble a cause as you can think of.

    What many people (including my fellow Jews) do not know is that had the Jews lost to the Syrian Greeks and Hellenized Jews; the flame of Judaism would have been snuffed out for good.

    No Jews; no Christianity, no Uzis, no Galils, no Mighty Morphin Power Rangers, no cellphones/smartphones and lots of polio to go around.

    Merry Christmas to all on this blog!

  6. Is there any accelerator included in this design?
    As far as I know, Johnson is using “pinch effect” on camming edges between carrier and bolt to unlock. With that, the major portion of primary mass momentum is retained in primary mass and that explains harsh recoil effect.

    Next question would be if MG34 is any better in that respect. My assumption would be it can, since transfer on motion is happening along longer cam track.

    • Do you mean a recoil booster like on a Maxim for example when you say an accelerator Denny- To give additional momentum to the barrels rearward travel, as oppose it just moving back as per a Remington model 8. I see the cocking handles movement against the angled surface on the ejector port as being responsible rotating the bolt,
      I don’t know if this is how the Johnson did it. If so, that perhaps explains the 7 lugged bolt head in that there’s little rotation required due to the reduced surface area between the lugs and there recesses. That Fosberry rifle had a similar arrangement, the Ak’s lugs a fewer but have a greater surface area so require more rotation as is my understanding “which might be wrong” I prefer fewer lugs myself though, I don’t like the cavern behind the recesses in front of the chamber mouth.

      • My note refers to transfer of momentum from primary to secondary mass, not to method of locking.

        AFAIK, there is not one recoil operating system, which would take advantage of moving mass momentum while reducing significantly felt recoil; be it short or long recoil.

        • The primary being – Bolt and barrel together, secondary – Bolt? The barrel returns separately, does that affect recoil positively or negatively…
          Thinking of the Ak 107 counter mass lark.

      • The Johnson had a gas assist thing nearer the chamber end, as oppose a muzzle mounted. I wonder if it could blast a bolt into the back of another bolt i.e. A Dror style rotary bolt with it’s own return spring, and a bolt behind/around that which resembles the ejector port shape- Cut out, for the bolt handle to pass through and engage with in this case. Which passes over the barrel to reach the gas assist plug on it, bolt is locked as per, upon firing gas pulls the rear bolt forward unlocking the bolt via the cocking handle being engaged it moves rearward but is impacted by the rear of the rear bolt, the bullet leaves at this point. Or something, counter recoil thing, I should buy a pen.

  7. Really cool looking weapon in its own way. It’s surprising that George Lucas didn’t copy it for a “Star Wars” Stormtrooper weapon. Seems like it falls in that category of weapons that “might have been good with just a little more development.” It does seem like the flaws, like the magazine, and maybe adding a dust cover over the ejection port, might have been fairly simple. A forward vertical hand grip would be nice, too. Rather than a light machine gun I think it really falls into what we’re calling an Infantry Automatic Rifle now. And you’re right that it would have been very handy for the sort of fighting going on at the time.

  8. The Dror is a very interesting piece; I’ve only spoken once to a collector on the Galil forum with one.

    However he said the Dror (2nd pattern) fires from the closed-bolt in semiautomatic and from the open-bolt in full auto, similar to an FG42, except hammer-fired.

    • As Seth C states, the Dror fired full-auto from an open bolt and semi-auto from a closed bolt. It was also hammer fired from both modes. Back in the early 80’s I bought many Model 2 Dror’s from Sidem International (interestingly one of the founding fund providers of Armalite and by association Eugene Stoner by the owner of Sidem, Jacques Michault).

      Sidem in the UK was a large surplus smallarms source and they had bought several hundred Dror’s (both the series 1 and 2) together with the Dror’s replacement the Israeli contract FN Model ‘D’ and it’s replacement the HB FAL. Sidem had a large inventory of ex IDF ‘kit’ including many MG34’s with the distinctive Israeli bipod, Mausers in every ‘flavor’, STEN’s of British, FN, and Israeli manufacture, as well as ‘odds & sods’ that were captured weapons.

      I converted several dozen Dror’s (both patterns) to semi-auto for the then UK, Canadian, New Zealand & Australian market (all now no longer allowed)! I have looked in my spares boxes and found several full-auto sears that were removed from guns as part of the conversion process. The .303″ guns were the most interesting but also the most crudely built but would have made an easy Johnson M1941 ‘copy’ with some work.

      As Ian mentions though, they both had the ability to mess up your face with the sharp edges around the stock cheek weld area when being fired!

      Ahhh, nostalgia (it’s not what it used to be)!

  9. The extra tube, stock… Seems a bit, odd “storage aspire” wonder if initially it was supposed to be like the mini me’s slide out one, and rotated into the position it is in now with a captive spring by a buffer as per an AR in one tube- The tubes being of that diameter also. The cocking handle would keep the bolt from dropping out, they’d be a muck hole though. Could have had a flap, operated like the Masden mag push aside feed allower thing.

  10. Thanks Mark and Ian.

    As I understand it, the functioning problems were related to light primer strikes, and difficulty in racking the gun afterwards. The slow motion photography clearly shows a separation between the bolt head and the carrier/extension. If foreign material entered the gap, in combination with the carrier mounted firing pin, firing pin protrusion would be reduced by the amount of gunk between the bolt and carrier. This also would prevent the hammer from going all the way forward, altering the angle at which the initial, static contact between the carrier and the hammer begins. This could tend to force the carrier up slightly, contributing to the binding.

  11. Guys,
    not wanting to blow my own trumpet (so to speak) BUT I am possibly the only person here that has shot the Dror in full-auto ‘open bolt’ configuration (albeit 30+ years ago). I used to test all the Dror’s (and other guns we converted and sold) BEFORE they were converted to semi-auto ONLY. The Dror had a peculiar firing ‘train’ so to speak… Picture this sequence of events:

    1. From an open bolt (FA mode), pull the trigger…
    2. Bolt moves forward into battery stripping a round from the mag as it goes (crappy BAR mag may cause stoppage here as the follower plate support/guide legs are too short)!
    3. Round chambers and the bolts rotates into the locked position… closely followed by the rear most bolt extension.
    4. The hammer trips, strikes the firing pin and FLASH, BANG, SMELL!
    5. Bullet exits barrel, recoil moves the barrel (and by extension the bolt) rearward.
    6. Bolt cam unlocks the bolt, the barrel stops moving rearward.
    7. Bolt continues rearward, extracts empty case.
    8. Empty case is ejected by pivoting ejector (forcibly as per Johnson design).
    9. Bolt continues rearward a short distance then is returned forward.
    10. If trigger is released, bolt is caught to the rear and firing ceases.
    11. If trigger is still pulled, repeat sequences 2-9 until 10 or magazine is empty (or stoppage occurs).

    All the guns I fired were cleaned meticulously before firing and functioned fine (unless parts were broken or missing) in the clean environment of an outdoor range in Northumberland in the North of England 😉

    The gun had a peculiar ‘stutter’ when fired due in some part to the open bolt, hammer fired nature of it’s mechanism. I find it amusing that the rate of fire is often quoted at 250-900 rounds per minute when it did not have an adjustable rate of fire! This must be down to variable ammunition types, manufacturing variables, and fouling due to powder residue and crap entering the action….

    The FN Model ‘D’ that replaced it was a much nicer gun 🙂



      • Maybe to change caliber to a smaller one, for example 5.56mm? Seems to me that the whole short recoil action is weak for a round used

        • I think that 5.56 mm was way ahead of the technology and time frame we are discussing but I know where you are coming from. My best bet would have been to try and make it belt fed but I think the energy available was not sufficient…

          • I was meaning in this timeframe (Ian also uses rebuilt semi),
            maybe downgrading the caliber would prove that this is not so shitty gun and concept as it was apparently when using original cartridges.
            This is a crude example, but; blowback in 7.92 VG1-5 – holy shit,almost too much;
            blowback used in pistol caliber – OK.

  12. I just wonder if there any detailed information on Japanese Type 62 NTK machine gun??
    Most interesting part is disassembly and parts description how they looks like,most unusual about it is feeding mechanism,it is most shortest compare with other machine guns…

  13. Thanks for this interesting video, I had heard of the Dror, but thought it was just a copy of the Johnson LMG.

    The actual gun reminds me more of the Swiss Stgw 57 (not as well made obviously), not so much a light machine gun as a heavy selective fire rifle with bipod, well suited to accurate longe range rifle fire or short range automatic defensive fire. If the Dror had been properly debugged prior to production it might have ended up as a very decent, if heavy, rifle rather than a failed light machine gun. As an LMG I think it was writing cheques it couldn’t cash.

Leave a Reply

Your email address will not be published.