L96A1 Behind the Scenes: Manufacturing Catastrophes and Exploding Rifles

Accuracy International’s L96A1 was a stunning success in British military trials, and became the basis for one of the most respected line of precision rifles in the world. However, it very nearly was abandoned almost as soon as the first rifles were delivered to the British military in 1986. Production had been subcontracted to a firm called Pylon Industries, a respected high-tech military equipment manufacturer. Pylon bungled the production so badly that the rifle was nearly recalled when improper materials use led to broken firing pins and out-of-battery detonations. One Royal Marine was seriously injured, and serious steps were taken to bring back the old L42 snipers…

Want to know more about the L96A1? You can get a copy of Steve Houghton’s book “L96A1: The Green Meanie” from Swift & Bold Publishing here:


    • Sorry for the necromancy involved but re-watching the above video I had a thought:

      Makes you wonder about the quality, reliability and safety of the *MISSILES* that Pylon was already contracted to make, given their apparent ignorance of fundamental materials science issues and lackadaisical adherence to specs…

  1. What a mess! That company must not have had any previous firearms’ manufacturing. experience. Just incredible.

    • You would be shocked at the hubris you’ll run into with a lot of people in any sort of manufacturing industry that think “Well, I know how to do “A”, how hard could “B” or “C” be…?”. Followed promptly by utter and catastrophic failure at “B” and “C”.

      Every industry and trade has its tribal knowledge, those things that the actual practitioners have built up over decades (sometimes actual centuries…) of experience. You don’t pick that stuff up as an outsider coming in laterally–At least, not easily. If you go back and look at a lot of case histories of failure in firearms manufacture, what you’re going to find will often be cases like this, where someone who literally knows nothing about building guns comes in and thinks that because they know how to build something else, making guns will be a walk in the park. It. Ain’t.

      It’s not only guns. I know of two cases in residential construction where people that knew what they were doing in one specialized area of that field completely screwed up when they moved into trying to do other things, or becoming a general contractor vice being a trade or specialty contractor. Both times resulted in catastrophe of truly epic proportions, the kind of thing where other contractors brought in to fix the issues were calling their friends and going “Man, I ain’t nevah seen dis befoah… Y’all gotta come down and take a look at this BS…”.

      Which, as we all know, is always a Really Bad Sign(TM). If you’re ever in a situation or crisis, and you note that the Very Experienced Warrant Officer you asked for “help and advice” is making phone calls to all his friends to have them come take a look at what is going on…? You know you’re screwed. That’s in a purely military context, but believe me, it happens everywhere, even in civilian life. It’s kinda like noticing that your mechanic is calling other mechanics to look at your engine, then pointing and laughing while looking at it, and then looking back at where you’re waiting with a serious expression and a furrowed brow. Again, at that point, you should know you’re screwed.

      • Yes, Kirk, I agree. It ain’t indeed! Thanks for your answer. I am aware of many similiar cases in the civilian field. Things one would expect to be easily mastered by non-skiled labour and yet, the results are terrible… Like a certain company active in residential construction whose bosses decided to subcontract a few blocks to a certain contractor specialized in doing road work. And yes, a few small concrete brideges. The result? Utter failure.

  2. Lots of pain (some literal) along the way, but also a great tribute to the ingenuity and determination of AI – who would probably never have gotten a chance under our “all or nothing” procurement system. The main upside of the latter seems to be producing a lot of cool fodder for this site!

  3. Hi Ian.

    Just a heads up – the thumbnail I see on YouTube shows the L96A2, not the A1 – I know, I know

    Loved the story in the original A1 video, but great to hear the full version. Hard to believe engineers were arrogant!

  4. A pair of academics actually got the whole

    “Fools rush in where angels fear to tread”

    Phenomenon named after them:


    There are also aphorisms from JFK and Donald Rumsfeldt, about there are things we don’t know that we don’t know, and believing in things that ain’t so.

    The stories of the “Walker” trigger and the “common fire control” in post wwii Remington guns are good examples.

    During and after WWII, DuPont parachuted production engineers from General Motors, into Remington. These were actually very good engineers, who did some very good work, for example “Mike” Walker’s development of button rifling.
    Walker was also a keen early competitor in benchrest shooting.

    Jack Belk’s “unsafe by design” covers the stories very well, from the several points of view, including forensic analysis of accidents with the guns. Lots of accidents.

    Even Walker didn’t have the long background in gun making to spot the dangers in making a trigger with a connector piece that could be misplaced due to dirt in the housing, and where there is no way of knowing or resetting the position of the connector piece.

    Similar things can be said of the common fire control, which has a safety that simply disconnects the trigger, it does not actually lock the sear, or block or retract the hammer. Both the parts and the working of the safety, bear an uncanny resemblance to a 1940s era GM car door catch mechanism

    Also the Nylon 66
    There’s a good reason why guns before the 66, avoided tension springs, and had the action in something more rigid than a piece of plastic.

    To give walker his due, when he did discover the problems with his trigger design, he did alert senior management with a memo, however the bean counters, who again didn’t have a deep understanding of guns, decided that changing the trigger would cost too much. IIRC, about 14 cents per rifle in the money of those days.

  5. Is it really any big surprise having a gov.contractor putting its bean counters to work finding cheaper methods and materials to boost profit margins? I wonder if there was accountability for the injured marine?

  6. Paul Mauser had developed the cam which retracts the firin pin tip when a bolt action is opened, as one of his modifications to the Dryse. This partially cured the problems of firing during bolt closure, that had gained the bolt action a reputation in some military circles, as being intrinsically unsafe.

    For example in Britain, following a serious injury due to a firing on feed in a “Bacon” bolt action rifle during military trials.

    Early experience with the Mauser 98, and a bad piece of drill, that called for dry firing,

    Resulted in the firing pins failing at the cut outs that form the clearances for the lugged connection with the cooking piece.

    The lugged cocking piece had replaced the screw thread attachment used in the Mauser 1889, that had been possible for ottoman conscripts to mis assemble.

    Failure of the lugged joint could result in out of battery firing.

    To correct this, two little lugs were formed on the front of the mainspring flange

    These interacted with two recesses cut in the bolt through one of the gas vent holes, so that the firing pin could only reach the primer if it was properly aligned by the bolt being locked.

    The 1915 modifications of the 98 designs both by Federle at Oberndorf and by steyr, to incorporate an arisaka style firing pin, that didn’t have a separate cocking piece, would have ended the need for the two little lugs.

    It is a great irony that having done so much to prevent out of battery firing, Paul Mauser, later lost an eye due to an out of battery firing of one of his semi auto prototypes.

  7. If you give me a choice between an extensively developed and iteratively redesigned weapon vice something that is the latest and greatest hotness out of the minds of the nation’s most brilliant engineers…? I’m taking that old POS, every damn time.

    When you get down to it, the Mauser 1898 really isn’t the Model 1898: What it actually is would be the model of 1889A9, or however many iterative changes there were between the one and the final product. Mauser knew what he was about, and instead of rushing out an entirely new model every time he found (or, more accurately, some lackwit Private out on the line…) a new problem or issue. He designed iteratively, fixing each new problem as it came up, refining his design until it became the definitive article for that class of rifle. People still use the Mauser 98 as a benchmark for excellence in bolt-action rifle design.

    What does all that tell you? Making guns ain’t all that easy, and if you go into it laterally? You best have some resources behind you. The only case I can think of where an outsider came into the field and didn’t meet with disaster was (maybe?) TRW with the M14 program.

  8. Excellent summary of the mess surrounding development of the L96A1! It put me in mind of the $125 million NASA Mars orbiter that crashed into Mars because the agencies involved did not convert from English to metric.

  9. One would expect English manufacturing industry to be versed in Metric system (this is more than linear units of distance, it is also threads and fitting system incl. standard gauges), after all they are part of European economy. True, this is some 35 years ago and that time it may have been different.

    My impression when reading automotive specs from Britain is that the metrics are well ingrained there now. They better be.

    • It actually should have come right from original designers, true to their name with word “international” in it. They should not mess with inches to begin with. It is old, obsolete system of measurement. I happen to be familiar with both.

      • Metric or SAE/Imperial/Standard… Whatever you use, it’s gonna be entirely arbitrary. And, unless it’s binary, the advantages and disadvantages of either system are still going to be there. With the decimal-dependency of Metric, it’s way too easy to misplace that decimal point and screw up by orders of magnitude. Not so easy with fractional systems.

        I’m an agnostic, at this point. The metric enthusiasts have had it their way for a long, long time, but the reality is that any system of measurement is essentially going to be capricious and arbitrary. It’s all what you know, and what you can work with instinctively. With the (admittedly, inconsistently…) base 12 SAE system, it’s a hell of a lot easier to naturally fractionate things. With metric? LOL… Yeah, you got halves and quarters of things. You can’t easily do thirds, at all.

        Every time someone goes onandonandonandon about the supposed superiority of the metric system, I just want to tell them to shut up: The majority of the “superiority” is only in their minds, and that’s there mostly because they’ve been propagandized endlessly by all the numpties who’ve fallen in love with the system.

        Ever notice how the most commonly used metric sizes of things are all basically rounded-off conversions from the old common-use measurements? There’s a reason for that–A pint of beer is about the portion size you can drink comfortably and manage easily, sooooo… There was a reason it was the “customary size”.

        Someone wants to do up a binary system predicated on something actually universal, like, say, the average orbit of an electron around an average hydrogen atom…? I might buy into that being the preferred and gold-standard system we should all adopt. So long as it’s this entirely arbitrary and nebulous concept that all our systems are based on, one line of idiocy is as good as another–So long as it’s all standardized, and we all agree on what it is.

          • The “system” is far too easily screwed up by the unknowing; that “ease” you talk about enables all too much confusion: Is he talking about distance, volume, or mass?

            With the “old system”, there’s no possibility of confusion. Feet, yards, miles? Distance. Volume? Quarts, gallons, pints. Mass? Yeah, that’s where the customary system breaks down.

            Distance = meter; volume = liter; mass = gram (and multiples). No possibility of confusion.
            MKS, or SI, is a choerent system of unit. In SI the unit of power, one Watt, is one Joule for one second. In the US customary system, that’s not coherent, the unit of power, the horsepower, is 550 foot-pounds for one second. Units are always derived from one-other, but you have to add a conversion value to obtain any of them and THAT generates confusion.

            However, that’s a dead topic. SI won decades ago. No industrial design had been designed in imperial anywere for decades.

        • The term “metric” which refers merely to unit of length is a part of common world recognized units system known as “mksA”.

          The “m” stands for meter, “k” for kilogram, “s” for second and “A” for Ampere. The first three combined give you unit for Force = kg x m x sec*2

          The units still somehow surviving in anglosphere are odd at best, in fact they are outright medieval (e.g. “country Mile), relegated to distant past/ history. I cannot fathom how anyone with sense of logic and order can bother with it. I was shocked to disbelief when I heard of BTUs. What are knots for? To tie and untie them as you move thru the air? To use them for airspeed is absolutely ridiculous.

          I give examples of Metric logic:

          – 1 liter (1000cc) of water at 22.5 deg. weights C exactly 1 kilogram,
          – amount of heat required to change temperature of same volume of water by one degree C represents 1 calorie,
          – 1 kilogram of force equivalent acting at 1 cm square area is 1 atmosphere. This happen to be a normal atmospheric pressure at the sea level.

          Based on this and other “metric” trivia you can calculate quantities without even having pen and paper.

          • So?

            The “system” is far too easily screwed up by the unknowing; that “ease” you talk about enables all too much confusion: Is he talking about distance, volume, or mass?

            With the “old system”, there’s no possibility of confusion. Feet, yards, miles? Distance. Volume? Quarts, gallons, pints. Mass? Yeah, that’s where the customary system breaks down.

            Point that I’m getting at is that no matter how you cut it, either system has its advantages and disadvantages. The customary system is chaotic and fundamentally illogical, but it does have the signal advantage of making you think and you have a hell of a time making a mistake between measurement categories because the units are so dissimilar and totally unrelated.

            I’ve seen horrendous mistakes made in metric, because the people who were doing the calculations got just a little bit too damn confident in where their decimal points were, and because they got fooled into thinking that they were working with one sort of unit when they were really working in another.

            I think the whole thing is a wash; the “logic” of metric is overcome by its disadvantages, and the “illogic” of the customary system leads to some error-correction features by sheer accident. It’s way too easy to make a mistake with the decimal point or in units with metric, and way too confusing to work in the customary system unless you really know what you’re doing. Either way, it’s all on the user. I don’t think there’s any one “superior method”, but I would be willing to agree that each has its superior points and negative ones, at the same time.

          • My base degree was a applied science one, through the two universities in which I studied, it always stressed that Metric was a useful scientific system, but flawed in practical use. Metrication was forced upon Australia in 1973 without any community consultation, and 43 years later the average punter in the street do not have a clue who to deal with it. You talk about Kilopascals a veil goes over peoples eyes, when asked about a Pascal, totally blank. Currently completing a major rebuild and extension of the family home, I am constantly amazed that the building trades do not use centimetres – just millimetres, and that the tradesmen cannot visualise measurement. We will not discuss Gigalitres, mass and volume is met with equally blank response, as for Kelvins, Moles and Candelas just forget about them even the French cannot comprehend these in daily usage! At sea the merchant navy still use nautical miles/knots, while the daily use of volume/mass does not fill you with confidence – last week the authoritative US web outlet gCaptain (for the merchant marine) had a article on the woes of ‘box boats’, with the constant slippage of containers at sea due to such resulting in incorrect loading of the boxes contents.

        • ‘Someone wants to do up a binary system predicated on something actually universal, like, say, the average orbit of an electron around an average hydrogen atom…?’

          Yes this had been done for modernized definition of 1 meter https://en.wikipedia.org/wiki/Metre

          The original 19 century definition was as 1/40,000,000 part of Earth equatorial circumference. The prototype is stored in Natural history museum in Sevres near Paris. There was a 75km error in it.

          • “Feet, yards, miles? Distance. Volume? Quarts, gallons, pints. Mass? Yeah, that’s where the customary system breaks down.’

            Oh man…..disaster on Earth. Oh please, G-d with us and Evil stay AWAY! 🙂

      • There’s not much difference between metric and SAE/Imperial/Standard/English measurements for things that fit in a laboratory. Outside a laboratory, the old measurements are just as good or better. Ships and airplanes use knots and nautical miles for measuring speed and distance because knots and nautical miles match up to latitude and longitude better than kilometers.
        Even old measures like the Russian “arshin” are better then the metric system, if you are used to stepping off distances.

        • I think that may be the one true flaw of the metric system… The arbitrary and “not-quite-right” nature of the standards they chose. The meter, for example? I think they’d have been better off going with something everyone could easily visualize, and looked to make it closer to the average person’s pace.

          The rest of it is just unnatural–The old measurements like acre and so forth were based on things like how much land a guy with a plow could plow in a day. Start talking to someone about “hectares” and the next thing you know, they’re talking to themselves and counting on fingers. Friend of mine grew up on a farm in Canada, and you don’t want to get him started on all the fallout from metrification. He had one deal where they screwed up the calculations of how much fertilizer/pesticide to apply to a field, and because they’d gotten the decimal point wrong in the calculations, they wound up killing an entire field by over-applying the fertilizer. In another case, same issue, they under-applied the pesticide, resulting in a huge loss to insects. Took years for his family and others around them to get used to the whole thing, and they’re still not doing things entirely in metric.

          There are a lot of nutty things encapsulated in the old system, things you’d never consider–But, which being the result of centuries of experience, make total sense once you get into them.

          I used to be a total metric enthusiast. Then, some experience and talking to a guy who laid out the logic of the old customary system for me in a way they never bothered to teach it in school led me to my current state of ambivalent feelings on the issue.

          • You are not talking of what’s natural and unnatural but of what you are used to or not. In metric countries, meters, kilometers, litres and hectares are what people visualize, while acres and gallons are completely unknown.
            BTW the meter is exactly “closer to the average person’s pace”. In imperial what’s “close to the average person’s pace” is the yard, that’s 0.9144 meters. Infact meters are usually measured in steps if an instrument is not available. Funnily the imperial “mile” (from “mille passum”, one thousands steps in latin) is not 1000 yards long, but 1,760 yards, so, again, it’s the kilometer (1000m) that’s a real “mile”.

          • There may be a few NBA giants whose steps are a meter long, but not classical-period Italians.

            A Roman mile was a thousand paces (pairs of steps), not steps. 5.28ft is much closer to a pair of normal (especially premodern) human steps than 2m.

          • What a “classical period Italian” has to do with the meter?
            A meter is as close to an average modern person step than a yard, and surely is closer than a yard was when the yard had been estabilished as a measure (do we want to talk about the “foot”?). And still, while a km is a thousand meters, and so a thousands steps, a mile is not a thousand nor two thousands yards, is just another weird measure unrelated with anything and incoherent with the others.

          • Classical-period Italians established the Roman imperial mile – which you brought up (using an incorrect definition of “pace”), remember?

            Not only did I not try to establish a linkage between Romans and meters (or yards, for that matter), but my point was that such a linkage makes no sense.

    • The European standard for pipe fittings is actually the old British Standard Pipe, with inch demensions converted to read in SI and 55° Whitworth V threads.

      At the other end, BA (British Association) threads for scientific instruments, originally came out of Switzerland and are SI dimensions converted to read in inches.

      • Fitting are subject to National standards with every nation practicing their own, although those in Central Europe are nearly coincident. Use Hungarian or Polish fittings in Austria are they will – fit.

        In country I come from the MAJOR break was time of German occupation. There were strictly imposed common standards for entire slate of industry. That allowed to build vehicles, planes or even submarines from components/ assemblies elsewhere and then complete them into units at final place of assembly. Convincing enough?

      • The Whitworth thread was known to us, but it was allowed new only for purpose of repairs. Btw., we had plethora of British cars on market in 1960-70s. I have no idea how repair were done though. With every bolt to be purchased from dealer, such a car probably came to a decent cost in long term. But, Ford Cortina for one enjoyed good reputation; they were reliable and durable.

    • In 1985 none submitted blueprints in imperial any more. Industry had switched to metric for decades. That had probably been part of the problem. It was incredibly backward coming form a couple of young engineers.

  10. Just a story: the great architect, Julia Morgan, was inspecting construction on some brickwork she’d designed. The contractor was ignoring her specs and shrugged off her objections. She pulled the building down with her bare hands, right on their heads.

  11. “Three blokes in a shed” is the genesis of much of the eccentric genius in the world. Witness TVR/Griffith, the Arial Atom, and the Wright Brothers (technically two blokes and a girl in a shed).

  12. Ian’s recent video on the transition from prototypes to production touched on this. Making guns is hard.

    Rock-Ola tried to make M1 carbines. They were awful.

    I have little affection for the Ingram M10, but the Powder Springs guns were actual functioning firearms, whereas some of the later stuff (RPB, Cobray etc etc) were junk.

    Although the AR180 was never truly perfected, the Costa Mesa and – especially?- Howa guns were well made. The Sterling ones weren’t. Despite Howa being basically a sporting arms maker, and Sterling a military one.

    IIRC, Springfield and TRW M14s were good, Winchester ones mostly OK, and H&R ones were initially rubbish and mechanically unsafe, despite H&R being a long-established gun manufacturer.

    And, of course, Colt, arguably the most famous name in American firearms, made a dog’s dinner of the early M16. Much was the fault of Ordnance, but Colt didn’t exactly shine either.

    • Japanese do not make shortcuts; they stick to the TDP all the way. But, they have one issue, the NEVER admit a mistake/ shortcoming. I heard this from several sources from which one was a Chinese man who knew them thru-out. This is something they should try to cut out – thru “personal attitude Kaizen”. It would boost their reputation even more; but then they would not be Japanese 🙂

      • Japanese does a lot of shortcuts in manufacturing. One of them famously led to the Tokaimura nuclear accident. A friend of mine that works in farmaceutic certifications says that’s it’s almost impossible to make them comprehend that a certain way of doing things is ineherently dangerous, because you are only a mistake away to make a disaster. They firmly believe that, since none will make mistakes, all is going to work.

        • A friend who worked with Japanese colleagues told me the same things. They always think a given effort will work out well because they firmly believe no one will make mistakes. They expect everyone will just do his/her part of the job well, without any mistakes. If these happen, as Denny remarked, they will never admit to them.

  13. I received my copy of this superb book last week from the UK, it is a fitting extension to the authors; HOUGHTON Steve. The British Sniper. A century of Evolution. Swift and Bold Publishing, Suffolk, England, 2018. Hard cover, dustjacket, 319p., photographs, drawings.

    You just cannot fault this book, and does the very effective weapons system justice, with a competent and authoritative text, the illustrations well reproduced, and they tell half of the story.

    Of course when you buy anything of quality its expensive, but, you get your monies worth and probably more.

    A recent text by a former sniper sergeant in the British Army who toured (multiple times) in Iraq and AFGHAN before being appallingly wounded, BAILEY Dean. Crawling out of Hell. The true story of a British Sniper’s greatest battle. Fonthill, UK, 2016. Illustrated card cover, 176p., photographs. (The greatest battle being his recovery and return to life from his crippling wounds). He stresses that in his opinion that the L96 consistently was the main killer on the battlefields, and that it provided 100% reliability on them and that it had the total trust of its users (interestingly he saying the same of the revamped L82A2, with or without scope).

    If you only buy one book this year, this is the one.

    In regard to the comments on Pylon Industries, this company was the beneficiary of a hostile takeover by a USA holding company. Asset stripping resulted in it also lost the bulk of its previous professional and technical production staff. The parent company of this takeover, also took over the ROF (Royal Ordnance Factories) doing similar “good” work in taking the maximum from the UK taxpayer (and it not just the L85A1 that they stuffed up, but such as the ammunition for the 51mm Mortar, fusing for the 81mm mortar and the 105mm Light Gun, amongst others.)

  14. I do not indulge squaring off in lowly subjects, but for purpose of diversion one anecdote. While visiting in the U.S. South long time ago I ran into a young man who spotted my VW (not very good car and I got soon after rid of it) and promptly saluted me.
    He turned out to be a German recently settled in the country. We exchanged couple of words and after he found I was from Canada he looked at me with discernable ENVY and said: “you are METRIC there right?” I admitted the obvious and felt like 100yards dash winner. No baby, no gallons for me, please 🙂

  15. My grandfather has his own system of measurements – a smidge, a wee bit, loads of, buckets of rain, tons of work, knots of laughing, that’ll do, a big mess, shocking funny etc. But then he’s a glass half full sorta fellah.

  16. Badly done subcontracting, where have I heard THAT before? Oh, wait, that describes the first batch of Browning M1918’s from Winchester (so badly dimensioned that they couldn’t interchange magazines), the Brewster F3A (a license-produced Vought F4U, but the Brewster version had a reputation for literally FALLING APART MIDFLIGHT), and M9 magazines made by cheap-as-dirt third party factories who DIDN’T EVEN FIELD-TEST THEIR MAGAZINES IN THE M9 IN THE FIRST PLACE. I could be wrong.

  17. I remember being in Canada when they brought in metric.All we did was convert our inches to metric as in 1/2 inch was 12.7mms. We called them metric inches. A side note here in France all the houses before 1914 were built in french inches slightly longer then the english one 2.54 cms and all our water fittings are whitworth gas threads given french metric designations

    • Might buy some of those, quite interested to see about blisters. I assume Romans were harder, unless its socks per se even with that ww1 stitch (Kitchener) salute peasant.

Leave a Reply

Your email address will not be published.