M1337
Cal. 5.56 / 300 BLK
The M1337 is a DIY Gatling Gun project, that utilizes commonly available AR15 components to achieve lockup, within a printed polymer housing that facilitates the function as a manually operated, rotary barrel firearm.
This results in what is possibly the lightest weight Gatling-derived firearm ever made in a true rifle caliber. Additionally, due to the reliance on AR15 components, calibers beyond 5.56 are possible, such as 300 BLK, which can even use the same belt links, and gives the opportunity for effective suppression.
Gatling guns have historically been expensive products, even in the current day. This project allows anyone with a desire to engage in an intermediate-difficulty project, to build what is certainly the most affordable Gatling gun ever made.
Development Overview
During the development of the RBC-9 the thought occurred to me, that surely it should be possible to make a rotary housing for AR15 Barrels and Bolt Carriers, resulting in essentially a gatling-type firearm that would actually be affordable to the average person. 3D Printing, and more so the circa 2023 advancements in hobbyist 3D printers that were generally available to consumers, allowed for this type of project to be developed in a way that wouldn’t take inconvenient levels of time and effort to build. I’m certain that I was not the first or only person to independently have this idea, but with the RBC-9 beta released in March of 2023, I had the available time and recourses to begin working on the design.
Initially starting from the same point as the RBC-9, that being the belt links. Determining the spacing and proper tension of the belt links, was the key foundation to design the rest of the firearm around. Generally, the M1337 disintegrating belt links are more loose-fitting than the usual steel M27 links used by the M249 and other weapons. Additionally, the spacing between rounds is wider, to prevent the links from being overly fragile
With the belt links now dialed in, the De-Linker mechanism was next to be designed. Prior to beginning this project, I did not have a very good understanding of how the M134, or similar rotary-barrel guns functioned, especially in the way which the belts of ammunition were distributed to the barrels throughout the gun. The Delinker, to me, was particularly a black-box so to speak. Primarily, the reference material I gleaned most information from on this matter, was from the World of Guns interactive model of the M134 minigun, and also Forgotten Weapons videos on the M134 as well. These sources combined allowed for a good general understanding of the way the Delinker should function. Essentially, there are six “pushrods” seated in slots within the Delinker Spindle. While the Spindle rotates, the Pushrods are moved forward and backward alongan eccentric cam path. These pushrods un-seat the rounds from the belt links, and move the unseated cartridge forward into the front of the De-linker, which acts as an elevator, with a set of ramps in the shape of a spiral, that gradually lift the loose rounds up into the Receiver of the firearm. This also is how the spacing of the rounds goes from the close-together spacing of the belt links, to matching the spacing of the barrels in the Receiver.
Following along the path of the bullets through the gun, the Receiver and Barrel Cluster was developed next. The Barrel Cluster is a 6-chamber design, to match the 6 sections of the Delinker, for ease of gearing the two components together, as opposed to having un-matched number of barrels vs sections in the delinker. The barrels are spaced 3 inches apart from centerlines of opposing bores, which led to the first major complication to the build- finding a large enough ball bearing to act as the main front bearing so the assembly could spin within the Receiver. It was surprising to see how expensive large ball bearings could be, with prices varying widely from several hundred dollars, to over a thousand dollars. This was discouraging at first, but ultimately I came across the bearing which is presently used- which ranged in price from roughly $70 to $300 depending on the vendor. At the time I only knew of the $300 dollar source, so that ended up being about 30% of the total build cost for just this single part, though now much cheaper sources have been found.
The Receiver functions largely the same as the Delinker, facilitating the movement of all six Bolt Carriers backward and forward, as they rotate along another eccentric cam path on the internal wall of the Receiver. This controls the timing of the feeding and extracting of rounds throughout the firing cycle. Parallel to this cam path, is a second cam path just behind the first, which controls the cocking of the Strikers within the Bolt Carriers. The two cam paths are evenly spaced from each other, with exception of a portion where the Striker Path diverges farther rearward, which has the effect of pulling the Striker to the rear, ultimately coming to a ledge of sorts, where the Striker in effect “falls off” the sear, and impacts the firing pin while the Bolt is in lockup.
The Bolt Carriers are printed, and utilized modified AR15 bolt components. The Bolt head has the “piston” portion at the rear cut nearly flush, and gas rings removed. The firing pin is shortened to match, as well as a slot being cut into the cam pin, which allow clearance with the Barrel Cluster. The ball bearings present at the top of the Carrier facilitate both its motion as well as the motion of the Striker assembly, as it follows along the cam paths within the Receiver
The Strikers are unquestionably the most tricky part of the build. They can be DIY made from simple round stock found at the hardware store, though this results in some issues with the “posts” needing to be fitted to the “bodies” using small pieces of tape to ensure a fit that is loose enough to remove for disassembly, but also tight enough to not come apart while firing. That isn’t to say that this method does not work however, as it can be done with careful attention to the fitment of the parts.
Alternatively, a set of pre-made Striker Assemblies such as the one shown above can be purchased through the Shop on this website. The “posts” are made from ball-detent pins, and interface with a groove modelled into the “bodies”, which are SLM printed in 304 Stainless Steel, to allow the parts to click into place, and still be easily removable.
It became clear at some point along the development, that the Barrel Cluster would necessarily need a way to spin independently of the Delinker, for the purposes of unloading the gun as well as clearing malfunctions. Working out how to the Clutch Assembly was not very straightforward, as there was little reference to how the M134 clutch system worked.
The Clutch for the M1337 consists of three parts- the Input Gear, the Clutch itself, and the Output Gear. The Input gear is the small inner gear, that is pinned to the Barrel Cluster’s central axle. When the clutch is disengaged/to the rear, the Input Gear spins independently of the Output Gear. When the Clutch Carrier, which is linked to the “Triggers” at the rear of the firearm, is pressed forward, two sets of teeth mesh with both the Input and Output gears, effectively linking them together to spin as one unit. The Output Gear is directly connected to the Delinker, and this is how the motion is transferred into the Delinker Spindle.
Being a Gatling gun, the firearm is hand-cranked. This is achieved by a simple rack and pinion gear setup, mounted to the right side of the Receiver. This assembly is directly geared to the Barrel Cluster at the front of the Receiver. There has been no shortage of people wondering if a power drill could be used as a means of actuating the gun, which while annoying illegal for most people to do, is also not possible, at least in that exact way. As seen to the right, the Crank Wheel is not connected in the center of the Housing, so there is no way to “attach a drill” directly in this manner.
I do hope that someone with the proper legal means to do so, works out a way to attach a motor to the Receiver, and run the system at its limits. Currently, under hand-crank operation, the rate of fire is around 550 – 650 rounds per minute. You may be thinking “well it should be any ROF since its dependent on how you crank it”, and while true, I find that it is actually quite hard to consciously crank the gun slower than ~550, and faster than ~650. Its an odd experience, but for me at least, I find that between just trying to aim the gun, and crank at the same time, its hard to focus on purposely controlling the speed, especially if only firing a short string of fire. Along this same line of thinking, I believe the gun is a uniquely thrilling thing to fire. Its not unsafe- but It isn’t so stable that its mindless to fire, such as with large several-hundred-pound real Gatling guns. Firing the gun requires a moment to “ride the lighting” in a sense which is very fun.
The rate of fire could potentially be increased through the means of a reworked crank to incorporate a gearbox, though its hard to say how much the gearing could be leveraged before it becomes impossible to crank, as the Strikers are under stiff spring pressure and it does take a fair amount of force to actuate the system when Strikers are present.
The tripod mount for the M1337 ended up being less straight-forward than I had suspected. The Tripod was effectively the last piece to be designed, and due to the lack of available real-estate on the outside of the Receiver, the options for making some sort of mounting solution became limited. Overall I think the Tripod is an area that could use some improvement, it has a simplistic elevation adjustment, and a semblance of a traverse lock, though it isn’t super effective.
I am however very happy with the way the firearm sits on the tripod. I think the “arm”/elevator which pins the firearm in place is a very aesthetically pleasing implementation. It curves and reaches under the Receiver and holds the firearm balanced over its hinge point on the Pintle. As best seen from the picture above, it is very unobtrusive and I think is a good balance of form and function.
Beta Version 1
Beta Version 1 was the initial public release version of this project. I was thrilled to see not even 24 hours after release that people were already staring to print out bits for their own builds. This initial version did have a number of issues which caused very hard-jams that required near full disassembly to remedy. Having to fully disassemble the gun multiple times every range trip was massively frustrating. Additionally the firearm had to be taken off of the mount to fully disassembly, which was arguably unsafe.
For a long time, it was difficult to diagnose the root cause of the issues, as the firearm effectively is a black-box while firing, and is nearly impossible to view what has gone wrong internally if the system is seized. It was clear that there were issues regarding both feeding and ejection, but narrowing down the scenarios that led to the jam was not very straightforward. To make matters more complicated, these sorts of issues only really happened under live fire. I found it very hard to purposely induce a jam while dry-cycling the gun.
Beta Version 2
Eventually, with the help of several other testers, the main cause of most of the jamming issues was discovered. The Barrel Cluster Body required more geometry to act as additional feed ramps to guide the bullet tips into the barrel extensions. Essentially the tips of rounds were getting caught in areas I hadn’t originally thought were possible for this to happen.
The second most common issue was a failure to eject. Ironically the main Ejector on the Receiver was the cause of this. A quirk of the M1337 is that empty casings eject at a different point in the rotation of the Barrel Cluster than full cartridges do. The design of the ejection port had largely been based around the theoretical ejection pattern of full cartridges, and so the pattern for empty cases ended up being a bit of an unknown. When empty rounds were bouncing outward from the ejection port, some would take an unlucky bounce, and take too much time to exit the ejection port before the next round would come and hit it from behind. This would bounce the second case back into the gun, jamming the Barrel Cluster by pinching the round between it and the Receiver. Omitting the main Ejector entirely, as well as a deleting some of the geometry around the ejection port, allowed for the spent casings to take one fewer bounces on the way out, and eject in a consistent pattern.
Beta Version 3
While the B2 changes made the firearm function mostly reliably, there still persisted some occasional issues that would seize the whole mechanism and require a disassembly to remedy. Version B3 changed many components of the Receiver and other housings, to greatly simplify the method of disassembling the firearm. There was an added component to the Receiver Cap, which could be separately removed to relieve tension on the Delinker’s axle, and remove it. In addition to this, the Delinker was reshaped to have 4 fewer screws, and also be removable without the need to dismount the firearm from the Tripod.
Internally, two main changes within the B3 version effectively resolved nearly all jamming scenarios. Firstly, the number of the feed ramps on the Barrel Cluster was increased, to further eliminate areas that bullet tips could get stuck. Secondly, the spiraling elevator ramp attached to the Receiver, was reshaped to prevent an issue that I believe was ultimately the cause of most hard-jams. Previously, bullet tips could sometimes fall on the inner edge of this ramp, and become stuck at an angle, which would then prevent rotation of the barrels in either direction.
Adding a lobe to the left side of this elevator ramp essentially eliminated this possibility, and with it, eliminated nearly all hard-jam scenarios that would require disassembly, which I find funny because now that disassembly is much easier, the need to do so it not there nearly as much.
Version B3 also added a Front Carry Handle, and points on the Clutch Housing for heat-set QD mounts. This allows the M1337 to be carried and fired, which is one of the only Gatling type guns ever that this could be reasonably done with, on account of its weight being so light.
Presently, I think Beta Version 3 essentially represents what will be the finished version. While I am aware of some remaining issues to be fixed, they are small details for the most part. Ultimately, I can with the B3 version, take it to the range and let someone else shoot it that hasn’t done so before, experience no stoppages, and return from the range with an expectation that the same will happen next time. To me, this is functional enough to warrant a proper non-beta Version 1.0 release.
Release Version 1.0
After years of total development, including nearly two years of Beta versions, in which countless wonderful people built their own examples, and provided invaluable feedback which was used to better the design, the full Release Version 1.0 of the project was ready. Publishing the version change, as well as the release video, was a rather surreal feeling; to cross a nominal ‘finish line’ for the project, and take a moment to reflect on it all to this point.
This update brought a number of small changes, mostly final tweaks to geometry. The most beneficial of which being a slight chamfer to the bottom edges of the Bolt Carrier rails, which allowed for the Carriers to be removed with only one missing Rail Cover, rather than two, but pivoting out of their slots at an angle. This sounds insignificant, but it drastically simplified the most irritating aspect of the design, that being the installation and removal of the Bolt Carriers. Previously, the Striker Posts needed to first be removed, before sequential removal of the Carriers, but due to the need of having both Rail Covers removed from any given Carrier, it inevitably left at least one Carrier partially unsupported, which would more often than not, cause a jam. Now with the ability to remove the first (or install the last) Carrier with only one Rail Cover removed, this jamming scenario is essentially impossible, and means the Striker Posts can remain in the Bolts. This cut down the time to install and remove the Carriers by more than half, and removed the possibility of causing a jam during the process.
Additionally this version contained the feed chute and belt box for the M1337. The feed chute is articulating of roughly up to 40 segments, and has a decently large range of motion. The max length chute itself can hold roughly 60 rounds of ammunition before reaching the belt box.
The belt box is that of a spiral drum, which can hold roughly 100 rounds, and is intended to be worn off the back of the user. I think this is still a point of future improvement, as the spiral, while done with the intent of addressing issues with a previous iteration, introduces a lot of friction into a belt of ammo, and can not feed while fully loaded. The intent is to unwind the full belt through the feed chute, leaving roughly 40 rounds in the first couple windings of the spiral, which is manageable for the Delinker to pull. Likely this drum will get a redesign, and be paired with a second drum better suited to just being placed on the ground next to the gun while firing.
Orbital Crank
After taking some time off from the project, following the release of the full version, I decided to delve into a possible drop-in replacement for the Crank, which would be geared to give a faster overall rate of fire.
I had it in mind to attempt something along the lines of a planetary gear mechanism, hence the name Orbital Crank, which would drive what would normally be the Crank Wheel’s bevel gear. The goal was a 2:1 ratio increase of speed at the barrel cluster, but because I can’t do math apparently, it ended up being a 3:1 ratio – 64 gear teeth arranged along the outer perimeter of the new Orbital Crank Housing, 32 teeth along back side of the new Bevel Gear, and 4 driven planet gears of 16 teeth. While on paper this gives a 3x speed increase, it effectively results in a 2-2.5x increase in real life, due to requiring roughly triple the effort to crank the mechanism.
Running the gun at between 1300-1700 rounds / min, really quickly put a spotlight on issues that could further be iterated, as the absurd speeds which the internals were operating resulted in new failure modes that wouldn’t otherwise happen. The locking lugs of the Crank Arm, easily broke off under the added input force to operate the gun, so they were made into a replaceable piece that can printed at a more advantageous orientation. A piece of the cam path within the Receiver chipped away after the first 50 rounds or so, caused by the Bolt’s speed and inertia being enough, at that ROF, to punch a hole through the thinnest sector of the path. Making this sector yet another replaceable part has seemed to fix this issue.
Most troublesome was that the higher operating speed was clearly causing feeding issues that wouldn’t normally be so prevalent. I had a hunch that there was now enough centrifugal force to make the ammunition ride along the outside shell of the Delinker Housing, rather than down the feed ramps, when being handed off into the Delinker, which I surmize would cause the rounds to be fed in such a way that they had room to bounce into funny positions before being pushed forward.
This forced me to look into the alignment of the Delinker more closely, and by a stroke of luck and laziness, the odd numbered 35 gear teeth of the Delinker’s Spindle Gear, which was only 35 due to simply not changing it from the gear model download, had the unforseen benefit of allowing for the Spindle’s alignment to be tuned to a certain degree. By offsetting the Delinker Spindle Gear’ alignment tooth by 12 teeth to the right, the relative position of the Spindle to the Barrel Cluster can be retarded by roughly -5 degrees. This I believe allows the rounds to be presented into the Barrel Cluster more closely to the (relative) left edge, and as such be better constrained from bouncing around before getting pushed forward into the barrel. Whether or not that is truly what is happening, I can’t say for sure, but I do know that the difference in reliability was profound, and immediately after said change, the next test saw 70 rounds clean with no issues, where previously reliability was around 1 failure per 20 rounds.
There will be a future Version 1.1 to reflect these few changes to the Receiver and other parts, alongside the eventual release of the Orbital Crank files
Testing on Nikolai Romanov’s PR.17 Malachit Suppressor will also continue. 300 BLK subsonic loadings work nicely with this suppressor, but this is currently all I have tested.
PR.17 Malachit – by Nikolai Romanov
The PR.17 Malachit is an awesome suppressor for the M1337 designed by Nikolai Romanov. It features an impressive internal geometry, which connects each baffle stack to an adjacent stack via passage vents to further increase the effective volume of the suppressor. It is a very large and hefty print, but well worth the effort, and isn’t very difficult to assemble. It is designed to work with either 5.56 or 300 BLK, of which I have only so far tested the 300 BLK setup. I intend to do further testing on it, but ejection issues specific of 300 BLK have put that on the back-burner for now, although I recently heard of a modification from another fellow who has gotten his to run reliably with 300 BLK, so that will be part of the next round of testing as well.