I've always wanted a 5.56 belt fed upper for my AR15, so I decided to design one myself. I'm about 30 hours into the design and I'm pretty happy with my progress. Here's what I've done so far:

Rough modeling:
Upper receiver
Top Cover
Bolt carrier
Feed lever
Feed Tray

Still need to figure out a latch assembly, charging handle, roller for feed lever, finish the bolt/carrier design, and iron out LOTS of details.

It will be DI gas operated with a non-quick-detach barrel (I'm trying to keep it simple, I figure I could always add something like a Dolos system if need be). It will work with both belts and mags. Charging handle will be fixed (reciprocating) on the right hand side.

Again this is just a starting point, so go easy on me.

Attached File


Attached File


Attached File

Hell yeah, is your goal to manufacture and sell these belt fed upper?

I’m sure more than just myself would be very interested in buying one.

Keep at it OP.   I'm sure if functions well and more affordable than some of the current offerings, it will sell well.

I wouldn't rule out selling them eventually, but for now it's just a personal project.



Yeah, prices are outragous. I'm done holding out for PSA to get into the belt fed game lol.

That’s really cool. I missed the “easy” beltfed days and think the market needs more.

Neat stuff OP.  Good luck!

Isn't this how the Shrike thing began?

Can you elaborate?

I'm tracking this project for sure!

Excellent! Give us a price when you are ready. I am certainly interested.

Sweet!

Unfortunately I don't have an M16 yet.

Please design it to have:

- More rearward bolt group travel than the MCR.
- Make certain the carrier cannot impact the rear tang of the lower and cause catastrophic failure.
- Short top cover to allow for optic mounting on a solid, immobile rail behind the top cover.
- Large enough to be scaled up into 308.
- Maybe use standard 249 barrels.

Much crying and nashing of teeth.
Circa 1998 a post much like yours appeared, people put down pre-orders. Seems it took about 15-20 years for promises to be delivered.

see : https://www.ar15.com/forums/ar-15/Whatever-Happened-to-the-ARES-SHRIKE-/12-769958/

Keep us posted OP! I'm sure most of us would love to see an affordable belt fed upper.

These are great suggestions, but most of these are not feasible with the constraints of an AR15 lower.

-The bolt travel cannot be altered without changes to the lower. However, the use of a Vltor A5 system or similar may give some reliability benefits.
-As for the carrier hitting the lower, I will check the geometry on that.
-A short top cover would be great. However, it would take significant additional engineering to make that happen, since the feed lever attaches to the top cover and needs to be the full length of bolt travel. My priority is to make something simple that works. Plenty of guns have pic rails on the top cover, and gives sufficient accuracy for a machine gun. As long as the hinge and latch are made with tight clearances and tolerances, it should be a non-issue
-I've thought about making a 308 version. I would be easier to make for an AR10 lower, but there would be a lot of benefits to a version made for an AR15 lower. For now I'm just going to focus on a 5.56 version, and I'll cross that bridge later.
-M249 barrels are very expensive, and would require major deviations from the AR15 platform to work. My design will use a slightly modified AR15 bolt and barrel extension, which is simple and inexpensive.



Gotcha. I am familiar with the Shrike but I wasn't sure specifically what parallel you were drawing between that and my project. As stated, this is currently a personal project, and I'm not taking orders. I don't even have a website or a platform to sell it.

Some background on me, despite the name Type7SOT I don't even have an FFL anymore, I made a career change a few years ago and now work full time as a manufacturing engineer for a metal fabrication shop (not mine). This upper is a project that I'm designing on nights and weekends. It's not a business venture and I'm pursuing it with my own time and money. When it's done I would consider selling them if the costs allowed it to be competitive. IF I were to sell it, I would not take deposits and end up in that nightmare scenario that we've all seen play out. I'd rather make small batches and only take money when they are ready to ship.

Got some more work done today:

Attached File


Attached File


Bolt carrier is coming along, but still need to refine the roller (part that engages with the feed lever) and figure out some sort of firing pin retention. The rear screw of the gas key exists right where the firing pin retention pin needs to go. I may try to integrate firing pin retention with the roller part.

Here it is on a stripped lower:

Attached File


I'm on track to be 3D printing a prototype in the next week or so. (it will be plastic and non-firing, just to test fit all of the components together).

The bolt travel can be increased as was done with the Surefire OBC.

I like the idea of a modified AR barrel and using the standard AR gas system. And of using a modified carrier and bolt.

Interesting, I just looked at the OBC and I see they accomplished the extra travel by shortening the gas key (which I was actually thinking about doing anyway). Only problem is the added roller for the belt feed mechanism needs to be in a particular spot, and would likely run into the lower receiver extension if extra travel was allowed.

It would actually be interesting to make a 5.56 belt fed for an ar10 lower, which would allow more bolt carrier travel, perhaps as much as the m249. I'd have to check the math on that.

I know you'll lose out on some people that only want one because they have a registered m16 lower, but I think any innovation is good. If you can make a product that is price competitive and demonstrably better, then people will adapt.

You just gave me an idea

@JoshNC

Regarding this, the fixed side charging handle would prevent the bolt carrier from impacting the lower if something went wrong. I just need to make the slot the appropriate length.

I'm working on the bolt design, and there is a pretty major issue.

The ar15 bolt has 7 lugs with 45 degree spacing (But we can think of it as 8 lugs, because really its an 8-lug pattern with one of the lugs removed to allow for the extractor).

The bottom two lugs also function to strip rounds from the magazine. Two lugs are used for this purpose, because the mag is double feed.

Herein lies the issue. With a 45 degree lug pattern, having two lugs on the bottom means you will have two symmetrical lugs on top. I need a single lug on top, right at the 12 oclock position, in order to feed from a belt. If I rotate the bolt by 22.5 degrees, it will allow for belt feeding but will no longer feed from a magazine.

The MCR got around this issue by removing one lug and changing the spacing to 51.43 degrees (360 degrees divided by 7). This allows two evenly spaced lugs on the bottom for magazine feeding, and a single lug at 12 oclock for belt feeding. Seemingly perfect solution, but with some caveats. The cam path will need to travel 25.7 degrees to unlock, compaired to 22.5 degrees for a standard AR15. I dont know if this causes any problems, but it might explain the required "break in period" and ammo pickiness that some have reported with the MCR. It also means that you cannot rely on published data and engineering specifications for the AR15 which have a proven track record for decades. It means going back to the drawing board and doing 10's of thounsands of rounds of testing to ensure that the cam path, dwell time, chamber pressures, etc., are all optimized.

For my puposes, I'm tempted to stay with standard AR15 specifications on the bolt, carrier, and barrel extension, with modifications only relating to feeding, but not cycling or lockup. This gives the best chance of reliabilty early on in the process. But this means giving up the ability to feed from magazines.

But there is a silver lining that I just thought of. I've been doing some research on the Surefire OBC bolt, which many people touted as having massive benefits for full auto use. But I learned that it was discontinued due to shearing off bolt catches due to the extra travel it allowed. Presumably, this would happen on the last round of the magazine, when the follower pushes up on the bolt catch and the bolt slams into it (with more force than usual since the extra travel). And here is my silver lining! By making my upper NOT accept magazines, I might be able to get that extra travel and not need to worry about damaging bolt catches.

Not being able to feed from a magazine kinda sucks, but lets be honest. The whole point of a belt fed upper is to shoot belts. They even make dedicated lowers for the MCR that accept m249 nutsacks and dont allow a mag to fit anyway.

I'm not 100% decided yet, but I think I'll start with a no-mag version first, and consider upgrading the bolt to allow for mag use down the road.

The possibility of breaking mag catches were brought up when it was first released but didn't it was officially discontinued now.  I tried to find it on Surefire's website and I don't see it other than in their 'news' section: https://www.surefire.com/news/?p=surefire-obc-ar-bolt-carrier-reliability-redefined
Do you have an official link saying it is really discontinued?
That is unfortunate, as I really like mine but I have also replicated almost identical results using an LWRC one piece carrier that I documented on my site here for my 'Ultimax' build: https://c3junkie.com/?page_id=1403
So back in the day..... Ares did have a belt fed only lower like Cyrostructure sells but I was told that they would only sell to LE / Mil....
The reasoning was that their was some possible legal issues making it belt fed only as the upper alone could be classified as a Title 1 item and he didn't want to do that.  I know there is the belt fed only LM7 22LR and the belt fed only 9mm FM9 but both of those came out later and we all know how consistent the ATF is.

Couldn’t the stud/roller for the feed arm be moved forward and then change the geometry on the arm to accommodate?

Extra bolt travel would really improve function.

I am just parroting what I've heard on a few forums. I don't know if that info about the Surefire is true.

Your website looks extremely helpful, I will be looking at that closely!




I am really leaning towards a beltfed only configuration and stretching the upper receiver to give LOTS of extra travel. Maybe use a shortened AR10 buffer to accomplish the travel. It should be more reliable and lower recoil, right?

If you are not trying to design it for an existing M16 you own I would suggest dropping the AR lower receiver all together. There is no run up length in the AR receiver to use as momentum for stripping rounds from a link. This is why the AR based beltfeds currently for sale are running on the ragged edge of being reliable for stripping links. They have to over gas the system and put larger recoil springs in. This makes them terrible shooters compared to purpose design beltfed guns in the same caliber.

Look how the Stoner 63a has a bit of a run up for the bolt before it starts to strip the round. This makes a lot of difference in reliability for a beltfed gun, specially in 5.56mm.

See my post directly above yours. I think I can do both.

If not using a magazine feed as an option, move the whole feed mech forward to give even more travel length.

Yeah, thats what I'm proposing. I think it could work.

OP, clever idea moving the feed mechanism forward.  I think the objective is to maintain compatibility w/ existing RRs, RDIAS, & RLLs?

If possible, maintain the existing AR barrel, extension, bolt, & gas block, and use a bent gas tube to get the gas to the side.  Or, use a bent gas tube to route gas to a long stroke piston that bolts on to your custom carrier or an upside down bolt carrier with an extended firing pin.

Your feed block does not have to be dead center in the middle of the gun, you can strip off rounds from the belt slightly off center, using a single bolt lug.

ETA:  Your barrel doesn't have to be in a conventional orientation either, you can rotate your feed ramps wherever you like.  Nor does your belt feed need to be on top, nor feed from the left side.

Jesus, Josh.  Let's give OP a set of impossible, conflicting design requirements from the get-go.  Were you in Army Procurement?

I want it to work on the AR15 platform because that's what I am most familiar with, there is a plethora of engineering documentation available, and excellent parts availability. Being able to use a transferable receiver is a possible benefit but not the main goal.

The goal was to use as many off-the-shelf ar15 parts as possible, but some of the things in your list are problematic:
-An AR15 barrel can work, but it will need a new gas port drilled (otherwise I need to make a custom bent tube as you mentioned, which I don't want to do).
-A standard AR15 barrel extension will work, but it will be rotated 22.5 degrees from it's normal orientation. This can be accomplished by simply changing the position of the slot in the upper receiver that accepts the barrel pin.
-The bolt needs to be custom made, because the cam pin is rotated by 90 degrees. It should otherwise be the same.
-It will use standard gas blocks, but it will be installed rotated 90 degrees. Gas tube will be standard
The bolt carrier cannot be upside-down. I did consider that briefly, but then I realized the gas key would run right into the hammer! One of thos "duh" moments

Yep, I've considered all of these options. I keep running into design issues where solving one problem creates another. This is the challenge though


I actually appreciate his input as well as everyone else's. I like to "think out loud" and put every idea on the table, it helps avoid getting tunnel vision. If there is one helpful idea for every 10 unhelpful ideas, it was worth listening.

If you go w/ a long stroke piston, you can bolt your piston mount into an existing bolt carrier and it can be low profile over the hammer rather than a standard height gas key.  If you are determined to use DI, then you'll have to mill your own carrier instead of modifying existing ones unless you don't feed from the vertical.

There are various left eject BCGs that might solve your cam pin issue.  You could also set up your feed mechanism on the left side of the gun & feed the belt vertically instead of horizontally.  You'd need to overhang the feed tray over the side of the lower.

I got the bolt modeled, which helped me better see the interaction between the bolt and feed tray (feed tray displayed in transparent):

Attached File


I had to make some adjustment based on my findings, but that was pretty easy.

Next step is figure out the roller design, and then stretch the upper receiver and associated parts for the extra travel. I'm feeling pretty good about the current status of the project.

Ha. Not in the Army.

Those are not conflicting requirements.

Anyone have suggestions on gas system length? I'm thinking midlength and 16" barrel to keep the factory dwell time.

I’d personally go with carbine gas.

If you go mid-gas or longer you're gonna need a larger than standard port.  Pulling up a belt & stripping rounds from it is gonna take a lotta oomph.  What does the Shrike use?

I plan to use a larger gas port. But I think a larger midlength port may be better than a standard size carbine port.

Pressure and flow are two different things. A midlength gas system should have lower pressure since its further away from the chamber, but I can make up for that with flow by having a larger gas port. Having the bolt locked until the bullet is further down the barrel should reduce chamber backpressure and heat in the receiver. At least, this is my understanding.

Not sure what the shrike uses, but they do have multiple barrel lenths available.

I can make the cam path whatever, but this is where I'm hitting the limitations of my engineering ability. Changing the cam path has a lot of implications that I don't fully understand. I feel that I'm already adding a lot of new variables with the feed system. When I make the first prototype, I want to be able to troubleshoot the feed system exclusively, without worrying about other variables.

Besides for the Surefire OBC, are there any other bolt carriers that have successfully improved the cam path? What specific changes do you recomend to the cam path, and what are its benefits?

Since you're machining your own bolt carrier, and moving the feed forward, you have all kinds of opportunities to improve on the geometry of the AR bolt carrier & reciprocating length.

You are constrained only by the location of the hammer, buffer tube boss, and the length of the rifle buffer tube, though even that can be extended w/ 1" steel pipe if needed.

Specifically to the cam path, you can have the bolt carrier move straight rearward for a time before beginning rotation of the bolt, increasing dwell time & allowing the gas pressure in the chamber to drop, and then have a shallower cam angle for rotation of the bolt, allowing more torque on the case & bolt as the bolt unlocks and causing the bolt to rotate more slowly for a given length of carrier travel.

Since you're not constrained by the AR upper length, you can have the bolt carrier recoil as far as you'd like into the buffer tube, perhaps even approaching constant recoil lengths, particularly as you're not concerned about the LRBHO catch breaking.  You just have to have it not bang your gas key.  You'll need a longer firing pin, which gives you the opportunity to add a firing pin spring &/or a firing pin safety if you want to get that complex.

You could also ditch the buffer tube entirely, or simply have a vestigial nub on the rear for the autosear folks, and put your recoil spring up front.  This again opens the opportunity to use a long or short stroke piston set up.

If you move the recoil spring into the upper, it’s highly likely ATF FTB is going to rule the upper a firearm and that will preclude its use on a transferable m16 lower.

Precedent?  The Olympic uppers weren't ruled as firearms.  PSA's Jakl hasn't been ruled a firearm.

Those examples all feed from the AR lower via a magazine. Remove magazine feed, add belt feed, and ATF will classify it as a firearm.

They declared Ernie Wrenn’s RPD and 249 uppers for the MAC series as firearms and thus couldn’t be used on a MAC lower. They were also open bolt and iirc FTB ruled that they were post samples. As I remember it, Ernie said FTB told him that the fact that the uppers were belt feed only and had the recoil asssmbly in the upper, they were a firearm.

The BRP mg34 upper for the m16 lower was declared a firearm. As I recall, it was because it was only belt feed.

I also recall a discussion about it as it related to the Shrike like 25 years ago on Subguns. It was back when Herring still posted there, before the taking of deposits and subsequent long wait.

Ah.  I remember the MG 34 uppers, didn't know ATF killed them too.

Not sure the length, but the MCR is a short-stroke piston system if that matters.

Random thoughts from a fatigue addled brain....
- Use a left handed bolt carrier group.
- Rotate the bolt carrier and barrel 90 degrees counterclockwise.
- That gives you off the shelf barrels, gas tubes, and only slightly unusual bolt carrier groups, potential ejection through the magwell, and something else I've promptly forgotten.

Off to bed with me.

If extending the receiver forward, I wonder if it’s possible to still design mag feed with a very extended feed ramp. The goal here would not be to actually feed from mags, but to lessen the ability of FTB to rule the upper a firearm.

And another thought. It could be possible to use a standard AR bolt group. Design a feed/delinker pawl that interfaces with the bolt carrier. As the carrier comes forward, the feed/delink pawl pushes the cartridge out of the link and into the feed ramp, positioning it for the bolt to fully chamber it. This would also allow the carrier to travel independently of the feed/delinker pawl so that upon forward return the carrier hits the pawl with more energy that if the pawl was traveling with the carrier. Not sure I’m describing this in a way that makes sense. This design would increase the height of the upper to accommodate the pawl design.


As backbencher mentioned, no reason the feed mech has to be top dead center. Look at the FN Evolus design with its feed tray at about 35 degrees to the left. It also seems like the Stoner 86 feed design, which ultimately became the KAC LAMG would be a good option in this case.

https://topwar.ru/uploads/posts/2020-03/1584893633_ares-lmg-us4942802-2.png


Another thought is to depart from the AR carrier and cam pin design and go to a more true AR18 design so that the cam pin limitation of bolt pre-engagement is accomplished via the bolt carrier rails.