Lowa Zephyr Mk2 GTX

NSWC-Crane Mid-Length Gas System Testing Shows Increased Performance & Service Life For M4 Carbines

Last week’s NDIA Armaments Forum ended with a briefing by Naval Surface Warfare Center – Crane, of what is this year’s most applicable topic, for both the US military as well as manufacturers of commercial AR variants. Last year Crane unveiled their findings regarding KeyMod vs M-Lok. This year it’s the performance of a mid-gas system on an M4 carbine.

NSWC-Crane, or Crane as it is commonly known, is located in rural Indiana. In addition to providing a wide range of acquisition services for the US Navy, they are also responsible for the test, evaluation, procurement and life-cycle management of SOF weapons. It’s in this role that they evaluated the mid-length gas system for United States Army Special Operations Command M4A1 carbines.

Gas System History

The M16 Rifle and variants use a 20” barrel and gas system. This rifle length gas system uses a gas tube 15” in length with gas port at 13”. The well distance is approximately 7”.


When the M4 was developed, research concluded the Army should utilize a 14.5” barrel for the M4 & M4A1 carbines

This necessitated redesign of the M16 gas system because a 14.5” barrel with a rifle length gas system had only 1.3” dwell distance. Consequently, they gas port was moved to 7.8” from bolt face on M4 offering 6.7” dwell distance.

This decrease in distance from the bolt face to the gas port resulted in an increased port pressure in the M4 carbine when compared to M16 rifle. The port pressure of the M4 at 7.8” from the bolt face is 17,000 psi, while port pressure at 13” from bolt face of the M16 is 10,000 psi.

Mid-Gas Testing included:

– Endurance
– Reliability
– Precision
– Muzzle Velocity
– Terminal Velocity (@100 yards)
– Bolt Speed
– Low Temperature (-60F)
– High Temperature (160F)
– Barrel Erosion

However, the briefing did not address every area of testing.

Endurance & Reliability

So far, Crane has put 30,400 rounds of M855A1 through three M4A1s equipped with 14.5″ cold hammer forged barrels and a mid-gas system with a gas block approximately 9.8″ from the bolt face.


They stated SOF M4A1s normally start to see accuracy degradation at around 6,000 rounds. But during testing of the mid-gas system, they’d hit 12,600 and still hadn’t seen any changes.

They also have only broken one bolt so far in testing, although I don’t think they’re ready to attribute the improved bolt performance to the mid-gas system.

The Crane team will finish testing up with 34,000 rounds per upper. It’s not that they don’t think the barrels can’t take more, but rather that they had to use the same lot of M855A1 to satisfy the accuracy portions of testing.

USSOCOM Accuracy Testing & Protocol

With this mid-gas system they are getting 5 MOA groups while the standard is 7 MOA. One of the three uppers was shooting 1 MOA, except for the tenth round which was still within limits.

Interestingly, USSOCOM tests accuracy differently than most others. They fire 10 rounds suppressed and another 10 rounds unsuppressed. They measure the extremes of the spread of impacts, rather than their closest points. Then, they do it again two more times and average the results to determine accuracy.

Muzzle Velocity

These measurements are averaged and validate what we know about the use of suppressors increasing muzzle velocity.


Terminal Velocity

The velocity at 100 yards for mid-length weapons is 32.6 fps or 1.2%, higher for suppressed fire and 41.7 fps or 1.6%, higher for unsuppressed fire.


Cyclic Rate

Mid-length cyclic rate of automatic fire was 62.7 rounds per minute (rpm), or 7%, lower than carbine-length for suppressed fire and 127.2 rpm, or 16%, lower for unsuppressed fire.


Temperature High & Low

960 rounds were fired at 160F for Reliability at High Temp and another at -60F for Reliability at Low Temp.

For carbine-length weapons, 5 out of 65 malfunctions occurred during high temperature testing. For mid-length weapons, 1 out of 30 malfunctions occurred during high temperature testing. For high temperature testing, carbine-length weapons had 576.0 mean rounds between failures (MRBF) compared to 836.1 MRBF for ambient temperature testing and mid-length weapons had 2800 MRBF compared to 1993.8 MRBF for ambient temperature testing.

For carbine-length weapons, 27 out of 65 malfunctions occurred during low temperature testing. For mid-length weapons, 16 out of 30 malfunctions occurred during low temperature testing. For low temperature testing, carbine-length weapons had 333.3 mean rounds between failures (MRBF) compared to 836.1 MRBF for ambient temperature testing and mid-length weapons had 562.5 MRBF compared to 1993.8 MRBF for ambient temperature testing. Approximately half of the total malfunctions recorded for both carbine-length and mid-length weapons occurred during low temperature testing, so the relative rate of malfunctions between carbine-length and mid-length remained similar to that of ambient temperature testing.


Although testing to 34,000 rounds isn’t yet complete, the conclusion is simple. Use of a mid-gas system significantly extends the life of the overall weapon system. It also offers increased performance over a carbine-length gas system.


This information is particularly important for the US Air Force’s Improved Modular Rifle – Blue program which templates off of upper receiver group improvements adopted by USASOC (Brownells is offering a similar package for reference). While USASOC will upgrade up to 15,000 carbines, the AF wants to modify around 50,000 guns. That could be enough to force a major Technical Data Package update applicable to all services and creation of a GOTS upgrade for all M4s, regardless of service.

This government testing also validates what many commercial vendors have been offering for years.


60 Responses to “NSWC-Crane Mid-Length Gas System Testing Shows Increased Performance & Service Life For M4 Carbines”

  1. Joe says:


  2. Luke says:

    Nice to see it all spelled out.

    Everyone likes 16″ midlengths, but there was a few holdouts voicing concerns about 14.5″ midlength reliability. Seems we can put that to rest.

    • Jake says:

      What were the supposed problems with the mid length 14.5?

      • James says:

        I think a lot of the problems that people were seeing were gas port diameter related. Some guns with smaller ports are just a little undergassed. Really like the .0810-0860 port size with an H1or H2 buffer but some guns have the .0760 or smaller size ports . Just makes them a little sluggish.

      • James says:

        Should probably clarify here that running 885a1 suppressed something like .0770 port size with H1 buffer would be better. Random ammo unsuppressed is the reason I like the bigger ports.

      • Luke says:

        As James said, I think it boiled down to undergass concerns.

      • Jake says:

        Thanks for the info, both of you. So there wasn’t a detrimental issue with 14.5s; they just needed to figure out gas port sizes.

  3. Kirk says:

    “When the M4 was developed, research concluded the Army should utilize a 14.5” barrel for the M4 & M4A1 carbines”

    Can someone produce a cite for this research? Asking for a friend, who remembers it differently…

    • SSD says:

      That was from the briefing.

      • Kirk says:

        One would hope that the briefer wasn’t just pulling it out of his ass, then…

        I gotta go dig up my references, but my memory is that there was damn little “science” that went into that decision–It was basically done in a near-complete absence of mind, and 14.5″ was just the longest they needed to make it so that the M203 didn’t extend past the muzzle, and the gas system length was set the way it was to enable a standard bayonet to fit…

        Memory could be failing me on this, but I’ve never heard of any real work done to validate this barrel and gas system length besides “Will the M203 and bayonet fit…?”.

        Which ain’t exactly surprising. The M4 was never meant to be the main rifle–Ever. When we first talked about getting these on general issue, it was supposed to be for the support bubbas, like the Artillery and us Engineers. The Light divisions MTOE kept the M16 for the Infantry, and the process by which the M4 morphed over into our standard infantry weapon was never planned for or really thought about, much–From my perspective out on the line, it was “Yeah, we’re getting this shorter, smaller rifle for you guys, ‘cos you don’t need to be hauling all that weight around, all the time…”. Followed by the “Ain’t those cool…” types up at the headquarters glomming on to the incoming new weapons, and finding out that they were a lot handier, followed by diversion over to the Infantry units and a lot more going on MTOE. The support branches that were the initial targets of the M4 fielding never did see “their” weapons–We were still carrying M16A2s into Iraq during the late 2000s…

        • DSM says:

          Pretty much how I remember it. The 14.5″ barrel pre-dates the M4 to the best of my knowledge as our GUU-5Ps had them back in the 90s. Given the time overlap both could’ve been concurrent developments however given some, but not all, GUUs wear/wore standard M4 barrels. But the “carbine” gas system had been the foundation of a short AR since the 60s whereas what would become the ultimate choice in length of barrel was more cursory to already standardized equipment, I.e. The bayonet.

          • James says:

            Everything I remember reading in the early 90’s was that they started with the carbine gas and then tested barrel lengths for that . 14.5 was a good compromise on reliability which fell off at 12.5 and velocity which started falling off at 16.

            • DangerMouse says:

              The old urban legend was that it was because they had to use as many existing parts as possible, so the XM177/CAR15/etc.. handguard and gas tube mated with the issued bayonet was what gave us the 14.5″ length. Just simple logistics and math.

              That’s been somewhat debunked by folks stating that through testing they decided that they needed to match the dwell time of the M16, which the 14.5″ barrel gets very close to the dwell time of the M16, based on roughly the same dwell distance from the gas port to the muzzle crown.

              Which story is true? I have no idea.

              • RSR says:

                Dwell time being equivalent also meant that the 14.5″ barrel would fit the current issue bayonet.

                Significantly, it also matched the length of the colt commando as well.

                colt commando barrel = 10-11.5 inches
                colt commando moderator = 4.2 inches
                assembled = 14-15 inches

                m4 barrel = 14.5 inches
                m4 a2 flash hider = 1.75 inches
                m4 w/ a2 flash hider = 15 7/8 inches

        • Canadian says:

          Not sure if there was. The Diemaco (now Colt Canda) carbine barrels are 15.7″, supposedly based on actual performance testing. Not sure about their methodology, but it would make sense that they had some plan, or they could have just chosen the same barrels as the M4.

          • Kirk says:

            Not all the C8s are 15.7″. The early ones were 14.5″, according to the literature. Only the models produced after the C8SFW had the longer barrels, and judging from the nice, even 400mm that works out to in the metric system, I’d about wager money that the basis behind that choice of length was about as arbitrary as the original decision for 14.5″.

            Every time I see nice, round numbers…? Yeah, that’s not a good sign that someone actually, y’know… Did any work, ‘cos mother nature and father physics both have a fondness for not finding human measurement systems particularly worth paying attention to, with things like constants and actual, real-world observed events. You tell me that a barrel that’s 337mm is ideal for M193, or whatever, I might take that number from you without giving you the stink-eye, because I’m used to seeing that sort of thing in real-world calcs. You tell me that it’s a nice, round 400mm which is the ideal platonic barrel length, and I’m gonna look at you like you think I’m stupid, or something.

            Of course, it may be that the ideal was 337mm, and someone decided to just round it off to 400mm because that would be easier to make, but if that was the case, then they should have said something like “Well, testing showed that the ideal barrel for this cartridge is 337mm, but we rounded it up 3mm in order to make it easier to build, and we then double-tested our work to ensure that the 400mm produced the same ballistics that the ideal 337mm one did…”.

            I’m super-effing tired of people pulling things out of their asses and then blowing smoke up mine that they did “scientific research” on something without showing me the work they did, and expecting me to just take their word on things… And, it’s not just in this sort of thing, either. There’s a lot of “…won’t show their work…” going on in general, whether it’s VW claiming they’re making their mileage figures honestly in their diesels, or some eedjit telling me that my electricity is gonna be cheaper after they spend a few million of my tax dollars building themselves some nice big windmills…

  4. Dustin says:

    Ive been saying mid length is better for years,especially on a 16″ barrel and I recommend the mid length gas system to anyone building or buying an ar15. I wont purchase any brands carbine length gas system ever again. Mids or rifle lengths only period. Carbine length gas system has also been an issue on some ar10 .308 versions with a 16″ barrel as well. Anything is better than nothing though, cheers to those who have had luck with carbine length gas system. I prefer to go with what i know. Long live our 2@!!!

  5. Bman says:

    So all this time that everyone swore how much more reliable all of the piston guns are/have been compared to the M4, they were basing it off of a carbine length gas system that practically everyone knows is the least reliable of the gas lengths in the AR platform?

    I thought the M4 or at least the A1 used a mid length system this entire time. So it’s only 2018 and they are starting to learn what the rest of the AR shooting community has known since several years to a decade ago. Now I would like to see a mid length system and modern bolt coating, if not an enhanced bold design tested against piston guns. I have a feeling the reliability results and weight would have even more people on the fence than before.

    • CAP says:

      Piston guns were more reliable, back in the early 2000s before we figured out how to make M4s run right.

      A current M4A1, with an H2 buffer, Crane extractor spring upgrade kit, Magpul Pmags, and full auto trigger group is a much more reliable weapon than an old M4. The new URG-I with the mid length gas system is even better.

  6. Darren says:

    You know how they figured out the barrel length? Bayonet lug.

  7. CAP says:

    Just awesome. I love seeing data that validates what most knowledgeable AR guys have been saying for years.

    Also, this isn’t just validation that the mid length gas system is more reliable, but also that CHF barrels last longer. The mid length CHF barrels showed more than double the round count before accuracy dropped off.

    • jason miller says:

      Ask random sheep what AR stands for, and its not Automatic Rifle! Wait for the argument…. “but the news said it stands for”….
      Those are the ones who are scared of a weapon that cannot fire w/o human action!

      • RT from UT says:


        In relation to the AR-15 it doesn’t stand for automatic rifle either…

        Pre m-249 there was ONE automatic rifleman (he who carries the clip on bipod)

        Everyone else just carried ARMALITE RIFLES

  8. Josh2 says:

    That accuracy standard is pretty crap .

    • Neil says:

      It is, because M855A1 has an acceptable SD of 6.8 MOA, and its getting worse as production rates go up. It’s really a byproduct of the production speeds of that ammunition.

      The guns do much, much better with better ammunition.

  9. CAP says:

    Does anyone know the gas port diameter of the mid length barrels they are testing or what weight buffer they have in the lowers? Would be nice to know as this is as close to a “mil spec” standard for mid lengths as we have ever had.

  10. patrick sweeney says:

    Is it willfull ignorance on the part of those supposedly in charge? The hi-zoot, upgraded, M4A1 of today is about on-par with an IPSC 3-gun rifle of 1995. Improved alloys, coatings, magazines, ammo, etc. all are years, decades, behind what the competition people have been using.

    Oh, and I love the adoption of the mid-length gas system, tested with the obnoxious, stupid-as-all-get-out M855A1 load, because “the rifle is more reliable” with a mid-length. how about not using M855A1 for starters?

    How about having people in charge who actually know this stuff?

    How about actually admitting that rifles are meant to be used to shoot bad people, and not just make noise until CAS or Arty solves the problem. (Yes, I know those are the real killers, but tell that to the PFC who has to stay alive long enough for the problem to be solved.)

  11. Joglee says:

    And the Army is chasing the NGSAR, when major upgrades are right there in their face and everyone around them is chasing them.

    • Kirk says:

      Incremental, evolutionary improvements are what they should have been moving towards from the beginning. I’ve said that for years…

      If they’d slipstreamed the CHF barrels, a mid-length gas system, maybe a full monolithic upper like Colt offered up lo these many years ago, or any of the stuff like LMT’s Monolithic Rail Platform, well… We would be in a lot better place than we are.

      Not to mention, if someone had done the work to balance the length of the carbine barrel with the propellant/projectile combination, as well as adjust the actual gas system to support that length…

      Reality is, I’m afraid, that the M16 and M4 are what they are not because of “scientific research and development”, but despite it. It’s actually pretty damn amazing that these weapons have been as successful as they have been, being as they were designed, procured, and fielded in what I’m afraid can only be described as an “absence of mind”…

  12. Correct me if im wrong, but i do believe and thought i read somewhere from a valid source, that the new m4s also run a one piece gas ring design similar to that of the Mcfarland type. I think this also contributes to reliability. Its a cheap upgrade that anyone can use.maybe that was concerning civilian use of pistol type uppers which would make sense as well. I also feel that the use of anti walk pins is very important even with semi auto.

  13. P.J. says:

    Would have been interesting to see the same tests done with a carbine system firing M855. Seems like a mid length system may solve some of the longevity concerns with adopting M855a1.

  14. patrick sweeney says:

    That’s the problem with “tests” such as this, which are generated only to “prove” one side or the other. I’m afraid to say it gentlemen (OK, I’m not, really) but if you don’t perform your test, any test, alongside a known standard, you’re just wasting time, money, ammo, rifles and reputation.

    As P.J. pointed out, performing the test with identical uppers, in the same program, with M855 would have actually told us something.

    • SSD says:

      There was zero point in conducting the test with M855. SOCOM kills with M855A1. They already know how it works in a carbine length gas System. This was a test to determine how reliability and performance differed between carbine and mid gas systems.

    • Neil says:

      This literally makes no sense.

      You literally have zero basis on which to criticize this test. I’ve reviewed it first hand, and it was valid.

      M855A1 is a known standard. Its not a great standard in many ways, but it is known, and it is the standard within the DoD from now into the future. So, what would one have to gain by seeing what the system would do with ammunition that will not (in time, there’s still stocks of M855 to burn up) be shot through it, particularly during a deployment?

      And if you fired M855 through it, it will wear on the guns less, but that’s about it. Dispersion specifications are the same as M855. Bolt carrier speeds will be a touch different, but still well within the window of regular operation.

      As much as you may not like it, M855A1 is here to stay. The solution now is to push for higher accuracy standards from lot to lot, and optimize the guns for it’s use as much as possible. The latter is exactly what happened here.

    • P.J. says:

      To be clear I wasn’t saying they should have done that test. I’d like to see it purely as a curiosity. The decision on M855a1 has been made. Those in charge have no reason to justify themselves to naysayers.

  15. Piet says:

    Simplification. Blow away all above BS. Do you want a gas system that blows carbon and filth into your chamber with every shot? Or, do you want a short stroke external gas system that does not? This baloney has been going on for FIFTY YEARS. About the length of time the Army took to adopt rifling. Nothing ever changes.-Piet

    • CAP says:

      In my experience, people who keep repeating the “but it sh*ts where it eats” BS are either , A. Trying to sell external piston ARs, or B. Have very little experience working with either direct impongement or external piston ARs.

    • P.J. says:

      The rifling part is blatantly false. The Army had riflemen units in 1775. There was a government contract for rifles in 1792. Harper’s Ferry Armory was producing rifles in 1803. Rifles were adopted as the primary long gun almost immediately after the Minié ball was developed. Prior to that rifles weren’t viable for army-wide use.

    • joglee says:


      The URG is blowing away the M27 across the board.

  16. Seans says:

    You want the gas system that works best. If you think fouling is a issue in a AR15s. You might want to do some more reading.

  17. J says:

    I wished they gave more information on those CHF barrels used, such as Government profile or Heavy profile or Mid-weight profile, that were used for the testing. I also wished they compared 14.5 inch carbine length barrels to 16 inch mid length barrels to see the differences.

    The Army should have went with the 16 inch barrel with a Mid-length gas system years ago. It shoots better with higher velocities and is more accurate at longer ranges. The Marines are totally in on the 16 in barrel length now with the implementation of the M27 (HK416) rifle in the 16 inch length for all of their infantryman now. They saw deficiencies of the M4 and M4A1 rifles and went for something better. Great information.

    • Neil says:

      Government profile is what was tested.

      The velocity jump from 14.5-16″ is nominal and not worth the increase in length, particularly when today you have bullets that are not nearly as velocity dependent as the SS109 bullet was.

      At no point has a lighter and shorter system been considered deficient on those basis.

      Know that the M27 effort had little to do with closing capability gaps, and far more to do with administrative and intellectual laziness.

      • James says:

        M27 was as much about a cooler running gun as anything. The difference between the sustained rate of fire for it vs. the m4a1 is significant.

        • RT from UT says:

          Citation needed…

          I’ve heard that since the dawn of piston AR’s, but never seen anything approaching proof of it.

          You hear this whole it cools the gun down thing from the parts of the piston posse that already know they’re busted on the reliability thing, but if anything, the piston posse was smart to lead with the reliability thing because it was at least true at the time sometimes.

          Realistically, adding a bit more weight to the cycling components gives you all the “benefits” of a piston without the downsides though.

        • joglee says:

          It’s actually not as significant as you think.

          Let me guess, you still think the M4A1 has a 12-15 rounds per minute sustained, despite that number came from the days of the 1990’s M4 using insulated polymer handguards and a thin barrel.

          • James says:

            15 is for the a1 with a heavy barrel. The numbers don’t lie. The test for the sustained fire rate is to 120°, it’s all about the temp not about fuctioning. A well built M4 will function well past that rate , but hard to claim it will maintain a 120° temp at the 36rpm rate of a M27. As for the source it was in the Marine testing of the M27. I’m sure the rate for the midlengths , will be a little better. How much better? Guess we’ll see.

            • Joglee says:

              15 is for the M16A2, M16A4, M4, and M4A1.

              It was never changed as the weapons were upgraded. This is shown in FM 3-22-9.

              You can’t honestly tell me you believe those 4 weapon systems all have the exact same sustained rate of fire?

              It mean it’s not possible for a M16A2, with polymer insulated handguards and a barrel diameter of .64″ to have the same exact sustained rate of fire as the M4A1 with a ventilated KAC RAS and a barrel diameter of .81″.

    • J says:

      Well, the M27 (HK416) 16 in barrel is CHF, which has at least a 20,000 round barrel life. I think the Marines did not want to have to change barrels ever 6,000 to 7,500 rounds with the chrome lined M4 or M4A1 heavy barrels. Velocities are not the significant between 14.5 and 16 in barrels, but the accuracy is with a mid length 16 in barrel opposed to a carbine length 14.5 in barrel.

      SS109 or M855 ammo is probably gone now in the US military with the advent of the M855A1 ammo. M855A1 has problems with barrel wear, throat cutting, stripping M4 feed ramps, and overpressure that I have read so a CHF barrel would be good for this round. The M855A1 round also has a significant amount of overpressure about 62,000 psi in the chamber compared to the M855 round having about 52,000 psi.

      The M16/M4 platforms weak area is the gas tube that was designed to break first when the gas tube is overheated to extremes. This can only happen when M16/M4 platforms are used in a continuous automatic fire till overheating occurs and the gas tube shears off. M16/M4 platforms were never designed for sustained automatic fire.

      I also wonder why they did not try chrome lined vs nitrite barrels to include in testing. I have read the main difference between chrome line and nitrite barrels is how long it takes for the bore to open up to be shoot out. Chrome line barrels tend to open up slowly and nitrite barrels do not open up and are just shoot when the bore reaches its limit. I forget the article I read about this, but will post it later when I find it.

  18. patrick sweeney says:

    Is there a link to the full test data? Because this test looks like the usual circular reasoning we get when “something must be done.”

    A mid-length gas system is better than a carbine. That’s like saying water is wet. That a mid-length works better when fed M855A1 than a carbine system does does not prove that M855A1 is a good choice. And “M855A1 is here to stay” like the M193 was here to stay, then the M855 was here to stay? Once the end-users and the maintenance people have had enough of busted guns from the over-pressure, M855A1 will be gone, for the next “great thing.”

    Question: How much better, and how does it do it, is the M855A1 over M855? And, second question, could the same have been accomplished without the excessive pressure and feed ramp destruction the M855A1 brings with it?

    • Joe_K says:

      Ya have one worn barrel exstension and the entire internet freaks out. At this point it’s about as bad as people claiming M-1 Carbines firing 110gr FMJ Ball won’t penetrate a Commie Chinaman’s winter coat!

  19. cimg says:

    Am I the only one who didn’t see a whole lot of difference between the two in this? Yes mid length is better, but is it gooder enough to justify the expense.

  20. patrick sweeney says:

    Maybe, maybe not, depending on exactly what you want out of the system.

    However, the one clever thing that is happening here is this; by hanging the improved reliability of the M855A1 round on a new, mid-length gas system, certain groups are boot-strapping other improvements, like the barrel steel, handguard, suppressor, etc. into reality.

  21. David Dale says:

    How painful. Unless Boyle’s Law has been rebutted, the pressure of a gas is inversely proportional to the volume. The pressure at the port is purely determined by the gas produced by the cartridge filling the volume contained by the chambered cartridge and the bore. Yes, you can control the amount of gas admitted into the gas block by altering the port dimension or altering the dimensions of the gas system. NSWC Crane has the data available from the MK 12 SPR development. This is not newly discovered information that a higher volume gas system is easier on the components and more reliable. Egads!