35 Muzzle Devices Tested – Recoil Mitigation Compared

For the last two years, Vuurwapen’s muzzle device article has stood as the seminal test of modern sporting rifle brakes. While detailed, the article has (sadly) not been updated with the latest and greatest that the industry has to offer.

Now competing for the mantle is Jeremy, who takes 35 5.56 modern devices through his own methodology. Its interesting to see how various devices compare in independent testing over marketing and forum hype.

My only complaint? Jeremy doesn’t tackle flash in low and no-light like Andrew’s original testing.

Nathan S

One of TFB’s resident Jarheads, Nathan now works within the firearms industry. A consecutive Marine rifle and pistol expert, he enjoys local 3-gun, NFA, gunsmithing, MSR’s, & high-speed gear. Nathan has traveled to over 30 countries working with US DoD & foreign MoDs.

The above post is my opinion and does not reflect the views of any company or organization.


  • JumpIf NotZero

    Good job, but this is not a replacment for Andrew’s testing.

    One of the best things about Vuurwapen’s was that it wasn’t a video. But more so, I’m not sure im feeling the “sled” technique for measuring rearward decrease. Like, at all. I’m pretty sure a plastic sled on the table like that with also a Sharpie activated trigger is flawed to give misleading results.

    He should been using an inflatable bag to activate the trigger, the sled was allowed to jump and rattle as well as move off target axis, where it while have held in on “rails”, there also should have been a return spring with a high marking indicator.

    What we have here is entirely uncontrolled movement. Qualities like reduction in actual muzzle rise, or effective felt recoil changes, have been tossed to see which hopped a sled to an arbitrary point.

    • Giolli Joker

      There’s the article as well on TTAG…

  • gggplaya

    I don’t agree with the sled testing, should attach a 3axis accelerometer to the rifle and log each shot.

    • MP

      Yeah, that would work.

      Alternatively I think a strain gauge as the point of contact behind the stock and another above the barrel and directly behind the muzzle device would work to capture recoil and rise forces independently.

    • The Real Teal’c

      That’s a real good idea!

  • yes and has been for some time.

  • McThag

    The problem with this sort of testing is the same problem that occurred with “seat of the pants” testing with all sorts of hot-rod parts. Lots of products seemed to increase power, but showed no effect on a dyno; meaning that seeming increase was illusory.

    Accurate and repeatable measurements with as many variables as possible eliminated is what’s required; average Joe is too apt to accept the appearance of impartial testing where it’s anything but.

  • JSmath

    Maybe we can get him to start a Kickstarter, so we can get the test to be re-ran (with more muzzle devices!), low lighting, a sensor setup to read the various different forms of recoil (recoil impulse force, recoil velocity, axial deflection like muzzle rise, drop, or sideways shift), and, something I found myself curious about halfway through the video, different barrel lengths. Obviously, that’s a lot of shots and time, which is why I’m suggesting a Kickstarter to fund it.

    I know, that’s a lot of stuff to mull over, but I think there’s enough people and manufacturers who would be interested in seeing how different products stack in more technical detail.

  • Nicks87

    Oh give me a break. Dramatic much?

  • raz-0

    The sled could be designed better perhaps. The real problem is that with a comp, you are usually looking for it to keep the muzzle as still as possible. This si something that is really hard to measure, and the sled method gives very little indication of what performs well in this regard. For a round that actually beats the snot out of your shoulder, this would probably be meaningful, for 223/5.56 it isn’t really.

  • I honestly don’t know. I don’t talk with him that often but Steve may be able to find out.

    • limasierra1813

      I think he writes for lucky gunner labs

  • The Real Teal’c

    BattleComp sent assassins to deal with him.

  • Hey guys, Jeremy S. here (the guy in the video). Thanks for sharing the video on TFB and thanks for the commentary! Not sure if it’s a faux pas to mention a “competing” firearms blog or not, but please check out the associated write-up on TTAG. It includes a lot more information and larger image files of the data charts and graphs. The recoil test was only supposed to add a little bit of color to the rest of the write-up. Obviously recoil in 5.56 isn’t a huge deal so it isn’t always/often at the top of a prospective muzzle device purchaser’s requirements list.

    I don’t purport to be a scientist (though I am educated and much of my professional background is engineering- and R&D-related) and I don’t claim that this is a valid, lab-quality test or anything. There are lots of variables that I did not isolate and there are lots of things I could have attempted to measure that I did not. I kept it simple to get an idea and an approximation of rearward recoil energy. This may or may not equate to “felt recoil” as sharpness of recoil impulse and other things can contribute to that independently of how much actual total energy ends up pushing backwards.

    That said, I’m VERY happy with the consistency of the results. The difference between the minimum and maximum recoil distance for each muzzle device was super tight. Running bare muzzle and A2 over and over again, including at the beginning and then at the end of the test to make sure the results still aligned, produced very tight results groupings. When I took the extreme spread (difference between shortest and farthest result for every muzzle device in the test plus the bare muzzle) for each device and then averaged all of them, I was surprised and happy to see a result of 0.19″. With many of the brakes I could run them 10 times and every single shot would come in within 1/8″. The rig, no matter how silly and unscientific and Bubba’d it may look, *was* extremely consistent.

    I have kept a few of the brakes from this test to use as baselines on future tests, and will be running a second one likely in late Q1 with some popular ones that were missing plus any I can get my hands on that were requested by folks in comments on YouTube, on TTAG, or via e-mail to me (GunsAndGearEJ20 [at ] gmail). I’ll upgrade the test rig a bit with a sandbag as a weight and a remote trigger firing device, but it will be substantially the same. I do have an accelerometer on order and if it’s ready in time that will be part of the testing.

    Also in late Q1 I’m intending to do a flash hider-specific test. It will involve a better high-speed camera to capture screen grabs of each FH at its “flashiest” frame. I may use a pistol length barrel for that but I haven’t decided yet. I feel keeping it simple and just visually comparing images of each is going to be better than trying to use a light meter. I know that would result in actual data, but I will not be able to properly control ambient lighting to the necessary degree for it to be truly meaningful. We’ll likely keep it simple again and I expect the photos to be pretty demonstrative.

    If you have feedback, ideas, something you’d like to loan me for the next test, etc etc, please shoot me an e-mail (GunsAndGearEJ20 [ at ] gmail)!


    Jeremy S.

    • BattleshipGrey

      Thanks for doing the test. I can understand other’s comments about how unscientific it was and that the results don’t relate actual felt impulses, but for the cost of 100 rounds and your time, (perhaps some shipping as well) your test didn’t do any harm. As said, even if the results don’t relate to felt recoil, you at least had consistent results, and I believe it was an informative test regardless simply based where the gasses were forced and how the firearm itself reacted.

      I think those that would completely disregard your results are acting in haste. I think if we’re serious about finding a non-standard muzzle device for a 5.56, this is only a piece of the information we should consult since it only deals with movement on one plane. I, for one thank you for covering that plane.

      • BattleshipGrey

        Forgot to add; great idea on using a pistol length barrel on the upcoming flash hider test.

      • Well for the record, if I can add just a little to my novel above haha:

        — I do personally find that the sled results equate to subjective “shoulder dyno” felt recoil. While I did state that there are factors of felt recoil that the test may not be able to quantify, when actually shooting many of these things from the shoulder I could guess where they would end up relative to each other on the sled with pretty good accuracy.

        — Muzzle rise or dip forces can be seen to some degree in the slow-mo footage. For example, you can see the BattleComp overcompensate as the barrel literally flexes downwards. Watch the couple inches of barrel behind the muzzle device in all tests and you’ll get a good idea for vibration and reverberation/oscillation and flexing of one device against the next. The really awesome ones go straight back and the barrel doesn’t budge up or down at all (watch JP Recoil Eliminator, Precision Armament M4-72, and even many of the ones that placed way down the list for recoil reduction but are very good comps). Many of them make the barrel wobble a bit like a spring. You’ll see it hop upwards and then oscillate down below midline and then back up, back down…. As mentioned, the BattleComp actually bends the barrel down and then it oscillates after that. This is consistent with testing done by other folks (that this comp and others over-compensate and cause muzzle dip). So… the test was NOT designed to show rise or dip, but it actually does in the slow-mo at least to a certain degree. When shooting from the shoulder, the ones that don’t budge the barrel up or down at all in the video hold rock freakin’ steady in “real life” also. The ones that lift the handguard off the rest or otherwise flex the barrel up or down upon firing are doing so because they legitimately impart force in that direction, and you can also feel this when shooting from the shoulder.

  • Andy Leach

    Need this in article form. I’m not gonna try to “reference” a 26 minute video to customers asking about Brake A. vs Brake B.