Computer Flow Simulation of a Firearm Compensator: See How They Function in 1/10,000 of a Second

Today I stumbled across a handful of the most underrated videos on YouTube. Published in 2013, the video below (and five others like it) showcase a CAD simulation of the volume of escaping gasses, and the amount of force enacted on the individual baffles, on a muzzle brake in ten-thousandths of a second.

We all know engineering simulations like these exist in the world, but proprietary data is secretive and good engineers (especially firearms designers) are hard to come by. YouTube user mpep556 has provided us with a rare glimpse into the world of advanced firearms design.

The numbers are interesting to say the least. Baffle 1, naturally, is carrying most of the load but I am surprised to see that subsequent baffles carry very little by comparison. I’m also surprised at the 0:39s mark when we see a dip in force immediately after -or during- the initial spike. My first hypothesis was that the dip is the relief on the baffle as the bullet passes through, but the math doesn’t add up. A “.38 Super” indicated by the title screen should have approx. 1,050 fps at the muzzle, and should pass through at approx. 0.004 thousandths of a second assuming a 5″ barrel like the one shown below. I consulted the help of a Senior Military Aerospace Fluid Conveyance Technician (yes, really) and he informed me that this means everything before 0.004 would be compressed air being forced down the bore by the advancing bullet. So that doesn’t seem to explain the dip, but it could explain the timing. To confuse the issue further, baffles 2 through 4 show this phenomenon at increasingly later times. Perhaps someone can enlighten us in the comments. (Please?)

My own confusion notwithstanding, I’m thrilled that mpep556 shared these videos with us. Unfortunately it appears this user is no longer publishing or is on hiatus. I like to imagine he or she is off somewhere doing great things for the industry.

Les Baer Custom Inc. 1911 Ultimate Master Combat Pistol – Compensated

Corey R. Wardrop

Corey R. Wardrop is the Museum Curator for the Institute of Military Technology in Titusville, Florida where he manages one of the finest, if not the finest, firearms collections in the country. Corey is a former OIF infantry Marine and has worked professionally in the firearms industry for nearly 20 years. In 2014 he obtained an unrelated Bachelor of Science degree from one of the nation’s leading diploma mills. Through his work at IMT he is currently studying CAD design with an emphasis in reverse engineering rare firearms.
Corey asks forgiveness for his novice-level photographs and insists they are improving dramatically thanks to certified rockstar Corey can be reached at and always appreciates suggestions for future articles.
For the record, Corey felt incredibly strange writing this bio in the third person.


  • There is a US military study on the effects of muzzle brakes, granted its for artillery though it applies to small arms, that shows optimum area ratios of the openings. The study revealed that more than 2 baffles were not effective enough to justify having them. I always chuckle at muzzle brake designs that you see after reading the study since there are some insanely stupid designs out there that are more weight than usefulness.

    • James Earl Jones

      Im sure most of them are banking on the weight providing the majority of the recoil dampening and proclaiming it a proprietary miracle of gasamascience spectrogramophy.

    • PK

      Is this study from later than the 1960s? I’d be interested in reading it, if you’d be willing to share the AD# so I can snag it from DTIC.

      • PK, all I have is the few pages that were from the section in a book about firearms design. It summarized several studies and reprinted graphs and the summarized study data from studies to condense it to the relevant information. I did get a copy of the reference page though and here it is.

        • PK

          Perfect, thank you very much!

        • PK

          I looked to see if I could find the book this reference page came from, but I didn’t have any luck. While I do now have more reading material thanks to the list, if you’d be willing to share the title of the text it came from I’d greatly appreciate it.

          • Unfortunately I did not get the name of the book copied with material. I was only after the one bit of information when I was doing muzzle brake design for AAC and it was not my book. I looked and there was not any name of the book on any of the pages as some books like to do.

          • PK

            Nuts! Well, thank you for double checking, I appreciate it.

            If you ever happen to get a chance to ask about it, I’d love to know where it came from. Or if you search phrases, perhaps Google books has it on file and can find the title.

            I’m always looking to understand all of this stuff more, and your willingness to share information is a great help.

          • send me an e-mail and I’ll send you what I copied.

          • PK

            I’ve somehow managed to not write your e-mail address down, or I lost it between backups when I reinstalled my OS.

            Here’s my e-mail address, instead: p.t.kovacs81 (at) gmail (dot) com

            Thanks very much for your offer!

      • There are quite a few. Just search ” muzzle brake” on Google while limiting the search to DTIC’s domain.

        • PK

          Oh, I’m aware. I have tons and tons of studies and publications from DTIC, various printed textbooks I’ve purchased when they’ve been referenced, and so forth. If there was one study in particular as the final word on brakes then I wanted to make certain I had it, as I have yet to see one quite matching the description.

    • Justin Galt

      This is exactly right. People start making wacky brake designs as if there has been no scientific study of muzzle brakes for the last 100+ years. With a little bit of research, one can leverage all those .gov research dollars and make a properly designed brake. Fortunately most designs work to some degree and most brakes aren’t terribly expensive.

      For anyone who wants to design a proper brake, get on DTIC and start searching around. We’ve pulled gigs of files off of there, and is part of the reason our company has the word “research” in our name, Kinetic Research Group.

      For those who don’t have time to sort through it all, we’ll be doing an article on the fundamentals of brake design that we’ll put up on our site and in our newsletter.
      KRG (Kinetic Research Group)

    • RealitiCzech

      JP’s tank brake exists for a reason. Simple but highly effective.

    • iksnilol

      I dunno, the extra weight of extra baffles might be justified in the sense that the weight itself keeps the muzzle down.

  • armatus rebellio

    Note that while the velocity at the muzzle may be 1050 fps, the INITIAL velocity is 0 fps. As such the average velocity over the 5″ barrel is 525 fps (assuming the bullet acceleration is constant.) This means the bullet arrives at the muzzle 0.008 seconds after firing.

    • Corey R. Wardrop

      I totally agree with you, but for the sake of the article I kept it simple. Also, I’m not smart enough to calculate the acceleration of a bullet down a bore. I did ask an engineer if he thought my omission was relevant and he said “for the purposes you’re trying to convey, we can assume that ‘acceleration is instant’. 🙂

    • JamesDrouin

      Your conclusion is incorrect, the kinematics formulas are as follows. Alternatively, there are a number of open-access engineering websites where the data can be input and results viewed:

      v^2 = vo^2 + 2as
      t = (vf-vo) / a
      a = (v^2) / (2*s)

      Where (in S.I. units):

      vf = muzzle velocity
      vo = initial velocity of the bullet
      a = acceleration (average)
      s = barrel length
      t = time

      Thus, with a barrel length of 5 inches and muzzle velocity of 1,050 fps (320.04 m/sec), the bullet accelerates at 1,322,998 fps (403,250 m/sec) and is only the barrel for 0.000794 seconds.

      Further, as you implied, bullet acceleration is not constant, and the computer simulations that I’m aware of show that it’s much like a sin wave that is truncated at the middle of its peak. I.e., the bullet starts at zero, acceleration increases rapidly as more and more powder converts to a gas, tapers off as the force exerted by the expanding gas becomes ‘balanced’ in the available barrel volume, and then acceleration stops as the bullet exits the barrel.

      • armatus rebellio

        I used d = 1/2(Vf + Vi)t and solved for t.
        This way acceleration doesn’t need be calculated.
        where vf = 1050 fps, vi = 0, d = 5″
        5/(525*12) = 0.000794s (your answer.)
        So, yes, my final answer was wrong (shifted a decimal in my head*) but is still based on average velocity of 525 fps.

        *Given that ignoring Vi gives 0.000396, Corey probably slipped a decimal in his answer too.

  • Muzzle velocity of a 38 super open gun is going to be a lot faster than 1050fps. I would say that 1,200 fps would be a minimum with many going 1,500fps (as most open shooters use 115gr or 124gr since that gives you more gas to work with).

    As with exception of steel and Bianchi open guns, the target power factor for an open gun is 165+ with many loading to 170 and even up to 175 PF for the people shooting IPSC.

    • Dave Re

      That’s right… 1380fps with a 124gr bullet is going to be pretty typical in the US – a little slower internationally (lower Open division power factor for IPSC vs. USPSA). Most folks are going to load 5-7 power factor higher than minimum (so, 165-167 for IPSC, 170-172 for USPSA – and most USPSA folks run their USPSA load when they shoot IPSC).

      • Actually it is the other way around. IPSC Major PF is still 170 minimum. While USPSA lower it to 165. Any American shooting internationally has to bump their loads up a tad.

        • According IPSC’s own website, the Major PF minimum is only 160 for Open Division. 170 is the minimum Major PF for Standard Division, but .38 Super isn’t legal for scoring Major PF in Standard Division.

  • JSmath

    The spikes and dips are explained by the wavelike nature of the pressure forces. Go to 0:24, turn playback speed to 0.25x, then watch each baffle a few times to see how the higher pressure zones form close to the baffle first, then “ripple” a bit with other zones forming earlier in the brake before eventually norming out.

  • atmar

    this could be misleading….. you’ll are assuming this is factory loaded super 38….. ipsc/uspsa open division shooters DO NOT shoot factory ammo…… factory ammo will not give them 165pf they need to be scored major. they shoot reloaded stuff w/ slow burning powders …… a lot of it.

  • roguetechie

    In other words the Russians really knew what they were doing when they designed the 5.45×39 / AK74 combination.

  • Road

    Solidworks rocks. After HPA passes, this is what I’m going to use to model my suppressor build. Such an insanely powerful program.

    • Dirk Dasterdly

      Please, dear baby Jesus, I wish people would stop waiting for HPA!! Sure, I’m a dealer and have a vested interest in it, but you guys have WAYYY too much faith in your government! They’d have to give up control and money. Sure, Trump will sign it. But it’ll never get to him. It was introduced Jan 9 and put into committee (and sub-committee) and that’s where it’s been for 6 months. And that’s where it will die like most other bills. I think when I searched there are 61 firearms related bills in Congress right now. Nothing ever gets done. In fact, if you had done your Form 4 on January 9 when the bill went in, you’d almost surely have your can in just a few months from now…long before everyone else waiting. So here’s the net results:

      1. You pay the $200, do the form, and then HPA passes the next day. You cancel your application, get the $200 refunded, and get your can.
      2. You pay the $200. Get your stamp in a few months. Then HPA passes. Both versions of the bill have a provision to refund your $200 (House says all stamps since 2015; Senate bill says since Jan 9 2017). You get your $200 back and already have your can.
      3. You pay the $200. Get your stamp in a few months. Get your can. Your friends continue to wait another 2 years for HPA which never goes through. Each hunting season, your friends are begging you to add them to your trust so they can borrow your can. You charge both of them $50 per season for 2 years and make your $200 back.

      Remember, the democrats took 5 years to pass an ANNUAL budget. The Republicans now dominate and they cant even repeal Obummercare or build a wall. Good luck getting the libs to not filibuster something they’d see as making military grade weapons easier for criminals to get…and giving up millions per year in unearned taxes.

      We have the silencershop kiosk now, too, which makes the fingerprints and photos even easier. Buying a can now is almost as easy as pre-41F and with everyone waiting for HPA, I bet new applications are approved in 4 months. We’re just now getting the last of the July 12 applications back so it should go very quickly from here out.

      • Scott Goofus

        You make a solid case. The RINOs in Congress are so busy cooperating with the DEMONcRATS in trying to stymie anything that President Trump wants to accomplish that they will let HPA fade away along with anything else that would truly help those who elected them. What a rotten pack of POS’s they are. They allow the DEMONs to set the agenda and lead them by the noses, even though we have worked our tails off to give them the presidency (can’t blame Trump, he’s doing his best), and both houses of Congress. All for naught, folks. Better get those cans while you can, for the DEMONS will be taking control again soon and all bets will be off. And you’d better buy lots of guns, ammo and bury barrels as well, and get ready for the confiscation assault that the Marxists will push full force when they get back in power. As for me? Molon labe … molon labe.

      • Road

        1) I want to build my own because that’s just the kind of person I am. I don’t like to give a nickel to someone else for something I’m capable of doing myself. On top of that, I’d have a lot of fun in the process.

        2) As for giving up control and money, I totally see your point, as politicians are power-hungry narcissists, but the same could have been said about every single state that passed constitutional carry; no way they’re going to give up all that revenue and control! Yet it happened 12 or 13 times (or whatever the number is), and will likely happen at least a few more times. I do not have too much faith in my government- I have zero faith in my government, which is exactly the correct amount. Sometimes good things can happen.

        3) Principle. I’ve jumped through all the NFA hoops for a SBR before and I’ll never do it again. I’m not going to subject myself to that crap again for one simple reason: I shouldn’t have to. No one should have to. I’d rather not have a silencer at all, than jump through hoops again like a monkey for some bureaucrat.

        4) Silencers are waaaaaay too expensive. While I can certainly afford one, again, it’s the principle. They are very severely overpriced. I’m not going to buy an add-on that costs the same, or more than the gun itself. It’s a tube with baffles in it.

        Maybe HPA won’t pass. I don’t know. But to assume it won’t is the wrong attitude to have.

  • Dave Re

    Appears to be a Brazos Custom Gunworks Thundercomp 2 that they’re testing (the oval shaped side holes suggest it, anyway). Brazos makes two more comps – one with one port less, and one with one port more. Having shot them side by side, I can say that even though the last couple baffles don’t take much force, there’s actually a noticeable difference between the smaller comp and this one that’s being simulated. The longer one, however, doesn’t really have a noticeable change vs. this one. It’s not a night and day type of difference, either way.

    You can also get a sense of which baffles take more force, etc, just by looking inside one after a few thousand rounds… and eventually, that second baffle will take more force (say, after 50,000 rounds) just from erosion at the first baffle… LOL

  • JamesDrouin

    Simulations can be interesting but, as the good folks at IBM stated many decades ago “garbage in equals garbage out”.

    Kinematics calculations show that a bullet with a muzzle velocity of 1,050 fps is only in a 5 inch barrel for 0.000794 seconds, a tiny fraction of the 0.004 stated.

    • UpChuck.Liberals

      i.e. much like the GCC nonsense.

      From direct observation, I had one brake that did nothing and directed ALL of the sound energy back to me. My current ones, actually do reduce muzzle rise on my AR’s and 300BLK significantly. They are very similar to the ones in the simulation.

  • Jason Adams

    Simple it is a simulation not the real thing. I wrote a boat hull design CAD program once and spent months researching Froude’s law and Reynold’s fluid dynamics with real world testing of the results from the software and found that while close it was not possible to simulate all effects encountered. Close is all you get and anomalies will occur. So don’t get hung up on it.

  • House MD

    This was done in SW Flow Sim. This piece of software is desperately trying to be CFD, but only comes with one turbulence model, can’t handle big grids and isn’t particularly fond of boundary layers either.

    Its sole purpose in life is to help electronics companies with cooling designs and going by the talk on CFDOnline – it’s just about adequate for that.

    Not remotely suitable for ballistics though