AR Sponsor
[ARCHIVED THREAD] - DPM Recoil Reduction System vs. Hydraulic buffer EDIT: Armament LARB Mod 2 and 3 added to the test. (Page 4 of 4)
|
Quoted:
Still needs to be mounted on the left side rail power button facing the muzzle for live fire tests. The fluctuations were even worse with over the bore mount. Remember that the Mantis device supposedly saturates on the X axis, the one we need data from. By the way I have 2 SeeAll Tritium sights on the way from the OEM. One with a crosshair and one with a delta triangle. They are interesting so I will send them around to a couple testers to give them a workout. The old design in the first post was pretty clunky but the new Mk2 looks good. Thread started by another member is here: SeeAll Tritium Sight Quoted:
Still needs to be mounted on the left side rail power button facing the muzzle for live fire tests. The fluctuations were even worse with over the bore mount. Remember that the Mantis device supposedly saturates on the X axis, the one we need data from. By the way I have 2 SeeAll Tritium sights on the way from the OEM. One with a crosshair and one with a delta triangle. They are interesting so I will send them around to a couple testers to give them a workout. The old design in the first post was pretty clunky but the new Mk2 looks good. Thread started by another member is here: SeeAll Tritium Sight Remember that the Mantis device supposedly saturates on the X axis, the one we need data from. Regardless, I'm wondering if the Mantis would be OK for something like the 9mm CMMG RDB. I've asked for clarification and waiting for a response. |
|
Quoted:
Got it.... As Mantis mentioned both are using the same Accelerometer hardware so that is interesting. Regardless, I'm wondering if the Mantis would be OK for something like the 9mm CMMG RDB. I've asked for clarification and waiting for a response. 
Should not be long before their tech nerds incorporate a raw data bluetooth download for users to evaluate on their own. I think they are keeping their proprietary data analysis methodology close to the vest for now. |
|
Quoted:
I used the app and anchored my phone to the stock since I was interested in felt recoil. I have the data for the LARB buffer tip. The image you posted on page one reminds me that peak values in raw data don't quantify felt recoil. Eric also pointed out it's the area under the curve that matters. To find the area in the charts on page 3 I would have to add all the + and - areas of the primary impulse. I am not that sharp with the math. The only conclusions I can make from my charts (without doing the math) is that the hydraulic buffer settles oscillations faster than the T3, and time to battery is the same. They felt the same in single shots. I picked the hydraulic over the T3 because it lowered the cyclic rate in full auto about 50 rpm. How can the rpm be lower with the hydraulic buffer when the timeframe from shot to battery is the same for both buffers? Attached File |
|
Not being a Mechanical Engineer, please take what I say we a grain of salt. I have a very different take on the data posted on the previous page. There is a slight negative G just before "A". To me that is the firing pin being struck by the hammer. The recoil "A" is as high or higher than "B", but "A" is a much longer duration than "B". To me "A" is the round going off. Also think about how the accelerameter is mounted. It is firmly attached to the upper. The upper is attached to the lower via push pins. There is bound to be some slop in that attachment. Then it is the lower which is actually held by the lead sled. The accelerameter is giving data on the upper, not the complete firearm. So to me the posted data has the slight negative acceleration of the firing pin being struck by the hammer. Then the big longer wave of acceleration back which would be "A". But like an op rod in an AR piston system, the push from the upper into the lower, causes the upper to rebound back in the slop between the upper and the lower. But the rebound has no where near the energy coming back. That is why the rebound is more of a quick spike because the duration is so short. And like the guitar string, the upper rebounds back and forth until it settles. Again I am not an engineer, but to me there is no way that the hammer strike could cause anywhere near the amplitude ("A") of acceleration that the round going off ("B") in the privious expansion could IMHO. To me the acceleration we are most interested in is up and down (muzzle rise) not so much the recoil back. Th recoil back is important. But staying as close to "on target" for follow up or burst fire I thought was the major point of this experimentation. Of course the further from the pivotal point (the back end of the stock) the greater the amplitude of the movement. I am not really sure how that movement should be quantified. YMMV. Scott |
|
Quoted:
Not being a Mechanical Engineer, please take what I say we a grain of salt. I have a very different take on the data posted on the previous page. There is a slight negative G just before "A". To me that is the firing pin being struck by the hammer. The recoil "A" is as high or higher than "B", but "A" is a much longer duration than "B". To me "A" is the round going off. Also think about how the accelerameter is mounted. It is firmly attached to the upper. The upper is attached to the lower via push pins. There is bound to be some slop in that attachment. Then it is the lower which is actually held by the lead sled. The accelerameter is giving data on the upper, not the complete firearm. So to me the posted data has the slight negative acceleration of the firing pin being struck by the hammer. Then the big longer wave of acceleration back which would be "A". But like an op rod in an AR piston system, the push from the upper into the lower, causes the upper to rebound back in the slop between the upper and the lower. But the rebound has no where near the energy coming back. That is why the rebound is more of a quick spike because the duration is so short. And like the guitar string, the upper rebounds back and forth until it settles. Again I am not an engineer, but to me there is no way that the hammer strike could cause anywhere near the amplitude ("A") of acceleration that the round going off ("B") in the privious expansion could IMHO. To me the acceleration we are most interested in is up and down (muzzle rise) not so much the recoil back. Th recoil back is important. But staying as close to "on target" for follow up or burst fire I thought was the major point of this experimentation. Of course the further from the pivotal point (the back end of the stock) the greater the amplitude of the movement. I am not really sure how that movement should be quantified. YMMV. Scott Always appreciate your input. I just checked my upper and lowers I tested and I don't have any detectable front/back slop. If I try to twist the upper and lower there is some slop there. I hear what you're saying but with a good fitting upper/lower, I wouldn't think it would be that big of a deal. I would presume that if the data is valid we would see what you are talking about but it wouldn't/shouldn't show up as some considerable spike. Regardless, as mentioned in my previous posts, I'm skeptical on the numbers...like you stated, I have a hard time believing the hammer strike could be so high in comparison to the round going off. I've been continuing to have conversations with Mantis and trying to get further clarification on some things but at this time, it seems that they think using the consumer grade accelerometers for measuring recoil is invalid data. They do say their gryoscope data which will give us muzzle rise is valid and repeatable data. So I'm looking forward to collecting the data from both devices at the same time to see if we can get something subjective. I'm supposed to get the accelerometer back from HDSledge today and depending on when/if I get it, will get some readings doing some dry fire testing on the MK18 and we'll see what that looks like. |
|
Quoted:
Not being a Mechanical Engineer, please take what I say we a grain of salt. I have a very different take on the data posted on the previous page. There is a slight negative G just before "A". To me that is the firing pin being struck by the hammer. The recoil "A" is as high or higher than "B", but "A" is a much longer duration than "B". To me "A" is the round going off. Also think about how the accelerameter is mounted. It is firmly attached to the upper. The upper is attached to the lower via push pins. There is bound to be some slop in that attachment. Then it is the lower which is actually held by the lead sled. The accelerameter is giving data on the upper, not the complete firearm. So to me the posted data has the slight negative acceleration of the firing pin being struck by the hammer. Then the big longer wave of acceleration back which would be "A". But like an op rod in an AR piston system, the push from the upper into the lower, causes the upper to rebound back in the slop between the upper and the lower. But the rebound has no where near the energy coming back. That is why the rebound is more of a quick spike because the duration is so short. And like the guitar string, the upper rebounds back and forth until it settles. Again I am not an engineer, but to me there is no way that the hammer strike could cause anywhere near the amplitude ("A") of acceleration that the round going off ("B") in the privious expansion could IMHO. To me the acceleration we are most interested in is up and down (muzzle rise) not so much the recoil back. Th recoil back is important. But staying as close to "on target" for follow up or burst fire I thought was the major point of this experimentation. Of course the further from the pivotal point (the back end of the stock) the greater the amplitude of the movement. I am not really sure how that movement should be quantified. YMMV. Scott The small peak at the beginning seems more representative of the hammer fall vs force of recoil at A and oscillation at B. On my system there is some play between upper and lower, not much but some. I didn't consider that to account for the extra data noise. (Edit: Wrong) The enclosure was on there tight this time since I tightened/blue loctite/then epoxy encapsulated the screws holding the enclosure to the bracket. Plus that side of the enclosure is relatively thick and not flimsy. All components are wedged in tight and hot glued in place especially the accelerometer module board, which is parallel to and only a quarter inch from the rail. Without serious number crunching X axis data only tells us how fast recoil settles down in that axis and how long to get back into battery. Calculating X axis recoil is more complicated than noting peak values. I'll post a chart of the Y axis (muzzle flip) charts in a bit. I suspect Amphibians' dry fire test will show you are right about the initial peak. Bolt rifle live fire test could also help to answer the question of whether the acc. mount or the AR system is causing the oscillations. There is more to these tests than I ever dreamed of. What an education I am getting. |
|
Thank you for the very kind words. That you would consider my musings is a huge compliment. Thank you. You have done so much for this community. I very much appreciate your insights. It is my understanding that an explosion inside the shell (the primer igniting the power burning making a large volume of gas) with the result being a projectile leaving the muzzle with approximately 1,000 ft./lb of energy. There is no way that the hammer striking the firing pin could come that close to accelerating the upper anywhere near what the round going off would do. Just like hitting a nail with a hammer, the hammer does bounce back. Even with what seems like a tight fit, the upper is accelerated back, hits the interface with the lower, and like hitting two hammers together, the upper bounces off the lower. I would think that the bouncing back and forth for several cycles would tend to happen. I don't claim to know what the data is showing. But the upper recoiling off the lower bouncing backs and forth would seem to me more plausible than the hammer striking the firing pin accelerating the upper forward with almost the energy of the round going off. It would be interesting to see both the dry fire and two rounds in auto to compare. My understanding of the graph is the area inside the peak is the total energy. The area inside the first big peak looks larger than the area inside the second peak, even though the peak is not as high. The explosion happens over time. Very little time, but I think that is why the first big peak is wider than the area under the second peak. The upper hits with increasing power behind it. Where as the recoil hits and rebounds back without continuing energy added on. I don't know as that is what the graph shows, but it makes a good story for the graph. Scott |
|
Quoted:
Thank you for the very kind words. That you would consider my musings is a huge compliment. Thank you. You have done so much for this community. I very much appreciate your insights. It is my understanding that an explosion inside the shell (the primer igniting the power burning making a large volume of gas) with the result being a projectile leaving the muzzle with approximately 1,000 ft./lb of energy. There is no way that the hammer striking the firing pin could come that close to accelerating the upper anywhere near what the round going off would do. Just like hitting a nail with a hammer, the hammer does bounce back. Even with what seems like a tight fit, the upper is accelerated back, hits the interface with the lower, and like hitting two hammers together, the upper bounces off the lower. I would think that the bouncing back and forth for several cycles would tend to happen. I don't claim to know what the data is showing. But the upper recoiling off the lower bouncing backs and forth would seem to me more plausible than the hammer striking the firing pin accelerating the upper forward with almost the energy of the round going off. It would be interesting to see both the dry fire and two rounds in auto to compare. My understanding of the graph is the area inside the peak is the total energy. The area inside the first big peak looks larger than the area inside the second peak, even though the peak is not as high. The explosion happens over time. Very little time, but I think that is why the first big peak is wider than the area under the second peak. The upper hits with increasing power behind it. Where as the recoil hits and rebounds back without continuing energy added on. I don't know as that is what the graph shows, but it makes a good story for the graph. Scott EDIT: The hammer fall impulse issue has me puzzled so I decided to chart both the 150ms data sets from both buffers together. The small first blip assumed to be the hammer fall is evident with the RB5005 but doesn't show up with the T3 buffer. More questions than answers so far. I am reminded that the buffers felt about the same to me at the range. Charts are very close with the RB5005 looking a lot smoother overall. (X axis) Attached File |
I got the accelerometer and did two dry fire tests w/ my MK18 as pictured below....it was done just as pictured below...on the concrete...sensor mounted to the side as instructed by HDSledge.
  And....  So nowhere near the -12 or -13 for hammer fall. I'm sure HDSledge will make nicer looking graphs than what I yanked out. I'm thinking to do a live fire and shut the gas off so it won't cycle and see what the data looks like. |
|
Quoted:
I got the accelerometer and did two dry fire tests w/ my MK18 as pictured below....it was done just as pictured below...on the concrete...sensor mounted to the side as instructed by HDSledge. http://c3junkie.com/wp-content/uploads/2019/12/MK18-dryfire-sled-1024x516.png http://c3junkie.com/wp-content/uploads/2019/12/MK18-dryfire1-1024x362.png And.... http://c3junkie.com/wp-content/uploads/2019/12/MK18-dryfire2.png So nowhere near the -12 or -13 for hammer fall. I'm sure HDSledge will make nicer looking graphs than what I yanked out. I'm thinking to do a live fire and shut the gas off so it won't cycle and see what the data looks like. |
|
Quoted:
Good idea. Now that I have seen the data, it is clear that what Eric said about the device sensitivity is correct. It catches everything in extreme high resolution. I am floored that hammer fall can generate so much acceleration in the weapon system. I'll have the chart up in a few minutes. |
|
Something is not adding up for me. Please help me with the math. The acceleration due to gravity is 32 ft. per second for every second of travel until an object reaches terminal velocity. We are only looking at one millisecond. According to Google 32 ft. per second (the acceleration due to gravity) for one millisecond would be 9.7536 millimeters in one millisecond. To me there is no way that just dropping the hammer would accelerate the upper 3 Gs. That would mean that in one millisecond the upper would go from a dead stop to moving over 29 millimeters or over 1.1 inches just from dropping the hammer. As pictured that gun didn't move 1.1". So would someone please explain how I got the math wrong, because this is not making sense to me. Scott |
|
Quoted:
Something is not adding up for me. Please help me with the math. The acceleration due to gravity is 32 ft. per second for every second of travel until an object reaches terminal velocity. We are only looking at one millisecond. According to Google 32 ft. per second (the acceleration due to gravity) for one millisecond would be 9.7536 millimeters in one millisecond. To me there is no way that just dropping the hammer would accelerate the upper 3 Gs. That would mean that in one millisecond the upper would go from a dead stop to moving over 29 millimeters or over 1.1 inches just from dropping the hammer. Scott Ok here is a chart of the dry fire test and a half assed explanation (sort of) for why my rig test charts show a much bigger initial acceleration. Chart is 100% hammer, firing pin, rifle, and sled interaction, with maybe a bit from finger or lever pulling the trigger. EDIT: Amphibian's Mk18 10 shot test today shows the same pattern as my tests. Big initial impulse. Equally large area under curve for return impulse but higher peak. First impulse = hammer fall + shot fired recoil. In any case the sled tests answered the question about hammer fall forces. They are bigger than we expected. Barrel vibrates up and down like a guitar string. In Amphibians' dry fire test the hammer fall impulse oscillations last about 70ms, half as long as the recoil impulse from firing a live round. My interpretation of the peaks: 7ms: Hammer in forward motion, pushing rifle back. 11ms: Hammer strike pushes rifle forward. 14ms: Sled pushes back against rifle, jarring it back. Etc. down the line exchanging momentum all the way. I'll bet that little jog at 12ms is the firing pin absorbing some energy on the return. I'm fairly confident now that vibration of the device is not causing much trash data if any. Attached File Ps. This is fun (even though several of my assumptions have been proven false.) 
|
|
Quoted:
You guys are awesome! Thanks for doing this stuff. Let me know how/where a schmuck like me can contribute. We have gotten a ton of assistance from Eric at Ammolytics on the use of his accelerometer which he provided free of charge. He has literally spent hours with me in back and forth comms coding the device and helping me to interpret the data. He would be grateful for your support through Patreon for a buck or 2 a month (you can cancel anytime). Google Patreon Ammolytics. |
|
First I would like to thank you guys for this effort. I think the concept of quantifying recoil/muzzle rise would be so cool. It would give a reference point and take some of the subjectiveness out of perceived recoil/sights remaining on target. I get that the firearm would absorb some of the recoil energy, but to me this data is way off. Compare the amount of energy involved. The accelerometer is saying the hammer dropping causes the firearm to accelerate at the rate of three times the acceleration due to gravity. And the round going off causes the gun to accelerate at 18 times the acceleration due to gravity. So are you saying that acceleration is a very nonlinear scale, or firing the gun with green tipped ammo has six times the energy of dry firing the gun? I can't believe that the hammer falling would create 1 ft/lb of energy and the bullet leaving the muzzle has over 1,000 ft/lbs of energy. So equal and opposite reaction would be 1,000 ft/lbs of recoil energy minus the energy absorbed by the mechanism and the inertia of the gun. That doesn't make sense to me. The gun and mechanism are not absorbing hundreds of ft/lbs of recoil. To me these events cause vibration which the accelerometer interprets as acceleration. So the round going off causes six times the vibration of the dry fire not six times the energy. Again I am not a mechanical engineer but there is no way in my mind that the gun firing has only six times the energy of dry firing the gun. Please someone explain to me how this works. Scott |
|
Before I post a data chart I have a question: Is the circled area a pivot point?
Attached File |
|
Quoted:
Before I post a data chart I have a question: Is the circled area a pivot point? https://www.AR15.Com/media/mediaFiles/374164/AMphibian_Live_Fire_tests_sled_edited_png-1190658.JPG |
|
I can try to answer some of your questions (I don't check this forum often)
Responses inline Response 1/2 (because of forum character limits) Quoted:
Compare the amount of energy involved. The accelerometer is saying the hammer dropping causes the firearm to accelerate at the rate of three times the acceleration due to gravity. And the round going off causes the gun to accelerate at 18 times the acceleration due to gravity. So are you saying that acceleration is a very nonlinear scale, The point is that 1G vs 16G doesn't mean all that much unless you also factor in the mass of the object being accelerated and for how long. You can easily find the equations for and examples of acceleration and velocity online. |
|
Response 2/2
Quoted:
... or firing the gun with green tipped ammo has six times the energy of dry firing the gun? I can't believe that the hammer falling would create 1 ft/lb of energy and the bullet leaving the muzzle has over 1,000 ft/lbs of energy. So equal and opposite reaction would be 1,000 ft/lbs of recoil energy minus the energy absorbed by the mechanism and the inertia of the gun. That doesn't make sense to me. The gun and mechanism are not absorbing hundreds of ft/lbs of recoil. Now that we're discussing energy instead of acceleration, I first want to clarify that the accelerometer does not record the former. Also, I think it's helpful to clarify the terms: One foot-pound of energy is defined as the kinetic energy of dropping a one pound weight from a distance of one foot. The variables are Mass, Distance, and Acceleration. An M855 round uses a 62gr bullet w/ an approx muzzle velocity of 2900 ft/s (16" barrel). Knowing this, the JBM calculator provides an initial energy of 1157.6 ft-lb for the bullet. It was a low-mass object accelerated quickly from an explosion. Because the bullet is decelerating through air, it's energy falls to 261.8 ft-lb at 600yd. Energy is mass and motion. The rifle's hammer, being steel and larger, has more mass than the bullet, but is accelerated slower -- it then impacts the firearm, which is not dampened, so the energy transfers directly (like one of those office toys with the swinging ball bearings). When a round is fired and the bolt-carrier moves back, it is dampened by the buffer weight and spring. The firearm, especially in a sturdy rig, is significantly more massive than the bullet. JBM has a recoil energy calculator and they provide the equations so you can do it by hand if you wish to learn more. I hope this helps! |
|
Quoted:
Yes...it is a pivot point....do I need to clamp that too??....this is getting crazy... If it pivots the arm and spring are probably playing a big part in acceleration. Add a 4 shot data run shoulder fired with a bag rest when you go out to familiarize with the Mantis. 4 rounds in the mag so the last shot holds the bolt open. No Gas no Cycle Tests, Sled: If I have it straight the bolt was jolted out of battery a bit in each of 3 shots. Here is one dataset: Attached File Looks like the bolt is still xfering momentum to the buffer even though it only moved a bit. Or the sled is adding vibration through the spring arm. Here is a chart of one of the 10 shots from your last outing. I'll compare it with the shoulder fired bag rest example when you get a chance to get that to me. Attached File A area (not peak) = hammerfall + ignition recoil rearward B area = return force forward as bolt is accelerated into battery C area = sled reverses system momentum to the rear again and bolt hits home D area = last big forward acceleration E area = last big rearward acceleration (The larger area at E may be from the bolt bouncing back a hair into the buffer.) Here is where the high speed videos of the cutaway tube could help us to see what is going on during the shot. |
|
Quoted:
@ammolytics , what’s your take on the oscillating that’s in the graphs? And why is it different from the graphs on your blog? Thanks in advance. The other thing to note is that these sensors always have a bit of noise no matter what. It could be sitting on a thick granite surface plate in a temperature controlled bunker and it'd still be registering a very small amount of "movement" on each axis. |
|
Quoted: I was firing a bolt-action and this chart is showing a semi-auto with a buffer weight and spring in the rear. You're likely seeing the bolt carrier cycle (initial impulse, secondary from spring bottoming, then sending it home) followed by the buffer weight bouncing against the back of the carrier after it returns to battery (or locks back, whichever). Springs don't just stop moving. The other thing to note is that these sensors always have a bit of noise no matter what. It could be sitting on a thick granite surface plate in a temperature controlled bunker and it'd still be registering a very small amount of "movement" on each axis. |
|
Amphibian sent me data from 5 shot tests with the sled arm clamped to prevent pivoting, and a 5 shot test shoulder fired using a bag rest. I suspected from a previous data set that the sled was prolonging and maybe amplifying vibration. It apparently does. Data appears to show that the sled causes the vibration to last longer while the shoulder fired bag rest method lessens vibration time (a lot) and magnitude (a bit).
Amphibian is planning to use the sled for videos since it's easier to frame with the Mk18 clamped down. I recommend shoulder fired bag rest tests of 5 shots each for data collection on the buffer comparisons. A firearm engineering friend of Amphibian recommended faster readings at 40khz (we are at 1khz) and the ability to record 1,200 G. I agree it would be nice to get faster readings but I disagree with the 1,200 G assessment. Sounds way too high to me but I am no engineer. Comments on the 1,200 G recommendation from you guys who have more experience with measuring recoil? Attached File Attached File FYI the folks at MantisX say their device records in the 8 G range currently. It saturates at 8 G in the X axis (like ours did until we tweaked it) so their data is invalid for measuring recoil into the shoulder. They are planning on improvements in future versions such as an upgrade to max G capability and perhaps an API to allow download of raw data so users can chart it any way they like. I'll be watching for that since their tiny device is bluetooth capable and the battery charge lasts 8 hours. It is a great product already to train with re. pistol use. It can help you determine flaws in trigger pull and other factors/habits that can put you off target. I'll post a closer look at a couple of these test shots tomorrow. UPDATE: A closer look shows the sled causes more muzzle vibration than with shoulder fired bag rest. Makes sense since both dampen vibration faster. About the same recoil force to the rear. Attached File Attached File |
|
We are finally ready for buffer comparison tests. Amphibian suggested the following 6 tests with the accelerometer, then we check the data to make sure it is good for relative comparisons. If so more detailed tests on the configs of DPM RRS, etc. follow.
@Amphibian chime in if anything has changed: ------------------------------- 1. Standard milspec config - H2 buffer and white sprinco spring 2. H2 buffer and 556 Tubb spring - to see if we can pick up any difference with the spring 3. H2 buffer w/ EAB tip and 556 Tubb 4. LARB w/ 556 Tubb 5. DPM w/ 556 Tubb 6. RB5005 w/ 556 Tubb All the above would be using the carbine buffer tube to minimize changes. Get cyclic rate in full auto without the sensor for all the above so I can get a feel for each one while recording the rate. ----------------------------- All the above shoulder fired with bag rest, 5 shots each. Amphibian says he will use the sled for high speed videos since it is easier to frame the action. |
|
Quoted:
Looking forward to the results! Merry Christmas everyone! |
|
Quoted:
Me too but we already know which buffer feels better so far with Amphibian's system. I'll be surprised if anything changes with further testing. I won't get ahead of Amphibian on the reveal but so far it's not a close call. Merry Christmas everyone! |
|
Quoted:
Yes...it is a pivot point....do I need to clamp that too??....this is getting crazy... Quoted:
Quoted:
Before I post a data chart I have a question: Is the circled area a pivot point? https://www.AR15.Com/media/mediaFiles/374164/AMphibian_Live_Fire_tests_sled_edited_png-1190658.JPG And I think you're starting to see some of why that never took off for me. 
|
|
Quoted:
I’ll be honest, I believe this too. Amphibian has already made it clear as to his bias even before any testing was done. I’m still curious if the results you record, but I can assume the outcome will be what he already decided. I can’t blame him, I’m biased as well. But it is what it is. Quoted:
I’ll be honest, I believe this too. Amphibian has already made it clear as to his bias even before any testing was done. I’m still curious if the results you record, but I can assume the outcome will be what he already decided. I can’t blame him, I’m biased as well. But it is what it is. We are both grateful for the support that OEM's like you have provided by supplying parts for the comparison tests. Each buffer has it's own advantages. I'll be trying them out myself in a stock M&P Sport when Amphibian is done with his tests. Quoted:
A few years back I looked into setting up an accelerometer to measure the forces on a scope and mount during recoil. And I think you're starting to see some of why that never took off for me. |
|
Quoted:I’ll be honest, I believe this too. Amphibian has already made it clear as to his bias even before any testing was done. I’m still curious if the results you record, but I can assume the outcome will be what he already decided. I can’t blame him, I’m biased as well. But it is what it is. |
|
Quoted: Amphibian isn’t affiliated with any buffer or buffer spring manufacturer, but performs exhaustive comparative testing of both and in combination with each other, documents the tests in text, pics and video, here and on his website, explains the methodology behind the testing, illustrates the tools and equipment used for testing, currently provides additional subjective impressions, and is in search of multiple measurement devices in order to remove the subjective impressions, so as to just provide the facts for others to make their own decisions. And you think he’s biased? That doesn’t make sense to me. That said, you are the manufacturer of the LARB buffer, right? |
|
Finally a decent explanation from an experienced member about different types of recoil forces and how the data ranges can be so different yet still be valid.
================ There are two models for calculating recoil, and there's two kinds of recoil to consider. 1) Weapon Shock - This is the high G stuff. 2) General momentum - This is the stuff that hits you in the shoulder, and is probably what you're measuring. Also you can use a basic model in which you just consider the total mass of the rifle, but that's only accurate to within about 50% and tends to underread. (SAAMI Free Recoil Formulae I found this and think that is what he is referring to.) Or you can calculate the shock moving through the rifle at the speed of sound ( in steel ) - which would be very accurate. Big rounds like 50cal are up around 4000 G's. (@Amphibian your friend is probably referring to this type of measurement.) ============== I am beginning to understand the myriad ways recoil can be calculated and that different methods yield different ranges with the same unit of measure. |
|
Quoted:
Anything new? There were some problems with one buffer when firing in full auto. May be out of spec or due to Amphibian's system quirk with that buffer. Until I get a chance to try it myself we've decided to wait on full auto results/videos. If I also have a problem with the buffer I'll contact the vendor to discuss it and get a replacement to try. Amphibian tested only the out of the box setup with the DPM RRS buffer, which has multiple configurations. Since it hasn't gotten a thorough test I will continue with that one using a stock M&P 15 Sport. I plan to run another round of short tests for each buffer to satisfy my personal curiosity to experience each buffer firsthand. I'll post the data charts later today. I haven't drawn any conclusions yet from the data and for now am leaving it open to interpretation. Many thanks to Amphibian for his work. |
|
Here are the charts from the 11 tests done with the MK18. There is much more data from Amphibian's tests as you can see from the schedule.
The standard blowback 9 tests with the standard carbine buffer vs. JC SCS vs RB5007 should be interesting to see charted. I'll start working on that data. The test schedule: Test Number 1MK18H2White Sprinco7:47 AM5LC M855 2MK18H2556 Tubb7:545LC M855 3MK18H2 w/ EAB Tip556 Tubb7:565LC M855 4MK18LARB556 Tubb7:595LC M855 5MK18DPM556 Tubb8:025LC M855 6MK18RB5005556 Tubb8:035LC M855 7MK18RB5000556 Tubb8:065LC M855 8MK18RB5000L556 Tubb8:145LC M855 9MK18A5 H2556 Tubb8:205LC M855 10MK18LARB Mod 3556 Tubb8:225LC M855 11MK18RB5007556 Tubb8:245LC M855This test was done twice, at 9:21 and 9:26, I did it a second time as I was concerned did test too fast. Previous test was done at 9:20. 12MK18RB5007556 Tubb8:285LC M855This was the same test as above done at 9:26AM 1312.5 Mid GasRB5007556 Tubb8:355LC M855 14CMMG RDB 9mm, mod 5.45RB5007556 Tubb8:465124 Gr Geco 15CMMG RDB 9mm, mod 5.45RB5007300 BLK Tubb8:575124 Gr Geco 16CMMG RDB 9mm, mod 5.45RB5005300 BLK Tubb9:015124 Gr Geco 17CMMG RDB 9mm, mod 5.45RB5005556 Tubb9:065124 Gr Geco 18CMMG RDB 9mm, std boltstd car bufferWhite Sprinco9:115124 Gr Geco 195" Straight Blowback 9mm9mm JP SCSNA9:215124 Gr Geco 205" Straight Blowback 9mmRB5007308 Tubb9:275124 Gr Geco3.5 oz steel Colt bolt weight added 215" Straight Blowback 9mmRB5007308 Tubb9:315124 Gr Geco6.1 oz tungsten SAW bolt weight added 22CMMG RDB 9mm, mod 5.45RB5005556 Tubb9:415124 Gr GecoModified carrier key to match Surefire SOB stroke length 23CMMG RDB 9mm, mod 5.45RB5000556 Tubb9:455124 Gr GecoModified carrier key to match Surefire SOB stroke length 24CMMG RDB 9mm, mod 5.45RB5000L556 Tubb9:473124 Gr GecoModified carrier key to match Surefire SOB stroke length, Stovepipe malfunction, matches CSV file 25CMMG RDB 9mm, mod 5.45RB5000L300 BLK Tubb9:505124 Gr GecoModified carrier key to match Surefire SOB stroke length 26CMMG RDB 9mm, mod 5.45RB5000300 BLK Tubb9:515124 Gr GecoModified carrier key to match Surefire SOB stroke length 27CMMG RDB 9mm, mod 5.45RB5005300 BLK Tubb9:535124 Gr GecoModified carrier key to match Surefire SOB stroke length Attached File Attached File Attached File Attached File Attached File More in next post... |
|
Attached File Attached File Attached File Attached File That's A5 not AS. Attached File |
|
Attached File Attached File I've noticed some interesting characteristics in the charted data. Although the differences in buffer performance may not be discernible to the operator during the short time period of recoil, it is obvious that some of the buffers handle dampening the energy better/quicker than others. Felt recoil varies a lot from person to person and system to system. Looking forward to a trip (or 3) to the range to compare notes with Amphibian. EDIT: Check out how the EAB tip affected the H2 buffer in charts 2 and 3. A definite improvement. Y'all chime in with observations. If anyone wants the raw data let me know and I'll post the data files somewhere. |
|
There is some difficulty associated with interpreting the charts as the millisecond intervals are often different.
I did a bit of basic Paint editing to get the first 3 charts to scale, here's what it looks like: https://imgur.com/a/xUVBTSD Results appear...inconclusive... At a glance, the overall x recoil impulse looks to be similar between all 3 setups, although their characteristics seem to be somewhat different, with the standard having sharper spikes and the Tubb having more of them. The EAB buffer tip seems to change the x component of recoil to something of a combination of the Tubb w/ ordinary H2 and the standard control, arguably better than the other 2. Gauging the y component of recoil is rather difficult..but it looks like the EAB may actually be the worst? Wut??? Edit: Upon further inspection, it seems the standard actually has the most desirable muzzle flip characteristics, which is kind of the opposite of what you'd expect. Edit 2: Actually, maybe not. The Tubb and the standard look to be roughly on par overall. |
|
Here are scaled comparisons of the Tubb w/ standard H2 vs LARB, and Standard spring vs LARB. I assume the "LARB" graphs were only for the buffer cap, and not the entire LARB buffer setup.
https://imgur.com/a/TjnRujZ https://imgur.com/a/eUlqQ3o However, the Tubb spring w/ LARB tip H2 is still only about as good as the standard spring & standard buffer. In sum, the tests would seem to suggest the following: 1. The standard spring is superior to the Tubb flatwire for staying on target, although the felt recoil is not necessarily better or worse. 2. The LARB buffer tip has little effect on felt recoil, but does reduce muzzle movement. 3. The EAB buffer tip softens felt recoil slightly, but at the cost of increased muzzle movement. @amphibian @HDsledge Do these conclusions seem consistent with your findings? Edit: Actually, I took another look at the graphs, and I'm unsure again. The Tubb and the standard actually look to be pretty similar overall on y movement, and I am also not sure the LARB has any substantive effect. In fact, it seems to increase the magnitude of the "aftershocks." My impression of the EAB buffer cap, on the other hand, remains. Also, every other shot from the A5 seems to practically be a squib load in terms of muzzle rise, even though felt recoil is normal and all the other shots are within accepted bounds. Edit #2: Here's two of the hydraulic buffers. 5005: https://imgur.com/a/qsVhead 5007: https://imgur.com/a/LSZMGty While both afford reductions in felt recoil, the 5005 has little, if any, muzzle movement reduction. I am starting to think there is a very high degree of error involved in the y aspect of the recoil measurements and/or there are confounding variables that differ noticeably from test to test. Some of these graphs seem kinda odd. |
|
The graphs of 4-5 shots don't allow for a detailed comparison. I'll work on single shot graphs for each buffer with a 100ms window. Can't conclude much from the charts as they are.
Some buffers split the recoil effect into 2 distinct impulses. Does that result in less felt recoil or is it too fast to discern over buffers that spread it out over a few more milliseconds? The H2 buffer showed the shortest impulse period, probably yielding the sharpest felt recoil. I suppose the theory with buffer tech is that the longer you spread out the impulse the less you feel it, hence reciprocating masses/springs, hydraulics, etc. The charts show that all the new buffers achieve the same effect and only yield slight differences in felt recoil according to Amphibian. How they handle the energy exchange is the key. It's not about how they cancel the impulse, but how they spread it out. No comment from Amphibian on ranking specifics other than hydraulic feels better. My curiosity won't be satisfied until I try them all myself though. |
|
A longer recoil impulse of equal overall momentum is generally a good thing, as it gives more time for gravity and shooter stabilization to counter movement. A spike of acceleration with equal height over a longer period of time would generally be detrimental, as it means the recoil velocity would continually increase over a longer duration. This is one of the difficulties when dealing with acceleration-it's not quite as intuitive as simple velocity. If acceleration drops from 10 Gs to 1 Gs, the recoil is not settling or slowing down yet: Rather, it is continuing to get faster and worse, it's just not worsening at as quick of a rate as it was before. But, because it was picking up speed the whole time, the gun is actually moving much faster after dropping to 1 G acceleration than it was at the 10 G peak. |
[ARCHIVED THREAD] - DPM Recoil Reduction System vs. Hydraulic buffer EDIT: Armament LARB Mod 2 and 3 added to the test. (Page 4 of 4)
AR Sponsor