Jump to content
Sturgeon's House

Recommended Posts

Most automatic weapons, with the exception of really weird designs like the Madsen LMG and Hino-Komuro, have a linear reciprocating breech member; either the bolt or a bolt carrier group.  This reciprocating member is supposed to move rearward (the recoil stroke) and pull the spent case from the chamber, and then rebound off of a spring to shove a new round into the chamber (the counter-recoil stroke).  After the counter-recoil stroke the reciprocating mass should come to a halt in its forward-most position; the "in-battery" position.  When the bolt carrier group is in battery the case is entirely surrounded by the walls of the firing chamber and the locking mechanism is fully engaged, so it is safe to fire.  Things do not always work ideally, however, and sometimes this reciprocating mass bounces instead of coming to rest.  This is called (somewhat erroneously in the case of gas-operated weapons) "bolt-bounce."

 

Andrew Tuohy removed the buffer weights from the buffer in an AR-15 to make the action bouncier for illustrative purposes:

 

 

There are two ways that bolt carrier rebound can be a problem.  In extreme cases the bolt carrier will rebound, but a combination of high friction in the action and weak return springs will mean that the bolt carrier gets stuck and does not go back into battery.  Hopefully the designer was smart enough to design the thing so that it absolutely cannot fire when it is out of battery, because out of battery cartridge ignition is an excellent way to convert a firearm into a pipe bomb.  If they were so wise, then there will be a failure to fire of some variety.  Generally speaking a weapon has to be unusually dirty, worn, or poorly designed for this problem to occur.  Return springs are usually strong enough to get the moving parts into battery even if they aren't fully compressed.  I have, however, witnessed this problem in German K43 rifles because they are a pile of suck and fail.

 

gewehr_43-4.jpg

 

But they're pretty.

The second, more likely problem only rears its ugly head in fully auto fire.  In most full auto weapons there is an auto-sear, which a secondary sear which releases the hammer as long as the trigger is depressed.  The auto-sear is tripped by the bolt carrier during counter-recoil, usually when or just before the bolt carrier goes into battery.  If the bolt carrier rebounds off the front of the receiver and the timing is just wrong, the hammer (or striker) will hit the bolt carrier when it is slightly out of battery.  Again, competent designs have means of preventing out of battery ignition and the attendant facial and manual reorganization that tends to go with that.  However, when the hammer or striker hits the out of battery bolt carrier its kinetic energy will be spent.  This means a failure to fire.

 

Early M16s had rebound problems, particularly during full auto fire.  Originally the buffer was intended simply to be a hollow spring guide, but a problem with light primer strikes forced a redesign of this component in 1966.

 

KDQXtGD.png

 

This image, from the patent for the improved buffer shows the series of sliding weights that were added to the buffer.  These work like the sliding pellets in a deadblow hammer and arrest the tendency for the bolt carrier to bounce.  The additional mass had the added benefit of slowing down the velocity of the bolt carrier, which reduced wear on the parts and lowered the cyclic rate of fire, which improved full auto control.

 

The HK roller-retarded blowback guns, owing to their extremely high bolt carrier velocities, have a strong tendency to rebound unless somehow checked.  The solution HK engineers hit on is an anti-rebound claw:

 

hk91boltgroup_2160_large.gif

 

Labeled as the "bolt head locking lever" in this diagram.  This is a spring-loaded claw mounted on the bolt carrier that grabs the bolt head as the bolt carrier group goes into battery.  The lever essentially ratchets into place with friction, providing enough resistance to being re-opened that the bolt carrier does not rebound.

DSCN0598_zpsnviefbcm.jpg

 

The FAMAS, which has a similarly insanely high bolt carrier velocity, solves the problem in a very similar way.  In this case, however, the charging handle is the anti-rebound device.  The arrangement is similar to the locking catch on an AR-15's charging handle, except that it's much more robust because the catch is responsible for arresting the rebound of the entire bolt carrier.

There are other ways still to arrest the rebound of the bolt carrier.  The Ruger MP-9 (the one designed by Uziel Gal, not the insanely over complex B&T product of the same name) is supposed to have a spring-cushioning pad at the front of the receiver which brings the bolt to a stop instead of bouncing.  The upcoming Desert Tech MDR has, by one account, "an asymmetrical [bolt carrier] face. This is accomplished with a protruding boss on one side of the carrier. As the carrier moves forward to go into battery, the asymmetrical face contacts the barrel extension first. Tolerances within the axial motion of the back end of the carrier group permit the energy to be redirected through a sideways movement. This micro-movement of the rear end of the carrier impedes the bounce and assures full function of the weapon, especially in select-fire operation."  Large caliber autocannons often have complex, articulated secondary locks that prevent bolt carrier bounce, since autocannon bolt carriers are enormous and have a great deal of residual kinetic energy.

 

So, when I read that the SIG MCX has some problems with full auto function that sound suspiciously exactly like the same problems the M16 had prior to the addition of the weighted buffer (the same weighted buffer the MCX does away with), I can only roll my eyes.  This is nothing new, and there are a half-dozen ways of fixing it.  Do your homework.

Link to post
Share on other sites

My sole critique is that the bolt bounce suffered by the early AR-15/XM16 was propellant induced, and not a direct result of a design fault.

The resultant experimentation did "improve the breed" though, so sometimes having to correct an obvious user induced fault can have benefits.

Link to post
Share on other sites

Per The Black Rifle the failures to fire in full auto prior to the buffer change could happen "irrespective of propellant used."  But it would make physical sense that it would be more common with the propellants that caused higher bolt carrier velocity.  Prior to the introduction of the plastic-tipped buffer the carrier group would ricochet off the back of the receiver extension, which would increase residual carrier KE and make it more likely to bounce during counter-recoil.  Also it caused other issues, like the bolt over-running the bolt catch on the last round.

Link to post
Share on other sites

All of which was caused by excessive breech velocities caused by that funky reformulated propellant.

The thing worked fine with the fodder it was designed for.

 

"could happen" and "will happen" are rather different in my mind.

I see it as someone buying a car that requires premium fuel, then complaining to the manufacturer after they've knocked holes in it's pistons after running it on a bunch of low octane fuel of dubious quality.

 

Again, I feel that it eventually resulted in the improvement of the design, but there is no doubt in my mind that it was to fix a user generated problem.

Link to post
Share on other sites

Again, per The Black Rifle, the carrier rebound problem was noticed in 1963, but the substitution of WC846 for IMR4475 only occurred in 1964, although experimentation had occurred in 1963 at Frankfurt Arsenal.

 

Bizarrely, there was a Colt internal memo which stated that the increased cyclic rate of fire would improve the function of the rifle.

Link to post
Share on other sites

Again, per The Black Rifle, the carrier rebound problem was noticed in 1963, but the substitution of WC846 for IMR4475 only occurred in 1964, although experimentation had occurred in 1963 at Frankfurt Arsenal.

 

Bizarrely, there was a Colt internal memo which stated that the increased cyclic rate of fire would improve the function of the rifle.

Again, the potential for it to occur is not an issue, it was brought to a head when the change occurred.

Stoner testified that the change to that propellant was the main reason for the troubles.

 

I forget the wording of the Colt memo in question, but I doubt the folks who designed the action agreed with that assessment.

Link to post
Share on other sites

The propellant issues of the M16 are actually more complex than even most folks with a good education on the subject know. Daniel Watters set me straight on this, here's what he said:

http://www.thefirearmblog.com/blog/2015/01/09/jim-sullivan-m16-vietnam/

 

" In April 1964, Colt’s senior product engineer Foster Sturtevant wrote in an internal report that the higher gas port pressures with WC846 were “in no way harmful to the AR-15” and would lead to more positive functioning of the rifle. -It strikes me that the increased cyclic rate was initially tolerated because some uninformed souls saw it as a bonus feature, not a flaw-."

 

Pretty much what I expected.

Link to post
Share on other sites

Colt began development of the current buffer after experiencing a rash of light strikes attributed to bolt bounce during testing of its HBAR variants.  Remember that most of Colt's acceptance testing was performed with IMR-loaded ammunition, not WC846.  Thus, if Colt was encountering bolt bounce issues in its own testing, it likely happened with IMR-loaded ammunition. 

Another thing is that that the Edgewater ring springs in the original buffer were prone to seizing when wet. So anything experienced in internal testing was likely going to be worse in the field.

Link to post
Share on other sites

Colt began development of the current buffer after experiencing a rash of light strikes attributed to bolt bounce during testing of its HBAR variants.  Remember that most of Colt's acceptance testing was performed with IMR-loaded ammunition, not WC846.  Thus, if Colt was encountering bolt bounce issues in its own testing, it likely happened with IMR-loaded ammunition. 

Another thing is that that the Edgewater ring springs in the original buffer were prone to seizing when wet. So anything experienced in internal testing was likely going to be worse in the field.

That starts into the whole "we have this neat core platform, lets see what we can do with it and who we can sell it to" cycle. Where you have to offer something new, in order to satisfy your market, but you have to offer it in sufficient volume and time to satisfy that demand.

 

The various carbines to the belt fed and LMG versions, and the early mounting of cans all had development and early issue problems from a lack of testing.

It was compounded by unreasonable expectations on the part of the DOD.

Link to post
Share on other sites

I have had to "cure" three M16s of bolt bounce. Factory setups are fine and run right, but when whoever gets ahold of a transferable MG and decides he needs a 7" barrel, to mess with internals, or add a piston setup I have seen hammer follow like crazy on the bolt's rebound (which usually results in about three rounds of full auto, then a chambered round with the hammer tripped).

 

Veteran gunsmiths I talked to said "Colt spent millions of dollars figuring out how much a buffer should weigh, and Jim Bob didn't". Thus, the solution was essentially to buy a case or two of ammo, 6 or 7 buffers, and find out which one worked best with the combo.

Link to post
Share on other sites

I have had to "cure" three M16s of bolt bounce. Factory setups are fine and run right, but when whoever gets ahold of a transferable MG and decides he needs a 7" barrel, to mess with internals, or add a piston setup I have seen hammer follow like crazy on the bolt's rebound (which usually results in about three rounds of full auto, then a chambered round with the hammer tripped).

 

Veteran gunsmiths I talked to said "Colt spent millions of dollars figuring out how much a buffer should weigh, and Jim Bob didn't". Thus, the solution was essentially to buy a case or two of ammo, 6 or 7 buffers, and find out which one worked best with the combo.

I've seen a couple where the end of the buffer tube endcap had nearly broken off the tube itself.  The springs were usually pretty mangled, and to top it off, the "users" mentioned how they'd "gone through a few buffer springs".

Link to post
Share on other sites

I've seen a couple where the end of the buffer tube endcap had nearly broken off the tube itself.  The springs were usually pretty mangled, and to top it off, the "users" mentioned how they'd "gone through a few buffer springs".

I have only seen this on 9mm guns. Straight blowback seems to rattle M16 pattern rifles pretty good.

Link to post
Share on other sites

" In April 1964, Colt’s senior product engineer Foster Sturtevant wrote in an internal report that the higher gas port pressures with WC846 were “in no way harmful to the AR-15” and would lead to more positive functioning of the rifle. -It strikes me that the increased cyclic rate was initially tolerated because some uninformed souls saw it as a bonus feature, not a flaw-."

 

Pretty much what I expected.

 

Yes; they thought that the more positive action would improve reliability.  And they were completely wrong.

 

I have only seen this on 9mm guns. Straight blowback seems to rattle M16 pattern rifles pretty good.

 

Not the first time I've heard of 9mm AR-15s tearing up parts.

Link to post
Share on other sites

I have only seen this on 9mm guns. Straight blowback seems to rattle M16 pattern rifles pretty good.

It's become a less common occurrence since most of the major parts rollers have got their collective shit in a pile.

Last one I saw it on was a shake-n-bake ~8" tubed flamethrower cobbled together with parts of questionable origin and rattlecanned in a garish JD green/yellow "camo" scheme.

The guy had two or three buffer springs that looked like they'd been put in a press.

Link to post
Share on other sites
  • 10 months later...
  • 2 years later...
On 11/22/2015 at 3:27 PM, Alex C. said:

I have had to "cure" three M16s of bolt bounce. Factory setups are fine and run right, but when whoever gets ahold of a transferable MG and decides he needs a 7" barrel, to mess with internals, or add a piston setup I have seen hammer follow like crazy on the bolt's rebound (which usually results in about three rounds of full auto, then a chambered round with the hammer tripped).

 

Veteran gunsmiths I talked to said "Colt spent millions of dollars figuring out how much a buffer should weigh, and Jim Bob didn't". Thus, the solution was essentially to buy a case or two of ammo, 6 or 7 buffers, and find out which one worked best with the combo.

I am having the problem you speak of with a piston upper on my select fire carbine rifle. Could you please explain to me the solution? I have tried a new buffer spring and H2 and H3 buffers. I can't get it to run properly.

Link to post
Share on other sites
On 3/6/2019 at 6:03 PM, hallnh said:

I am having the problem you speak of with a piston upper on my select fire carbine rifle. Could you please explain to me the solution? I have tried a new buffer spring and H2 and H3 buffers. I can't get it to run properly.

 

 

Do you have a high speed camera?  Those are surprisingly cheap now.

 

If it is bolt bounce, the problem is that your bolt carrier is moving forward too quickly.

Link to post
Share on other sites
  • 1 year later...

Question: In the first video, is it the bolt or the bolt carrier group that is bouncing around ? Assuming its only the carrier group and not the (now locked) bolt itself, couldnt you design the bolt carrier and hammer so that it doesnt matter if the carrier group jumps around a bit ?

Link to post
Share on other sites
On 12/3/2020 at 3:22 PM, R_K_M said:

Question: In the first video, is it the bolt or the bolt carrier group that is bouncing around ? Assuming its only the carrier group and not the (now locked) bolt itself, couldnt you design the bolt carrier and hammer so that it doesnt matter if the carrier group jumps around a bit ?


Yes the AK basically does this. It has very generous underslide and excess carrier travel forward after the out of battery safety clears the bolt stem, which is exactly the solution you describe. Several of my rifle designs use this same solution.

Link to post
Share on other sites
  • 1 month later...
On 11/22/2015 at 3:48 AM, Collimatrix said:

Most automatic weapons, with the exception of really weird designs like the Madsen LMG and Hino-Komuro, have a linear reciprocating breech member; either the bolt or a bolt carrier group.  This reciprocating member is supposed to move rearward (the recoil stroke) and pull the spent case from the chamber, and then rebound off of a spring to shove a new round into the chamber (the counter-recoil stroke).  After the counter-recoil stroke the reciprocating mass should come to a halt in its forward-most position; the "in-battery" position.  When the bolt carrier group is in battery the case like they are show on this site https://primearmor.us/ is entirely surrounded by the walls of the firing chamber and the locking mechanism is fully engaged, so it is safe to fire.  Things do not always work ideally, however, and sometimes this reciprocating mass bounces instead of coming to rest.  This is called (somewhat erroneously in the case of gas-operated weapons) "bolt-bounce."

 

Andrew Tuohy removed the buffer weights from the buffer in an AR-15 to make the action bouncier for illustrative purposes:

 

 

 

There are two ways that bolt carrier rebound can be a problem.  In extreme cases the bolt carrier will rebound, but a combination of high friction in the action and weak return springs will mean that the bolt carrier gets stuck and does not go back into battery.  Hopefully the designer was smart enough to design the thing so that it absolutely cannot fire when it is out of battery, because out of battery cartridge ignition is an excellent way to convert a firearm into a pipe bomb.  If they were so wise, then there will be a failure to fire of some variety.  Generally speaking a weapon has to be unusually dirty, worn, or poorly designed for this problem to occur.  Return springs are usually strong enough to get the moving parts into battery even if they aren't fully compressed.  I have, however, witnessed this problem in German K43 rifles because they are a pile of suck and fail.

 

http://tonnel-ufo.ru/foto/oruhie/snaiper_3/germani/gewehr_43-4.jpg

 

But they're pretty.

The second, more likely problem only rears its ugly head in fully auto fire.  In most full auto weapons there is an auto-sear, which a secondary sear which releases the hammer as long as the trigger is depressed.  The auto-sear is tripped by the bolt carrier during counter-recoil, usually when or just before the bolt carrier goes into battery.  If the bolt carrier rebounds off the front of the receiver and the timing is just wrong, the hammer (or striker) will hit the bolt carrier when it is slightly out of battery.  Again, competent designs have means of preventing out of battery ignition and the attendant facial and manual reorganization that tends to go with that.  However, when the hammer or striker hits the out of battery bolt carrier its kinetic energy will be spent.  This means a failure to fire.

 

Early M16s had rebound problems, particularly during full auto fire.  Originally the buffer was intended simply to be a hollow spring guide, but a problem with light primer strikes forced a redesign of this component in 1966.

 

http://i.imgur.com/KDQXtGD.png

 

This image, from the patent for the improved buffer shows the series of sliding weights that were added to the buffer.  These work like the sliding pellets in a deadblow hammer and arrest the tendency for the bolt carrier to bounce.  The additional mass had the added benefit of slowing down the velocity of the bolt carrier, which reduced wear on the parts and lowered the cyclic rate of fire, which improved full auto control.

 

The HK roller-retarded blowback guns, owing to their extremely high bolt carrier velocities, have a strong tendency to rebound unless somehow checked.  The solution HK engineers hit on is an anti-rebound claw:

 

http://www.hkparts.net/shop/pc/catalog/hk91boltgroup_2160_large.gif

 

Labeled as the "bolt head locking lever" in this diagram.  This is a spring-loaded claw mounted on the bolt carrier that grabs the bolt head as the bolt carrier group goes into battery.  The lever essentially ratchets into place with friction, providing enough resistance to being re-opened that the bolt carrier does not rebound.

http://i13.photobucket.com/albums/a254/joeplf/4%20Bullpups/DSCN0598_zpsnviefbcm.jpg

 

The FAMAS, which has a similarly insanely high bolt carrier velocity, solves the problem in a very similar way.  In this case, however, the charging handle is the anti-rebound device.  The arrangement is similar to the locking catch on an AR-15's charging handle, except that it's much more robust because the catch is responsible for arresting the rebound of the entire bolt carrier.

There are other ways still to arrest the rebound of the bolt carrier.  The Ruger MP-9 (the one designed by Uziel Gal, not the insanely over complex B&T product of the same name) is supposed to have a spring-cushioning pad at the front of the receiver which brings the bolt to a stop instead of bouncing.  The upcoming Desert Tech MDR has, by one account, "an asymmetrical [bolt carrier] face. This is accomplished with a protruding boss on one side of the carrier. As the carrier moves forward to go into battery, the asymmetrical face contacts the barrel extension first. Tolerances within the axial motion of the back end of the carrier group permit the energy to be redirected through a sideways movement. This micro-movement of the rear end of the carrier impedes the bounce and assures full function of the weapon, especially in select-fire operation."  Large caliber autocannons often have complex, articulated secondary locks that prevent bolt carrier bounce, since autocannon bolt carriers are enormous and have a great deal of residual kinetic energy.

 

So, when I read that the SIG MCX has some problems with full auto function that sound suspiciously exactly like the same problems the M16 had prior to the addition of the weighted buffer (the same weighted buffer the MCX does away with), I can only roll my eyes.  This is nothing new, and there are a half-dozen ways of fixing it.  Do your homework.

The premise of this question isn’t entirely correct because all self-loading firearms which have reciprocating bolts experience bolt bounce to some degree. When the bolt (or bolt carrier) on a firearm slams forward into the barrel to chamber a new cartridge, the bolt (or bolt carrier) and the barrel elastically rebound off each other, sometimes multiple times, before coming to rest. Bolt bounce can be particularly pronounced in designs with very high bolt velocity or minimal (or nonexistent) locking mechanisms. Bolt bounce can be very difficult to observe in designs with low bolt velocity and locking mechanisms which require a great deal of travel to lock and unlock.
 
 
Link to post
Share on other sites
On 1/18/2021 at 6:51 AM, Mostright60 said:
The premise of this question isn’t entirely correct because all self-loading firearms which have reciprocating bolts experience bolt bounce to some degree. When the bolt (or bolt carrier) on a firearm slams forward into the barrel to chamber a new cartridge, the bolt (or bolt carrier) and the barrel elastically rebound off each other, sometimes multiple times, before coming to rest. Bolt bounce can be particularly pronounced in designs with very high bolt velocity or minimal (or nonexistent) locking mechanisms. Bolt bounce can be very difficult to observe in designs with low bolt velocity and locking mechanisms which require a great deal of travel to lock and unlock.
 
 

If you're going to be pedantic, you should at least try to be correct.  This is the real fucking world we're talking about here.  There isn't some infinitesimal dribble of rebound in all cases because there are plenty of sources of damping, like friction from locking, or any of the several deliberate ways to damp carrier motion as it comes into battery that I detailed in the OP.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Similar Content

    • By Beer
      I haven't found an appropriate thread where to put some interesting rare stuff related to WW2 development, be it industrial one or makeshift field modifications. 
       
      Let's start with two things. The first one is a relatively recently found rarity from Swedish archives - a drawing of ČKD/BMM V8H-Sv tank. The drawing and a letter was found by WoT enthusiasts in Swedish archives in 2014 (the original announcement and the drawing source is here). The drawing is from a message dated 8th September 1941. One of the reasons why this drawing was not known before may be that the Czech archives were partially destroyed by floods in 2002. Anyway it is an export modification of the V-8-H tank accepted into Czechoslovak service as ST vz.39 but never produced due to the cancelation of all orders after Münich 1938 (for the same reason negotiations about licence production in Britain failed). Also later attempt to sell the tank to Romania failed due to BMM being fully busy with Wehrmacht priority orders. The negotiations with Sweden about licence production of V8H-Sv lasted till 1942, at least in May 1942 Swedish commission was present in Prague for negotiations. The tank differed compared to the base ST vz.39 in thicker armor with different front hull shape (armor 60 mm @ 30° on the hull front and also 60 mm on the turret; all sides were 40 mm thick). The tank was heavier (20 tons) and had the LT vz.38 style suspension with probably even larger wheels. The engine was still the same Praga NR V8 (240-250 Hp per source). The armament was unchanged with 47 mm Škoda A11 gun and two vz.37 HMG. The commander's cupola was of the simple small rotating type similar to those used on AH-IV-Sv tankettes. It is known that the Swedes officially asked to arm the tank with 75 mm gun, replace the engine with Volvo V12 and adding third HMG to the back of the turret. In the end the Swedes decided to prefer their own Strv/m42. 

      Source of the drawing
       
      The second is makeshift field modification found on Balkans. It appears Ustasha forces (and possibly some SS anti-partizan units) used several Italian M15/42 medium tanks with turrets from Pz.38(t). There are several photos of such hybrids but little more is known. On one photo it is possible to see Ustasha registration number U.O. 139.

      Few more photos of such hybrid.
       
      It appears that the source of all those photos to be found on the internet is this book, Armoured units of the Axis forces in southeastern Europe in WW2 by Dinko Predoevic. 
       
    • By pizza654
      Hi as most of you know who are in the gun community a bunch of AR 70/90 kits came into the country and theirs still no barrels or receivers in production. Since I cant find anything I decided that I'm just going to make my own barrel and I've found the measurements from a guy on reddit who lives in Italy. 
       
      https://imgur.com/gallery/WUtxuAP
       
      https://imgur.com/gallery/1ydhDUS
       
      As I was getting the measurements a curiosity ran through my head, how do you mathematically figure out the proper diameter of the gas port hole for the gas block in the barrel?
       
      Much help will be appreciated! 
       
       
    • By CharlieAlphaVictor
      This may have already been answered, but why are so many modern assault rifles gas-operated, when blowback-operated designs are (generally speaking) simpler/cheaper to manufacture and require less maintenance? I've been doing some research and can't seem to figure out why for the life of me. Any assistance is greatly appreciated.
    • By Curly_
      This came up in a recent discussion I had with a friend over Discord, concerning a hypothetical near-peer conflict (particularly in an urban environment). My friend is of the belief that in such a conflict there isn't no reason to field any small arms (not just service rifles or light machine guns) in anything larger than something like 5.56 NATO, the rationale being that a.) you can carry more of it compared something like 6.5 Creedmoor or 7.62 NATO, b.) you're mostly going to be using it suppressing the enemy so that you can drop high explosives on their heads, and c.) since most combat takes place under 300 meters the extra range would be unnecessary. Is there any merit to this line of thinking? What cases can be made for using more powerful, longer ranged cartridges in SDMRs and (tripod or vehicle mounted) machine guns?
×
×
  • Create New...