Milspec and beyond

[Shihan]

Very informative thread. Is it safe to say that in 2009 an M4-series carbine is as mechanically reliable and has comparable longevity to an M16-series rifle?

On alot of the parts no where close.

[Sttrongbow]

My understanding is, Armalite's engineering analysis predicts (maybe proves???) that the assymetric lug design on the AR's bolt results in a disproportionate loading on the two lugs adjacent to the extractor cutout, and it's because of this higher loading that lug failure almost always occurs on either of these two lugs.

What Armalite does is, they shorten the lug opposite the extractor to the point where it's no longer load bearing. What this does is change the loading pattern from two heavily loaded lugs plus five lightly loaded lugs to six symetrically spaced, evenly loaded lugs, with each having significantly less loading than the two heavily loaded lugs in a conventional design.

Sounded interesting to me, but Armalite is one of those brands where you never hear much anecdotal evidence one way or another to compare observed results with modelling/theory.

From an engineering point of view, it makes sense. You want to distribute load as much as possible. I'm curious as to whether this plays out in real world evidence.

[Sttrongbow]

The theory is that by HP testing something, you are removing one of its lives. Now the question is, how many lives does the barrel or bolt have in it? No one knows this for certain.



C4

Exactly Grant. If you do an relaibility analysis, virtually all parts follow a "bathtub curve" for failures. There are a high number of failures early, then relatively few for the majority of the life of the part, then a high number of failures at the end of the part's life. These early failures are referred to (rather morbidy) as "infant mortality". For companies who are very concerned with reliability, you want to try and deal with those early "infant" deaths. One way, of course, is to try and refine the design and process of manufacturing to reduce such defects (which are usually attributed to faults in materials or manufacturing processes, or BIG design flaws). Another is to inspect the hell out of them.

But the last, is to ship the part after you've clipped off a big portion of the left hand side of the "bath tub." Computer companies, for example, "burn in" their computers to reduce those early failures. This sort of HPT testing does the same thing, but in an accelerated fasion. 25% reduced life? I'm not sure about that. III has tons more experience with ARs than I do, but as I understand it, late failure in AR15 bolts are mainly due to fatigue at the locking lugs. Early failures exposed by HPT would be mainly due to poor hardening, or faults in the material.

But although a mechanical engineer... I am very new to the AR world. This is a fascinating discussion.

[taliv]

but as I understand it, late failure in AR15 bolts are mainly due to fatigue at the locking lugs. Early failures exposed by HPT would be mainly due to poor hardening, or faults in the material.

just my anecdotal experience and memory here, but aren't most of the late bolt-failures cracks at the cam pin hole? while broken lugs are typically associated with overpressure events?

[Bill Alexander]

The degredation of component life from overload events is well documented and can be supported from a significant number of case studies as well a a wealth of analytical research. It is not a question of does an overload such as a high pressure cartridge reduce the bolt life but rather it is a judgement of how much. The stress concentrations that exist withing the bolt are difficult to characterize exactly as both the loading and the form are complex.

In considering the bolt, the analogy of electronic burn in is not very representative. Correctly the life of a fleet is described by a Guassian distribution curve but an overload event will not in effect remove the early failures and leave only the better units, it acts to shift the whole curve to the left. It will remove defective units from the fleet, but these should be identified from the process data and in most cases, break in of the unit with standard load cycles.

The theoretical foundation for Armalites bolt design is solid and the stress transfer to the two lugs adjacent to the extractor can be argued especially as enough float exists within the AR to allow this to take place. However as both the plane of the lugs and the plane of the barrel extension lugs are maintained to a tight specifications, then the gain in the real world must be considered in the light of the removal of +14% of the load bearing members. Those familiar with the LMT bolt will note that a different approach is to add a greater degree of elasticity to the lug plane and thus allow the lugs to all share the load evenly.

Agressive chemicals used for cleaning and/or a less than true receiever will probably have more influence on the bolt life.

Cam pin hole failures can ususally be traced to the swaging on the bottom of the cam pin hole, which initiates the crack. Cam pins moving under the leverage of the cam path will effectively bend the bolt and diagnosis of the fault in failed parts is easy to see from two bright spots at the top and bottom of the cam pin hole. Cam pins should be replaced when wear is seen on the surface and when they become loose in the bolt.

Bill Alexander

[Sttrongbow]

Mr Alexander,

Thank you for your excellent analysis!

I did not mean to suggest that bolt life is not shortened by the HPT, just that I thought 25% seems intuitively high to me. But like I said, I am new to the AR, so I could very well be wrong.

Also, you did a much better job describing my thoughts on how HPT could weed out the bad bolts early.

I have one question: If failures of the bolt at teh cam pin hole can be traced to a particular spot, is there an inspection would could identify the the ones likely to fail early?

Again, thanks for taking to time to educate us here!

The degredation of component life from overload events is well documented and can be supported from a significant number of case studies as well a a wealth of analytical research. It is not a question of does an overload such as a high pressure cartridge reduce the bolt life but rather it is a judgement of how much. The stress concentrations that exist withing the bolt are difficult to characterize exactly as both the loading and the form are complex.

In considering the bolt, the analogy of electronic burn in is not very representative. Correctly the life of a fleet is described by a Guassian distribution curve but an overload event will not in effect remove the early failures and leave only the better units, it acts to shift the whole curve to the left. It will remove defective units from the fleet, but these should be identified from the process data and in most cases, break in of the unit with standard load cycles.

The theoretical foundation for Armalites bolt design is solid and the stress transfer to the two lugs adjacent to the extractor can be argued especially as enough float exists within the AR to allow this to take place. However as both the plane of the lugs and the plane of the barrel extension lugs are maintained to a tight specifications, then the gain in the real world must be considered in the light of the removal of +14% of the load bearing members. Those familiar with the LMT bolt will note that a different approach is to add a greater degree of elasticity to the lug plane and thus allow the lugs to all share the load evenly.

Agressive chemicals used for cleaning and/or a less than true receiever will probably have more influence on the bolt life.

Cam pin hole failures can ususally be traced to the swaging on the bottom of the cam pin hole, which initiates the crack. Cam pins moving under the leverage of the cam path will effectively bend the bolt and diagnosis of the fault in failed parts is easy to see from two bright spots at the top and bottom of the cam pin hole. Cam pins should be replaced when wear is seen on the surface and when they become loose in the bolt.

Bill Alexander