PDA

View Full Version : Developing a reasonable testing protocol



AllenLash
03-26-11, 15:29
I have been given a piston driven AR to T&E. I don't want to "name names" just yet so I'll just call it "Brand X". The LE rep for Brand X stopped by the range earlier this week and dropped off a piston driven AR and asked us to see what we thought. In the interest of full disclosure, while I don't have decades of experience with the AR platform, I do have a few years as a patrol rifleman, instructor and armorer. In that time, I have come to believe that the idea of a piston driven AR is a solution for a problem that largely isn't there--more proprietary parts to break, heavier and more complex to keep running in large numbers (we have over 600 rifles in service).

Yesterday, myself and our assistant armorer took the rifle out to the range and did a "stress" test-see how long it would shoot before it started puking. The plan was to fire as many rounds as possible, as quickly as safety and our loading ability allowed. We had nine magazines available and around 720 rounds (mostly Federal 55 grains and a few Winchester Ballistic Silver Tips, 55 grains).

First, we got a decent zero at 100 yards. The groups for both of us were larger than our norms with our own guns but I thought acceptable over all. Probably 3-4 inches for both of us. The trigger seemed heavier than my Colt or Rock River. We both had issues with short stroking the trigger on reset. Were going to look at that closer-check pull weight and do a side by side comparisons. Next we started the "stress" test in earnest.

We got through 6 and a half magazines, 28 rounds each, before we had our first failure, the weapon failed to extract an empty case. Before we started, I checked the bolt assembly and looked at the extractor-spring and black insert only. We cleared the stoppage and continued another 3.5 magazines and another failure to extract. Similar failures continued to occur at an increasing rate until we would have two failures on every magazines by the time we were done. I don't have the notes with me, but I believe all but one failure was an extraction issue. The other was a double feed, two rounds trying to enter the chamber at the same time. The Winchester ammo never failed to extract or feed, but it only accounted for around 5 percent of the total rounds fired.

We collected most of the cases that failed to extract (most required the "mortar" extraction method). We noted some signs of higher that normal pressure as all the cases had flattened primers.

Another issue we noted was the rear sight assembly came loose and all but fell off.

A few questions and thoughts: Was this a reasonable start for the testing process? Hindsight being what it is, I wish I had checked the chamber and headspace before starting, checked pull weight, and snugged that rear sight down. I think we shouldn't have to do those things because it should be "right" from the beginning. I know, that's not reality and I probably should have done those things before we started. I played Devil's advocate with my coworker and asked if a rifle should be able to fire 700 rounds as quickly as we can fire and reload magazines? Is that a valid test? How many rounds should it be able to fire? Less or more? In this case, I think it was, because I would hope that I didn't have a rifle that started out clean and well lubed to puke after only 180 rounds. When you put yourself in the shoes, well boots, of a Marine at Chosin Reservoir, facing ChiCom hordes this becomes very reasonable for a combat rifle. Ok, I'm not a Marine, I'm a cop and most LE shooting last only a few rounds. Typically. Then I think of Beslin. If (or more likely-when) that happens in the United States it will be cops who first engage an enemy who may be well equipped, well trained, and have mindset to create as much suffering and death as possible (that we in the West have trouble comprehending) before they become martyrs. Again, this test seems reasonable to me. What say you?

What are some additional tests? Put it in a freezer overnight and see what happens when you start shooting it in 70-80 degrees (replicating freezing cold gun going inside for a fight). Dunk it in mud? Sand/Dirt? Drop test? I don't necessarily want to break the gun. It's not ours after all. Although, the last rifle this rep gave us was sent back to the factory with the action inoperable. So, I guess he should have known better than to give us another rifle. ;)

My goal here is to develop a testing protocol for this (and other weapons) that will be defensible. So we can explain to our officers why they can't carry whatever they want and show the command staff and other weapon challenged folks why Brand A may be cheaper but not nearly as good as Brand B. And why quality matters in terms of cost of repairs and in lives.

Input, please.

Thanks in advance,

Allen

RAM Engineer
03-26-11, 16:05
I'd love to see what is required of a lot of M4 Carbines by the DCMA inspectors before they get DD-250'd. Why couldn't Wikileaks use their power for USEFUL things?

Amicus
03-26-11, 16:10
This is a very good question, and a good learning experience for people who want to develop skills as an evaluator.

First off, not to be a jerk, but what is Beslin? From the context I can't work out if you mean "Berlin" (as in battle of in 1945) or "Beslan" (as in school hostages in 2004).

Back to the question. I would first develop a rigorous checklist for establishing a baseline. All items on the list must meet the minimum standards for the baseline or the test will be skewed (and that should be noted). I believe you approached that when you checked headspace, etc; but, unless you have the baseline thoroughly documented, it become hard to evaluate any wear or changes that occurred, or to compare to other rifles.

Once baseline standards are met, then each branch of testing will attempt to evaluate one thing, or, one set of things. It appears you attempted what some might call an ammunition stress test. You appear to have done that well, stopping and documenting as you went on. Did you completely document everything in the system though? What ammo failed, when, fed from which magazine, and retained for further evaluation the case?

If you then want to evaluate temperature effects, then clean the carbine, bring it back up to baseline (including repair or replacement of parts) and start over. Otherwise, you are only evaluating the effects of temperature after an ammunition stress test. Of course, you could do that as well, but it would have less meaning than if you attempted to isolate each factor that could affect performance.

In short, each evaluation should be separate. I'd use the scientific method myself. Set up your experiment, make a prediction (hypothesis), conduct the experiment, and evaluate the results (incorporating the reasons why your prediction did or did not occur).

Just stuff off the top of my head. I look forward to other responses to learn more about this subject myself.

AllenLash
03-26-11, 16:17
Opps: "Beslan"

http://en.wikipedia.org/wiki/Beslan_school_hostage_crisis

We started marking magazines, but it quickly became evident that the issue was independent of magazines. It happened to most of the magazines (known good).

Iraqgunz
03-26-11, 16:58
Here are some of my thoughts.

1. Ask for a spec sheet on the gun to determine what parts and materials were used to build it.

2. How do you plan on checking headspace?

3. I would do a detailed examination of the components as it is shipped from the factory and point out whatever deficiencies you find.

4. When using various lots/types of ammo I would try and document lot numbers and make sure that you fire the different type together (don't mix and match).

5. Accuracy testing from positions that would be encountered. Not shooting from the bench.

6. If they claim it doesn't need lube, then run it w/o and see when it starts to puke.

ZRH
03-26-11, 17:33
You also have to know what you want (criteria) before you can really establish a standard. That is, you have to know what performance you expect before you can say "this does not perform as expected." You could run individual tests on a couple rifles you already have to establish an expected performance and compare that to test samples. Log book would be useful.

As for the specific tests that is completely dependent on end use.

markm
03-26-11, 17:42
Can't be an LWRC.... you wouldn't have made it through 6-1/2 mags if it were. :p

Send it back and quit wasting ammo on a hunk of shit Piston. It's not meant to be.... despite what the Piston Apologicians keep spewing.:rolleyes:

graffex
03-26-11, 17:43
Can't be an LWRC.... you wouldn't have made it through 6-1/2 mags if it were. :p

Send it back and quit wasting ammo on a hunk of shit Piston. It's not meant to be.... despite what the Piston Apologicians keep spewing.:rolleyes:

Haha, I agree.

Bimmer
03-26-11, 18:51
I have been given a piston driven AR to T&E.

We collected most of the cases that failed to extract (most required the "mortar" extraction method). We noted some signs of higher that normal pressure as all the cases had flattened primers.

Another issue we noted was the rear sight assembly came loose and all but fell off.

A few questions and thoughts: Was this a reasonable start for the testing process?

...

I don't necessarily want to break the gun. It's not ours after all. Although, the last rifle this rep gave us was sent back to the factory with the action inoperable. So, I guess he should have known better than to give us another rifle. ;)

It sounds like the chamber is tight, and it sounds like the rear sight doesn't attach well, or whoever assembled the gun botch attaching the rear sight — if it were loose when you started (shooting groups), then you'd have noticed it.

Either way, these are two failures that essetially put the gun out of action.

As for breaking it... of course you shouldn't break it on purpose or abuse it, but this guy gave it to you "test," and your test seems entirely normal and reasonable to me.

Either this guy wants you to buy a bunch of these, or he wants you to beta test it for him.

Either way, if you break it in normal use, then giving it back to him broken seems reasonable to me.

skyugo
03-26-11, 19:14
Can't be an LWRC.... you wouldn't have made it through 6-1/2 mags if it were. :p

Send it back and quit wasting ammo on a hunk of shit Piston. It's not meant to be.... despite what the Piston Apologicians keep spewing.:rolleyes:

piston AR's are by definition a hack. the DI gun is a known and well tuned entity at this point. Really the DI system of an AR15 is pretty ingenious.

to the OP-i think you've talked me out of buying that particular brand of rifle. any decent DI AR will run longer than that one did. Maybe you got a lemon- so there's always sample size to consider. Though I'd say having a lemon make it through QC to an end user is a black eye for a manufacturer.

CumbiaDude
03-26-11, 19:59
Take a look at this to get an idea of how military T&E testing might be.

http://www.biggerhammer.net/sigamt/550/550techinspection/

Saw that in a thread a few days ago. It shows all the specified minimums demanded of the SG 550, shows how they get into specifics for each test, etc. Could give you a good idea on how to set it up.

Also, with you talking about firing 720 rounds out of it as fast as you could, it reminded me of the tests they did back in 2008/2009 with the M4 and M4A1 barrels. They fired full-auto for 535 and 900+ rounds before failure (not jams, the weapons were destroyed).

Doesitgobang
04-03-11, 14:21
Howdy all, I am new here and decided to register because of this thread.

After reading your post and looking at what was going on there are a few thing to consider when doing a testing protocol. As Amicus points out a baseline is needed to allow for the evaluation/comparisons between different platforms. Some common goals in testing are:
Safety – What can be safely done with the system?
Reliability – Does it go bang when needed? – will it let you down in a fight?
Performance – Will it hit the target? If it misses you have compromised your position and have a whole other list of issues to deal with.
The original “stress test” of nine magazines of a mix of ammunition has some “Safety” concerns that need to be addressed before the testing not after. The “mortar” extraction of the fail to eject cases was a sign that the test was getting into an unsafe area. The high pressure signs you were seeing were real. The pressure is bumped up due to the hot camber heating the powder and primer to a level that makes them burn quicker giving the over pressure condition. Each type of powder/primer react differently to the heat and gives a mixed result to the test. If continued you will get a blown case or worst yet physical damage to the weapon/shooter. Lets step back for a minute and look at how this testing can be performed without a hospital visit.

Testing of a new system (weapon and ammunition) needs to start with a cookoff test. This test is explained in TOP 3-2-045 section 4.2. http://www.dtic.mil/cgi-bin/GetTRDoc?Location=U2&doc=GetTRDoc.pdf&AD=ADA481861 .

This test is basically firing X amount of rounds as fast as possible then chambering a live round to observe the time it takes to “cookoff” or fire without use of the trigger. The TOP uses a 30 min time. Depending on your level of comfort a shorter time could be used. Start with a low X number and work your way up. Thre is a round count that will cause the round to fire by its self (note how unsafe this event would be outside of a controlled range setting) The number of rounds will be different for different ammunition types or weapons. This test give the maximum number of rounds to fire between cooling.

Lets start there and other aspects of TOP 3-2-045 may be of interest.

Stickman
04-03-11, 15:01
My goal here is to develop a testing protocol for this (and other weapons) that will be defensible. So we can explain to our officers why they can't carry whatever they want and show the command staff and other weapon challenged folks why Brand A may be cheaper but not nearly as good as Brand B. And why quality matters in terms of cost of repairs and in lives.

Input, please.

Thanks in advance,

Allen



Allen,

We developed a check list that is used before any testing ever starts. If the weapon doesn't meet certain the specs, its not accepted. Simple things like a mil spec buffer tube weed out a massive chunk of the ARs on the market. Staking the castle nut and BC are another that weeds out a good amount.

With a solid check list in place, you are able to get rid of wasted time, otherwise you are out testing a weapon looking for failures when you know they aren't up to par to start with.

If you aren't actively looking to transition to piston weapons, do the testing as simple familiarization as opposed to a formal T&E.

CumbiaDude
04-03-11, 17:03
Howdy all, I am new here and decided to register because of this thread.(...)
Reliability – Does it go bang when needed? – will it let you down in a fight?I think that's the best username/first post combination I've seen :D

Doesitgobang
04-09-11, 21:31
Surfing the net I came across some torture testing videos that may be of interest. In the videos a M4 and M4A1 are taken to distruction shooting full auto. Find videos at : http://atwar.blogs.nytimes.com/2010/01/12/m4-and-m4a1-guns/?hp

Enjoy the light show - remember when push comes to shove something is going to give!