Minimum # of rounds through a gun before you trust it??

[ClarkEMyers]

I once carried guns that I wouldn't expect to suvive these tests

I have had aluminum frame Stars that I wouldn't expect to do well after these tests but I once carried them in confidence they would last my basic load.

I long ago had a 5 shot revolver that I loaded with Jeff Cooper's then suggested loads to achieve full service revolver performance from a short barrel and I never supposed that gun would last the this kind of test with that kind of load - in fact it shot loose pretty quickly. The guns are tougher today one way and another.

There are some very well thought of guns out there that fall into the carry a lot shoot a little category that I would carry if at all with a far lower round count than folks here are talking about.

Still today I wouldn't over stress my guns the same way and I'd hope my carry guns (all steel) would be as reliable as a competition shotgun.

Have requirements and expectations changed over time?

 

[Candiru]

Have requirements and expectations changed over time?

I think the customer base who buy guns has changed. 25 to 50 years ago, people bought guns as tools, made sure they worked, and then just kept them around in case of need. This was where gun sales went. Nowadays, I imagine most gun sales go to enthusiasts who enjoy shooting their guns a lot and value durability and longevity that permits them to do so. Improvements in design, manufacturing, and metallurgy permit hardier guns, from which all benefit. The downside is that when only hobbyists buy guns, the base of people who support RKBA is limited to an easily marginalized group.

 

[Malamute]

"I think the customer base who buy guns has changed. 25 to 50 years ago, people bought guns as tools, made sure they worked, and then just kept them around in case of need. This was where gun sales went. Nowadays, I imagine most gun sales go to enthusiasts who enjoy shooting their guns a lot and value durability and longevity that permits them to do so."


I would respecfully disagree.

I've been shooting for longer than 25 years, and most people I knew that bought guns 25 or 30 years ago shot them as much as most I know or hear about today. Same for my father that used to shoot competition shotgun and hunted a lot in years past. There will always be those that don't shoot much, and those that do. We perhaps larger total numbers of gun buyers and shooters today, but we will have both groups represented.

Many older guns seem to last as well as any of the newer guns.

 

[SouthpawShootr]

My general feeling would be 500 rounds to take care of any break-in issues and completely familiarize myself with the gun. Before I trust it, the last 200 rounds should have gone through without a hitch.

 

[Whatsit]

One's enough!

 

[faustulus]

I usually take it to an IDPA or IPSC match using my normal match ammo. Then feed a mag or two of my carry ammo through it to make sure it likes them. Then I feel good to go.

 

[brickeyee]

Can we put a little engineerig and science on this? The Chi-square test is used to establish failure rates and confidence intervals for equipment. It provides both a probability of failure, and a measure of the confidence (odds the calculation is correct).

If you fire the weapon 100 times, there is ~1% chance of a failure on the next shot.
Fire it 1000 times and the odds go to about 0.1%.
The number is slightly different for a real Chi-square calcualtion at 60% confidence, but off only by fractions.

If you want 90% confidence in the number the multiplier for 'shots without a problem goes up by ~2.5x times.
To be 90% confident the gun wil be 99.9% reliable you need to fire ~2500 consecutive shots without a problem.
95% confidence in a 99.9% reliability takes ~3300 shots.
These are the types of calculations performed for 'hi-rel' systems, but instead of shots fired it is hours of use.

Since the requirements for correct operation are pretty stringent (when you need it you really NEED it) the 95% confidence and 99.9% reliability are a sort of bottom line.

 

[ClarkEMyers]

Chi Square is flat wrong in the sense that very strong assumptions are not met

Chi Square is flat wrong in the sense that very strong assumptions are not met. For the less inclined to follow think normal distribution. We can treat the gun as a black box and make assumptions on that basis but they will be wrong or perhaps someone can show the assumptions of Chi sqaure are met or can refer me to a technical discussion showing the assumptions are met or alternatively how the errors are handled.

The glaring example of course is that it's not the same gun in any way shape or form for successive tests. FREX there are the proverbial carry a lot shoot a little guns which have an increasing chance of failure as the number of shots fired without failure increases. Some say a certain examples of very samll .380 chambered pistol fall into this category. Or try higher pressure loads in many guns - compare expectations and test results over time. Even the 1911 in some versions will require repeat require springs replaced in both the pistol and magazines from time to time or a record of fine performance will be ruined.

Quite true that treating the pistol as a black box may require large numbers of shots for high confidence - fortunately for us the pistol is not a black box.

See e.g.

Deming, a disciple of Shewhart was. later to rename the types of variation as common cause. and special cause. ..

 

[brickeyee]

Chi square is used to asses the reliability of hardware all the time.
Every satellite launched uses chi square to predict the reliability of the components used, both mechanical and electrical.
Arhenius has problems with establishing activation energy for mechanical items to try and accelerate testing, but chi square remains a perfectly valid technique to determine the confidence intervals of any piece of equipment.
The deviation used in engineering to determine confidence intervals and assess reliability is the assignment of a failure in the next test. While arguably incorrect from a purely statistical view, it allows the rates to be bounded without actually having a failure (a common occurrence in the real world of testing).
Normal wear and tear is a legitimate part of the reliability of equipment, and the wear accumulated is realistic wear if the test is conducted under realistic conditions.
For a discussion of how chi square and arrhenius are used in eqipment testing I would refer you to almost any semiconductor manufacturers web site. Arhenius is used to accelerate testing, and chi square is used (with failures) to determine MTBF (mean time between failures). Analog Devices has a very good discussion of the methodology.
It is not the same as using chi square to asses the significance of a test outcome - a more common usage.

 

[Magnum88C]

For autoloaders, 200 rounds, mixed brands, of FMJ ammo. Then 100 rounds of carry ammo. Yep, and I carry Hydra-Shoks which run about $1/round here. So whatever I pay for a gun I have to automatically add $100 to the price. I currently have 1 autoloader.

Revolvers, shoot one box of intended carry ammo, then load 'er up and two speedloaders and put a smile on my face.

 

[Gunsnrovers]

I also tend to stick with the 500 + 100 round count for my weapons.

Just did that with my Commander I got back from the gunsmith today. 250 rounds of PMC, 250 of S&B, and 100 rounds of Hornady TAP. Very satisfying afternoon.

 

[TimboKhan]

I say a five or six times through each of your magazines, with a range of different ammo to make sure they work, and there you go. If its a revo, then I would trust mine right out of the box. Now, thats not to say that I don't prefer to shoot my guns a few times before they go into rotation, but I started carrying my current carry revo right out of the box, and it was a week before I could actually take it out and shoot it (I had no problems whatsoever once I got it out). Again, not advocating not shooting it first, but if your skilled enough to be familiar and confident with the functioning and controls (easier yet on a Revo), and it's either your new gun or no gun at all, then I say right out of the box is fine.