•••Load Testing Procedures•••?

[the Black Spot]

When I test handloads, I use a good benchrest set up. I usually do 3 shots, swab the barrel twice with a brush and cleaner, then run patches through till they come out clean. Then I let the barrel cool a little, then test the next 3 rounds of a different load. Then repeat process. Am I wasting time doing it this way? Trying to eliminate all variables in the testing.
thanks!

 

[Ru4real]

Not a waste of time if you want a clean barrel for each group.

Definitely a waste of time if you are looking for the best (least cost in time and money) way to find the best (tightest group) loads for your rifle.

The best way will be along shortly.

 

[NMexJim]

Depending on how you feel about Hornady, you might watch the Hornady Podcast 50 and 52, the latter being a follow up to the first. These are looonng and at times the important stuff is at the end.

Personally, I’ve worked 3 and 5 shot groups over the years. I’ve cleaned as you and allowed barrels to cool. If there is a ladder system, I’ve tried it.
I did get away from chasing lands.

I have seen some weird results with ladder systems. I’m a careful handloader that puts a lot of effort in case prep and measurements in order to have consistent loads.

I worked with statistical analysis in my job. For some reason I did not equate sample size (duh!) with extreme spread and standard deviation for ladder work. The larger the sample size, the more accurate the data.

I found those podcasts answering some of my head scratcher moments. Put aside an hour or two and watch these and see if they make some sense to you.

I think 3-shot is still useful for quick elimination of some combinations. Then going to 10-shot begins to tell you what you really have. Not many of us can run sample sizes as large as Hornady does.

Anyway, interesting.

 

[Varminterror]

When I test handloads, I use a good benchrest set up. I usually do 3 shots, swab the barrel twice with a brush and cleaner, then run patches through till they come out clean. Then I let the barrel cool a little, then test the next 3 rounds of a different load. Then repeat process. Am I wasting time doing it this way? Trying to eliminate all variables in the testing.
thanks!

Cleaning like this is introducing variables, not eliminating them.

 

[JFrank]

When I test handloads, I use a good benchrest set up. I usually do 3 shots, swab the barrel twice with a brush and cleaner, then run patches through till they come out clean. Then I let the barrel cool a little, then test the next 3 rounds of a different load. Then repeat process. Am I wasting time doing it this way? Trying to eliminate all variables in the testing.
thanks!

Yeahhh, your probably introducing a layer of complexity you could do without as from a clean barrel we have extra friction, I like a few fouler shots and get the barrel warmed up etc. from there I have my test rounds ( and wind flag ) set up and fire my test in consecutive manner at a medium pace to capture the same conditions for each set of rounds and not wait for the barrel to cool off unless it’s a pencil barrel hunting rifle and if that’s the case I just use less rounds but the same method.
Best wishes
Jim

 

[AJC1]

I test the same bullet, same powder just at different volume, and clean when I get home. Bullet jacket material and coatings are different and I start with a clean barrel to prevent any issue with that.

 

[Varminterror]

Reiterating here - the inherent variability of a population, a data set, dictates the sample volume required to establish whether the sample set is truly representative (statistically valid) of the population, or not (statistically invalid). @denton mentioned the decreasing value in additional shots fired between 5 and 10 in another thread this week - entitled “5 shot groups are near worthless” or something of the sort.

We do have the complication that not all error sources in rifle shooting follow Gaussian/Normal distributions - not all errors present a “bell curve” - ESPECIALLY not mechanical errors by the shooter (these can be “on/off,” bi-nodal, with a relatively bell shaped node happening at each position, for example, flinching vs. not for some shots, or slapping trigger vs. follow through, OR they may be directionally biased, for example, ignoring wind, or jerking a trigger which is always in one direction, or subconsciously favoring an imbalance sight picture, or parallax via improper head position).

So Gaussian statistics don’t always offer applicable predictability, but we do know, if a shooter is putting up big groups, they need more shots to define their variance and confidence intervals. We really have to identify which are the dominating sources of error, and analyze the system to determine the appropriate number of shots - in general, we know it’s more than 1 shot per group, but stating a validity threshold of 3 or 5 or 10 or 50 really isn’t sensible - because what is valid for one population (data set) may not be valid for another. I’ve had systems which required more than 50 samples to approach differentiated data between step changes. We really can’t assign blindly that “5 isn’t enough, you need 10,” just the same as we shouldn’t blindly assume “10 is always better than 5.”

 

[Varminterror]

***of note: when load testing 22LR, brushing out and refouling with ~5-10 rounds of the new ammo is usually necessary. 22LR behaves very differently when fouled with its own lube vs. what amounts to contaminated fouling from other lube from other ammo. Centerfire doesn’t typically have this issue (some guys talk about extreme detail of varying jacket alloy, or different powder fouling, but I shoot smaller than most folks, and I’ve tried and can’t shoot the difference between two jacket alloys or powders. MAYBE the moly coating guys have a point).

 

[AJC1]

Reiterating here - the inherent variability of a population, a data set, dictates the sample volume required to establish whether the sample set is truly representative (statistically valid) of the population, or not (statistically invalid). @denton mentioned the decreasing value in additional shots fired between 5 and 10 in another thread this week - entitled “5 shot groups are near worthless” or something of the sort.

We do have the complication that not all error sources in rifle shooting follow Gaussian/Normal distributions - not all errors present a “bell curve” - ESPECIALLY not mechanical errors by the shooter (these can be “on/off,” bi-nodal, with a relatively bell shaped node happening at each position, for example, flinching vs. not for some shots, or slapping trigger vs. follow through, OR they may be directionally biased, for example, ignoring wind, or jerking a trigger which is always in one direction, or subconsciously favoring an imbalance sight picture, or parallax via improper head position).

So Gaussian statistics don’t always offer applicable predictability, but we do know, if a shooter is putting up big groups, they need more shots to define their variance and confidence intervals. We really have to identify which are the dominating sources of error, and analyze the system to determine the appropriate number of shots - in general, we know it’s more than 1 shot per group, but stating a validity threshold of 3 or 5 or 10 or 50 really isn’t sensible - because what is valid for one population (data set) may not be valid for another. I’ve had systems which required more than 50 samples to approach differentiated data between step changes. We really can’t assign blindly that “5 isn’t enough, you need 10,” just the same as we shouldn’t blindly assume “10 is always better than 5.”

A sample that is valid for one inquiry might not be valid for other reasons. I consider a 10 shot group valid, because it directly reflects the 10 shot strings in metallic Silhouette. That may not be statistically valid, but I'm not a scientist claiming that it is. Maybe running tests that help accomplish or validate your needs are good enough....

 

[the Black Spot]

(Btw, savage 110 in 25-06, hunting rifle as test example)
So if I start with a clean barrel. Fire 3 shots of my test loads, let’s say I get 1.5” group at 100yds.
I can let barrel cool and fire my next 3 test rounds without the fouled barrel scewing the data? What if this set gets a 2” group, but would have been a 1.25” group if I had cleaned the barrel?

 

[kmw1954]

When doing OCW I do 3 shot groups shot in a round robin fashion just to get a baseline. Look for two that are next to each other as accurate. I then reload those and shoot again to see if they repeat. If so continue if not move on. If they do repeat I will load again and fill in the middle and repeat once again. If the testing for me is not repeatable then it is useless.

 

[Bazoo]

Benchrest shooters clean between groups or number of groups depending on how their guns behave. So it's not something that's weird. But it's not normal for anything other than the most extreme accuracy to clean between groups when testing loads.

 

[Blue68f100]

All depends on your application. If your working on a load for a hunting rifle, your 1st shot is the most important. Most always your shooting a cold barrel. You have to decide if you want your barrel fouled or clean.

I always shoot a few fouling shots. This allows me to check out my scope zero, crony working .... For my bench rifles I normally only shoot 3 shots unless something stupid happened with the old man pulling the trigger. When testing I monitor barrel temp, if too hot to hold onto, I allow it to cool. I normally allow 5 min cool down between 3 shot groups. During the summer time I used a forced air cooling fan to cool the barrel.

 

[Varminterror]

I consider a 10 shot group valid, because it directly reflects the 10 shot strings in metallic Silhouette. That may not be statistically valid

I’m gonna build on this, hopefully let somebody here make decisions about how they design their load development methods. Because this is a great example of observer bias in experimental design, a pretty common trap folks fall into - it’s the same justification as the guys which say “only the first cold bore shot counts, because I’m a hunter, so I don’t shoot groups.”

No industry lives like this: we always include some other confidence factor. Whether we call it safety factor, design factor, proofing factor, engineering factor, confidence factor, whatever you want to call it, it’s a means of creating confidence that our future results will OUT-perform the necessary expectation of the task to be accomplished. Folks don’t design systems to ONLY deliver that which is expected during normal operation. Elevators with maximum occupancy ratings of 12 people don’t collapse with 13. Cars which need to travel at 75mph on the interstate don’t run out of power at 76mph. We overdo our tests to ensure we will get the performance we need.

So we see a lot of benchrest shooters shooting 20 shot groups, so when they send 5 or 10 for score in competition, they know every shot will fall within the group, and so they know if their load throws occasional flyers, if they send a handful of 20 round groups without flyers, they’ll be quite confident their 5 or 10 shot competition groups won’t send flyers either. Interpolation is always greater confidence than extrapolation.

 

[Varminterror]

Benchrest shooters clean between groups or number of groups depending on how their guns behave. So it's not something that's weird. But it's not normal for anything other than the most extreme accuracy to clean between groups when testing loads.

Clarifying though - benchrest shooters don’t start match strings without fouling, either intentionally or via sighters.

Cleaning then directly firing groups for load development is not the same thing, and is introducing variability to the OP’s process.

 

[the Black Spot]

Ah, I see. So I could clean, fire a fouler, then test the next group.

 

[JFrank]

Benchrest shooters clean between groups or number of groups depending on how their guns behave. So it's not something that's weird. But it's not normal for anything other than the most extreme accuracy to clean between groups when testing loads.

Good eye’
The short range Benchrest crowd have the advantage of not pulling pit duty, allowing time enough to clean and reload between relays, starting off with a clean barrel each relay. @Walkalong might chime in with his first hand experience. We long range fellas build our loads a little differently because of the pit duty for example, some guys go two days without cleaning while others clean at the end of the day. That just depends on how your rifle reacts.

 

[jmorris]

My procedure depends on a number of variables, I don’t treat them all the same or have the same expectations.

I do often remove as much “human” from the equation, as I can.

 

[Varminterror]

Ah, I see. So I could clean, fire a fouler, then test the next group.

Fire a few foulers each. Most of my rifles will take 3-8 foulers to stabilize after cleaning, and most will send that first few rounds out of the rest of the group, often by as much or more than the rest of the group size.

 

[barnfrog]

(Btw, savage 110 in 25-06, hunting rifle as test example)
So if I start with a clean barrel. Fire 3 shots of my test loads, let’s say I get 1.5” group at 100yds.
I can let barrel cool and fire my next 3 test rounds without the fouled barrel scewing the data? What if this set gets a 2” group, but would have been a 1.25” group if I had cleaned the barrel?

I think the only way to answer this for your gun is to try both. If that second group is 2" without cleaning, try doing your cleaning routine and fire another group with that second load combo. Keeping in mind all the statistical caveats about small sample size, if the load prints smaller groups after cleaning, maybe your gun shoots best with a clean barrel. My guess is you won't find much difference, though.

One bit of wisdom I've heard a lot is "the gun will tell you when it needs to be cleaned." In other words, when barrel fouling reaches a certain point, it will start to affect performance. In order to find out when your barrel reaches that point, you keep firing the same load without cleaning until the group starts to open up. Could be 50 rounds, could be 500. I'm betting it won't be 5.

 

[AJC1]

I’m gonna build on this, hopefully let somebody here make decisions about how they design their load development methods. Because this is a great example of observer bias in experimental design, a pretty common trap folks fall into - it’s the same justification as the guys which say “only the first cold bore shot counts, because I’m a hunter, so I don’t shoot groups.”

No industry lives like this: we always include some other confidence factor. Whether we call it safety factor, design factor, proofing factor, engineering factor, confidence factor, whatever you want to call it, it’s a means of creating confidence that our future results will OUT-perform the necessary expectation of the task to be accomplished. Folks don’t design systems to ONLY deliver that which is expected during normal operation. Elevators with maximum occupancy ratings of 12 people don’t collapse with 13. Cars which need to travel at 75mph on the interstate don’t run out of power at 76mph. We overdo our tests to ensure we will get the performance we need.

So we see a lot of benchrest shooters shooting 20 shot groups, so when they send 5 or 10 for score in competition, they know every shot will fall within the group, and so they know if their load throws occasional flyers, if they send a handful of 20 round groups without flyers, they’ll be quite confident their 5 or 10 shot competition groups won’t send flyers either. Interpolation is always greater confidence than extrapolation.

Well I'm not in industry so I don't have their capitol resources or profit off my output. A test procedure for those at a much higher level with money for travel, hotel, entrie fees, sponsor expectations etc, probably won't be using my plan. One has to operate within the limitations of their resources, and the equipment. My 308 is a production gun, I could test much more extensively but the result isn't getting custom build small. My levers aren't either... I wish there was more detail in the op to give clarity on goals and platform.

 

[Varminterror]

@AJC1 - my explanation above was rather to point out a broader paradigm, not referring to your personal application.

BUT…

The principle applied in your personal case would work like this: Quite likely, if you worked load development LARGER than your competition strings, instead of matching them, your scores would likely improve. So when you do spend money on hotels, travel, meals, entry fees, etc, your investment at the bench pays back more.

 

[AJC1]

@AJC1 - my explanation above was rather to point out a broader paradigm, not referring to your personal application.

BUT…

The principle applied in your personal case would work like this: Quite likely, if you worked load development LARGER than your competition strings, instead of matching them, your scores would likely improve. So when you do spend money on hotels, travel, meals, entry fees, etc, your investment at the bench pays back more.

I still believe that more detail would significantly help refine the responses. I don't respond to BR posts because I don't shoot at that level. I'm inclined to believe this guy is shooting a bolt in some level of compotion, no idea what level...

 

[thump_rrr]

I think the only way to answer this for your gun is to try both. If that second group is 2" without cleaning, try doing your cleaning routine and fire another group with that second load combo. Keeping in mind all the statistical caveats about small sample size, if the load prints smaller groups after cleaning, maybe your gun shoots best with a clean barrel. My guess is you won't find much difference, though.

One bit of wisdom I've heard a lot is "the gun will tell you when it needs to be cleaned." In other words, when barrel fouling reaches a certain point, it will start to affect performance. In order to find out when your barrel reaches that point, you keep firing the same load without cleaning until the group starts to open up. Could be 50 rounds, could be 500. I'm betting it won't be 5.

I'm with you on this one.

 

[JerkyFreak]

I don't care which type of competition you shoot, but if 5 rounds from clean negatively affect the group size, you are way overdue for a new barrel.

My load development is done with a fouled barrel. And my cleaning regime is set up so that the clean barrel poi is as close to the fouled barrel poi as possible.

Lets also differentiate between clean barrel and cold barrel. Clean bore I get a POI shift; cold bore no shift.