45 Case length affect on velocity

Withdrawn.

Sorry to trouble you with a quick experiment I found intresting.

What are we supposed to do with it?

All I can see is that if you use less of the same powder, the bullet is slower. Just as expected.

Case length has little to no bearing on much of anything, the differences are minuscule to begin with.
If each id of the case is the same dimension, each bullet seats at the same depth then case volume will be near exactly the same for a given set of ammunition components. A better way to measure case length would be to measure from the floor inside to the case lip. then you know what the volume may be for each case ? Yes ? No ? Just a two cent version here #dont flame too hard.

The thing i took away us the change in standard deviation

Are you talking about the two examples of 3.1 gr E3? 

That's the only comparison you have in that data set. Everything else is either a different powder altogether or diff weights of the same powder.

How about giving a reason for all the other data you posted?  It's hard to sense out of what point you're trying to make if you don't tell us, especially given that your data is barely related at all.

And how about telling us how many shots you fired for each of the data you posted. It's hard to make sense out of this unless you post more details. The details matter.

I take the data to demonstrate no appreciable deviation affecting accuracy.
Thanks for the effort.

A couple of winters ago when I was snowed in and bored I sorted out the shortest and longest cases I could find loaded them up and measured crimp variation.  Then I took them to the range (25yd) used a ransom rest to check group size, I also checked out their std. deviation with my LabRadar.   

Was the effort worth it?  Well yes because it kept me from being bored.

I do think experimenting to see what does or doesn't have an impact on accuracy is good because (hopefully) it keeps you focused on the important things and eliminates possible obsessions with things that don't matter.
- Dave

The velocities you presented may or not be significantly different .

If you or others would like someone to more rigorously compare the data you presented, you also will need to provide n=number of observations, e.g., 742 fps, sd= 14 (n=7).  Of course comparison of data is dependent on care and technique of collection.

If you are interested and can't do the comparisons on your own, others mght step up to help.  Statistical tools are available on many calculators and on Excel.

Thanks for offer shark doctor, but ive got pretty technical background and can do more in depth analysis if I want to. My thought was to present a short data set for others to review. I thought my post was already getting verbose so I didn't include every little detail. As this wasn't recieved well, I've learned my lesson, and will keep my data to my self I the future. 

PS, n=10

Well, 10sandXs, considering some of the many posts we receive on this list, your post was not at all verbose, and actually contained some useful info for those who ask for load suggestions and resultant velocities, so don't keep your data to yourself.  Don't let comments bother you - this is the Internet!

10sandxs wrote:Thanks for offer shark doctor, but ive got pretty technical background and can do more in depth analysis if I want to. My thought was to present a short data set for others to review. I thought my post was already getting verbose so I didn't include every little detail. As this wasn't recieved well, I've learned my lesson, and will keep my data to my self I the future. 

PS, n=10

Asking you questions about your data translates into not being well received?

S148, its more about the tone, which I took as a bit derogatory and somewhat condecending. 

My intent wasn't to do a 3 factor 3 level RSM with full ANOVA and paired t-tests to indicate the probability of stastical significance. I was preparing some different loads based on previous data and decided to throw in the case length test and it's affect on velocity distribution as an after thought based on a recent post relating the two. My take away was that velocity may be affected by length due to more or less crimp, but probably not enough to matter, especially at the short line which is what this will be used for.  

Thank you for your input, I'll keep it in mind should I decide to post data in the future.

Now we get to see some of the reasons why you posted the data. Having this information included with your data would have been useful. It would have explained the point of posting it and what you would want readers to look at, think about and discuss. Not including that information means people are left guessing what the point is. See what I'm saying?

How about posting it again and see what readers think about it. Are there any other things you would like to readers to think about? 

I'm sorry if direct statements offend you.

Actually, I got what he was saying. That there is little statistically valid velocity variations with cases having random lengths vs cases with uniform and/or uniformly long lengths. Much as I suspected.

The negative tone was perhaps brought by a thin explanation of the data and perhaps, that a sacred cow was being slain. I mean everybody just knows that random case lengths means random crimps and random headspace. That can’t be good for consistency, can it? It turns out that it might not have much effect.

The deletion of the data was unfortunate.

Wobbley wrote:I mean everybody just knows that random case lengths means random crimps and random headspace.  That can’t be good for consistency, can it?   It turns out that it might not have much effect.  

Hard to say without it being repeated, of course; and a sample size of 10- well, not much to say there, don't you think? A Students-t test would, bare min, required 20 samples ea., compared against others of same size, right? 

It turns out it might have a great deal of effect (contra your speculation)- but we don't know because of the paucity of data.

Joe Morgan wrote:

Wobbley wrote:I mean everybody just knows that random case lengths means random crimps and random headspace.  That can’t be good for consistency, can it?   It turns out that it might not have much effect.  

Hard to say without it being repeated, of course; and a sample size of 10- well, not much to say there, don't you think? A Students-t test would, bare min, required 20 samples ea., compared against others of same size, right? 

It turns out it might have a great deal of effect (contra your speculation)- but we don't know because of the paucity of data.

20 samples in specially prepared barrels of minimum “Headspace” to maximum headspace. Say 3 barrels each tested in a barrel tester after determining the “best” load. So it can get to be quite the exercise. Especially when a simple test in your gun can give a valid test if enough shots are fired. Say 3 strings of 15 should give a great indication. Will it pass a full-up statistical battering? Probably not but it’s still a good indicator. Maybe that what the OP intended here.

Certainly I’m not gonna bother. I’m still working my way through a 20mm ammo can full of once fired 45 TZZ brass.

I have proven to myself the validity of the testing that Jerry Keefer and Col. Bacon did with shortening .460 Roland cases to fit match chambers. 

Longer in my Ransom Rest tests equals better. At the Sharpshooter level the improvement is marginal. But I know High Masters who debur flash holes for the long line. Ironically they tend to win.