Primer differences

[TRG65]

Please let me know if these results make any sense to the rest of you.  I tested 3-30 round groups of 9mm loads.  All 3 sets were loaded same day, same powder, same OAL, same box of bullets and of course same press.  The only differences were the primers I used and I used mixed headstamp brass.

147 gr West Coast Gold bullet

3.0 gr VV-320

FSPM

Avg 879.8 fps

SD       9.6 fps

ES     32.1 fps

WSPM

Avg 900.1 fps

SD    10.4 fps

ES    37.2 fps

WSR

Avg 890.2 fps

SD     10.0 fps

ES     40.0 fps

I didn't expect 20 fps difference between brands of SPM primers but I can accept that.  I thought that small rifle primers would be hotter than pistol primers even magnum primers just because they are expected to ignite larger charges of slower burning powders.

[Loves2Shoot]

Your test proves one thing only, it could be the primers, or it could be the brass. You have to use same headstamp, same weight brass to issolate the primers. Mixed brass won't cut it.

[TRG65]

If I was trying to make a case for the SD or ES being great then yes I would agree.  But with the SD's and ES's all being fairly equal.  I felt each sample had an equal amount of variation in brass.

If all samples see the same random variable doesn't that variable cancel out when comparing large samples?

(Edited by HighTechRedneck at 9:25 pm on Feb. 19, 2003)

[bhg]

I believe for a true random sample you would have to start with equal amounts of each headstamp.  ie: say you have 300 cases, 100 would have to be win, 100 rem, and 100 PMC.  Then you could draw 30 rounds for FSPM, replenish equal amounts and draw the next 30 for WSPM and so on...

But then it has been a while since that statistics course...when will I forget this crap? :-)

[kimel]

Quick and dirty stats:  If the mean +/- the SD overlap then your data is probably not statistically different.  It might "look" significant but it probably isn't.

Looking at your results it does not appear that you have statistically different samples.  Some interesting trends, but nothing that will stand up to a true test.

Your sample sizes are probably too small to pull out any differences that might be there.

As others mentioned: go with single headstamp brass and try again.  Bigger sample sizes are almost always better.

You also need to be careful about pistol condition.  Dirty barrel vs clean barrel.  Hot barrel vs cold barrel.  All these things will have an effect on velocity.  Don't forget temp of ammo, temp of air, humidity etc.  Chrono everything under as close to the same conditions as possible.  The name of the game is to eliminate as many variables as possible so that any difference can be totally attributed to the one variable you are interested in.  Easier to do in laboratory conditions than in the field that is for sure.

All that having been said...I believe it is fairly well documented that various primers will yield a difference in velocity all other things being equal.

I have known people that will rechrono and adjust whenever their primer or powder lots change because they are convinced that the difference MUST be accounted for and removed.  I may be Type A but I am NOT that much of a Type A.

Hope that helps!

Cheers!

Kevin

[Singlestack]

My very limited experience with reloading has led me to think that WSR primers give me less velocity than WSP primers. I started loading .40 with WSP and then switched to WSR's and my results were very simular to HTR's. I use N320 for .40.

I just worked up a load for my 38 Super and had the same results. I use N350 for 38 Super.

[DogmaDog]

Kimel is right.

I messed around on Excel (bored at work today), and crunched some "slow and dirty statistics" (Excel is sucky for stats).

So, given the real, no kidding TRUTH, that the mean velocity is 890.0 fps (the mean for all 90 of your shots), and the std. D really is 10.0 (you may want to calculate the Std. D for all 90 shots--it ought to be lower than any of the Std. Ds for 30 shot strings), and the distribution really is normal (as in the mathematical function "normal distribution", not as in "the kind of thing I normally expect to see"), the probability that you'd sample 30 rounds, and measure a mean velocity of 879.8 fps or less is about 15%.  

Scientists generally don't consider a result "significant" until the probability is only 5% or less that you'd see that great a difference.

In any case, 15% is likely enough that we might just be seeing random variation within one population, rather than systematic variation between three populations.  

Collecting more data would help resolve the issue, as would more sophisticated statistics (ANOVA would work well)...but really, is there any need?  

Why not just pick one primer and go with it?  

Good luck,

DogmaDog

[TRG65]

I remember reading somewhere very important that thou shall not covet.  That being said I've been coveting a nice AR-15 for about the last year and a half.  I have shot some less expensive ones but don't want to spend 600 to continually want to upgrade.  But eventually I will get it and then I'll do what Dogma suggests, pick one and go.  I'll probably use the WSR's.  In the mean time I'm going to use up the 9,000 FSPM's I have and drop the powder charge.  If I'm getting 880 and 10 fps SD I'm too far away from 125 pf.  I need to drop back to the low 860's....I'll save .1 grains of powder.