jwhittin

Cuz, To answer your original questions, the extreme spread is not really useful for reloading (you can't use it to make any useful conclusions about your reloads). For meeting PF, you should ALWAYS use the Standard Deviation NOT the average velocity converted to PF. You can read the article to explain how to use your SD. Its very simple. You simply add at least 2 twice your measured SD to the minimum velocity required to meet PF for your bullet. This is now your desired average velocity for your loads. So a 180 gn bullet requires 917 FPS to make major. If your SD = 20 then 917 + 2*20 = 957 fps. That's the average velocity you should load for. You can use 2.2, 2.5 or even 3 time your SD as needed for unknown environmental differences or to adjust for reliability and accuracy. As people have pointed out, you won't be able to tell the diff between 168 and 172 PF. This method gives you a high confidence of passing an official PF check using only your first 3 rounds so you don't have to sweat it! From my experience, most people end up with SD measurements in the 10 to 20 range. That's very reasonable. I shoot mixed brass 45 ACP and my current reloading process gives me a consistent 15 fps. Some people are able to consistently get below 10 fps. Using the method above, it doesn't matter as long as you check your SD from time to time. Don't listen to the people saying you should use the ES or measure 20 rounds, or no velocity should be below X, that's all a waste of time and money. You only need to use 8 to 10 rounds to check your SD and velocity. Hope this helps. Shoot with any questions.

Great discussion points Memphis and Brooke! I agree, you have to build up your load for reliability and reasonable accuracy too. No need to go into the minutia, just work up a load that is reliable and reasonably accurate and focus on training and improving your skills (instead of whining) (I completely agree!). The statistical approach is just a place to start which is based on science instead of hearsay (that was my impetus to begin with). We still need to make adjustments for the unknowns or the reasons just mentioned. I do know people that consistently measure SDs in the sub 10 fps and they are just not comfortable running that tight so they bump it up to 3*SD. I agree, most people can't feel the diff between 125 and a 132 PF. Brooke – Sounds like you do check your SD regularly. That’s important. I disagree with your notion that use of the SD is only a point solution. In fact the SD is the sample variance and is a proven way to make sound judgements (inferences) about the performance of your reloads. Then we have to apply our knowledge and experience to make adjustments based on the differences between our baseline SD and match conditions (difference in temp, humidity, altitude, etc.,). I think we are saying the same thing just in a different way of looking at it. Also note, there will be very little difference in SD between chronos (typically 1 to 2 fps but many are within tens of fps). http://www.appliedballisticsllc.com/Articles/ChronographChapter.pdf Off to practice!

Brooke, Why don't you use your STD as explained above? As I point out, if you don't know your STD you can fail at any time if you just add a few PF points OR your chrono is off. Just curious.

Once boxed, I just swipe a sharpie across each row. Takes less than 5 seconds per 100 rnds. Ive used more than one color too.

GLOCKED, In your case, you need a minimum of 850 fps to meet PF. Your STD is 14 fps so 850+28=878 fps (say 880 fps) which is now your desired load velocity. As a general rule, using 2*STD gives you a 90% confidence level you will meet an official PF test using only your first 3 rounds. If there are additional unknowns then add 2.5 or even 3*STD to ensure you meet PF.

GLOCKED, See this posting which will give you a simple way to know you will always meet PF. Bottom line is you should always use the STD to calculate the average velocity needed to make PF. I recommend adding a minimum of twice your measured STD to the average velocity required to meet PF for your bullet weight. There are two reasons why you DON'T want to just add a few power factor points. First is using just the average velocity measurement of your chronograph can introduce significant error as you don't know how accurate your chrono is (see this article http://www.appliedballisticsllc.com/Articles/ChronographChapter.pdf ). Whereas the difference in STD between chronographs is very small because it is a variance. Secondly, if your reloading process changes, it can cause you to fail PF. If you don't use your STD, you would never know (see example in the EB post given in the link above). Shoot with any questions!

I'll toss out another option. I use the SpringCo management systems in several of my comp guns. The design makes a lot of sense, they are made in the USA, and it doesn't require any proprietary main springs.

Excellent resource! Thanks for posting it. I knew there were differences but didn't know how much they varied.

RandomB The connection between the two is quantified by the confidence level of passing PF which corresponds to the value of Z you use (see article above). So in your case, if the conditions when you checked you ammo and when it was tested for PF are the same, then its reasonable to use Z=2 and add 2*STD to the velocity required to meet PF for you bullet weight. Selecting Z=2 gives you about a 90% confidence that you will pass PF using the first 3 rounds of the PF test. If there were unknowns and you wanted more margin, then increase Z. Z=2.5 bumps up the confidence level to about 95%, Z=3 is about 97%. Your 650 load data looks good. Maybe a tad hot unless you want the extra margin. If I use z=2 I get 1044 or higher as a load velocity for a min 90% confidence. Again, that's for all things equal. Adjust from there.

Thanks for the info.

I'm running a JP adjustable gas block and Im getting some black residue on my support hand (just surface residue not hot gasses). Mapped it with white paper around the hand guard and it appears to be coming equally from each side of the gas tube pin. It seems to me it would be hard to seal the pin area completely. I will double check to ensure the pins is installed properly. Just curious if this is normal. thanks

Yep, minimum 2SD. Use 2.5 or 3SD if more margin is needed.

Your average velocity is always determined using 3 rounds, never 7. If you fail the first attempt using 3 rounds, the 3 highest measurements are used out of 6 rounds. If you still fail to meet PF, you have a choice of using best 3 out of 7 rounds or recalc based on a new bullet weight. I much prefer to pass first time and not have to worry about it.

If you do this and use 165, your chance of failing chrono is about 44%.

Here is the link. Forget about PF, what's your SD? http://www.brianenos.com/forums/index.php?showtopic=229005

Did they happen to explain why they don't recommend Titegroup? I see the comment about Titegroup and Vihtavuori N310 on their FAQ section but no details. I assume it is because they both have fast burn rates?

Johnbu, My method does not account for variations in bullet weight. Mainly because a 1 or 2 gn change in weight doesn't impact PF much. So if you are adding 2 to 3 times your standard deviation of your velocity, you don't have to worry about the weight. If you are concerned, stay in the 2.5 to 3*SD range. You can check worst case bullet weight but even at 3 or 4 gns you should have good margin.

They designed it to a specific accuracy based on a marketing/business decision. The design, technology, and techniques in large part determine the accuracy. Probably the most important is the clock speed they use to estimate projectile flight time across the sensors. There are other contributors too like the precision of the sensor distance, sensor performance, and even the manufacturing process.

I thought 20 fps sounded a bit high, but after checking I found that the CE Pro Chrono accuracy is only +/- 1%. So at pistol velocities of about 1000 fps, that can certainly happen if one chronograph reads high and another low. Some manufacturers don’t even state their accuracy, hmm, I wonder why. Others that state their accuracy typically are no more than +/- 0.5% such as Shooting Pro Chrony and Oehler.

Red, I use the average velocity and I don’t worry about the PF except to know the minimum velocity needed. So for a 200 gn bullet, the minimum velocity to make major PF is 165000/200 or 825 fps. So what we want to do is ensure that our average load velocity is at least 2*SD above 825 fps. This gives you about 90% chance of passing the official chrono the first time. So the math goes like this: So 825 + 2 X 16.3 = 857.6 or about 858 fps. As long as your 8 round average velocity is at or above 858 fps, AND your SD remains at or below 16, you are good. If you want more margin you can use 2.5 * SD or even 3 * SD. The same calculation with 3*SD is 825 + 3 X 16.3 = 873.9 or about 874 fps. So as long as you verify your 8 round SD is 16fps or lower AND your average velocity is somewhere in the range of 858 to 874 fps, you are good! I hope this helps.

Good comments! Yes, the rules allow for best 3 of 7. I just wanted to pass using the first 3 rounds and not have to worry about it. It was just the approach I chose which made sense to me. I guess I could update my analysis to use the best 3 of 7. But as several people point out, this result is more conservative.

Brooke, I didn’t claim fewer samples yields a “lower result”. I said more samples is statistically “better”. My point was that more samples always yields less uncertainty (i.e., less variation around the average which is the STD). On that we agree. Your statement “no matter how many rounds” is false. The number of samples does in fact matter. We use samples to estimate sample statistics and make inferences about a much larger population (in this case, the total population of all rounds of the same recipe, the same process, components, etc.). If the Normal distribution applies (assumption #1), it is well known that for most applications, 8 to 10 samples (assumption #2) yields very good approximations for the statistics of the Normal distribution. IF both assumptions are met, THEN use of the Normal statistics is correct. However, once the number of samples used to calculate your sample statistics goes below about 8 to 10 samples, the results diverge from the Normal distribution and the T-distribution should be used. My example of 95.3% versus 85.2% illustrates the correct results using 8 sample versus 3 samples (at 2STD as I recall). “One excellent recommendation is to fire 20 rounds and get a result with an SD of 10 or less.” The SD is a result of the overall process, you can’t “get a result” by using more samples. You only remove uncertainty about your ESTIMATE of the sample statistic by using more samples. That is all. Your recommendation to use 20 samples is unnecessary as the minuscule accuracy gained in going from 10 samples to 20 samples is irrelevant for our needs. The difficult part is in relating our reload chrono results to the USPSA chrono results which is the point of the Z value and the corresponding chance of failure.

Cross post. This seems like a better fit. In preparation for several major matches in the area, a club performed a chronograph check at a local match last weekend. About 6% of the competitors failed to meet their declared PF. This included several experienced competitors. So remember this! The average velocity (and hence PF) you measure at the range will always be better than the official USPSA chronograph results. Yes, it is true! The two measurements are like comparing apples and oranges. It has nothing to do with the small differences between chronographs and of course all the laws of physics apply equally in both cases. So how can I make that claim? It is strictly because of the number of rounds used to determine the average velocity in each case. To pass the official PF test the first time, only 3 rounds are used compared to 8 or more rounds we typically use at the range. The bottom line is that velocity is random in nature and the best way to understand and manage it is to use statistics. So statistically speaking, the term better means less uncertainty. We can quantify this uncertainty in terms of probability which can easily be determined from statistical tables. My post "Reloading to Meet PF with Confidence" located here ( http://www.brianenos...opic=229005&hl= ) illustrates the problem and provides a very simple solution. Also in this post, there is an example which compares the probability of being at or above the same average velocity in each case (at the range versus a USPSA chrono). Using 8 or more samples the result is at least 97.3%, but using 3 samples the result is only 85.2%. And it is important to note that the two results can be much farther apart. A seemingly obvious solution is to just chrono your ammo using 3 rounds. The problem is that using only 3 samples causes wild variations in the results and you will drive yourself crazy trying to get any type of consistent measurement. The good news is that the link above provides a very simple solution to avoid this problem. For a better explanation and more examples, see page 70 of the Jan/Feb 2015 edition of FrontSight Magazine.

Hi Red. What's your bullet weight?

In preparation for several major matches in the area, a club performed a chronograph check at a local match last weekend. About 6% of the competitors failed to meet their declared PF. This included several experienced competitors. So remember this! The average velocity (and hence PF) you measure at the range will always be better than the official USPSA chronograph results. Yes, it is true! The two measurements are like comparing apples and oranges. It has nothing to do with the small differences between chronographs and of course all the laws of physics apply equally in both cases. So how can I say that? It is strictly because of the number of rounds used to determine the average velocity in each case. To pass the official PF test the first time, only 3 rounds are used compared to 8 or more rounds we typically use at the range. The bottom line is that velocity is random in nature and the best way to understand and manage it is to use statistics. So statistically speaking, the term “better” means less uncertainty. We can quantify this uncertainty in terms of probability which can easily be determined from statistical tables. My post "Reloading to Meet PF with Confidence" located here ( http://www.brianenos...opic=229005&hl= ) illustrates the problem and provides a very simple solution. Also in this post, there is an example which compares the probability of being at or above the same average velocity in each case (at the range versus a USPSA chrono). Using 8 or more samples the result is at least 97.3%, but using 3 samples the result is only 85.2%. And it is important to note that the two results can be much farther apart. A seemingly obvious solution is to just chrono your ammo using 3 rounds. The problem is that using only 3 samples causes wild variations in the results and you will drive yourself crazy trying to get any type of consistent measurement. The good news is that the link above provides a very simple solution to avoid this problem. For a better explanation and more examples, see page 70 of the Jan/Feb 2015 edition of FrontSight Magazine.