Results 1 to 17 of 17

Thread: Does the Eley Lot Analyzer predict ammo performance in custom .22lr BR rifles?

  1. Default Does the Eley Lot Analyzer predict ammo performance in custom .22lr BR rifles?

    For those who just want to know the answer -

    It is doubtful.

    I see little or no evidence that the information provided by Eley correlates with the performance of their ammo in custom .22lr rifles using match-grade barrels.

    To learn how I came to that conclusion, read on. In Part I I’ll discuss my analysis of reported Analyzer data. In Part II, I’ll continue with a comparison to measured performance of custom match-grade barrels.

    Part I – Tenex and Match ammo
    First I performed a general survey of Tenex and Eley ammo available in the USA during 2016 (prior to the release of the Analyzer). I don’t have a list of all lots for that year, but starting with available lists of inventory at Zanders, I identified 52 lots of Tenex and 116 of Match whose performance was reported. Although it is likely I missed some, I suspect these two groups are representative.

    Figure 1 indicates the number of rounds in each lot of Tenex. The average size was 28,500. The largest and smallest lots contained 39,000 and 6,000 rounds, respectively.

    Click image for larger version. 

Name:	Fig1.jpg 
Views:	9 
Size:	65.6 KB 
ID:	2741

    Figure 1 – Lot sizes of typical Tenex ammo.

    The average velocity was 1062.8 fps. As shown in Figure 2, the range spanned 1040 to 1091 fps.

    Click image for larger version. 

Name:	Fig2.jpg 
Views:	9 
Size:	77.3 KB 
ID:	2742

    Figure 2 – Reported velocities of typical Tenex ammo.

    In order to compare an equal number of lots of Match and Tenex, I randomly selected 52 lots of Match. The number of rounds in each of those lots is shown in Figure 3. The average size was 42,000. The largest and smallest contained 75,760 and 18,200 rounds, respectively. Based on the number of lots I found, and the average number of rounds in each, it appears Eley makes (exports to the US?) as much as four times as much Match as Tenex per year.

    Click image for larger version. 

Name:	Fig3.jpg 
Views:	9 
Size:	57.2 KB 
ID:	2743

    Figure 3 – Lot sizes of typical Match ammo.

    The average velocity was 1061.7 fps. As shown in Figure 4, the range spanned 1040 to 1084 fps. These values are similar to Tenex.

    Click image for larger version. 

Name:	Fig4.jpg 
Views:	6 
Size:	58.2 KB 
ID:	2744

    Figure 4 – Reported velocities of typical Match ammo.

    The above summary did not require the Analyzer. Although it would have been a little more challenging, lot size can be determined by how many rounds initially appear in Zanders inventory.

    New Analyzer data
    In principle, all you need to know about a lot is the mean radius of its impact distribution. All other forms of measurement are likely redundant or irrelevant. I’ll return to this later, but first let’s focus on what Eley reports in the Analyzer. Presumably these are some of the factors used to insure that ammo meets their standards.

    One possible criterion can be based on the edge-to-edge size of 10-shot groups, measured at 50 meters. Shooting five for each of the four test barrels produces a total of 20 groups (200 rounds). The average of all 20 isn’t reported, and is tedious to calculate for 100 lots.

    The “Target Rings” graphic supplied by the Analyzer (Figure 5) appeals to me. It is a coarse measure of the impact distribution. Intuitively, “better” ammo should produce more impacts near the center. In addition, for a given distribution shape, the percentage of impacts within a given distance from the center should be proportional to the mean radius. It is a reasonable candidate for screening lots.

    Click image for larger version. 

Name:	Fig5.jpg 
Views:	12 
Size:	73.5 KB 
ID:	2745

    Figure 5 – Example of Target Rings representation of impact distribution.

    Figures 6 and 7 show the reported impact performance of Tenex and Match lots, respectively.

    Click image for larger version. 

Name:	Fig6.jpg 
Views:	14 
Size:	90.7 KB 
ID:	2746

    Figure 6 – Tenex lots ranked by the percentage of rounds impacting within 0.2 inch of target center.

    Click image for larger version. 

Name:	Fig7.jpg 
Views:	11 
Size:	60.9 KB 
ID:	2747

    Figure 7 –Match lots ranked by the percentage of rounds impacting within 0.2 inch of target center.

    Based on Analyzer data, as a rule, Tenex appears to perform better than Match. The average number of rounds impacting within 0.2 inch of target center for Tenex and Match are 89.6% and 85.3%, respectively. About half of the lots of Tenex lie above 90% and none below 87%. In contrast, few lots of Match lie above 90% and nearly half are below 85%. The data from both plots are combined in Figure 8.

    Click image for larger version. 

Name:	Fig8b.jpg 
Views:	15 
Size:	97.1 KB 
ID:	2752

    Figure 8 – All lots of Tenex and Match ranked by the percentage of rounds impacting within 0.2 inch of target center.

    Figures 6 - 8 suggests that Eley generally knows how to control ammo performance. When they start with raw materials that meet more rigorous requirements, they usually produce ammo that meets Tenex performance standards. That isn’t always the case. Sometimes during final qualification, some lots fail to meet test criteria and are sold as Match.

    There is some overlap between Match and Tenex rankings. Despite starting with materials that meet less rigorous requirements, some Match ammo not only meets Tenex performance standards, a few lots appear to perform exceptionally well. However, those lots still are sold as Match.

    I was surprised to see that 12 of the 52 lots of Match were intended to be Tenex. If we include them in the analysis of Tenex, the spread shown in Figure 6 increases (Figure 9). The average number of rounds impacting within 0.2 inch of target center decreases slightly from 89.6% to 89.1%. And 7 of those 12 lots were below 87%.

    Click image for larger version. 

Name:	Fig9N.jpg 
Views:	5 
Size:	96.7 KB 
ID:	2755

    Figure 9 – All intended Tenex lots ranked by the percentage of rounds impacting within 0.2 inch of target

    Similarly, designating those same 12 lots as Tenex changes Figure 8 to Figure 10.

    Click image for larger version. 

Name:	Fig10N.jpg 
Views:	4 
Size:	44.5 KB 
ID:	2756

    Figure 10 – All lots of intended Tenex and Match ranked by the percentage of rounds impacting within 0.2 inch of target center.

    Eley’s control of performance doesn’t look as favorable. Their manufacturing processes influence the outcome, providing an edge to those intended to be Tenex. But the spread and overlap of results indicate that other variables are significant and out of their control.

    Performance compared to competitive criteria
    The percentages shown in the Target Rings should raise concerns among bench rest shooters. At 50 yards, for a rifle that shoots an aggregate of 2200, 95% of impacts should lie within 0.2 inch. That number increases to 98% for one that aggs 2300. In order to compare to Eley’s reported numbers, expected impact distributions must be adjusted to 50 meters. For example, rifles that agg 2200 and 2300 at 50 yards can be expected to land 92% and 96%, respectively, of their impacts within 0.2 inch. The data from Figure 8 reappear In Figure 11 along with dashed lines indicating expected ARA scores.

    Click image for larger version. 

Name:	Fig9b.jpg 
Views:	15 
Size:	50.0 KB 
ID:	2754

    Figure 11 – Lots from Figure 8 ranked by the percentage of rounds impacting within 0.2 inch of target center. Dashed lines indicate three levels of expected ARA aggregates.

    When shooting even the best lots, Eley’s test barrels likely will score below average against custom rifles at a typical match. A large fraction would achieve very disappointing aggs.
    For comparison, the most probable indoor ARA agg is around 2200. (See Post #6 in http://www.rimfireaccuracy.com/Forum...-Probabilities). Some competitors even achieve seasonal indoor aggregates above 2300.

    If match performance actually met Eley’s qualification results, most people I know wouldn’t buy their ammo.

    Since custom bench rest rifles score significantly higher than Eley’s results suggest, I conclude that the barrels they use to qualify their ammo do not meet bench rest standards. If test barrels are inferior, I don’t know if they started at that level, or if they degraded to that point due to the many hundreds of thousands of rounds they’ve seen. Despite claims made on their website, I have seen no convincing evidence that accuracy is unchanged by excessive use. However, accuracy may show little change after performance decreases below some threshold.

    Analysis Methodology
    Ideally, Eley would supply the impact coordinates for each round. If we had that data, we would calculate the standard deviations of the distributions along x and y axes (and the related mean radius).

    Even though the raw data is unavailable, indirectly, Eley supplies the standard deviations. A tedious, but straightforward, method to determine them is to measure the shapes of the Gaussian curves they supply, as shown in Figure 12.

    Click image for larger version. 

Name:	Fig10.jpg 
Views:	14 
Size:	59.4 KB 
ID:	2750

    Figure 12 – Typical representation of impact distributions.

    As I looked closely at these curves, I realized something was amiss. Reported standard deviations for windage and elevation were always the same! I know from simulations and analysis of real data that they almost never are.

    Take a closer look at the above figure. The elevation curve fits the data pretty well. But the windage curve does not. In Figure 13, the red curve is my rendering of what the distribution should be. Note that both center and sharpness were off

    Click image for larger version. 

Name:	Fig11.jpg 
Views:	20 
Size:	63.3 KB 
ID:	2751

    Figure 13 – Typical representation of impact distributions with better fits (red curves) to distributions.

    In addition, the curve representing the radial distribution function is wrong. The function is well known, so I am puzzled why Eley didn’t use it. Its shape only depends on the average standard deviation of the impact distribution. The correct function is represented by the red curve at the lower right. Note how much better it fits the distribution. Consequently, values in the Target graphic likely are incorrect. Using the correct function could increase expected scores. On the other hand, I cannot be sure there aren’t other flaws in their analysis.

    I don’t mean to offend anyone. People make mistakes (me included!). But these systematic errors are too important to overlook. They affect the results displayed in the Target Rings.

    Group size
    The reported size of 10-shot groups are as close as Eley gets to supplying raw data. For a given distribution shape, the average of all 20 groups should be proportional to the mean radius of the impact distribution, allowing an ARA score to be calculated. Because entering the data is tedious, I analyzed only 53 lots. The conclusion is unchanged. The performance of Eley’s four test barrels is inferior to custom match-grade barrels.

    Will the Analyzer work for you?
    The discrepancy between the reported performance of Eley’s test barrels and actual ARA scores makes me wary of using the Analyzer to infer performance in typical custom match-grade barrels. On the other hand, if your barrel is similar to those used in Eley’s qualification testing, you might see a correlation.

    Even though the expected scores for the test barrels are lower on an absolute scale, custom barrel performance may be proportional. The only way to know is to measure the performance in specific barrels.

    Next: performance in typical custom .22lr BR rifles.

    Thanks for reading.
    Albert
    Last edited by ahighe; 1 Week Ago at 01:42 PM.

  2. #2
    Join Date
    Aug 2015
    Location
    Ohio - Land of the "Mogs"
    Posts
    187

    Default At the Triple Crown...

    Some background on the testing...
    My memory isn’t what it once was but back in May the Eley CEO was at the Triple Crown and participated in a Q & A session.
    He was asked what barrels were used... we were told three brands Walther, Anshutz, and ?.
    How many rounds fired till cleaned? This is where my memory fails... I believe the response was ~500 - 600 rounds.
    The Cleaning protocol was unknown.
    You can't change the past. Only the future.

  3. Default Does the Eley Lot Analyzer predict ammo performance in custom .22lr BR rifles?

    I'm neither saying yes or no but here is my personal experience. I ordered 12 lots of Tenex to test earlier in the year. Of those 12, three were fairly good and two seemed very good. I had never heard of an Eley Analyzer, until I got home. That's when I saw the post about the analyzer and, of course, I read it. Of the lots I tested the three best lots were the top three in the analyzer just not in the same order. I was torn between the two better ones and ended up making the choice based on the analyzer because it showed a slightly smaller average group. That was probably a mistake but not a big one. I still believe the other lot was a little better. So of those three I chose 2 and 3 as the best and ended up buying 2 and probably should have bought 3. Keep in mind the testing and my decision was done without knowing there was such a thing as an Eley Analyzer.

    So, with that said. I think it's a tool but not an ultimate tool. What I did notice and what I believe is the most important graph they present is the windage graph. Every lot I tested was skewed slightly in the same manner as the graph. That does not mean the groups were good or bad, just that they followed the graph. The composite groups did not necessarily follow the actual groups.

  4. #4
    Join Date
    Apr 2015
    Location
    Lisbon
    Posts
    146

    Default curious enough...

    Ok... I got curious enough to dig on the analyzer regarding my Eley test (2 guns) on last March 2015 in Germany.

    At that time I had bought 3 lots with 50%-25%-25% quantities, only looking at the results the lots did on my rifles.

    So I pick the pdfs and compared to Eley analyzer... man, the lots I picked were the worst ones on the analyzer... but they allowed me to break some National records and won several important matches. So, I'm puzzled here...

    Do I continue to shoot them? No. I went last year to Germany again but to another manufacturer and brought only 1 lot (lucky enough it was the very best among the 23 on display) that for chance worked the best on both my rifles. Thing is both rifles shot much tigher groups with this ammo than with Eley (one rifle from 16.6mm to 10.9mm, and the other from 17.8mm to 11.4mm).

    This des not mean, Eley is worst or better, means you have to go through many lots from different brands, to choose the best you'll find... and of course have luck to have on display that special lot...

  5. #5

    Default

    Quote Originally Posted by ahighe View Post
    Hi Marty,
    Thanks for sharing! I'll have to study the targets and what you've said.

    It struck me that a wider audience might want to read your posts, and see these targets in particular. Have you considered moving them to a new thread with an appropriate title rather than be hidden in a thread that some might not bother to read?

    Albert
    I agree. I moved all the posts pertaining to martys targets.

    stiller

  6. Default

    Albert,
    I always enjoy reading your ongoing testing topics and, with apologies, as usual permit me a question.
    Recent comments in various places seem to put their cleaning regimine somewhere between 500-600 shots which would seem unlikely in a properly maintained match barrel in the US.
    While I understand the requirements of a production facility drive this, any thoughts as to how much this may impact test data, in particular shot percentages that tend to fall outside optimal scoring as opposed to raw grouping ?

    Thanks,
    Tim

  7. Default

    Hi Tim,
    No apologies necessary. Ask away.

    In my limited experience, I would say it depends on the barrel. I suspect people form differing opinions about cleaning because their experiences are different. For example, when I bought my first custom barreled action almost two years ago, I knew almost nothing about rifles and had almost no experience cleaning them. Opinions ranged from "never clean a .22" to clean after every card. Of course, those competing at the highest level were closer to the later. Their experience taught them that a dirty barrel would decrease scores. But I never saw any statements about how large a penalty insufficient cleaning would cause.

    Claims that improper cleaning procedures could damage a barrel dictated that I take a cautious approach. I didn't want to risk damaging such expensive gear. I bought bore guides and good cleaning rods.

    At my first match with my brand new 20-inch Shilen barrel and 2500X action (smithed by Mark Penrod), every target was over 2000. I was thrilled. I practiced at the range a few times. At the next two matches my aggs were 500 points lower! I doubted that wind accounted for all of that. Long story short, I bought a bore scope and discovered the throat had a substantial carbon ring build up. My cleaning regimen was insufficient to remove it. After every range session, I now routinely short stroke the throat with a VFG pellet covered with bore paste. The bore scope confirms when I've done enough and identifies when I've done too little. My scores recovered.

    I verified that the carbon ring significantly lowers scores. After around 600 rounds and cleaning using my initial process (without scrubbing the throat) scores plummeted. They immediately recovered once I removed the carbon ring. I didn't determine a critical limit when accuracy loss was significant. I feel I'm pretty safe shooting four cards without a comprehensive cleaning. I begin to worry after five or six.

    This year Mark Penrod smithed a 24-inch 3-groove Benchmark for another 2500X. Same smith, same chamber. Yet the carbon ring forms slower and to a lesser extent. I feel I could go longer between scrubbing the throat. Because my hands are weak, cleaning is a chore I have to take some time at. At matches I now push 2 dry patches after every card, and clean well after every range session (typically 150-200 rounds). But I think I need to do something more during a six card match, especially with the first barrel. I haven't determined when the carbon ring affects performance in the Benchmark.

    These observations are largely anecdotal. The only way to know if cleaning is sufficient is to conduct controlled experiments shooting a lot of rounds. I would be surprised if Eley didn't clean each barrel after every lot. I would, unless I had solid evidence it wasn't necessary.

    My 2 cents.
    Albert

    Quote Originally Posted by tim simbari View Post
    Albert,
    I always enjoy reading your ongoing testing topics and, with apologies, as usual permit me a question.
    Recent comments in various places seem to put their cleaning regimine somewhere between 500-600 shots which would seem unlikely in a properly maintained match barrel in the US.
    While I understand the requirements of a production facility drive this, any thoughts as to how much this may impact test data, in particular shot percentages that tend to fall outside optimal scoring as opposed to raw grouping ?

    Thanks,
    Tim

  8. Default

    Albert have you read this article?
    http://www.shootingtimes.com/ammo/am...n_eley_101405/

    today is eley meeting that standard? thanks marty

  9. Default

    Hi Marty,
    Excellent question!
    I had forgotten about that article.

    I assume you are referring to the specification for the percentage of 10-shot reference groups whose size is less than 15 mm at 50 meters. The article reported that current (as of Jan, 2011) Eley TenEx Ultimate EPS ammo averaged 50 percent.

    Because averaging the Analyzer-supplied group sizes is tedious, I’ve calculated them for 26 Tenex lots so far. Only three (12%) met the specification of 15 mm if measured center-to-center. The average outside-to-outside group size for those same 26 lots was 0.885 inch (22.6 mm).

    I hope my math is wrong, or they used higher-quality barrels for previous testing, because the performance doesn't appear to meet their former levels.

    Albert


    Quote Originally Posted by marty hammond View Post
    Albert have you read this article?
    http://www.shootingtimes.com/ammo/am...n_eley_101405/

    today is eley meeting that standard? thanks marty
    Last edited by ahighe; 4 Days Ago at 01:02 PM.

  10. #10
    Join Date
    May 2015
    Location
    Tannersville S.W VA
    Posts
    140

    Default Downhill

    Any one with a couple hours to spend with the analyzer can plainly see. The black box of 2015 is now on par with the Red box produced today!

  11. Default

    Quote Originally Posted by striperfisher View Post
    Any one with a couple hours to spend with the analyzer can plainly see. The black box of 2015 is now on par with the Red box produced today!
    One would wonder about that, making it somewhat more difficult is not knowing if their barrels are swapped out and when they might have been or what the frequency might be .

  12. #12
    Join Date
    May 2015
    Location
    Tannersville S.W VA
    Posts
    140

    Default

    I'd say it had to do more with being bought out by a investment firm. Focusing more on production vs quality.

  13. Default

    Quote Originally Posted by striperfisher View Post
    I'd say it had to do more with being bought out by a investment firm. Focusing more on production vs quality.
    That is really only partially true, it was sold to the former management financed by a private equity firm, hard to find the exact ownership % . Still a pretty small international business.
    Everybody thinks this is a large global oufit but it's total global revenues are still barely $20mil. Still pretty small and they could disappear in a heartbeat.

  14. Default

    It's very small. It would easily fall into the SBA's tables for small manufacturing businesses. Tiny really. Usually the only way places like that make it is to be bought out or just created by a larger parent company or find a niche. Until now the niche has worked. But they seem to be losing ground fast. The majority of their revenues come from target rimfire ammo. I wonder what percentage of NAMMO's revenues come from the same.

  15. Default A more critical look at the specification

    After carefully re-reading the Eley article, it appears the 15mm (0.59 inch) goal in the specification refers to measuring group size using outside edge to outside edge.

    It also said 50 percent of all 10-shot groups would meet that spec. It does not say average 10-shot group size of a particular lot would meet that spec.

    This is a subtle, but important distinction. I consider it a bit of sleight-of-hand that likely creates a misleading impression of performance. For example, see the comparison between Figures 14 and 15 described below.

    Using the Analyzer, I downloaded the group sizes for the remaining 52 lots of Tenex ammo from 2016. The red bars in Figure 14 indicate the fraction of the 52 lots whose 20 groups have a particular average size (measured outside edge to outside edge). The red curve is the Normal distribution corresponding to an average of 15.8 mm and a standard deviation of 0.4 mm.

    Click image for larger version. 

Name:	Fig14.jpg 
Views:	3 
Size:	31.1 KB 
ID:	2768

    Figure 14 – Distribution of the average of 20 10-shot groups for 52 lots of Tenex ammo (2016).

    The vertical dashed black line sits at 15 mm. Clearly almost none of these lots have average group sizes below that value. Does Figure 14 prove 2016 Tenex ammo doesn't meet the 2011 level of performance? No.

    In contrast, take a look at Figure 15. The red bars represent the fraction of individual 10-shot groups that have a given size. These are the same groups used to calculate the averages shown above. The average of the distribution, 15.8 mm, still is the same. But the curve is much broader. Its standard deviation increased to 1.9 mm. This is a direct consequence of using fewer points (10 vs. 200) to sample a population.

    Click image for larger version. 

Name:	Fig15.jpg 
Views:	1 
Size:	49.8 KB 
ID:	2772

    Figure 15 – Distribution of the individual 10-shot groups for 52 lots of Tenex ammo (2016).

    The fraction of groups below 15 mm isn’t 50%, but it is considerably larger.

    Based on Eley’s reported performance in 2011, 2016’s Tenex doesn’t appear to have quite measured up. But it isn’t far off. However, more importantly, I find the standard of performance to be insufficient. A single group of 10-rounds tells you almost nothing. An average of 20 such groups is a fair start.

    Thanks for reading.
    Albert
    Last edited by ahighe; 1 Day Ago at 09:34 PM.

  16. Default Eley’s specification compared to ARA score

    Let’s compare Eley’s 2011 performance specification to the accuracy needed for competitive bench rest scores. I calculated the expected size distribution of 10-shot groups at 50 meters for a rifle/ammo combination that produces an ARA aggregate of 2200 at 50 yards. That distribution appears in Figure 16.

    Click image for larger version. 

Name:	Fig16.jpg 
Views:	7 
Size:	39.5 KB 
ID:	2770

    Figure 16 – Expected distribution of 10-shot groups for a rifle/ammo that produces an ARA agg of 2200.

    The average size is 12.6 mm with a standard deviation of 1.6 mm. The fraction of groups below 15 mm is greater than 90%. This is a very high hurdle for all lots to meet. But the plot provides a reference to judge performance relative to a familiar measurement.

    While I was at it, I thought we might as well look at the expected distribution for a rifle/ammo combination that produces an aggregate of 2300. It is shown as the green curve in Figure 17.

    Click image for larger version. 

Name:	Fig17.jpg 
Views:	11 
Size:	42.7 KB 
ID:	2771

    Figure 17 – Expected distributions of 10-shot groups for rifle/ammo combinations that produce ARA aggs of 2200 and 2300.

    The average size is 11.2 mm with a standard deviation of 1.4 mm. The size of almost all 10-shot groups is expected to be below 15 mm.

    Thanks for reading.
    Albert

  17. Default

    Quote Originally Posted by wsmallwood View Post
    It's very small. It would easily fall into the SBA's tables for small manufacturing businesses. Tiny really. Usually the only way places like that make it is to be bought out or just created by a larger parent company or find a niche. Until now the niche has worked. But they seem to be losing ground fast. The majority of their revenues come from target rimfire ammo. I wonder what percentage of NAMMO's revenues come from the same.
    I seem to recall in a conversation with Jack Neary( sponswered CF shooter) a few years back that it was pretty small as well, well below their aggregate level of CF ammunition, brass, and powder in combination.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •