Trapshooters Forum banner
Status
Not open for further replies.
1 - 9 of 9 Posts

· Registered
Joined
·
9,941 Posts
Discussion Starter · #1 ·
"Fast shells spread more"- do they?

“The effect of shot speed on pattern width is known by all but properly exploited by but few. When you need just a bit more spread and don’t want to bother with changing chokes, just slipping in the next “speed group” – say substituting “heavy” for “light” – can give you just the edge you need, with the bonus of extra pellet energy at the target that changing chokes won’t accomplish.

But you’d best not try this without regard for the physics involved, as thoughtless changes can be more trouble than benefit. Here’s why: pattern development is dependent on air drag, and that drag is not a simple monotonic function of pellet speed, but rather is a function of the square of the speed. So the effect of pattern-widening is magnified at the upper reaches of pellet velocity. Here is an illustrative figure:



And it’s that huge increase in pattern-width at the right side of the graph that can lead the unthinking claysman into trouble. Trying for that last bit of pellet energy for a 60-yard crosser, he loads up with 1400 FPS and pays twice for his mistake: By the time the pattern gets out there it’s often too spread out to even clip off a “visible piece” and the recoil – which he has suffered for no benefit - will affect his shooting for the rest of the day.”

I pen that last period and lay down my quill, letting my eyes drift back appreciatively over the text. Very good – very good indeed! Beyond just repeating what every reader will agree with, I offer not just scientific causation but even surprising graphical evidence, as well as both support and warning to my readers, which about covers the bases, to my mind. OK, that “instrumental” flourish on the x-axis of the graph might be a bit over the top, but it lends such an air of erudition to the enterprise it’s just too good to pass up. If the response to this article is favorable I may be able to string this out for a couple more issues, dreaming up wad and powder effects, interactions with shot size, whatever comes to mind; you know the drill.

Time again to reflect how good I have it. An evening in my den in front of the fire, building castles in air for yet another shooting publication. My ever-more-grey-muzzled retriever curled at my feet, napping and no doubt dreaming about the day in the field I promised him next month when I will pretend to test a brace of matched Purdeys. How can you beat it?


Admit it – you were well into this before you first began to smell a rat. The first parts are so much like what you read everywhere, hear at every club, I dare say they slipped past your BS detector without raising any alarm whatever. So now backtrack and see how far you have to go to find some fact –

Back 1: the “60-yard crosser” paragraph. Well that was a giveaway, wasn’t it?

Back 2: The graph. Sure, that’s a graph of squared numbers, but is it possible? I hope you don’t think so!

Back 4: This is the support for the graph. If it’s nonsense, so is this.

Back 5, right to the start. Unfortunately, there’s hardly a word of truth in it either. In fact the first 12 words, “The effect of shot speed on pattern width is known by all” are as questionable as any of the other fibs I told.

Introduction:

I’ll bet that every reader knows that when you push shot faster the pattern “opens up.” But what happens when I now ask you “How much?” On the off chance that you have an answer to that, I keep another question in reserve: “How do you know?”

This article will cover an experiment involving shot speed and pattern width. For a limited set of data, it answers those questions which I’ll wager stumped most of you: “How much?” and “How do you know?” Based on data from 10 patterns each from shells measured at 1030, 1130, and 1230 feet per second, it will report some statistically significant differences, and some cases where the “expected” differences failed to appear.

Properly, this is the first test of patterning which should be done, since it answers the question of how well speed has to be controlled in comparing shells, chokes, powders and so on. If a difference is found is it due to the change in components or equipment, or could shot velocity be responsible for the whole finding? Conversely, if no difference if found, is it speed making a change in one direction and “canceling out” a change in the other, thus masking a real effect?

The best experiments make you re-examine your ideas, perhaps even change your mind or behavior. This experiment has led me to put some of my MEC bushings away in favor of bigger-numbered, larger-holed ones. With luck, it may change your outlook as well.

Equipment:

The critical piece of equipment making this experiment possible was the software program “Shotgun-Insight”, an application which reads digital photos of patterns and analyses them, making it possible to read enough patterns to draw reasonable conclusions from them. The rest of the equipment used: pattern paper, a digital camera with flash and tripod, PC and the rest, act only as support for the program. Kaleidagraph software produced the graphs.

Method:

A MEC 9000H was used to load once-fired STS hulls with Red Dot – 15.7 grains, 17.7 grains, or 19.7 grains. With the components used, the resulting speeds were about 1030, 1130, and 1230 feet per second, as measured by an inductive chronograph, and the 10-shot standard deviations of all were under 5, which is consistent ammo indeed. The shot is Remington magnum 8’s with a count of about 440 per load as dropped by a 1 1/8 oz. MEC bar. All shot is from the same bag.

On a warm day at Metro Gun Club in Blaine, Minnesota, patterns were produced firing a “factory bigbore” Perazzi which has produced consistent full choke patterns in other experiments. The distance used was 34 yards, in an effort to make more use of the “75% diameter” statistic offered by Shotgun Insight. Ten patterns were shot at each shot-speed.

The digital photos were intentionally overexposed by one stop and a fill-in flash was used; all the resulting JPEG’s were usable.

Results:

Here, in bare-bones form, are the results of the test.



Based on averages (mean), the patterns from the 1030 FPS loads were denser in the 0-10 inch and 10-20 inch diameter rings than the others and the 20-30 inch diameter ring was less dense. The innermost-ring differences were not statistically significant (I.E., roughly, they could reasonably attributed to chance) while the others were significant.

Look, even I see this as something of a letdown. Hours of work resulting in 12 numbers only four of which mean anything? I plan to go on with “How much?” and “How do I know?” to show what’s behind those numbers, but in the end, those are the “experimental results.”

Let’s look at the data as it comes in, that is, shot by shot.



Looking at the traces from the top down on the graph, the first, the black line and open squares, shows that the pellet count is stable, pattern-to-pattern.

The green dots, connected by a green line, are the pellet counts in the 10-to-20-inch ring. This count too is quite stable.

Going on down, we meet the red trace, the pellet count in the inner 10 inch circle, and the blue trace, the 20-to-30-inch ring. They are more variable than the others, and in addition, vary inversely, that is, when one goes up the other goes down. When there are few pellets in the center there are more on the edge and visa-versa, while the 10-20 inch ring just tracks along. No, these pellets are not jumping from the inner to the outer rings, it’s just that even when the total number of pellets in a 30-inch pattern is the same, some patterns are more spread out than others. “Spreading out” means some 1-10 drifting into the 10-20 area and, at the same time, 10-20 pellets drifting into the 20-30 area. The sector with the most pellets is the one from 10 to 20 inches, but since it is 3 times the area of the 10 inch circle, the pellet density is only about 2/3 that of the center. Once again, these patterns are “hot in the center” like all the others intended for trap shooting.

So what does a significant difference look like?



The shots have been ordered according to total pellets in 30 inches, increasing left to right. The 1030 patterns are the solid markers and lines; the 1130 are outline markers and dotted lines. We see some real differences here. The 1030 FPS 0-10 and 10-20 rings are more dense, less dense in the 20-30 inch ring. So, in fact, increasing the speed 100 fps with this slow start has widened the pattern, much as opening a choke tries, but mostly fails, to do.

And what do non-significant differences look like?



Here the traces cross and re-cross each other in random ways and are often just the same. In this case, starting from a faster baseline than the first graph, increasing shot speed by 100 FPS did not widen the pattern.

Another way of looking at this is the “75% diameter” statistic. This is a calculated rather than counted datum which predicts a circle diameter which would contain 75% of the shot.



This is the same story as told by the “rings” analysis. The 75% diameter of a 1030 FPS shell is smaller, that is, the pattern is smaller.

Just a couple more things. Could I have gotten these results by other means, specifically “by eye” or “side-by-side inspection” or fewer patterns?

No, and here’s reason 1. There’s too much overlap:



Most of the time most of the rings look about the same across all speeds. You can’t possibly tell more than a couple apart, much less several at once.

And here’s reason 2:



If I’d shot just the first 5 I’d say that 1130 FPS had a hotter center than 1230. If I’d shot just the second 5, I’d say that 1230 had a hotter center than 1130. It’s only by shooting all ten can I see that there’s really no difference between them at all. You need at least 10 patterns to avoid being misled.

Discussion:

Based on this, I’m going to speed up my first-shot-doubles shells to about 1150 FPS from the 1050 FPS range I’ve been using. They are 1-oz. anyway; why not give myself a better chance? I’m going to try Extralites at handicap. Just to see if they are denser – denser enough, that is, to help.

Is the odd effect seen here related to sub-sonic v. super-sonic? Someone here, maybe it was HMB, made a big deal of that in an earlier thread. Could he have been right? Some evening I’ll settle down by the fire with Kyra at my feet and explain it all for you.

Yours in Sport,

Neil
 

· Registered
Joined
·
9,941 Posts
Discussion Starter · #12 ·
"Fast shells spread more"- do they? (Winston)

Thanks for all the comments. Though Harold has decribed what's going on, I'll fill it in a little more.

The first few paragraphs are a bit of a trick. Using the literary form "dramatic monologue" it was intended to entertain the careful reader and cue him (and her, BB) to sharpen up his/her critical skills for the rest. Though not exactly Browning (Robert, not John or Babe) , I see that it was accurate-enough parody to misdirect a few a while and they, at least, will concede me journeyman's competence in the genre.

A few paragraphs of equipment and how it was used follow. Then the results and here's something I should have pointed out: They are all the same, but presented three ways. The first is the table with 4 significant numbers, the 1030 ones. The next section of ring analysis tells the story a different way, and the last, which Harold (and I'll bet Andrew) thinks is the best version, just does it again. My aim was to show that in previous and forthcoming experiments all these ways are valid analytical tools. (aside: my choice of 34 yards finally made the 75% diameter statistic useful, which hasn't worked as well at 40 yards since too many pellets are "lost.")

The discussion points out that this is not just "all talk;" actual choices of shells flow directly from it.

The last part is a preemptive strike. I was expecting some objections in the form "I shot several patterns and I could clearly see that..." and these two graphs prove, to me, that "seeing" any of these results (without 10 patterns, without counting) is simply impossible.

Hap, I referred in the article to a "limited data set" and that's what it is. I make, as yet, no great claim to the generality of these findings. At the Grand I bought Light, Heavy, and Nitro 27 Remingtons and they will be tested next. Then I'll have more to say about this subject. The Shotgun Insight program can recalculate the patterns to other distances. Though I'll have to confirm it's accuracy someday, it sure looks like it works so far.

N9FZX, I owe Old Cowboy a test of soft vs. hard shot and I'll do it this way, based on your suggestion. Thanks for the observation.

Andrew, when I loaded the shells my pellet counter gave me 438, 440, 438 pellets and a post test gave me 435, 436, 440.

I'm headed to the Calf and Steer Shoot so I can't post the exact pellets on paper, but figure 420 for the 1030 shells. the same for 1130, maybe 5 fewer for the fastest ones. This loss of about 20 pellets is a constant at 35 yards.

Well, it's time to dash. I'll be back...

Neil
 

· Registered
Joined
·
9,941 Posts
Discussion Starter · #32 ·


Pat, reading the Ansi/Saami manual, I get an entirely different take on the relationship between shell speed and "what's printed on the box."

First, mode is out the question. It's got to be mean. Second, according to the voluntary standards, all a test batch has to do is produce an average (mean) within 90 fps of what is intended. So a factory tried to make 1200 and get something close enough for their own standards - I'm sure it's less than 90 but it's bigger than anyone posting here seems to realize. That's what goes in the box. You can't standardize a chronograph with factory shells; they vary too much.

The industry standard is a 30-inch test barrel, full choke. They are all the same within anyone's ability to measure them and yet they will produce different speeds. If you visit Federal, you will see a stack of barrels and a number painted in yellow how much to correct the reading by - it'll be -15, +20, generally in that range.

My barrel is cylinder bore, that raises the speed-reading I get by about 20 FPS. This is the opposite effect from what you get with a light-operated chronograph, where a full choke reads faster. So the number I provide (1230 for example) is faster than "factory" but closer to what the home-user will get.

The speeds used in this experiment are _far_ more different that you will run into with any usual factory range. Going from light to Handicap you maybe jump 100 FPS, I covered twice that.

The shot-string (pellet variation) discussion is going right by me. What's it about?

There are hot spots in patterns. They are in the center. That's about all I can say about them which can be experimentally supported, and I'm sure Andrew will agree. The other use of the term "hot spot" may be some congregation of pellets here or there; they mean nothing since they won't be there the next shot.

More later,

Neil
 

· Registered
Joined
·
9,941 Posts
Discussion Starter · #36 ·


Lorenzo, your test of 1000 and 1200 is not twice what I tested, it's the same as I tested. And I found greater density at 1030 than either 1130 or 1230, just as you did with 1000 vs. 1200.

The interesting thing to me was what so few have noticed. There's no difference between 1130 and 1230. None. Everyone is talking about speed. With these 2 - equivalent to light and handicap, speed has made no difference.

Neil
 

· Registered
Joined
·
9,941 Posts
Discussion Starter · #41 ·
"Fast shells spread more"- do they? (Winston)

Pat, the equipment everyone uses, the Oehler 71, comes with a description of its principles of operation and the request that it not be further disseminated.But it's not the mode of the velocity it reports, that's for sure.

The utility of using a light operated (first pellet) chronograph depends on the choke used. If you use a cylinder choke and two chronographs to measure each shot, you get a result which is every bit as useful as the industry model. The mean and SD are both OK. With a full choke the speed will read higher but still will provide a valid mean, though the SD will be pretty worthless, even if you use two chronographs.

As my report says, my speeds _were_ measured by the inductive chronograph, the one credited with measuring the speed of the center of mass of the charge. I did not state the assumption about the velocity of the shot cloud, since I instead wrote:

"With the components used, the resulting speeds were about 1030, 1130, and 1230 feet per second, as measured by an inductive chronograph, and the 10-shot standard deviations of all were under 5, which is consistent ammo indeed. "

Neil
 

· Registered
Joined
·
9,941 Posts
Discussion Starter · #48 ·


Quicky, there are three common ways to express the average: the mode, median, and mean. Each has its uses.

The mode, which is the "most frequent" figure, roughly the "hump" in a graph of frequencies. It is mostly used when the scale used is "nominal", for example the responses to a poll.

The median, which divides that graph of frequencies in the middle, placing the "average" at the point where half are above, half are below. It is appropriate for numbers on an ordinal scale. Income stastitics are best described by the median, since they are less influenced by extreme values than the

Mean, the "average" mostly used, the sum of numbers (on a ratio scale) divided by the number of data points. Like gas mileage, expecially when calculated over a number of fill ups.

And Pat, I hate to keep hammering on this,, but inductive chronographs _do not_ work the way you think they do. They do not measure the speed of individual pellets. Nor do they measure the mean velocity. They are close to measuring the median, in principle anyway.

Neil
 

· Registered
Joined
·
9,941 Posts
Discussion Starter · #58 ·


zzt, there isn't anything in your reports to make me think people can tell the difference.

But they do raise a couple of interesting questions. In the first one, you said 2/3 were right when they were different - what did they say when they were the same? This is a necessary test for any further work.

The speed of the shells has to be known. Tested, that is.

I've made a couple of stabs at this with some real basic "start of the road" tests. Nothing has worked yet, but I have to get beyond the basics and see what happens.

Neil
 

· Registered
Joined
·
9,941 Posts
Discussion Starter · #61 ·


zzt, I asked about the "shells-the-same" because I assumed some said they were different. Your 2/3 right, assuming a run of say 10 shells, is best explained by guessing.

You make more of the differences than you need to, especially if they are guessing, and they are in your tests.

Actually, I did exactly as Andrew suggests. I weighed a number of loads (Pact dribbler, it doesn't take that long) and set out to see how sensitive 2 subject were to speed differences. The test really is double blind, and the order of presentatioin is genuinely random. Some pairs of shells are the same, and the shooters' only task is to say "same or different." The two subjects, I and my friend Roman, performed exactly at chance level. We were completely guessing. And there's more - I _knew_ I was guessing. Try it sometime the way I describe it - after 20 shots you will surely see that these two speeds, 1140 and 1200, are impossible to tell apart. Since the speed difference in shells was a tested 60 ft/sec, so far I can say that _if_ anyone can tell any difference between fast ans slow powders, it's not recoil they are using. And we (Roman and I) aren't insensitive to recoil at all. We can tell our 1 oz. loads, or Winchester's, from extra lites, if not heavy papers from handicap papers in some lot numbers.

I personally also think that the push/shove comparison is completely wrong as well. There's just not enough difference, particularly if the speed test is such a failure. If you can't tell 4% in recoil, how can you differentiate timings in the range of 3/10,000 second?

So recoil is out, time is out , but vibration and sound are definitely in play, as far as I'm concerned. Particularly sound. That's the next place I'm going, and if the results are positive in a very tricky experiment I have in mind, I think we can put all the rest of this powder difference to bed.

The 1 oz results are in. They will be in new thread. I look forward to your comments and participation.

Yours in Sport,

Neil
 
1 - 9 of 9 Posts
Status
Not open for further replies.
Top