Evanix Blizzard S10-Part 4

by B.B. Pelletier

Part 1
Part 2
Part 3

Today, I’ll show you the results of testing the Evanix Blizzard S10 with the barrel shroud removed. I tested it against the CB cap-firing Winchester Winder musket and will show you those results, as well.

You may remember that in Part 3 I was disappointed with the accuracy of the rifle and thought that the baffles in the shroud might be touching the pellets on their way out. In fact, I implied that rather strongly. Today, I’m not so sure. I think I have discovered what was wrong–because when I fixed it, the rifle suddenly became much more accurate.

So, we have a lot in store for today’s report. Sit back and enjoy the ride!

Pride goeth before a fall…
I was SO CERTAIN that the pellets were touching one or more baffles on the way out of the shroud that I couldn’t think of anything else. I fully expected the first group at 50 yards from the unshrouded barrel to be a pleasant surprise and a welcome relief to this problem. I selected the new 18-grain JSB Exacts because I also “knew” that they would be the most accurate pellets in this rifle.


With the barrel shroud removed, the Blizzard S10 barrel is free-floating.

The machine worked fine until we turned it on!
The day was breezy, but the breeze came in gusts and it was possible to wait them out. I took great care to shoot only when the wind was calm. But even with all that care, my first group of 10 with the JSBs measured 1.658″. That’s only a little better than the last time I tested it, which surprised me a lot. I expected a sub-inch group.

Beeman Kodiaks
Switching to Beeman Kodiaks, the groups got larger. Ten shots went into 2.25″. Sorry, Kevin, but that’s the truth.

Eun Jin
The big Eun Jin dome shrank the group back to 1.716″, but that’s not what I was hoping to see. After that, I just sat for awhile and examined the rifle. Was the scope tight? Yes, it was. Was the barrel well-anchored in the receiver? Yes, it was, but when I checked that I felt a bump of something else shifting.

The action was loose in the stock! I felt like such a rookie for not checking the stock screws–except that’s a problem you normally associate with spring guns, not PCPs. And, when I tried to tighten the stock screw, it was already tight. But the action was definitely rocking in the stock.

There was no way to tighten the action, so I reverted to an old-time standby–shimming! I folded a foot-long length of duct tape into a two-inch pad and stuck it between the front of the forearm and the reservoir tube. It went in and stopped when there was no more room. Now, the action and stock were tight–no rocking movement.


As you can see, my shim was not a precision fix. But it worked, and that’s the lesson to take from this.

Success!
The next group of 10 JSBs went into a group measuring 0.862.” That was exactly the level of accuracy I had been expecting from the Blizzard S10. Remember, these are 10-shot groups and will be larger than a 5-shot group from the same gun.


Unshimmed group of 10 18-grain JSBs on the left. Shimmed group on the right. Pretty conclusive evidence that the action needs to be tight in the stock.

Before and after
The difference before and after the field fix is obvious. From 1.658″ down to 0.862″ for the JSBs is nearly a 50 percent reduction in size! The Eun Jins dropped from 1.716″ down to 1.106,” another significant reduction. Kodiaks and Air Arms domes had reductions as well, though not as large. Kodiaks went from 2.224″ to 1.735″ and Air Arms pellets dropped from 1.125″ to 0.973.”


Unshimmed group of 10 Eun Jins on the left. Shimmed group on the right. Could the results be any clearer?

These reductions in 10-shot group sizes are both immediate and quite dramatic. And this is just with my quickie field fix. If I bed the action correctly, we could expect at least this and maybe even more.

I contacted Pyramyd Air. After they researched it, we learned that they had received an initial lot of guns that were missing a washer for the stock screw. All were loose in their stocks. The Pyramyd Air technicians added the necessary washers and the rifles tightened up just as mine did. So, I went to the hardware store and bought some washers to fix my rifle permanently.

This has been an interesting test for me. First, I was surprised by how quiet the Blizzard is, even though it puts out a lot of power. Second, the accuracy thing was a learning experience. I always default to my experience–which in this case was with baffles that nick the pellets on their way out. This strange turn of events really surprised me, which is a good cure for hardening of the attitudes.

Lest I forget–the Winder musket and CB caps
I also had the Winder musket at the range to test the accuracy of CB caps against the Blizzard. I wasn’t able to fool you guys about CB caps. Most of you already knew their capabilities and their shortcomings. And this is not a report on them–just an update as they play against the Blizzard S10, since I introduced that topic last time.

At 50 yards, the CB cap bullet drops a lot more than a pellet from the Blizzard–even a heavy one. While the advertised velocity is 710 f.p.s., that has to be with a short barrel. The Winder has a 28″ barrel, so it should rob some velocity from that bullet. CB caps are so low-powered that I can test them in my office, just like pellets from a powerful rifle, so that’s what I did.

They averaged 696 f.p.s., with a spread from 659 to 720. They were faster than I thought they’d be, so apparently the long barrel doesn’t slow them down much. But they’re also quieter than most PCPs. The impact of the 29-grain bullet makes more noise than the muzzle blast.

CB caps are not a match for an accurate air rifle like the Blizzard. They’re lower in power and far less accurate, plus they cost a lot for anyone with access to a powerful air rifle. I suppose that if all you own is a .22 rimfire, they make sense as long as thousands of shots aren’t in order. However, with the huge difference in price over pellets, you could buy a Discovery and soon make up the difference.


Ten short CB caps from the Winder musket went into this group at 50 yards that measures 1.837″ between the centers of the two widest shots. It’s not bad for CB caps, but don’t even think of competing with an accurate PCP!

CB caps did pretty much what I expected at 50 yards, but that’s not the end of testing them. I’m going to try them in a number of accurate .22 rifles, including the CB longs that fit long rifle chambers. When I’m finished, we should have a pretty good idea of where CB caps fit into the big picture.

46 thoughts on “Evanix Blizzard S10-Part 4

  1. HI B.B.,Please excuse this off topic question,but no one Ive asked seems to know.Because Leapers scopes are built so well to handle the recoil of air rifles, do you think they would survive on a 12ga. shotgun for hunting.There specs are perfect for up close deer hunting.Thank you.Jersey Boy



  2. BB
    Glad you found the problem.
    just goes to show what precision machines
    these rifles can be,when a misplaced
    washer can cause 50% loss or gain in
    accuracy.
    Also glad you're testing the CB caps.
    I've used them as quiet critter gitters,
    but only at 25yds. or less.They work well for not waking the neighbors,even on
    'possum and coon sized pests up close.
    Easy to get head shots at that range but
    I've never tried them at longer dist.
    will be interesting to see how they
    stack up.
    For people who already have a RF around
    (doesn't everybody?):)and only need to
    remove the occasional pest I think these
    are more practical than any PCP.BUT if
    hunting or doing lots of target shooting
    PCP's get my vote,even though I don't
    have 1 yet.Cost and logistics are the
    big draw backs for me.I would have to
    make a special trip of more than 30 mi.
    one way to get HPA or CO2 bulk fills.
    What I'd like to get is a detuned Disco
    doing @ 650 for lots of shots and easy
    pumping with hand pump:)Disco's seem
    very reasonably priced but I haven't
    checked on the cost of this type of
    tune,and I'm not confident enough to
    do it myself having never even handled
    one yet.

    1st frost of the season for us this A.M
    so I'm thinkin of rabbit stew and fried
    squirrel.Cheerios just aint gonna taste
    as good today:(

    JTinAL


  3. Jersey Boy,

    Yes, Leapers scopes should be no problem on a shotgun. The recoil is actually less harsh. More of a slow push than a sharp snap. It doesn't feel like that, of course, but to the scope it does.

    B.B.


  4. I suddenly feel much better about not finding the loose stock screws on my 92 for several weeks. Rookie I definitely still am, but it's nice to see that Mr. Pelletier can still have some of the same problems (although it turns out to be caused by something entirely different, I'll ignore that part.)


  5. Tom what in your opinion constitutes a noteable difference in pellet weights? To get specific, I'm referring to the difference between .22 Cal. Beeman Kodiaks at 21.1 Gr. Vs. the JSB at 18.1, is three grains a meaningful difference or insignificant? T.I.A.


  6. It's a true range rat that carries duct tape in his shooting kit.

    After all the recent talk about a few marauders that were having trouble with pellets clipping the baffles I was sure that was the problem with the S10. It's good news that it just needed a washer under the stock screw since there were a few folks that are considering purchasing an S10 and were put off at the thought of sending the gun back to PA for a fix.

    I really appreciate the reminder of checking the basics when diagnosing accuracy.

    Surprised that after shimming the stock the kodiaks group was twice as large as the jsb 18.1's (1.735 vs. .862). Seems that these groups would shrink even more after installing the washer. Maybe even shrink a bit more after cleaning the barrel.

    Not a bad gun considering price and power.

    Really enjoyed the "twist" of incorporating the windsor musket into this series. Great fun.

    kevin


  7. Savagesams,

    I don't know. Three grains at this low level probably is a very significant difference. But when I shoot my .43 Spanish, whose bullet weighs a nominal 370 grains, I have them sorted into piles that are within two grains of each other. Ay that level, three grains is probably insignificant for most general shooting.

    B.B.


  8. savagesam,

    If the pellet doesn't group, like the kodiaks vs. the jsb 18.1 gr in the S10, then the difference is irrelevant.

    But…if you have two pellets that shoot with the same accuracy in your gun (like the jsb 15.8 gr & jsb 18.1 gr in my S410) then a heavier pellet groups better in wind in my experience. A heavier pellet is also better for hunting.

    Those are two "meaningful differences" that I can think of.

    What does T.I.A. stand for?

    kevin



  9. I figured it out to be a 16.58 percent difference. Which imho is a difference, not HUGE but not insignificant either. Thanks JT. Kevin I know what your talkin about. My reason for the question was that sometimes Kodiaks are O.O.S.



  10. RE: Statistics on Groups

    Found another article on web about the statistics of groups.

    Group Therapy
    © 2007 Denton Bramwell
    http://www.snipershide.com/UserFiles/Image/articles/group_size_oct_27.pdf

    What is very nice is Figure 4 which contains a histogram of 3 shot groups. You can see that the distribution is not Gaussian. The mean and mode (most probable value) are different. The distribution is skewed on the right.

    An average of 3 5-shot groups would be much more Gaussian. In all reality you'd probably never be able to actually shoot enough groups with a rifle to show that the distribution is non-Gaussian. The distribution shown is for 20,000 simulated shots. In shooting that many shots, something would likely change.

    Groups of 2 would be even less Gaussian. Groups of a larger size get more and more Gaussian like.

    In Item 2 of the conclusions he has a table to convert sizes between 3,4, and 5 shot groups. The conversion factors agree with the paper:

    APPROXIMATE PROBABILITY DISTRIBUTIONS FOR THE EXTREME SPREAD
    M. S. Taylor, et al
    http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA006586&Location=U2&doc=GetTRDoc.pdf

    For a reason that I do not understand, using the factors in the Long Family paper produce slightly different results. Since the Long paper converts groups size to MOA, you must use the inverse of their factors. I've compared the values in the Long paper to the factors in Taylor's paper. There is a linear relationship with a non-zero intercept. There should be a direct proportional relationship. I suspect that it the Long's results which are slightly off. I'm looking for other sources to verify my (educated ?!?) guess.

    Herb


  11. savagesam,

    You have to be careful with these abbreviations. They can lead to a great deal of misunderstanding, and you don't want to create a FUBAR SNAFU and then be SOL.

    Slinging Lead


  12. ROTFL your right…O.O.S.=Out Of Stock. P.A. seems to be having a hard time keeping up stock levels of Beeman Kodiaks and Beeman can't seem to import enough from H.& N. which leads to hording.





  13. B.B.

    The loose action is a quality control issue vaguely reminiscent of the B30 although it seems that there is one quick fix, not the multiple problems of the B30.

    Oh no! There it is!! If my 10 shot group is X then what would the equivalent 5 shot group be? The very question about which much ink has been spilled. At last a purpose. Well, you've come to the right people. State of the art research suggests that 5 shots will be a fairly large fraction of 10. The "Long" Paper

    http://www.the-long-family.com/group_size_analysis.htm

    predicts that the five shot group will be 75% of the 10 shot group or .645 inches.

    By the way, I can attest now that Umarex has fabulous customer service. I got my Walther Nighthawk back and it is shooting much better than it has in a long time. Nice folks too.

    On the subject of pistol shooting, the movie with Crystal posted on the PA site is very entertaining. The lipstick smile is nice touch. That's right, take a deep breath before shooting balloons filled with talcum powder and push over a tree to get a tactical opening. Ha ha. Anyway, very enjoyable. That desert eagle is an impressive gun. Too bad it's a gas hog and that I already have a Nighthawk.

    Herb, regarding my experiment, I thought that the beauty of the number 30 is that it constitutes a representative sample of a larger population regardless of what that population is–whether it is five shot groups or individual shots. If true, then 30 five shot groups should carry the same weight as one 30 shot group. The average of two 30 shot groups will be a little different, but surely I wouldn't need to shoot 30 30 shot groups to get a usable result!

    You are correct that there was no outside aimpoint for the 30 shot group which is a problem. On the other hand, there was no outside aimpoint for the 5 shot groups which mostly obliterated their aimpoints. I was aiming for where the aimpoint should have been in the 30 shot hole and was not spraying all over it. And while shooting, the rounds generally didn't nibble away at the edges. They mostly went into the middle with only the occasional wide shot consistent with statistical variation.

    The big problem with the experiment is the tearing of the paper. This is reminiscent of the hanging chads of the 2000 election. I just measured all breaks in the paper regardless of what they looked like.

    You are correct that there were fliers in the one inch five shot group as well as the 30 shot group. But the question then is "What is a flier?" I knew these shots didn't feel quite right when they went off, but where do you draw the line on statistical variation and a called flier? To be consistent, my solution was to count everything except actual mistakes like touching off the shot while getting into position, and there were none of these.

    It is harder than it seems to run a proper experiment. I'm reminded that even the U.S. Army with all of its resources is now under scrutiny for incorrectly testing body armor. You would think they would get that right.

    Anyway, I'm satisfied that either the Long paper or the missing link paper are fairly close to correct and nothing short of very rigorous laboratory conditions will distinguish between them.

    Matt61


  14. BG_Farmer,

    Take care. :-) I've made another radical new discovery. I seem to be on a roll here. I have been pushing the David Tubb approach method for offhand which consists in purposely moving in to the target from the same direction. My innovation is to start with the reticle on target, then move out a distance, then back in. What a discovery. So, it's not exactly the Whitcomb area rule, but it's useful to me. I found that before all of the uncertainty and anxiety of a shot were getting shoved to the beginning leading to pauses before shooting, and this uncertainty was undermining my technique. Now, I have a definite starting point.

    Just as people share their load recipes, I will share my shooting technique recipe with the world for what it's worth.

    -Inhale and release half of the breath.

    -Take up the trigger.

    -Reticle on target.

    -Reticle a distance above the target, not too much, not too little.

    -Reticle down on target in a decisive motion.

    -Hold for a barely perceptible moment.

    -Release the shot.

    -Eyes on the target as the reticle jumps with recoil.

    This will not make you an elite shooter in which case you wouldn't need any advice from me. But it is a good 90% solution which should eliminate truly bad shooting. What it has to recommend it among all the shooting techniques out there is that each step is easy and the connection between them is easy. Most shooting techniques emphasize a steady hold and trigger control to produce a surprise break. Both these require a lot of skill. Each one of the steps above is easy, and if you do each one distinctly with no additional pause between each step, you cannot fail. For me, the accuracy is somewhat better than before and consistency is much better.

    Most of this is copied from David Tubb, and equally important, which I like, is a decisive attitude. Once you've worked out your method, you want to go in with confidence and no second-guessing. As he puts it, once he gets his position and preps the trigger, it is "Front sight–rear sight–front sight–gone."

    Anyway, BG_Farmer, I'll be unleashing this next month, and we'll see what it's good for. :-)

    Matt61


  15. Matt,
    I did a bit of practice myself this morning. Perfect conditions, light fog, 35 degrees, little or no wind. The first two shots of the first two groups went into the same (just elongated) hole! Now, if Herb would just give in and prove 2 shot groups are all we need, I'd be on top of the world, but unfortunately the other shots widened the groups considerably, and I was getting tired by the third group:). I've got three 5 shot groups I can use if I don't get out more in time, although they aren't my proudest moments (aforementioned shots excepted) — you definitely have a shot:).

    I have a similar procedure and it works like a charm — on some days:). The muzzleloader has really been educational in terms of forcing me to slow down between shots and "forget" the group, which is where I always screw up, i.e., see there's a good group going and then tense up so much that crashing and burning is the only choice:). On the other hand, I'm not good enough to relax completely and still have everything work:). Add to that, I'm still mucking around with patches, lubes, tamping force, and there's a million ways I can screw up.


  16. BB,
    The S10 looks like what a lot of people are asking for, and that accuracy (once you fixed the problem) is quite good.

    I think JT had the right twist on CB caps. Maybe the right question for some people is PCP or rimfire…which is a much more complicated question:).


  17. Another acronym for you all – TARFU – with respect to pellet ordering.

    Anyway, any news from WWB (Wacky Wayne Burns)? The Nationals were this weekend past.

    Fred



  18. BG_Farmer,

    I've been enjoying listening to you and Matt going back and forth. Have a suggestion for you concerning tamping force.

    Have you thought about using a bathroon scale to measure said force? That's the way the Lee Loader folks control wad pressure.

    Just a thought from a nbon smoke pole shooter.

    Mr B.

    Blogger didn't like sggestion. Word verification is dogrank


  19. Mr.B,
    Great suggestion. I've thought about ways to do it, but I'm holding out hope that I can get good enough consistency without more equipment, which distracts my feeble mind:). Even at this point, the rifle is shooting much better than I am, just because I'm still settling into a rhythm:). I just got a handle on cutting patches (thanks to BB and TwoTalon, although I'm not necessarily doing it the way you said:)) and am satisfied with the rest of my procedures, lubes, ammo, etc.. Perhaps after a bit of practice unimpeded by uncertainty, I'll tweak a bit with tamping force.



  20. "Tip" and a gift.

    This is for you pcp guys.

    As we all know B.B. is always preaching about the importance of keeping dirt out of your connections since this is a leading cause of seals failing.

    If you use foster quick connects you're probably aware that there are neat caps to prevent dirt from entering the male end of the quick connect. But what to use to cover the female quick connect?

    Heard a neat suggestion. Use a rubber finger tip (the kind librarians put on their finger tip to turn book pages). They come in different sizes but the large works best. The downside is that you have to buy a dozen.

    First 5 people that send me their address via email will get one free. My email is klentz4@comcast.net.

    kevin


  21. BG_Farmer,

    Okay, hang on to your groups. 35 degrees sounds pretty cold, and I would find that distracting myself. One advantage of blackpowder–depending on your load–is low recoil. When I fired a rifle at my blackpowder workshop, it felt more like a push than a blow. I think I'll use my Savage 10FP against you with the 40mm TruGlo red dot sight that I'm planning to get. Last time, I believe excess power on the scope was freaking me out. Three 5 shots groups with a muzzle-loader is a lot. I could never have done my experiment with a single-shot. It had to be the IZH 61.

    Again on the subject of relative group sizes, I can't help thinking back to a blog post of long ago which got me interested in the problem in the first place. It was 3/5= 50% and 5/10= 70%. After all that has been said the 5/10 ratio is just about dead on. The 3/5 seems off of what statistical models predict as 3/5= 86%, but actually not. If, as I'm guessing, the 3/5= 50% ratio is based on experience rather than a mathematical model, it could be interpreted as reflecting the higher variability you get with smaller groups. Understood this way, 3/5= 50% is actually a better predictor from one group to the next. 3/5= 86% indicates the probabilities operating at a deeper level, but it won't appear without a ton of shooting under highly controlled conditions.

    So, we return to the beginning. I suspect in cases where there is a large body of experience that science will often validate what has already been discovered by trial and error rather than reveal something new.

    Matt61


  22. Barrel cleaning,

    I have noticed that cleaning the barrel of a spring piston gun often will shorten the break-in period. However, it isn't necessary. Back 20 years ago, I just shot the guns and they cleaned themselves as I shot.

    So you can do it either way.

    B.B.


  23. Matt,

    You're absolutely right that a "good" experiment is hard to run. You have to make certain assumptions, and then do an experiment. The experiment then raises other questions which lead to more experiments and so on ad infintum. You’re always peeling the onion, never getting to the center.

    ——————–
    Information in 30 5-shot groups vs 1 30-shot group

    30 5-shot groups is 150 shots. One 30 shot group is 30 shots. There is more "information" (as in information theory) in 150 shots than 30 shots. The 30 5-shot groups also give you 30 group size data points. The one 30-shot group gives you one group size data point.

    Using the tables in the paper by Taylor et al, is a very useful technique to wring as much information as possible out of your group sizes. It would obviously be better to use radial standard distribution, but that is a LOT (!!) more number crunching work.

    Using Taylor’s tables, the Relative Standard deviation (RSD) for a 5-shot group is:
    0.828 / 3.066 *100% = 27.0%

    for a 30-shot group
    0.635 / 4.788 *100% = 13.26%

    But there are 30 5-shot groups, so the RSD of the average is
    27.0% / SQRT(30) = 4.93%
    (We must use the normal approximation since Taylor gave no better way to pool measurements.)

    In order to get a RSD as small using 30-shot groups, you’d need about 7 groups.
    13.26% / SQRT(7) = 5.01%

    ——————-
    What is a flier?

    The question "What is a flier?" is a difficult question to answer. The only way is really to adopt some arbitrary, but well defined criteria, and be consistent in applying that criteria. It is common to use a 95% confidence interval. A few fliers won't change the mean very much. But a few fliers could change the standard deviation quite a bit. Trimming the outside +/-2.5% won't change either to any significant amount. Also since we know that the group size distribution is skewed towards higher group sizes, it helps make the distribution more like the Gaussian distribution.

    Of course it is quite proper to throw out what you know to be “abnormal shots.” For instance someone slaps you on the back to say hello and causes you to make a “wild” shot. But you can’t decide, that you don’t “like” a shot and just throw it out. (Like I did in throwing out one of the 30 shots in your 30 shot group. I didn’t do the proper tests, I just arbitrarily threw it out to fit my notion of about what the group size for 30 shots should be.)

    The other weakness of course is that a single group size value is very sensitive to fliers. The radial standard deviation would be much sensitive to one flier in 30 shots.

    ——————-
    What does 30 mean?

    Even if you had the X-Y coordinates of every shot in a 30 shot group, 30 shots does not give you an exact measure of the distribution. It gives you a pretty decent average, and a "fair" standard deviation. Look at the Student's T tables. See for example:
    http://en.wikipedia.org/wiki/Student%27s_t-distribution

    For a 95% 2-sided CI, one degree of freedom (two measurements used to calculate mean and std dev) has a T-value of 12.71. For 29 degrees of freedom (30 data points used to calculate mean and std dev) the T-Value is 2.045. An infinite number of degrees of freedom has a T-value of 1.96. So for most practical purposes, 30 is "close enough" to infinity to be useful.
    [2.045/1.96 = 104.3%]

    The Student's T test is used to test the distribution of individual values about the mean. The F-Test is used to compare standard deviations. The point is that the mean and std dev determined from 30 points is much better for predicting the variation of individual points about the mean that it is at predicting the "true" standard deviation.


  24. B.B.,
    I would not expect any better accuracy from the CB caps than you have seen so far. I took my 2 most accurate rifles to the range to test them out. I used CCI CB longs in my Rem. 40XB heavy and a Cooper 57M LVT. Both were on par with your results avg. 2" at 50 yds. Considering the 40XB will shoot .2" at 50 yds with Eley and the LVT does .4" with regularity with Wolf Match Extra maybe the Rem 504 with a faster twist barrel would help.


  25. Im worried about debris left in the bore of my rifle from when it was made, is that a reason to clean a bore out of the box and what should i use to clean the bore of a .177 break barrel


  26. Matt,
    I was being truthful when I said 35 was pleasant, but not everyone feels the same way:). It was really not bad at all shooting 3×5, but that's getting close to the edge, at least for me; it actually felt pretty good, and I expect it will get even easier as I eliminate variables one by one. I've talked to other muzzleloading shooters, and it doesn't seem to be too odd not to shoot 100's of rounds. BP recoil up to 70 gr's at least (as far as I've seen the need to go) is not a problem offhand, but you have to be careful on the bench.

    10FP is fine — I won't feel so bad losing:). How low does your scope go — you may like it at 3 or 4x and not need the red dot? The 0.75" 3-shot group at 100 yards with my .30-06 was at 3x, so its not a huge handicap, but I've got a target that seems to work really well and, obviously, the rifle was rested:). Maybe some nice open sights:).

    Regarding the group-size discussion, I think you make a good point about experience versus pure science.


  27. Not to beat a dead horse, but I think the pistol arena is where CB caps shine. I have yet to meet an air pistol that really thrills me like CB longs in a single or double action revolver.

    Now there are a few air pistols I’ve liked in the dozen I’ve owned, but they were never in that ultimate status. Each has had its own drawback whether catapult, spring, pneumatic or CO2. The combination of power, speed of use, accuracy, along with reasonable and realistic size still needs filled by a yet to be invented PCP pistol.

    They also sell 20 grain BB cabs at 500 fps if the CB’s are too hot for your needs….


  28. Volvo
    The Aguila Colibri at @ 350fps or the
    Super Colibri at @ 500fps are super fun
    indoors with a good back stop.Use something like the Ruger Single Six or
    Bearcat for cowboy action fun.SA shooters
    are the best feeling revolvers I've found.
    Too bad I haven't found a CO2 SA style in
    good working order yet.They all seemed to have their own brand of problems from
    what I've read.
    My HI-Standard Double-Nine DASA would be a great model for a co2 pistol,
    wonder who could build one:)

    JTinAL


  29. Herb,

    Your calculation of Relative Standard Deviation is intriguing and is the most promising method I've heard yet of relating the value of sets of groups of different sizes. But if the criteria is information and each shot represents an equivalent amount of information then if RSD equates thirty 5 shot groups with seven 30 shot groups, then 150 shots worth of information is equivalent to 210 shots worth of information. This doesn't sound right.

    BG_Farmer, my scope is the Leapers 6-24X50. I had it on 10X last time at 50 yards and couldn't hit anything.

    Matt61


  30. Sam,

    I am impressed. A 40X? Wow. When you want to move up from a Winchester 52, that's the way to go.

    Thanks for that grouping info. Of course I will shoot some groups, too, because I'm trying to build a data base for a Shotgun News article. But You told me what I need to know.

    Thanks,

    B.B.


  31. Debris,

    There is no reason to worry about debris left in your barrel from manufacture, but if you want to clean it, go right aherad. Like I said, cleaning will speed up the break-in period a lot.

    Clean with a solid rod and a brass or bronze brush. Load the brush with JB Non-Embedding Bore Cleaning Compound and run the brush through the bore in both directions 20 full strokes.

    Following that, clean out the compound thoroughly with dry patches. That is it.

    B.B.


  32. Volvo,

    I'm going to listen to what you said about pistols and CB caps, but I'm not going to put it in my report. I just traded for a Heritage Rough Rider .22 single action that will make a fine testbed.

    However, I must know if you have ever shot these air pistols:

    S&W 78G

    Crosman Mark I

    These two pistols seem as nice as any .22 rimfire to me, and they group just as tight at 50 feet. Both have some recoil, though not as much as a rimfire shooting long rifles. CB caps may be another thing. We shall see.

    B.B.


  33. Matt,

    RE: RSD & 30 5-shot & 7 30-shot

    The problem is that group size isn't 100% efficient. If you calculated the radial standard deviation, then that would be 100% efficient. But instead of 5 differences from the average, a 5-shot group gives one group size measurement. Likewise a thirty shot group doesn't give 30 differences to mean, but just one group size measurement.

    I looked quickly but couldn't find efficiency for more than 10 for a one dimensional range (ie range for Gaussian Distribution).

    n Relative Efficiency
    2 1.000
    3 0.992
    4 0.975
    5 0.955
    6 0.930
    10 0.850

    So back to our group size problem, the relative efficiency of 30 shots compared to 5 is:
    SQRT(7/30)*100% = 48%

    ———–
    I'll beat a dead horse again, and point out that the efficiencies don't consider fliers.



  34. Anonymous,

    RE: cleaning pellets instead of patches or a bore snake.

    Yes, but it depends on what you are trying to do. Cleaning pellets are not very effective in removing lead mashed into groves. they do get out dirt and oil. Some pellets are graphite coated which leaves the graphite in the barrel. It makes patches look awful, but it is OK.

    BB generally cleans a new steel barrel with bore paste and a brass brush.

    A pellet rifle isn't like a firearm that needs cleaning after every use. In general I think BB would say less cleaning is better, unless you are having problems with group sizes increasing.

    So what are you trying to clean out?



  35. Matt,

    Found a paper by some Japanese authors that allows calculation of efficiency of one range measurement as opposed to calculating the standard deviation using all the measurements. This isn't exact for the "group size" measurement which is a variation in two dimensions, but it will give you some idea how the efficiencies vary.

    The following table is for N measurements. Assumes you calculate mean and std dev from the measurements.
    N-Measurements / %Efficiency
    2 100.0
    3 99.2
    4 97.5
    5 95.5
    6 93.3
    7 91.1
    8 89.0
    9 86.9
    10 84.9
    20 70.1
    30 60.6
    50 49.0
    100 34.9

    So for the Gaussian distribution the range for 5 vs 30 (measurements not degrees of freedom as is usually done…) is:
    60.6/95.5*100% = 63.5%

    The paper is here:
    http://www.takamatsu-u.ac.jp/library/06_gakunaisyupan/kiyo/no48/063-077_mendori.pdf

    You need to take the ratio of V(a(n)s)/V(b(n)s from tables 1 and 2. the values for 2 to 10 are given in Table 3.



Leave a Reply

Your email address will not be published.


+ 4 = 6

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>