Load Workup for Subsonic, Heavy-for-Caliber (60+ grain) Bullets in .22 Hornet

Started by gitano, July 16, 2017, 10:27:45 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

gitano

I've been trying to get a quiet .22 caliber round for use 'around the yard'. I have the Aguila 60-grain SSS cartridge working pretty well, but I can't get it up to it's 'factory' muzzle velocity of 950 f/s. In my attempts to achieve that MV, I have come around once again to the .22 Hornet, also known as the "reloadable rimfire". (You can read about my initial efforts in all of the above, here: http://thehunterslife.com/forums/showthread.php?t=19507 .)

Since I have now started to "get serious" about reloading heavy (60 grains and above), subsonic, or near subsonic loads, for the .22 Hornet, I have decided to start a new thread devoted to that endeavor.

At the moment, the rifle I am reloading for is a Contender with a 26" barrel with 1:9 twist. Given that twist rate, I can hover around the speed of sound (~1125 f/s) and still keep the Speer 70-grain semi-spitzer gyroscopically stabilized. I like this bullet, because at ~1125 f/s, it has about 200 ft-lbs at the muzzle, and ~175 at 50 yards, (about the max range I would use it). The Aguila 60-grain SSS, doing about 850 f/s has a ME of about 100 ft-lbs and only 80-ish at 50 yd. So you can see why I like the 70-grain bullet. Significantly, I LIKE being able to RELOAD for this application. A major negative for the SSS is being able to BUY them in Alaska!

The "bad news" with the Hornet was initially that it was difficult to keep the projectile subsonic, STABILIZED, AND keep the pressure up enough. Finally, I thought about using black powder (actually, BP substitutes), and voila'!, that shows promise! This thread will be about heading down the black powder (substitute) path in an effort to get a good, PRECISE, subsonic or near subsonic, RELOADABLE load to be used for 'plinking'. The thread cited above goes into the preliminary look down 'the black powder path'. The following is the first purpose-driven steps down that path.

I have two black powder substitutes (hereafter BPS). The first is Pyrodex's RS powder. The second is Alliant's "Black Powder Substitute" (hereafter ABPS). (I happen to LIKE that name. Straight to the point. No ambiguity.) A full-up-to-the-case-mouth charge of ABPS is 13.0 grains, and in a sample of one, gives a MV of 1325-ish f/s. A full case of RS is 10.0 grains, and gives a sample of one MV or 1127 f/s. After cleaning and prepping 20 cases, I loaded 10 of them with 10.0 grains of RS, and 10 of them with 10.0 grains of ABPS. I used CCI 450 (magnum small rifle) primers and of course the Speer 70-grain bullet. (Speer #1053.) I seat the bullet 0.49" in the case. Since 10.0 grains of RS is a full case, his compresses the RS load significantly, and the ABPS load "a little bit".

I installed the MagnetoSpeed chronograph on the rifle and set the target at 35 yd. I shot 5 rounds of ABPS first, then 5 rounds of RS, then 5 rounds of ABPS, and finally the last 4 rounds of RS. (One of the cases had a bad primer. :mad:) Here are the targets with the shot sequence and MV for each bullet.
The first ABPS group:


The second ABPS group:


The first RS group:


And the second RS group:


Overall, I was relatively pleased with this showing. The ABPS was pretty poor, but the RS wasn't bad. Mostly.

The MVs were all over the place. Here they are in tabular form, in the order in which they were shot.

RS:
1 - 890
2 - 1051
3 - 997
4 - 817
5 - 1021
6 - 1208
7 - 1107
8 - 992
9 - 1073

The ABPS:
1 - 940
2 - 959
3 - 1224
4 - 1235
5 - 877
6 - 1020
7 - 802
8 - 943
9 - 1225
10-785

As you can see, that's some SERIOUS variability in MV! No wonder the shots were 'here and there'.  So, I needed to find some reality in all of this. The first RS group looked VERY good except... for that one REALLY far flier! Dang! Could that be explained by variations in MV?

First, I wanted to know if there was any way to OBJECTIVELY (honestly) remove any of the shots based on the statistical variance of the MVs.  If I was a gun-writer (ptooey) like Bryce Towsley, I wouldn't let a little thing like objectivity get in the way of saying what I WANTED to say about these loads. However, I am NOT like Bryce Towsley, and I care most about not fooling MYSELF, (to say nothing of everyone else). Here are graphs of the MVs:



And


Looks like we might be able to 'reject' some of the shots based on MV. Standard statistical analysis says that we can look at the standard deviation and if - based on the SD and sample size - a given value is farther than a certain distance away from the average, it can be removed from the data set. Doing this without using a recognized standard procedure is called "cherry-picking". It is a common tactic used by charlatans, hucksters, con men, and other ne'er-do-wells like gun-writers (ptooey). (Sadly, it has become a VERY common tactic in science in general these days.)

So, in the case of the RS powder, we can legitimately remove the lowest two MVs (817 qnd 890), and the uppermost MV (1208). That leaves 6 rounds, the average of which is 1040 f/s with a SD of 45 f/s.

For the ABPS powder, we can remove the bottom two (785 and 802), and the uppermost three (1224,1225, and 1235). That leaves 5 rounds, the average of which is 948 f/s with a SD of 51 f/s.

The SDs of the two powders are similar at 45 and 51 f/s, indicating that within those MVs most central AND consistent, they have similar variance.

Still, these are very small sample sizes to start with  - 9 and 10 - and to reduce them by essentially half may be statistically "allowed", but 1) it is dissatisfying to say the least, and 2) doesn't really 'help' much. If the point of impact for half of the bullets fired can't be fairly closely predicted, being statistically 'justified' is little comfort.

However, that's only half the MV analysis. I needed to see if there was a correlation between MV and point of impact elevation. Here are those graphs:

RS:

"R-Squared" value of 0.3201 (correlation coefficient)

and ABPS:

"R-Squared" value of 0.9422 (correlation coefficient)

For those not familiar with linear regression, the correlation coefficient "explains" the variability in the variable. In other words, in the RS powder, 32% of the variability in the point of impact elevation is 'explained' by the muzzle velocity. In the ABPS powder, 94% of the vertical distribution of the points of impact are due to muzzle velocity.

At first glance, one might assume something "weird" with the RS powder. However the reality is that the more precise a load is, the less the elevation is correlated to the muzzle velocity. In other words, if you have a small little group, there is very little chance that the points of impact will be associated with the muzzle velocity of each shot in the group. On the other hand, the bigger the group (bad precision), the more likely that the vertical displacement of the points of impact will be closely associated with their muzzle velocities. HOWEVER, that said, a correlation coefficient of 0.9422 is VERY high. Essentially ALL of the variation in elevation is due to MV. What that means is; if I can diminish the variability in MV, then I should be able to get more precision.

So... after taking out the "bad" MVs, what do the groups look like?
RS:


ABPS:


Eh... Nothing to write home about. Still, maybe some promise. I think the RS shows more promise, BUT, maybe if I put more of the ABPS in, it would 'straighten up and fly right". However, the BIG deal about this to me is the SERIOUSLY HIGH variability in MV. At the moment, I have only two 'explanations':
1) Crappy primers. Maybe the "magnum" primers in this small case actually result in VARIABLE ignition. I wouldn't have thought so, but I think I should try "normal" primers, and
2) Black powder (and BPS) is not particularly 'uniform' in burning characteristics. Most muzzle-loaders have low expectations when it comes to 'precision'. Maybe one of the reasons is highly variable burning characteristics of the powder. ???

In the mean time, I think I will load up some more RS, and use non-magnum primers. If that gives me more consistent MVs, I THINK I might get this idea - subsonic BP loads in a Hornet - to 'work'.

Paul

PS - I forgot to mention: Even the slowest of MVs didn't exhibit much key-holing. That IS good news!
Paul

PPS - There's something else I forgot to mention: Look at the slope and intercepts of the linear models of elevation based on MV. They are almost identical.
RS Slope = 0.0033
ABPS Slope = 0.0037
Only 0.0004 units apart! In other words, for every increase of 1 f/s in MV, the difference in elevation change between the two powders is FOUR TEN-THOUSANDTHS OF AN INCH!
RS intercept = -3.3577
ABPS intercept = -3.4811
Again, VERY close! This value reflects the relative difference in energy density between the two powders.
That means that the mechanism (in this case, burn characteristic) that causes elevation to be correlated to MV is THE SAME FOR EACH POWDER. That's a big deal. It suggests that the linear models are "good" (real) representations of the phenomenon being looked at.
Paul
Be nicer than necessary.

gitano

So j0e_bl0ggs was asking how 'dirty' the BPSs were. After the first day's shooting, I noticed that the MagnetoSpeed bayonet looked 'greasy'. I wiped my thumb across it, and it WAS. It was almost 'wet'. Today, after shooting 19 rounds, the bayonet was definitely "greasy". Actually, given the texture of the stuff, the proper pronunciation should be "greezy", because that is what it is! Here are a couple of pictures:





Don't take the above as complaint. It is not! This 'moist' fouling is VERY EASY to clean up, and in fact, I think it adds a bit of "lubricant' to the bore. I find this 'interesting'.

Paul
Be nicer than necessary.

gitano

Based on yesterday's range session, the fundamental question was "What can I do to get the MV more consistent?" In order to address that question, I did the following:
1) Seated bullet 'out' further until it was just touching the lands, and
2) Changed primers to Remington 6 1/2 Benchrest.

The cases were full-length resized, and the bullets were crimped using the Lee "Factory Crimp" die. That is about all I could think of that was within my control to adjust in an effort to reduce the variance of the MV.

Since I was changing multiple variables, I only loaded 5 cartridges. (I wanted to make sure I didn't have to do too much 'unloading' if the results went the wrong direction.) Normally, I am loathe to change more than one variable at a time when I am working up a load, but since in this cartridge for this specific application I don't care what the reason is - I only want 'results' - I decided to change more than one variable at a time. The 'results' were interesting. Here is the target with shot sequence and associated MV:


"Velly interrressstink". Windage variation went to essentially 'zero'. Unfortunately, elevation variation was mediocre at best. However, the correlation between MV and elevation was even greater than for yesterday's loads! Look at this graph:


96% of the variation in elevation is explained by MV. The reality being that the variance in the MV didn't decrease at all. :( The range was 868 f/s to 1200 f/s. A spread of 332 f/s. So...

The change in primer and seating depth produced an improvement in the variance of the windage component of Point of Impact, but did not improve the variance in the MV and associated elevation component of PoI.

There are some other observations to note:
1) The average MV went down to 1006 f/s even though the charge was the same - 10.0 grains. The change of primer from "magnum" to "benchrest", and the reduced seating depth, had the expected effect on MV.
2) The slope of the model of elevation with MV is almost identical to the slopes of the models from yesterday. Further verification of the appropriateness of the model.
3) The model's y-axis intercept is also almost identical to the previous graphs. More validation.
4) The rather extraordinary correlation (0.9621), between MV and elevation of PoI (along with a VERY small variance in windage), strongly suggests that if I could get the MV 'stabilized', precision would be very good.

So the question essentially remains the same: How do I get the MV to be consistent?

I am beginning to think "black powder" just ain't gonna give consistent MVs. However, there is one remaining qualifier to that conclusion: I don't have a case trimmer for the .22 Hornet. Therefore, there are small variations in the length of the cases I am using. I wouldn't THINK that would cause the kinds of fluctuations I am seeing here, but when you don't know, you can imagine all kinds of explanations. And that is precisely why I go to all the trouble of the analysis/number crunching - eliminating imaginary 'solutions' whenever possible.

The big question is now; "what next?" I think I may go to some duplex loads using "smokeless" and a smaller amount of BPS. The idea being to rely on the consistency of the nitrocellulose powder, and use the BPS for 'filler' to keep the small volume of the smokeless in the right 'place' in the case, (so I stay near speed of sound). There's some number crunching that has to be done to come up with some semblance of experimental reality for those duplex loads. When I have something to report, I will.

Paul

Be nicer than necessary.

gitano

Sometimes I get 'focused' on the details, and miss a bigger picture. I was looking at the target "overall", and decided to calculate the "max spread" (the Euclidean distance between the two shots that are farthest apart). It is 1.27". That's almost "acceptable" precision. Theory would dictate that at 50 yards, (my max range for this cartridge and application), the max spread would be (50/35)*1.27 = 1.81". That's just a little bigger than I want the CONSISTENT group size to be. I'd like it 1.5" or smaller at 50 yd, which means 1.05" at 35 yd. But this group isn't as bad as I was 'seeing' it.

Paul
Be nicer than necessary.

gitano

So as you might imagine, I'm still 'fiddling', but I gotta admit that I'm about to cry 'uncle'. I decided that I would try some IMR 7828. QuickLOAD suggested that it would give a good MV of about 1050 f/s even if it did only burn about 30% of the powder in the barrel. That little adventure was INSANE!

The first MV was 736 f/s and hit about 4" low of the PoA. The second shot went "pffffffffft". When I broke the action open, the case ejected and hit me in the chest with enough force to 'smart' a little bit. When I looked down the bore, it was of course plugged. Here's a picture of where the bullet was. That's the muzzle.

The bullet's tip was right even with the muzzle. :mad:

It gets better. The third shot went BANG! Really loud. MagnetoSpeed said 1235 f/s! The fourth shot was a hang fire! Pffft...Bang! Didn't even hit a 12x18 inch target at 35 yards! Last shot was 900 and something f/s. INSANE!

The issue is inconsistent ignition and or burning. It's a tiny case in a LONG (26" barrel). Keeping the MV down and the pressure up may simply not be possible. However, I thought there still might be two possible explanations:
1) Flash-hole size, and
2) Primers.

I didn't really like the magnum primers, (they gave inconsistent MVs too if you recall), and I didn't like these Remington 6 1/2s either. They were fairly old. Maybe they were "bad". So...

I removed the Remington 6 1/2s from the remaining 31 cases. I then drilled ALL the cases' flash-holes to a uniform 0.088". A little larger than the recommended 0.065 (1/16th of an inch), but I wanted PLENTY of 'breathing room'. I then primed the cases with Remington 7 1/2s, (Remington's Small Rifle Benchrest primers), hoping for some extra consistency.

I spent about 10 hours working on several duplex loads, but really couldn't come up with one that was particularly satisfying. I finally decided to try the black powder substitutes again now that I had a different primer and the flash-holes were uniformly large.

I had noticed that there was some difference in case capacity between cases. Since black powder "likes" to be compressed, and it is very common to make the BP charge "a case full", AND since that would ameliorate to a large degree any differences in case capacity, I decided to load three charges:
1) A case full of TrailBoss. It had produced nothing but bullets left in the bore the first time, but maybe that was due to primer and flash-hole inconsistencies I thought.
2) A case full of Pyrodex RS. It too had worked fairly well the first time. Just had large max spread in MV. Maybe due to primers and flash-holes.
3) A case full of Alliant Black MZ (heretofore BPS). It had also done "ok", again, maybe the wide variance was due to the primers and flash-hole inconsistencies I thought.

I set the target up; got the chronograph hooked up, and 'let fly' with one of the TrailBoss loads. KRAP! Bullet in barrel AGAIN!. That was all for TrailBoss!

I shot the Alliant Black MZ loads next. Here's the target with the shot sequence and associated MVs:

Not bad group. (I'll come back to the stats in a minute.) But look at the MVs. They're all between 1429 and 1511 f/s. Not exactly "quiet".

Finally the Pyrodex charged rounds. Here's its target:


Good MVs, but look at the max spread - 989 to 1240 f/s (251 f/s) - and look at the stringing. AGAIN the windage is GREAT, and AGAIN the Point of Impact is highly correlated to MV! Grrr...

So... I think I may be resigned to an average MV somewhere in the 1200 f/s range, and having the 'live with' the loudness. :( I am beginning to believe that there is no way to get a subsonic load in a .22 Hornet with a 26" barrel.

The Alliant Black MZ group was 0.94" max spread (2.57 MoA). That's tolerable. Group size at 50 yd would be ~1.35". Average MV was 1461 f/s with a 32 f/s standard deviation. Not 'bad' other than too loud. Nice 'square' group.

The Pyrodex group was 2.02" max spread (5.5 MoA). That group size would be 2.9" at 50 yd. Not acceptable. Average MV was 1100 F/s with a SD of 93 f/s. Good mv; unacceptable SD.

So... I think I'll try some fast powders - Blue Dot, Unique, Bullseye, Herco, Li'LGun, Power Pistol, and 2400 - WITH some 'filler'. Maybe I can get one of them to work.

Paul

Be nicer than necessary.

Hunterbug

Were these charges weighed or measured? What about cleaning off the bayonet of the chronograph after every shot? I wonder of it's getting so dirty and interfering?
Ask not what your government can do for you. Ask how your government can go away and get out of your life.
 
 
The unarmed man is is not only defenseless, he is also contemptible.
Niccolo Machiavelli

gitano

Prior to the LAST batch, all charges were weighed to better than 0.05 grains. The last batch was "case full".
A "dirty chrono" is a good question, but the MagnetoSpeed works on the principle of induction. As a metal object passes through a magnetic field, it changes the induction (and capacitance). Measuring that change is what triggers the chrono. I could put a thick layer of mud on the bayonet and it might (but probably not) affect trigger sensitivity, but it wouldn't affect the value of the velocity reading.

I'm not sure I have defined what "working" means to me in this application. First, defining the application:

1) .22 Hornet cartridge, (because it's reloadable),
2) Range of use between 10 and 50 yards.

So to "work", that .22 Hornet must do the following.
1) Be quiet,
2) "Quiet" means slow, near subsonic,
3) "Slow" means a heavy-for-caliber bullet is necessary; must be 60-grains or heavier. Since I don't have any molds for .22 caliber bullets heavier than 45 grains, I am using the Speer 70-grain  bullet.
4) "Heavy" means long. "Long" means fast twist barrel for gyroscopic stability. Barrel has 1:9 twist. Just fast enough for the 70-grain Speer bullet (#1053).
5) Precise (AKA "accurate"). Must be able to shoot a 1" group at 35 yards. That translates to better than 2.73 MoA.

So I have numbers 3 and 4 'working'. I don't seem to be able to get number 5 AND numbers 1 & 2, to work at the same time. I can get the needed precision, but I can't do it 'quietly'. I can get 'slow and quiet', but when I do, I lose the precision. Since precision is "what it's about", it is looking like I am going to have to give up "quiet". At least quiet in the sense of being near the speed of sound. It's looking like I can't get the desired precision at MVs less than about 1200 f/s. And that may be too optimistic. I may have to go to 1300 or more.

Like I said, I think I'll give one more try to some fast powders WITH FILLER, and see what kinds of MVs AND precision I can get.

It - the problem - isn't exactly the cartridge. It's the barrel length. The longer barrel either ups the MV or lowers the max pressure. In other words, in a shorter barrel I would have to up the pressure to get the same MV. Upping the pressure would improve the burn characteristics, (consistency), and thereby improve precision. With the long barrel, low pressures are too erratic to get good precision. If I up the pressure, I up the 'loudness'. One can also up the pressure without upping the MV by upping the bullet weight. However, in a given caliber one cannot up the bullet weight without making the bullet LONGER. At a point, the bullet gets too long for the twist rate of the rifling to stabilize it. About the only way I could up the bullet weight is to go to tungsten! I think everyone can appreciate that getting tungsten bullets might pose a few extra challenges. ;) Too bad really. Tungsten has 1.7 times the density of lead. Therefore, a tungsten bullet shaped exactly like the Speer 70-grainer would actually weigh 119 grains! Wouldn't that be a cool bullet!

Anyway, tungsten bullets are a pipedream. I have to live with what I have. What's that's looking like is something north of 1200 f/s MV.

Paul
Be nicer than necessary.

j0e_bl0ggs (deceased)

Turvey Stalking
Learn from the Limeys or the Canucks, or the Aussies, or the Kiwis, or the...
                   "The ONLY reason to register a firearm is for future confiscation - How can it serve ANY other purpose?"

j0e_bl0ggs (deceased)

Left to right;

   • Cut wire #8 (~0.156") x 1.125L
   • Squirted core (60gn)
   • Herters swage
   • My point form to reduce the shoulder.
   • Cannelured for lube

Total length 0.620"









...
Turvey Stalking
Learn from the Limeys or the Canucks, or the Aussies, or the Kiwis, or the...
                   "The ONLY reason to register a firearm is for future confiscation - How can it serve ANY other purpose?"

gitano

I think I'm on to something. There's good news and 'annoying' news.

I started reloading when I got a Lee Loader for my 16th birthday. My Dad was in Vietnam, but he told my step-mother to get it for me. It was for my .270 Win. I just followed the directions meticulously, and loaded some of the most precise ammo I ever have. There was no expensive press. I did later get a separate RCBS scale, but at first the cases were charged using only the dipper provided in the kit. Standard 150-grain bullets. (I don't recall the brand.) The only additional tool needed was a hammer. It was used to decap, resize, and seat the bullet. Resizing was accomplished by pounding the case into the resizing die with the hammer, and seating the bullet - to "standard overall length" - meant only pounding the 'handle' of the priming head into the bullet seating die. I honestly cannot imagine a simpler device or process. Simply GREAT ammo was produced. It wasn't long though until 'I wuz thinking'...  So began a lifetime of 'experimenting'... with bullets, and just about everything else I could get my hands on.

Loading the .22 Hornet cartridge for this particular application in 2017 could hardly be more different than loading for that .270 50 years ago in 1967. Almost every element is both complex and convoluted. I think I have found a PROCESS and a load that is going to 'work'. But DANG! It is the reloading antithesis of 'simple'!

As I mentioned above, my 'last grasp at straws' was to return to some fast-burning powders and significantly, FILLER. I am happy to report that preliminary results indicate that it seems to have worked. But... Let me take you through the process.

As mentioned above, I paper-whipped (QuickLOAD) charges for 8 "fast" powders: Blue Dot, Unique, Bullseye, Herco, Li'LGun, Power Pistol, W296, and 2400. I had to first modify the "ratio of specific heats" for each powder in QL per the instructions of the author of QL for use in the .22 Hornet case. After 'creating' these 'new' powders in a separate file, I turned the crank on QL and looked at the charges suggested that give a MV of 1120 f/s. (I had given up on getting down in the 1050 f/s range.) According to QL, only 5 of the 8 powders produced 'viable' loads: Blue Dot, Power Pistol, Li'lGun, 2400, and W296. The charges and the Max Pressure generated for each of those powders were:
Li'lGun - 9,775 PSI from 5.4 grains,
W296 - 10,695 PSI from 5.6 grains,
Blue Dot - 11,269 PSI from 4.0 grains,
2400 - 11,412 PSI from 4.4 grains, and
Power Pistol - 11,503 PSI from 3.1 grains.

On one hand I wanted the most pressure I could get, (remember the long barrel), but I also wanted the most powder weight/volume I could get too. If you look at the above values, you can see that pressure and charge are essentially inversely related. Meaning that once again, what I 'wanted' was the opposite of what there 'was'. Those with the highest pressure had the lowest charge weight. Since I had read and heard that Li'lGun was often used in the Hornet case, I decided to start with that powder. It had a high charge weight but the lowest max pressure. Still, the 'experts' often used Li'lGun, and at this point, I'm clutching at straws in the wind.

Based on the experiences of the past few days, I decided that I was tired of "unloading" cases with loads that 'didn't work'. I was going to load ONE case, fire that cartridge, check the 'output', and go from there. I also decided to do "everything" I could to ensure uniform ignition and burn. Here's the procedure in all its gory detail:

1) Remove the existing primer, but with the LEE "Universal Decapping Die", NOT a .22 Hornet resizing die. This meant that the spent primer was removed, but there was no resizing performed on the case.
2) I resized ONLY the neck using the LEE "Collet Neck-sizing Die" specifically for the hornet.
3) I primed the case with a CCI 450 primer - Small Rifle Magnum. (I wanted all the oomph I could generate.)
4) I charged the case with 5.4 grains of Li'lGun powder.
5) I filled the case to the mouth with cornmeal. (I'll discuss details associated with the cornmeal, below.)
6) I seated the Speer 70-grain bullet using a very old Lyman bullet seating die, and
7) Put a crimp on the case mouth using LEE's "Factory Crimp Die" for the .22 Hornet.

The only thing I didn't do was check the case volume.

I took this one cartridge to 'the range', and fired a shot. I didn't bother to set up a target because I didn't care about precision for the moment. I just wanted to make sure the cartridge actually got the bullet out of the barrel, and see if the MV was somewhere close to what QL said it was 'supposed' to be. That first Li'lGun 5.4 grain charge produced an MV of 1490 f/s. Hmm... Not surprising due to the added weight of the cornmeal, and the magnum primer. I returned to the reloading bench.

I ran through the above sequence of reloading THE SAME CASE. This would ensure that there was no "between case variability". This time I used the same charge, but changed the primer to the Remington 7 1/2 - Small Rifle Benchrest. This load produced a MV of 1414 f/s. Good! Moving in the right direction. I decided to try another powder.

Since I was 'off' the Li'lGun, I decided to use the powder that had the most weight AND a pretty good max pressure: Winchester 296. I have very little experience with W296 powder. I had bought the cannister of it for use in the .223 Rem, but ended up using CFE223 instead. So, I started with the QL-suggested load of 5.6 grains and the Remington 7 1/2 primer. It produced a MV of 1385 f/s. Good! Moving in the right direction! Back to the reloading bench.

I lowered the charge to 5.0 grains. That charge - everything else the same - produced a MV of 1140 f/s. Ah. That's close to where I want to be. Now I'll load 5 of them and see if I can get consistent MVs. I selected 4 more cases at random and the 'original' case and loaded 5 rounds. Back to the "range" AND a target set up at the normal 35 yd.

The first shot - using the 'original' case - produced an MV of 1141 f/s. (Remember that the first shot from that case was 1140 f/s. THAT'S "consistent"!) The second shot was 1019 f/s. The third was 1068 f/s. The fourth was 878 f/s. DANG! There went the consistency! The fifth shot was 1026 f/s. Hmm.... The first shot of this 5-shot group was 'right on' what the first shot from that case had produced, and three of the remaining four were pretty close together: 1019, 1068, and 1026. THAT'S consistent. Maybe the 878 could be explained.

Here's the target with the shot sequence and each shot's associated MV:


Notice two aspects of this target:
1) Again, the variation in windage is 'tiny', and
2) Elevation is correlated to MV.

Hmm... There's only one thing that could "explain" the 878 f/s MV. I hated to even consider it, but it was the only thing left to account for - case capacity. Grr... I went back to the bench and measured the case capacity of each of the cases. Here's a graph of case capacity regressed on MV:


Clearly case capacity 'mattered'. :mad: 71% of the variability in MV is explained by case capacity. That's significant. Here are the case capacities (in grains of water) of the 5 cases:
1) 13.65 - the 'original' case
2) 14.35
3) 14.60
4) 15.40
5) 14.50

Note particularly the first and fourth capacities. The first one is the smallest by a significant amount. It's the one that produced the highest velocity. The fourth one is the largest capacity: It produced the lowest velocity. The other three are close, and they produced velocities similar to one another. The 'verdict' is in: In this small case, I HAVE to pay attention to case capacity if I want consistent MVs. :mad:

So once again, (the .17 Remington in the first instance), I have to 'pay attention' to all the little nuances of reloading in order to get reloads that 'work'. While I enjoy the 'chase', the fact of the matter is that I often finding myself longing for the simpler reloading days of my youth when all I had to do to get really good loads was "follow the instructions"...

Paul
Be nicer than necessary.

sakorick

Win 296 was the standard .410 powder for years. It was replaced by H110 when Hornady took over the Winchester powders. When I ran out of the 5 pound keg I had for 30 odd years, I picked up some H110 and it looked just like W296. I called Hornady and the techie told me that it looked the same because it is eggzactly the same thing. H110 is a rebranded Win296. Why they just didn't call it H296 is beyond me!:Banghead:
Talk to yourself. There are times you need expert advice.

gitano

Very interesting, Rick. I purchased the 296 I have not more than 2 years ago. Maybe it was "new" old stock.

So, the data from shooting the fast powders, specifically W296, topped off with cornmeal, clearly indicated that MV was correlated with case capacity. Given that information, I weighed and measured the case capacity of all of the Hornady .22 Hornet cases I have on hand - 23 of them. Here is that data:

By the way, the case capacity was measured using W296 powder.

Graphed, that data looks like this:

Clearly there is a correlation, but it's not particularly strong - R-Squared = 0.2858. Based on their weight, at least 5 of the case capacities are not consistent with the model. Five out of 23 is actually a pretty big number when MV is highly correlated to case capacity.

Given that QL predicted very closely the MV of the one case that I looked at in the "fast-powder-plus-cornmeal" experiment, (MVs were 1140 and 1141 and QL predicted 1138), I decided to enter the exact case capacity for each case and have QL suggest a specific load for each case. When I went through the exercise, it was clear that I could put cases into batches of 0.1 grain. In other words, I could lump cases with capacities of 13.20, 13.25, and 13.30 together and get one charge for those three case capacities. Doing that gave be four batches of case capacities among those 23 cases. There were seven cases in each of the cases with capacities centered on 13.25 grains and 13.50 grains. I took 5 cases from each of those two batches and loaded them with 4.85 grains of W296 for the 13.25-grain batch, and 4.90 grains for the 13.50 grain batch. Each case was topped off with cornmeal. Each case was primed with a Remington 7 1/2 BenchRest primer, and the bullet crimped with the Lee Factory Crimp die. The QL predicted MV was 1050 f/s.

I chose to shoot the cartridges with the 4.90-grain charge first. Here are the MVs from those cartridges:
891
1327
1252
1192
You'll note that there are only four.

Here is the target from those four shots, with the shot sequence and associated MV:


As you can see, the first shot was 891 f/s and 0.9" low of the PoA. The second shot was a squib, but the bullet had enough energy to not stick in the barrel. No MV was recorded. The third shot was 275 f/s higher than predicted. The fourth shot was 202 higher than predicted and the fifth shot was 152 f/s higher than predicted. Overall, pretty much a pile of krap. :mad:

In spite of the 'poor' showing of the first 5-shot group, I decided to shoot the second batch of 5; those charged with 4.85 grains. Here are their MVs:
1249
1218
1237
1245
1308
Average = 1251 f/s, Standard Deviation - 34 f/s. While the average was 201 f/s faster than predicted, at least: 1) They all SHOT!, 2) they had a relatively small SD. The better news was the PoIs.


That's not bad. In fact:
Euclidean Max Spread = 0.76" (The group size at 35 yd; the range at which it was shot.)
Minutes of an Angle = 2.08
Group size at 50 yd. = 1.09"

This is what I was hoping for. However, this really can't be taken 'in a vacuum'. (Unless of course you're a lowlife scumbag gun-writer (ptooey). And 'you' know exactly to whom that description refers.) Taken IN TOTAL, 10 rounds were fired. Two were squibs, and four of them of one charge were "all over the place".

I am going to take the cases from the "good" shots and reload them exactly the same: Resize neck only with collet resizer, prime with Remington 7 1/2 BenchRest primers, charge with 4.85 grains of W296, fill case to mouth with cornmeal, seat bullet for OAL of 1.786" (~0.407" seating depth), crimp bullet with Lee Factory Crimp die, and shoot again. If those same cases perform "nicely" this time, I'll reduce the load and try to get down closer to 1100 f/s on average.

News at ll.

Paul
 
Be nicer than necessary.

gitano

It occurs to me that there MAY have been some error introduced in using W296 for measuring the volume of the cases. I am going to remeasure the capacities of the 10 cases used in the last 'shoot' using water instead of W296 powder. Maybe that will explain the craziness of the ones charged with 4.90 grains.

Paul
Be nicer than necessary.

gitano

OK, I'll spare you the yelling and swearing and cut to the chase scene...

I thought I could get down to near 1050 f/s, so I 1) reduced the charge to 4.5 grains, and 2) reduced the cornmeal filler to "just at the bottom of the seated bullet". First shot went Pfffft. The barrel was plugged as usual. I THOUGHT it was plugged with the bullet "as usual". However, I BROKE a cleaning rod trying to get "it" out. Again, sparing you the gory details, what it turned out to be was a plug of cornmeal about 3/8ths of an inch long!. And it was practically 'welded' to the bore. The plug was about 8 inches from the muzzle, BUT THE BULLET HAD EXITED THE BORE! :eek: All that 'work' had me looking closely at the MagnetoSpeed bayonet. It had a very hard CRUST of cornmeal all over it! I had to scrape it off with a knife! OK... No more cornmeal!

Last week I ordered some .22 Hornet brass from Cabelas. It came in the mail today. First thing I did was weigh and measure the case capacities on 10 cases. There was one capacity of 13.85 grains; one of 13.75 grains, and eight of exactly the same 13.80 grains! Measured with water. Now THAT'S uniform case capacities!

While I was out getting the mail, I stopped at Sportsman's Warehouse and bought, among other things, some CCI Small Pistol primers. I primed 5 of the new Winchester cases with the new CCI Small Pistol primers, put in a charge of 5.0 grains of W296, stuffed 1/4 sheet of toilet paper on top, crimped the bullet in, and 'let fly'. Here is a picture of the spreadsheet and the target:


Now THAT'S what I have been looking for!

Even the first shot high can be explained. Because of all the hoohah with cleaning up the cornmeal, that first shot was from a very clean barrel that was slightly 'lubed'. Nevertheless, the average MV is 1134 with a SD of only 15.4 f/s. I happy as a clam with that! Group size is 0.91", 2.48 MoA, and 1.3" at 50 yards. I'd bet dollars to donuts that the next group will be about the size of the last four of this group. I'll shoot that group tomorrow and refine the aim. (The group center is 0.35" left and 0.52" high of the PoA: 3 clicks down and 4 clicks right will give me 0.2" high at 35 yd. That's 0.2" low at 50 if the .219 BC for the Speer bullet is correct.)

I want to lower the charge a bit to get the average down to about 1100, but I'm a little afraid to mess with it. We'll see if I get braver over night.

Ahhhhh... I think I've got a load. And I'm not even remotely worried about residue from toilet paper 'wadding'.

Paul
Be nicer than necessary.

sakorick

Talk to yourself. There are times you need expert advice.

Tags: