Range report for high velocity test of Bullshops NOE 107s and Lotak lube

[DrB]

Bullshop had posted a thread a while back, linked below, that reported some high velocity (3650 fps) results with a 37 grain 22 caliber NOE 107 bullet out of a 221 Fireball, and lubed using his soft Lotak lube.

http://www.castboolits.gunloads.com/...elocity+usable

Bullshop's initial reported accuracy was so-so, but what captured my interest was the velocities he was reporting and the lack of lead fouling. Bullshop was kind enough to send me a small sample of bullets and a stick of his soft Lotak lube to experiment with. In exchange I promised Bullshop this range report with my results.

To cut to the chase, I found that in my k-hornet, I was able to achieve velocities of around 3200 fps average with Bullshop's NOE 107 bullets and Lotak soft lube. In my gun, I encountered no observable lead fouling after 80 rounds fired. The only apparent fouling was light carbon fouling that brushed/patched out quite easily. The best load I tested produced a measured five shot accuracy of just under 3 MOA, at 3217 fps with a standard deviation of 28 fps. The loads were tested at 25 and 100 yards, with no apparent change in the angle of dispersion as a function of range.

Rifle preparation:
I had previously used this rifle to shoot both jacketed and more recently, exclusively molybdenum disulfide coated jacketed bullets. Consequently, I wanted to be sure to remove as much copper fouling and/or any moly “pebbling” from the bores interior as possible prior to trying the cast bullets. I started out with a liberal spray of a penetrating oil, followed by about a half hour period of time for the oil to coat the bore, then bore brushing, a brake cleaner rinse and dry patches. This approach was repeated a couple of times, after which it still appeared to me some fouling was present. At this point, I utilized a patch wrapped bore brush coated with flitz polish and gave the bore fifty strokes, followed by penetrating oil, a brush, brake cleaner rinse, and dry patches till clean. After visual inspection which seemed to suggest the bore was free of fouling (as far as I could tell without better inspection tools) I then ran a lightly oiled patch through the bore.

Cast bullet processing:
I visually inspected the bullets Bullshop sent me, and found them to be quite nice – well filled out with sharp edges, and no visible defects. The bullets were from the same lot Bullshop had used in his earlier test from range scrap, and had not been measured for hardness. I checked the hardness and arrived at 23 BHN. (I used the camera methodology I have previously posted on for measuring hardness in combination with a Lee tester.)

Slugging the bore throat on my k-hornet, I found the bore to be about .225 groove to groove. The bullets I found to be .2252 to .2258 on their diameter. I didn't wish to decrease the as-cast diameter of the bullets at all, so I seated Hornady gas checks onto the bullets using a Lyman gas check seater with a RBCS lube-a-matic 2, such that the checks were just seated and crimped without sizing the bullet driving bands. Upon seating some checks, I noticed a couple of things.

First, there was a small crescent-shaped shiny spot of varying size on one side of the rear driving band where the bullet was apparently “kissing” the side of the die during check seating. Sometimes this shiny spot would extend up partially onto the forward driving band.

Second, upon fitting and playing with the checks I observed they were not seating flush on the bullet shank with initial finger seating. I measured the bullet shank and arrived at a dimension of .216-.217, and measured the Hornady check ID at the base and found the dimension was .214. Because of this mismatch in diameter, the checks would not seat flush with finger pressure.

To evaluate how badly they might be seating out of flush, I used a micrometer to measure the overall length of the bullets so I could find minimum and maximum OAL before and after check seating and estimate out-of-flatness for the check. What I found was that for this run of NOE 107 bullets, the nose of the bullets from the two mold halves were about .002 to .003 offset along the bullets axis of symmetry. I measure several bullets and got similar results, and confirmed that it appeared to be the result of a mold half offset by dragging my fingernail across the parting line on the nose from one side and then the other. As expected, it caught on the ~.002 ledge one way but not the other. Bases of the bullets appeared to be “perfectly” flat, having filled out to the sprue plate.

After seating gas checks, I found that typically there was not a variance in the minimum and maximum OAL as far as I could detect, so as near as I can tell the checks ended up square, at least to within a thousandth of an inch.

I then pan lubed the bullets with the soft Lotak in a pie pan/double boiler arrangement. While this mostly worked, the lube tended to want to pull out of the lube groove during cutting out and handling of the lubed bullets, and I did not find this to be an ideal method to apply the soft Lotak. While this formula is supposed to be “low-tackiness” (and perhaps was relatively so in stick form), after melting and resolidifying, the lube was both soft and quite tacky/sticky. Perhaps the melting/resolidifying altered the tack of the lube. Lotak would work well for hand application by smearing into the grooves, or by applying as intended without remelting the stick in the lubrisizer.

Brass processing:
The cases used for these loads are fireformed, annealed, and trimmed Remington Peters (RP headstamp) cases. The RP cases I have found have significantly greater case capacity (and thinner walls) than the Winchester WW headstamp cases. Most of the load variants were only neck-sized for about half of the length of the neck, such that the cartridge shoulder and bottom of the neck would align with the chamber and throat. A variant was tried in which the neck was unsized and the lubed bullet crimped into place after seating using a lee collet crimp die.

Loads Tried:
I tried several load variants ranging from 15 to 17 grains of Lil'Gun. Of course I do not recommend this load for others, etc., etc.... actually, without the right brass, a K-hornet or similar improved chamber, and fluidizing the charge I doubt others will even fit so much powder in a hornet case.

If anyone has an interest I'll be happy to post particulars on all the load variants and their results, but will omit it for now. Besides charge weight, the variants I ran also investigated bullet seating depth (out to engrave the lands and seated deeper with the lube groove inside the case neck), uncrimped vs. crimped with a lee collet crimp, and settled or unsettled powder charges fluidized with a headless ultrasonic toothbrush.

The brass were primed using Wolf small pistol primers. I have not done an exhaustive primer study, but use these wolf small pistol primers in my jacketed bullet loads in hornet with reasonable results, and so used them here as well.

Each charge was individually trickled and weighed on a powder scale. The heavier loads were settled in the case using a powder funnel and an ultrasonic toothbrush without the bristle head (which works great).

Bullets were seated in inside chamfered but unflared case necks using a Hornady concentric bullet seater. When crimped, the lee collet crimp die was set for one half turn of crimp.


The best results were obtained from the following load:
RP case fireformed and neck sized for half of the neck length
Wolf Small Pistol primer
NOE 107 bullet of 23 BHN mystery alloy

• Hornady gas check
• Unsized (.2252-.2258)
• Bullshop Lotak bullet lube

Lil'Gun charge of 17 grains settled with an ultrasonic toothbrush sans brush head (again, I do not necessarily recommend this load for others).
Bullet seated to just cover the lube groove, and crimped in place with a lee factory collet crimp die with crimp at ½ turn.

This load, as previously mentioned, produced 3.0 MOA in my rifle at 3217 fps average velocity, and 28 fps standard deviation.

Some possible sources of (in)accuracy:

• The ~.002-.003 apparent offset in the mold halves produces a bullet that is slightly heavier on one side than the other.
• The gas check shank was too large for the Hornady gas checks, a check that started out seated flush might end up more consistently flush (though I was unable to detect any out-of-flatness in the ones I spot-checked)
• Wolf primers (a different primer might produce more consistent velocities, though I consider <1% SD variation to be pretty good)
• Lotak lube buildup in the bullet seating die. I observed that bullet seating depth tended to become progressively deeper (measured variation was about .005 in OAL) due to the stickiness of the pan lubed Lotak bullets and the smears/pieces of Lotak stuck to bullet noses getting transferred to the die. A cleaner application of the soft Lotak such that the lube doesn't get into the concentric seating/nose portion of the die would ameliorate this. A more consistent bullet seating depth might tighten this up.
• It's a ruger model 77. I've gotten less than ~1.2 MOA from this ruger with jacketed loads before, but it is probably not the most precise of rifles... off a bench, it does not appear to shoot better than I can see, hold, & squeeze even with the best jacketed loads I've developed.
• Bullshop thinks an increase in the alloy hardness might have a beneficial effect at this velocity.

Fouling:
The first picture below was after firing twenty rounds, and shows a patch wrapped on a bore brush and pushed through once from chamber to muzzle. There was zero evidence of any gray colored or metallic fouling on the patch, after this one pass or after several passes through the barrel. The patch picked up a loose dry carbon fouling, which seemed to clean easily from the bore. The same fouling was observed after eighty shots, and cleaned up with several passes of a patched brush soaked in penetrating oil and followed with brake cleaner. No dull or streaky patches, or filling of the lands were observable after shooting – with a little cleaning, the bore was bright, shiny, with sharp dark rifling, and apparently fouling free.

After 20 rounds, patched brush passed from bore to muzzle


Several scrubs of same patch...


After 80 rounds, patched brush scrubbed several strokes.


You can see the toolmarks from the rifling cutting process in this one... clean and crisp rifling.




Some carbon/powder fouling toward the chamber end of the barrel. This scrubbed/patched out easily.



In summary, lead fouling at 3200 fps was a complete non-issue for these bullets and Lotak lube in my rifle over the course of 80 shots. There's some work to be done to improve accuracy, and I'd also like to try some different lube/bullet/load combinations to see where/when things fall apart. But really, getting these NOE 107 bullets with Lotak lube to go 3200 fps without lead fouling was completely painless.

Thanks, Bullshop, for the lube and bullets. I'm looking forward to playing with this combination more in the future.

Best regards,
DrB

[Bret4207]

Interesting. For the pressure you aren't using a super hard alloy, the design is pretty simple, a good lube of course helps. Kinda makes me wonder if it's just a matter or some sort of case capacity-bore ratio thing? I dunno. Keep at it!

[Larry Gibson]

DrB

Well done, a detailed excellent report. Consider if you backed off on the velocity and accuracy improves then the possible sources of inaccuracy you list are adversely affecting the bullets accuracy in flight as the velocity and RPM go up. If that is a 14" twist barrel you are pushing over the RPM threshold. If it is a 16" barrel twist you are on the ragged edge of the RPM threshold. Either way the 3 moa suggests something isn't flying right (pun intended).

Larry Gibson

[DrB]

DrB

Well done, a detailed excellent report. Consider if you backed off on the velocity and accuracy improves then the possible sources of inaccuracy you list are adversely affecting the bullets accuracy in flight as the velocity and RPM go up. If that is a 14" twist barrel you are pushing over the RPM threshold. If it is a 16" barrel twist you are on the ragged edge of the RPM threshold. Either way the 3 moa suggests something isn't flying right (pun intended).

Larry Gibson

Larry, thank you. I recalled your concern from prior discussion. Spin rate for the 3200 fps loads is running about 165,000 rpm. You had previously proposed, I believe, an rpm limit for this caliber of around 120,000-140,000 rpm? So it seems we are a fair bit above that here.

I'll add a post with moa vs. velocity. As mentioned, I did run lower powder charge variants. This load had both the highest velocity and the smallest grouping of those tried.

Our range is limited to 100 yds, so while I couldn't shoot these at greater ranges, I did shoot a target at 25 and 100. I was unable to determine a significant change in moa group size between the two ranges.

As near as I can tell, grouping here appears to be principally the result of barrel pointing at departure plus whatever lateral velocity component the bullets are getting at departure. I'm not seeing anything that looks like an "in-fight" effect.

Best regards,
DrB

[DrB]

Larry, here you go...

It would be interesting to collect more data for just two different velocity loads at 25 and 100 yards, and with more than five shots each. Note that the data points plotted represent five different load variants and five shots each. For each load variant, more than just powder charge/velocity were changed (crimp, seating depth, and powder settling were changed). A more correct way to analyze/plot these points would be using multiple linear regression, vs. just the x vs. y plots below. While I've done this before I'm still getting my software tools back in place to enable me to do it again... and really to have done it right I should have done a more formal experimental design that would have required more test points and work to collect/reduce the data.



In the above on the left, you can see that the group size generally trends down with increasing velocity (higher velocity is generally giving smaller groups, opposite to what you are suggesting should be expected). Note that the correlation coefficient (R^2) is poor, indicating that much of the variance in the data is not explained by the equation shown. In the plot on the right, you can see that there is generally a positive correlation between velocity standard deviation and increasing group size (group size goes up with increasing muzzle velocity variability).

While these plots don't show great fits to the data, again, keep in mind that there are other hidden variables in these loads that almost certainly account for some of the variability. Also, much of the poorness of the fit in both plots is a result of the Variant #4 data point, which I've removed below.



In the above plots, you can see that there is actually a pretty good linear fit for both trends... it's a bit iffy given that there are only four data points to which the line is fitted (which reduces the statistical likelihood that the trend is real). Also, when you have one outlier among five data points and you eliminate it, that's a questionable action in as much as you are throwing away a fair bit of a limited data set...

Nonetheless, the data I have available are not consistent with the proposition that accuracy is decreasing as a function of increasing velocity -- if anything, the opposite appears to be a significant trend. It also appears that their is a correlation between the velocity variance and the group size, which makes sense if the principle source of inaccuracy is from the pointing of the muzzle at departure plus any lateral velocity component the bullet may be picking up at the time of departure of the bullet from the crown.

These data are also compatible with the jacketed data I have -- I suspect this is a trend principally related to the case capacity becoming filled, and the powder geometry/combustion/and velocity all becoming more consistent from shot-to-shot as a result.

Best regards,
DrB

[DrB]

Interesting. For the pressure you aren't using a super hard alloy, the design is pretty simple, a good lube of course helps. Kinda makes me wonder if it's just a matter or some sort of case capacity-bore ratio thing? I dunno. Keep at it!

Bret, I wonder if it may not be the combination of the smaller caliber and also the length/to bearing length of the bullet. Both of these things I think will tend to reduce the balloting/rifling loads on the bullet as it spins down the barrel as compared to a larger caliber/longer design.

RE case capacity vs. caliber -- Bullshop wondered something similar in his original post, and was using a 221 fireball for his shots. I really don't know. With a ball powder I wouldn't be surprised if you had a cloud/slug of powder following the bullet at least initially and acting kind of like (burning) cream of wheat. While I don't think it would seal the bore directly (it would produce hot gas at the base of the bullet), I do have an idea I'll be posting later on in the bullet base vs. accuracy sticky as to why a filler may help reduce gas leakage/cutting. A lightbulb went off while I was looking at one of 303Guy's recovered paper patched bullet pictures.

Best regards,
Caleb

[Bret4207]

FWIW- that boolit looks to me like a shrunken 311316 Lyman, one of the easiest boolits I've used to achieve high velocities in the 30 cals. Of course my "high velocities" were more like 23-2400 fps, but maybe it does have something to do with bearing length.

[DrB]

FWIW- that boolit looks to me like a shrunken 311316 Lyman, one of the easiest boolits I've used to achieve high velocities in the 30 cals. Of course my "high velocities" were more like 23-2400 fps, but maybe it does have something to do with bearing length.

Yes... Aside from the crimp groove it does, doesn't it? But it is 112 gr vs 37gr... do you have any suggestions for similar larger and smaller caliber bullet designs?

Found a picture of the 311316 below.

http://www.three-peaks.net/bullet_molds.htm

[Larry Gibson]

DrB

Once again, let me remind you and all others who think the RPM threshold is a "limit", it is not a limit. It is a threshold that may be moved up or down depending on the load and componants used. The RPM threshold generally runs in the 120 - 140,000 RPM range. You are indeed pushing it upward as I have and others have. Not a limit, ok?

"I suspect this is a trend principally related to the case capacity becoming filled, and the powder geometry/combustion/and velocity all becoming more consistent from shot-to-shot as a result."

I concur. Lil'Gun burns best at max charges. In the smaller capacity you are dealing with you may have hit case capacity before reaching the optimum pressure. Your charts/data is showing the groups tightening up as the pressure increases. What were the actual SD and ES for the progression of those loads? Switching to 2400, 4227, 4759, 4198 or 5744 might well bring better accuracy at a lower velocity or perhaps even as high as 3200 fps.

It is unfortunate also that you are restricted to 100 yards. The non linear dispersion difference between 25, 50 and 100 yards is not near as telling as between 100 and 200 yards.

Keep in mind that while a 5 shot group is "standard" fare these days it still is not really statistcally valid. A minimal sample of 8 is with 10 being preferred to get a 90% + assurance the data is valid.

Please do not misunderstand me, I am not being critical here at all. I am trying to help. A bump of the RPM threshold up to 165,000 is not that difficult if we find 3 moa accuracy to be acceptable. As I noted in Bullshop's thread, with such a rifle for varmint shooting 3 moa is not acceptable to me. It is however quite interesting and by no means and easy feat at 3200 fps with no leading.

Larry Gibson

[Bullshop]

Caleb that is a most excellent write up. Thank you for all the time and effort that you put into it.
Some of the other graph stuff is way over my head and I dont really even comprehend it.
I am pretty much just a dumb dumb trying to have fun shooting.
I am posting a picture of a couple targets I shot with the second batch of bullets I cast with the 107 mold and the 6/1 ww/monotype alloy and quenched. This is my favorite high velocity alloy.
The two targets at right are from the 221 fireball at 3650 fps from the 1/14" twist barrel. They should be somewhere near 196'000 rpm.
That is the best accuracy I have been able to get so far at this velocity.
The problem is I cant get it consistently and there are still the occasional wild flyer that goes 5" to 6" out of the group.
The target at left is very interesting also. I wanted to see if I could get any kind of accuracy from a much longer boolit in the same rpm range.
That target was fired using the NOE 72 grain rn boolit from a 22 BR with a 1/9" twist. The velocity was 2460 fps so should have an rpm of about 196'800 so is exceeding the short boolit by a little bit.
If you look close you can see the order that the shots were fired in. The #5 shot went out about 2" to the right so I fired a sixth shot that went right back into the group. I cant say why #5 went out but figure it was a bad boolit. I dont weigh or sort them other than to look for defects when size/lubing them.
Anyway there is a new batch of boolits from the 6/1 alloy on the way to you. It did seem to improve my average with the 221 at least a little.
OH BTW yesterday I had to dispatch a sick chicken. Since it was in a safe position about 40 yards from my shop I thought I would use the fireball with the 107/3650 fps load. It did just exactly what you would think it would do, it blew chicken all over the place. A most excellent load for ridding the garden of pests!
Oh and I forgot to mention that these are 100 yard targets and the coin on the plate is a nickle.

[DrB]

Once again, let me remind you and all others who think the RPM threshold is a "limit", it is not a limit. It is a threshold that may be moved up or down depending on the load and componants used. The RPM threshold generally runs in the 120 - 140,000 RPM range. You are indeed pushing it upward as I have and others have. Not a limit, ok?

Larry, I think maybe I got in on this conversation late, because I really doubt I understand what your RPM threshold theory is. It seems to me your theory either makes a testable prediction or it does not?

If your theory just boils down to "you can spin a bullet too fast" I don't see how that is very interesting. Yes, it is obviously true (all solids have ultimate strengths), but if you aren't going to say something empirically or analytically about when it becomes true, or why, how is it useful to anyone? I'm not trying to question your observations here in any way, I'm just wondering about how you have chosen to interpret them?

As I thought I understood it, you were making two testable assertions:

1. accuracy would go to pot beyond 120k-140k rpm for a 22 caliber, and
2. the mechanism causing accuracy to go to pot was (some unspecified mechanism) occuring in freeflight (external ballistics) and would result in a non-constant angular dispersion (nonlinear increase in group size).

Accuracy improving with increasing RPM, and constant angular dispersion over range are both observations in my results that are not supportive of your predictions as I understood them.

1. This load is above your previously stated "rpm threshold" of 140k rpm
2. it exhibits no measured increase in MOA as a function of range
3. MOA in fact is decreasing with increasing RPM.

You also have asserted that the rpm threshold would be lower for smaller caliber projectiles, as the force due to a given rpm would be greater, and I demonstrated to you that this was the opposite of what was physically true -- that for a given rpm, acceleration, force, and stress would all go up with increasing caliber. A table, a plot -- something illustrating the proposed variation of your "threshold" as a function of caliber, etc., would be helpful.

What were the actual SD and ES for the progression of those loads? Switching to 2400, 4227, 4759, 4198 or 5744 might well bring better accuracy at a lower velocity or perhaps even as high as 3200 fps.

SD (Standard Deviation) was noted in the previously posted chart on the right. I shouldn't have used the word "Variance" in the independent axis label, as I should have said variability. Typo.

RE alternative powders, perhaps, but as far as I've seen to date, folks don't report as high velocities with powders other than lil'gun in the hornet. At maximum loads, they also typically report pressure signs -- which I have yet to encounter even using pistol primers at these maximum velocity loads. Maybe someone has relevant experience in the hornet they could share that would suggest otherwise. I'd be interested to hear loads.

It is unfortunate also that you are restricted to 100 yards. The non linear dispersion difference between 25, 50 and 100 yards is not near as telling as between 100 and 200 yards.

OK, but Larry -- there is no observed dispersion in the data. I am open to taking further data at 25 and 100 yards, but when do you propose it would be enough? Also, if I kept going out to greater and greater ranges, how would you suggest I distinguish between the effects of air convection/turbulence and this bullet spin rate/range dependent phenomenon of yours? To discount the effects of near ground air speed over the course of the bullet from shot to shot, the effect would have to be fairly dramatic, and the wind obviously pretty calm, right? Well, nothing dramatic was apparent here out to 100 yards.... just saying.

Perhaps you can offer some plots from your load experiments of constant shot number group sizes as a function of range? The shape of those might help us here with discussing what this effect of yours should look like in the present data, if actually present.

Keep in mind that while a 5 shot group is "standard" fare these days it still is not really statistcally valid. A minimal sample of 8 is with 10 being preferred to get a 90% + assurance the data is valid.

Larry, your statement is incorrect (and vague). The statistical adequacy of the population sample size is dependent upon both the nature of the population(s) and the statistical inference that is being made.

For example, here's an example of how your statement is untrue. If I wish to make the statistical inference that one population's mean was greater than another's, and the real mean value of one population was a million with a standard deviation of one, and the real mean value of a second population was one, with a standard deviation of one, and I took a sample size of three from each population, computed the mean of each, etc., are you really arguing that if I repeated the statistical test using samples of three from each population a large number of times, I wouldn't come to the correct decision more than 90% of the time (or less than 99.9999...%)? Or that the computed theoretical statistical confidence from a single test of a sample of three from each wouldn't be WAY in excess of 90% for just three samples each (I could run that number for you, but I think the answer is intuitively obvious as I've laid it out)?

There is not a fixed "right" sample size for a statistical inference. The necessary size depends on the inference and the populations being sampled... it also depends on the model.

Please do not misunderstand me, I am not being critical here at all. I am trying to help. A bump of the RPM threshold up to 165,000 is not that difficult if we find 3 moa accuracy to be acceptable. As I noted in Bullshop's thread, with such a rifle for varmint shooting 3 moa is not acceptable to me. It is however quite interesting and by no means and easy feat at 3200 fps with no leading.

Thanks Larry -- I do appreciate your feedback, and love your wealth of observations and personal experience. The range report of yours I've read so far was very well written.

I really think a compilation/plot of your data that you used to deduce your rpm theory would be helpful if you ever have the opportunity and inclination to put one together.

Best regards,
DrB

[DrB]

Caleb that is a most excellent write up. Thank you for all the time and effort that you put into it.
Some of the other graph stuff is way over my head and I dont really even comprehend it.
I am pretty much just a dumb dumb trying to have fun shooting.
I am posting a picture of a couple targets I shot with the second batch of bullets I cast with the 107 mold and the 6/1 ww/monotype alloy and quenched. This is my favorite high velocity alloy.
The two targets at right are from the 221 fireball at 3650 fps from the 1/14" twist barrel. They should be somewhere near 196'000 rpm.
That is the best accuracy I have been able to get so far at this velocity.
The problem is I cant get it consistently and there are still the occasional wild flyer that goes 5" to 6" out of the group.
The target at left is very interesting also. I wanted to see if I could get any kind of accuracy from a much longer boolit in the same rpm range.
That target was fired using the NOE 72 grain rn boolit from a 22 BR with a 1/9" twist. The velocity was 2460 fps so should have an rpm of about 196'800 so is exceeding the short boolit by a little bit.
If you look close you can see the order that the shots were fired in. The #5 shot went out about 2" to the right so I fired a sixth shot that went right back into the group. I cant say why #5 went out but figure it was a bad boolit. I dont weigh or sort them other than to look for defects when size/lubing them.
Anyway there is a new batch of boolits from the 6/1 alloy on the way to you. It did seem to improve my average with the 221 at least a little.
OH BTW yesterday I had to dispatch a sick chicken. Since it was in a safe position about 40 yards from my shop I thought I would use the fireball with the 107/3650 fps load. It did just exactly what you would think it would do, it blew chicken all over the place. A most excellent load for ridding the garden of pests!
Oh and I forgot to mention that these are 100 yard targets and the coin on the plate is a nickle.

Dan, neat results!

I look forward to playing with the new batch of bullets.

Lay off putting yourself down... (unless all this sandbagging is how you win at poker ). I was surprised when you called the reduction in velocity I saw going from jacketed to cast and I saw the numbers I was getting. I haven't gone back to my jacketed data to calculate it, but your prediction of about 200 fps reduction with the change to cast is pretty darn close. (btw, this reduction also makes me wonder if perhaps a source of the velocity variability is the initial confinement of the charge. Seems like the in-bore resistance should be less with cast, and yet the velocity is lower with slightly more powder -- maybe the powder isn't getting ignited as completely? A hotter primer may be in order).

Anyway, so far, the products of yours I've tried (bullplate lube, your NOE 107 cast bullets, and your lotak) have each done exactly what they were billed to do. From everything I've read about cast bullets over 1800-2400 fps and fouling, I really expected at 3200 fps there to be some major trial and tribulation, tears, yelling, and gnashing of teeth with fouling, but there was no grief whatsoever with any of the load variants I tried. Slick as a whistle for eighty shots, as far as I can tell, with just a cursory dry scrubbing at 20 rounds fired. There are plenty of overeducated folks out here (I definitely include myself ) who don't have as good a track record as I've seen from you so far.

I've got a NOE 107 load coming from Al. I'm interested to compare the grouping of my few remaining first BS run bullets, the new ones you've sent, and a run from the new mold. That mold half offset on the first batch makes me wonder how much of the group size might be due to bullet asymmetry, and how much the group size may change with the new run? Have you ever tried clocking your bullets so the same mold half was up to see if it changed your grouping?

Also, regarding terminal effect, I haven't tested it yet. With my 35 grain V-maxes at ~3400 fps, they make a neat round hole in the front of a full milk jug, shoot the top about 15 yards into the air, burst the jug about flat, but make not a mark at all on the back face of the jug. I suspect these 37 grain NOEs at 23 BHN and 3200 fps will penetrate to make a hole on the back side, but with similar hydrostatic effect. What do you think? You've already shot some jugs, haven't you?

Best regards,
DrB

[Bret4207]

Yes... Aside from the crimp groove it does, doesn't it? But it is 112 gr vs 37gr... do you have any suggestions for similar larger and smaller caliber bullet designs?

Found a picture of the 311316 below.

http://www.three-peaks.net/bullet_molds.htm

Nope, but maybe some bright engineer type could figure out the diameter/length ratio or whatever the number is called and add in the CG and ballistic coefficient and a mess of other numbers to try and come up with a number that might lead to something. Darn sure I can't do it.

I will note that IME the blunt RN or FN Loverin designs were always easy to get shooting at high speeds. Don't know if that helps or not.

[DrB]

Nope, but maybe some bright engineer type could figure out the diameter/length ratio or whatever the number is called and add in the CG and ballistic coefficient and a mess of other numbers to try and come up with a number that might lead to something. Darn sure I can't do it.

I will note that IME the blunt RN or FN Loverin designs were always easy to get shooting at high speeds. Don't know if that helps or not.

L/D would be fineness ratio, I believe.

To come up with a similarity parameter it would be nice to know the shape of what works and doesn't... Seems to me there're so many constraints at play, it won't be just one.

[Bullshop]

We have shot some quart bottles and yes there is always an exit hole in the back but you have to be good with puzzles to find it. The water show is very impressive.
I have not yet tried to determine the percent of meplate to caliber of the little 107 but it will be high as the meplate is the largest of all my 22 cal molds.
I feel I am getting good enough results to call my 221 a good garden gun. For shots not exceeding 100 yards a chuck even if hit on the fringes will be messes up badly. I doubt there will be any cripples to crawl off and suffer.

[Larry Gibson]

Larry, I think maybe I got in on this conversation late, because I really doubt I understand what your RPM threshold theory is. It seems to me your theory either makes a testable prediction or it does not?

If your theory just boils down to "you can spin a bullet too fast" I don't see how that is very interesting. Yes, it is obviously true (all solids have ultimate strengths), but if you aren't going to say something empirically or analytically about when it becomes true, or why, how is it useful to anyone? I'm not trying to question your observations here in any way, I'm just wondering about how you have chosen to interpret them?

DrB

I’m not exactly sure what many of your points are here. It is obvious you do not understand the RPM threshold or how it works. I have explained this many times on this forum. Also I suggest you study ballistics a little more, seems some confusion there also, especially in exterior ballistics. No criticism there, just pointing out that some of your arguments are not correct in that respect.

As I thought I understood it, you were making two testable assertions:
1. accuracy would go to pot beyond 120k-140k rpm for a 22 caliber, and
2. the mechanism causing accuracy to go to pot was (some unspecified mechanism) occuring in freeflight (external ballistics) and would result in a non-constant angular dispersion (nonlinear increase in group size).

You are making erroneous assumptions of ascertations you acredit to me. In our previous conversations discussing this I said it was more likely for a smaller caliber to be more difficult to shoot accurately at high velocity/RPM. I did not say “accuracy would go to pot” as a certainty. I have always said (as I said in this thread) that the RPM threshold could be raised or lowered with any caliber, including the .224.

The “mechanism” is not “unspecified” as I have explained it before. For your edification I will quote it again;

“the bullet is unbalanced or becomes unbalance due to obturation in the bore during acceleration. The unbalanced bullet is forced to conform while in the barrel and its center of mass is revolving around it's geometric center. When the bullet is free of the barrel's constraint, it will move in the direction that its mass center had at the point of release. After exiting the muzzle, the geometric center will begin to revolve about the center of mass and it will depart at an angle to the bore (line of departure). At 54,000 RPM to 250,000 RPM, depending on velocity and twist, the centrifugal force can be tremendous. It will result in an outward or radial acceleration from the intended flight path (line of departure) and will try to get the bullet to rotate in a constantly growing helix.”

Accuracy improving with increasing RPM, and constant angular dispersion over range are both observations in my results that are not supportive of your predictions as I understood them.
1. This load is above your previously stated "rpm threshold" of 140k rpm
2. it exhibits no measured increase in MOA as a function of range
3. MOA in fact is decreasing with increasing RPM.
They were not supportive of your own misunderstanding of the RPM threshold. If the bullet is not exceeding the RPM threshold it does not apply. Accuracy improving as pressure builds so the powder burns consistently and efficiently is a common and well known fact of reloading when working up loads. Most often there is a direct correlation, given a proper and well made bullet, between improved SD/ES and accuracy. The adverse affects of the exceeding the RPM threshold when the bullet begins a helical spiral in flight occur after the RPM threshold is exceeded, not before.


You also have asserted that the rpm threshold would be lower for smaller caliber projectiles, as the force due to a given rpm would be greater, and I demonstrated to you that this was the opposite of what was physically true -- that for a given rpm, acceleration, force, and stress would all go up with increasing caliber.

That is correct and I agreed with you but you are misunderstanding and again acrediting me with something not quite what I said. If the defect is the same in relation to the mass of the bullet then what you say is indeed correct. If, however, the defect is simply the same (say a .3 gr defect) then the ratio of the defect to the bullets mass gets much larger in a smaller caliber bullet. The adverse affect of that larger ratio makes the smaller caliber more difficult to get past the RPM threshold. If we are weighing bullets and we use a +/- .2 gr that +/- .2 gr is a much larger variance to a 35 or 55 gr .225 cast bullet than it is to a 180 – 200 gr .30 cal cast bullet or a 500 gr .45 cal cast bullet. A better understanding of ballistics will also demonstrate that where that imbalance is located (in front of the center of pressure, center of gravity, in line with the center of form, for & aft center and the center of spin) can and does have an adverse affect on nutations and yawing of the bullet in flight. Those nutations and yawing also cause inaccuracies which are why we shoot groups instead of all bullets following exactly the same flight path. Nutations and yawing are present before during and after the RPM threshold.

If there was a previous misunderstanding of what I said I hope this clears it up.

A table, a plot -- something illustrating the proposed variation of your "threshold" as a function of caliber, etc., would be helpful.
Quote:
What were the actual SD and ES for the progression of those loads? Switching to 2400, 4227, 4759, 4198 or 5744 might well bring better accuracy at a lower velocity or perhaps even as high as 3200 fps.
SD (Standard Deviation) was noted in the previously posted chart on the right. I shouldn't have used the word "Variance" in the independent axis label, as I should have said variability. Typo.

In my case the tables/plots you’ve posted are hard to read and I can not enlarge them. I understand you like “plots” but a simple list of the actual SD/ES figures with the load in grains would be easy to understand. I have posted similar “plots” in my original RPM threshold thread and found others could not read them well enough to understand having the similar problems with enlarging that I have. Thus i stick with simply posting the actual numbers and making comments. Seems more here understand that way.

RE alternative powders, perhaps, but as far as I've seen to date, folks don't report as high velocities with powders other than lil'gun in the hornet. At maximum loads, they also typically report pressure signs -- which I have yet to encounter even using pistol primers at these maximum velocity loads. Maybe someone has relevant experience in the hornet they could share that would suggest otherwise. I'd be interested to hear loads.

You’re probably aware that I use a M43 Oehler PBL to measure pressures with. I have measured the pressures of numerous .22 Hornet loads using a 21” test barrel with a 12” twist. 12.5 gr of Lil’gun under a 45 gr jacketed bullet produces less psi than factory Remington/Winchester ammunition. It also seems to take at least a 40 gr jacketed bullet to make it burn efficiently. That's why I asked for the actual SD/ES fps data in lieu of a "plot". Loads with H110, 2400 and 4227 will get you into the 2800 – 3000 fps range with 40 gr jacketed bullets at less than CIP MAPs (SAAMI does not list a piezo psi MAP for the .22 Hornet). Using those powders with a lighter weight cast bullet will produce even less psi. Hey, if you don’t want to test with another powder at a lower velocity to find out then that’s ok with me.

Quote:
It is unfortunate also that you are restricted to 100 yards. The non linear dispersion difference between 25, 50 and 100 yards is not near as telling as between 100 and 200 yards.
OK, but Larry -- there is no observed dispersion in the data. I am open to taking further data at 25 and 100 yards, but when do you propose it would be enough? Also, if I kept going out to greater and greater ranges, how would you suggest I distinguish between the effects of air convection/turbulence and this bullet spin rate/range dependent phenomenon of yours? To discount the effects of near ground air speed over the course of the bullet from shot to shot, the effect would have to be fairly dramatic, and the wind obviously pretty calm, right? Well, nothing dramatic was apparent here out to 100 yards.... just saying.

The farther past the RPM threshold the greater the adverse affect of the helical spiral will be. If a load is on the ragged edge of the RPM threshold at ranges of 25 -100 yards it may not be that noticeable. As range increases the helical spiral increases non linearly. Thus there will be a much greater disparity on non linear group size between 100 and 200 yards than there is at ranges shorter than 100 yards.

For example; I ran a test with 311291 in a 10” twist .308W test barrel. I tested group size at 50, 100 and 200 yard. I shot 10 shot groups at each range with a standard load that was below the RPM threshold at 136,400 RPM. I then pushed that bullet above it’s RPM threshold (note I said “its RPM threshold” not THE RPM threshold) at 179,400 RPM. The standard load’s groups at 50, 100 & 200 yards were .7”, 1.3” and 2.5” respectively. We see there the groups expanded in a pretty straight forward linear manner. The 50, 100 and 200 yard groups with the load that exceed that set of components RPM threshold were 2.55”, 4.75” and 14.5” respectively. We see here while the groups were larger at 50 and 100 yards there really isn’t a non linear dispersion. However the very much non linear 14.5” group at 200 yards tells the tale. There was sufficient range for the helical spiral to become very apparent. In the case of your and Bullshop’s tests at 100 yards the “flyers” may very well be those that are over the RPM threshold and the helical spiral is more pronounced, even at 100 yards.
No need to “plot” as the group sizes tell the story sufficiently.

Quote:
Keep in mind that while a 5 shot group is "standard" fare these days it still is not really statistcally valid. A minimal sample of 8 is with 10 being preferred to get a 90% + assurance the data is valid.
Larry, your statement is incorrect (and vague). The statistical adequacy of the population sample size is dependent upon both the nature of the population(s) and the statistical inference that is being made.

Sorry but my statement is not “incorrect” if we are talking ballistics here, not some abstract thought. It is well known among ballisticians that a 5 shot sample is not a sufficient sample. Let me simply quote from the Speer#14 Manual;

“….load seven to ten rounds…….A single five shot group is seldom adequate to give you an accurate picture of the quality of your ammunition. Even though a load produces a five shot group of 0.75 inches, you don’t know if the next group will be 0.55 inches or 1.75 inches. For an industrial level of statistical confidence, four groups of five shots each is considered minimum. ……..A statistical analysis performed by the CCI-Speer Quality Assurance Section, showed that seven shot groups gave the highest degree of statistical confidence with the fewest shots.”

Note also that SAAMI and CIP use a minimum of ten shot groups/test strings.
Now I could be wrong here but I don’t think the entire ballistician industry is wrong.

Quote:
Please do not misunderstand me, I am not being critical here at all. I am trying to help. A bump of the RPM threshold up to 165,000 is not that difficult if we find 3 moa accuracy to be acceptable. As I noted in Bullshop's thread, with such a rifle for varmint shooting 3 moa is not acceptable to me. It is however quite interesting and by no means and easy feat at 3200 fps with no leading.
Thanks Larry -- I do appreciate your feedback, and love your wealth of observations and personal experience. The range report of yours I've read so far was very well written.

I really think a compilation/plot of your data that you used to deduce your rpm theory would be helpful if you ever have the opportunity and inclination to put one together.

If you search for my original RPM thread you will find the charts/plots.

Interesting discussion. Just remember that there is a lot to ballistics that affects the flight of the bullet. Many of these may or may not be apparent at 100 yards or less. And.....the RPM threshold is not a "limit", just a reminder.

Larry Gibson

[DrB]

Larry, frankly, you have just begun to irritate me. By all means share your observations, share your data, even share (but don't push) your pet theories. Please do suggest any means of proving or disproving your theories that would actually be experimentally feasible (and the results of which you would consider if they didn't go your way).

But please, dear lord, stop trying to explain the meaning of the word "threshold" in the english language to me. I have several dictionaries, and apparently, I am more acquainted with how to use one than you are. If threshold meant in the english language what you have abused it to mean, then if I were visiting your home I would be upon the threshold to your living room from the time I was in the middle of your front yard until I walked out your back door. 120,000 rpm? No. 140,000 rpm? No. 165,000 rpm? No. 197,000 rpm? No. For goodness sake, the idea of a threshold is that there is some threshold value at which something changes when you pass it! Give the RPM theory a rest in the context of the present discussion in the face of contradictory physical evidence. If nothing else, park the RPM theory and share a map/spreadsheet based on real physical data. We could all add data points to it and together begin to get a more complete picture.

Also, please do not tell me I am misunderstanding what you said RE acceleration vs. caliber when I have your IM in front of me. Your understanding may have evolved, and you may have new theories, but I can read quite clearly what you typed:

(DrB)"i would expect a larger caliber to have a lower rpm limit."

(Larry)Actually I'm finding just the opposite. Reason being on smaller calibers given the size or amount of defect it represents a larger percentage of the bullets weight/mass. Also the farther from the center of spin a defect is the less effect as the rotational velocity is less, hence less centrifugal force.

While you have just represented that you told me you agreed with my assertion, the fact is that after this IM from you I responded by IM with the equation for rotational velocity as a function of radius, and centripetal acceleration as a function of that, and you did not acknowledge it -- you just quietly went away, and I have not received an IM from you since. No biggie -- just please don't always ascribe my disagreement with you to my "misunderstanding." If I say that, it's often because I am being polite. When you say it, particularly when you misrepresent the content of a discussion, it is easy to interpret as you being disingenuous and condescending.

As far as the state of the evidence goes: Dan posted on remarkably high 3600+ fps velocities with cast bullets, no fouling, and so-so accuracy and your reaction was "slow-down, slow-down it's RPM." I was able to replicate Dan's lack of fouling and essentially the same accuracy with the same bullets and lube at lower RPM and velocity, and your reaction was "slow-down, slow-down it's RPM." Dan was able to get one hole accuracy at WAY HIGHER RPM in the same caliber with a different bullet, and you take no notice of this remarkable (but inconvenient to your theory) result. So far the results here span 145,000 to 196,000 rpm, and 2400 to 3600+ fps with a negative correlation between RPM and MOA, and you still say it's the RPM. Well, if our so-so accuracy is the result of your RPM theory (in spite of the mold-half offset I measured, or the mismatch in gascheck/bullet shank diameters, or the negative correlation in the data (MOA is decreasing with increasing RPM)), and as you say, you can observe this nonlinear group expansion effect easily only after 100 yards, then why in heck are you so stuck on our constant MOA accuracy at less than 100 yards being principally attributable to your pet theory?!! How is there anything in the data we have posted that is primarily consistent with your theory as distinct from the umpteen different suspect inaccuracy causes that occur for loads in general or boolits in particular?

Another thing, please do not presume to know my resume regarding ballistics. I'd hazard to guess there's only one of us here (between the two of us -- I make no assumptions about the membership at large!) who's written Runge-Kutta ballistics integrators, performed CFD on a bullet design, written a navier-stokes flow solver, performed thermochemical equilibrium calculations, knows what an "ensemble statistic" is as it relates to calculating high temperature properties of a gas, has solved for the principle rotational axes and moments of inertia of a body, performed statistical experimental design, performed multivariate regression, etc., etc., etc. If I politely say I do not understand your theory, you might want to consider I'm saying it because your theory as explained doesn't make physical sense. I will be polite and say this even if I think your theory makes no physical sense because I am humble, conversational language is imprecise and so the theory may be difficult to exactly convey, and even if the theory isn't all right there may be a good idea or two within (and because I would prefer to be polite).

I am not, by a long long long shot, the smartest guy Larry. There is an unbelievably huge universe of things I know nothing about, and a small one I have some passing familiarity with. I am always eager for new data and observations as experiments are costly and all of our observations together can facilitate a far better understanding than our limited efforts alone. I strongly believe in profiting from the experience of others, and I strongly believe in the scientific method.

Now, if you would like to continue to politely expand upon your theory in a gentlemanly way, I'd love to talk to you further about it. Depending upon what you mean by "a helical path" I think we can actually physically prove or disprove that part of your theory with one of your load/gun combinations that you are confident exhibit this behavior, using an inexpensive and simple experiment (on your time and dime). Please describe your theory, exactly, as to what is moving in a helical path, roughly what the diameter of this path corresponds to, and what the forward progress per revolution around the path is. We can experimentally take the lateral deviation due to wind and vertical stringing due to muzzle velocity variation right out of the equation. I'd also love to discuss the internal pressure data you have with lil'gun and what primers you may have data with in the hornet.

Just please be so gentle as to leave anything that sounds like condescension out of our conversation, and discontinue the "appeal to authority" arguments... I tend to overreact (badly) to those.

Best regards,
DrB

[Bullshop]

I do hope this works out well! It seems we may be on a precipice of new understanding of the limits of boolits and we can now either fall or climb. I hope we climb!

[DrB]

Bullshop, the two biggest questions running around in my mind are how much we can improve accuracy, and what to attribute this non fouling performance to?

What is different that facilitates a ~1000 fps increase? Is it caliber, lube, bullet geometry, gradual acceleration with a reasonable twist, etc.? What combination of the above? You've already tried a different bullet design in 22, have you data yet in 30 cal? It might be interesting to try the lyman 311316 bret suggested, especially in a slower twist barrel...

Thanks again for posting your results and providing the materials for me try to replicate them.

[leftiye]

Whatever else is said here, the velocities of both Bullshop's and DrB's results have waaay beyond doubt put us in to another level of possibilities concerning high velocity with accuracy still being possible. As for RPM theory, these results stand so prominently as having "raised the threshold" as to obviate argument. Along with velocity, rpms have gone sky high, and what has been said before must be evaluated in that light. "Resistance is futile".