Positive Compensation

[DaveMc]

David, Don't get me wrong, I am not labelling you with anything and we just need to nut out the truth. Between what you have done and what we have there is a huge amount of information.

A few things to clarify

1) I believe positive compensation exists.
2) We all have a great deal of respect for your work and knowledge.
3) The original poster is trying to get to the bottom of what is required at various ranges - and I think this needs a balanced discussion with scientific backing on any part - This thread is about positive compensation and is a discussion. I would like to present my views even if they differ from yours - no one should take offence at that.

Just because I challenge the results does not mean I don't have a lot of respect for what you have done. I just maintain that it is very hard to achieve enough compensation at or beyond 1000 yards to truly make a big inroad into the overpowering effect of velocity variation. My observations in the equipment we have used it is a ratio of about 1:10 at 1000. This is similar to what Peters graphs and figures show above and so totally agrees. If you have some plots from different equipment that shows another order of magnitude then please show us as it would be worth discussing. I had some photos of very flat groups of 15 shots at 300 yards over 80-100 fps. and similarly 110m groups with dips in the order of 0.25 minutes over the same velocity range. If you look at Peters graphs above this fits closely with his projections of compensation required at these ranges. At 1000 yards we need 2 minutes for 80 fps. I still have not seen that sort of level of compensation in any data but would welcome it.

So I keep saying and emphasizing it is MORE important to tune for small groups and low velocity spreads than compensation at the long ranges (especially with standard equipment). It does not mean I don't believe we are not capable of picking areas where they both can help. I think it is vitally important we nut this out with the above data and get readers on the right understanding and there is no need for anyone to take offence.

The challenge I put out a few years ago wasn't directed at anyone and I dearly want to see some evidence of significant long range compensation and how it can be achieved. The work you have done would theoretically put us closer than anything. We need to create a significant upward thrust of the barrel at a short enough wavelength but high enough amplitude to overcome a drop rate of 1 minute every 40 odd fps (faster in 308 heavies - but even a rate of 1/2 this would be fantastic). I can see this could be investigated with weights under cog etc and you have done more than anyone on this.

BUT my firm belief is even if we could achieve this level of movement (And I believe it could be achieved with some engineering, barrel profiles and weight placement) -
it will come at a cost in two forms - 1) what goes up must come down and imagine wandering out of the positive area when you travelled or your barrel sped up and then doubling the effect of velocity spread. 2) Alternatively if you can achieve this level of compensation then your short range groups would have the converse effect - ie at 300 yards you would have to have a significantly lower poi for a high speed shot (even 20fps would send you out of the 6).

So - either way if you achieved long or short range compensation you still need a very tight velocity spread to have an accurate rifle from 300 through to 1000 and beyond. Of course if you want to use multiple barrels to achieve same thing then go for your life - I see nothing wrong with that either.

[jasmay]

So, correct me if I am wrong Dave's, but from following a lot of these threads we could put a few key factors together that we are looking for.

1) Positive Compensation, in particular when tuned for a given range in this case 1000yds, will help with group size.

2) Minimizing trajectory height i.e. flatter trajectory will help minimize POI deviation with slightly higher velocity spreads, not the most scientific method, but a key player in the hunt for flat plots at 1000?

3) Tight velocity spreads, low SD & ES

If we focused on these 3 principles, not in the order laid out we should be on the path to a good setup.

Question: if we follow the above principles, with a tuner added to the mix and in particular focused on tight velocity we should be able to shape the group for best 1000yd outcome for any given barrel, is this a good starting point for development?

I wont lie, I think most of us have our own method of madness in achieving the success we have, but a formula to follow would be very handy to the many new comers wondering where the heck to start ( I put myself in this class still ).

[williada]

Dave, I have had a second heart op this year and a few problems since. Like your mate Pete S, I am tired so my contributions have dropped off anyway. I don't know whether the brain tumor was worse. Our compensation stuff with one rifle across the course has the same outcome. I couldn't reply in detail because it would mean breaking the commitment and confidentiality of champions seeking to compete in individual events and other team disciplines at the international level in the near future who like to think they have an edge and whose approval I have to get because I have learned and they have given me a lot too. Its a wonder someone did not tap you on the shoulder in private it was not about the challenge because I like to share. So my goal before I fade out has been to stimulate the thinking of others and let them take the path of self discovery with the right tools. That was the whole purpose of a train the trainer program we started. I can no longer do it and just want to enjoy my club shooting without hassel these days. It has been so time consuming responding to requests from America, Britain, New Zealand and from every state in Australia as a result of posting on this great forum. You guys paddle on and learn something for yourselves.

It really hurt that I had to pull out of coaching in the Vic champs at the last moment. But sometimes you have to lie down to get up again and put priorities in perspective.

One last tip. Everything has cause and effect. You need a separate rifle for 1000 yards where you can see through the optical centre of your scope. This effects centre of gravity in the achievement of the set up and as !Peter! has recognised and I spoke about on the post about rifle design and recoil, will effect barrel whip particularly with skinny long barrel and weights at the muzzle. A central reason why I think one rifle across the course is at a disadvantage.

[!Peter!]

1) what goes up must come down and imagine wandering out of the positive area when you travelled or your barrel sped up and then doubling the effect of velocity spread.

DaveMc, I really don't see how this can be a concern as things would have to wander a long way.

Williada, Thank you for the information you've shared in this post and elsewhere.

[DaveMc]

David - yes we all thank you for your contribution and it has stimulated a lot of activity. I also do not have much time to contribute but will push ahead a little here and try and present a few of our findings. I wish you the best health wise and thank the marvels of modern medicine for keeping you and Peter both up and going and contributing to the shooting world. We are all richer for it.

I will use some of my rifles so as not to upset anyone. Feel free to comment. (Peter - hopefully I can present this in a way that will help - and clarify my above comments - and see where this can start to work against you - over time I have come to the conclusion that we are better off designing stocks, rifles and barrels that reduce vibration effects at time of muzzle exit rather than increase - wide forgiving nodes are paramount to consistent success. That does not mean there is no compensation in them - there generally is - just a small amount rather than trying to go for massive effects to compensate at 1000). Muzzle weights generally do this and act as a dampener to slow the upward thrust required for compensation.

Lets start with one of my old rifles - then we can't break any ones confidentiality - this data was presented in 2012 in a post by Tony Berry. viewtopic.php?f=5&t=4318&hilit
I have reinstated my account with imageshack so see if the pics return http://img28.imageshack.us/img28/4999/284sh.jpg.

be265524e5d21342c79b527da6b89c4e.jpg

All these modified ladders were generally (not always though) shot in round robin fashion in very good conditions and I can't remember how this one was done.

I will post the spreadsheet data with velocities shortly when I get time and figure out how to do it. https://docs.google.com/open?id=0B6Fwvt ... 1JONzgwLTA

I will also attempt over time to model these particular shots into 100 and out to 1000 yards to fit in with Peters graphs and the real world observations. I also have pletnty of 1000 yard data for this rifle somewhere to show the models meet the real world.

This rifle was a very accurate 284Shehane that served me well for a few years and won a couple of Queens and several leadups - It was Set up to shoot from 300-1000 and with a 20moa rail and March scope was 15moa below centre at 300 and 6 above at 1000 - still well in the middle of the scope and with plenty of adjustment to go. Travelling around on planes to attend Queens is hard enough with one rifle and a heavy rest rather less 2 or more but I accept others may wish to set up a few rifles to approach different distances. I am a one gun (or at least one caliber with a few actions and barrels) man.

Anyway - This rifle consistently performed will between 2800 and 2850 fps. I was confident enough to fly in and shoot and it always shot well straight off the bat at the leadups due to a wide and forgiving node. It also demonstrated a consistent SD in the 6fps range with the equipment we had back then. This particular ladder test was one of many (I put over 10,000 rounds down range through the chronographs over those 2 years and mapped every shot) 9 shots fell within 0.25 minutes over 90+ fps range. Whilst this is an extremely small sample size and could be subject to sampling errors this pattern was repeated many times for this rifle and similar patterns observed in quite a few rifles since. It is clear to see why the 2800-2850 area has become so popular with 284 shooters and the like. Most using heavy straight or HV tapers or the like. All my rifles showed a similar pattern even the HV profiles with the heavy weights on muzzle.

David I don't post this to challenge you but respect your opinion and ask for your input. I think we are 99% on the same page I haven't seen anything you have really presented that is contradictory to this - ... and by the end of this may find total agreement? who knows.

[!Peter!]

DaveMc,
Thanks for the spreadsheet. I like the approach using a trend line.

!Peter!

[DaveMc]

Hi Peter,
I had to do a bit of work to validate the trendline approach. Basically I did a montecarlo simulation again with thousands of randomly generated numbers and then sampled them in 3,5,10,15 and 30 shot groups. The montecarlo simulation proved the average of 10*3, 6*5, 3* 10 is for all intensive purposes as statistically significant as a single 30 shot group. By inference this also means fitted lines can be used with some confidence as predictors - providing theere is no significant bias at either end of the sample.

Now to the next phase - I modelled up the predicted impact point of each shot. This can be done using any number of ballistic programs to gain a similar result and Peter Smith first did this for me back in 2012 with similar data before we went out and plotted the results at 1000 yards.
https://photos.google.com/photo/AF1QipNPWg7JGchMN6irpmctLxoF3kRTPkZjPEWVX8x1

My first comment would be not to take too much notice of the 100 yard graph. As williada correctly pointed out a lot of these models don't take initial launch angle into account. This will have a bigger effect the closer we get to zero. When Peter Smith and I were first discussing this we of course came to the conclusion that we must verify this with actual 1000 yard plots. Along with the aid of Marty and the magnificent 1000 yard Ponderosa Testing laboratory we put down many a 1000 yard ladder and came up with several dozen strings that had very strong correlation of velocity to heigh tas in the predicted graph for 1000 yards. Since that time and even more so recently with the ability to plot Labradar readings against 1000 yard shooters at many ranges there is a huge body of evidence to show the relationship at 1000 is very linear with the type of gear we use.

[DaveMc]

I just removed some old sketches and thought I would replace with one of the latest 25m targets on a new barrel. I am learning to use 25m load development with its hang ups but believe it can be a valuable tool. Sorry the data obscurred the loads and one was cut out accidentally but ladder i (LtoR) 52 (cut) to 57 grains with first one in picture 52.5 and in half grain increments.

SAUMIX2209HYBRID.jpg

This target is done at 25m. And yes I take it on board that the bullets are still suffering from some epicyclic swerve and should stabilise a bit better further out but still we can get an idea.

from 52-57 grains there was no detectable rise or fall in the group. All 33 shots when overlaid were within a half minute circle. The little horizontal mark you see just to the left of the groups is 40mm below the aim point.

This is a standard Barnard in one of our stocks, 30inch x 1.25 straight Bartlein barrel and aluminium bedding block.

[williada]

Dave we are on the same page, but I have moved on to a quicker, simpler and a more effective means of identifying a tune to which only a few have been exposed to in terms of vibration analysis. It is based on understanding the different tunes, moments of inertia and compensation profile for very good reason. I have done more than most on compensation, barrel profiling and internal sizing, chamber design etc for barrels since the 1980's being associated with manufacturing and firing thousands of rounds from a machine rest to take out mirage and light factors and discovered that normal distributions don't fit the model as well as they could at 1000 yards because the relationship of the Magnus force with the gyroscopic twist IMO changes significantly at 1000 yards and can't be seen in a short range evaluation which if we were serious about writing an empirical paper we would take into account because that is better science.

When the wind oscillated above and below 10 mph, strange things happened at 1000 yards. I did the machine rest testing over a two year period and in that time the data was skewed and or suffered from kurtosis. Some of the thinking is behind my nodal pyramid which we have revisited with a Qld boy since. Also the electronic data from top shots we now have at our disposal is telling us something by way of group shape and that the 1/4 minute rifle at short range struggles and that the best shooters are consistently producing groups over 10 inches.

For a single rifle your approach is correct. Given. But I am concerned that at an elite level the 140 yards must be used in preference to a shorter distance unless you have the experience of bigger data sets to know what counts because a good rifle may be discarded on the day if it does not meet criteria when the distance is insufficient to test that properly.

I have always used trends as indicators and that doesn't matter if we use statistical smoothing methods or not. Yes, I have advocated 25 yard testing as you well know. I remember Brad being skeptical saying he would put them all through the one hole. I remember Tony saying the same thing at 140 yards. Ah but we change the loads up. You have to walk people through this for them to understand. That's why I build on concepts and refine detail as personal growth occurs.

I see things as a divergent thinker not a convergent thinker i.e. ask me a question and I immediately think of other questions because of experience that need to be distilled too.

Because I see the compensation thing as an integrated system, naturally I can't dismiss other relationships and how they impact such as bloop tubes, barrel weights, centre of gravity and torque etc. My views on the Purdy system based on pure harmonics are tempered by what the discoverers of the OCW say about OBT and vibration patterns. You have to understand your barrel profile to get the most out of it. Sometimes we can get the best of all things happening in union rather than independently.

But the guts of my work feels the it is the outlier that is a problem and it is always on and many a match is lost today by very tiny margins. So my thinking is focussed on that statistical outlier and reducing barrel wear and tear to find it because I don't think a normal distribution where SD's can only apply is a way of solving it. The answer lies in knowing what caused it and what things do we have up our sleeve to fix it or mitigate it.

We know nodal tunes are strictly velocity dependent, OCW tunes less so and compensation tunes compliment nodal tunes. The SD is determined at your test range but we know atmospheric density can change velocity up, so there goes your tune at times. Sure we choose the best tune area, but we should also choose a better way of finding that which I can not elaborate on publicly for strategic reasons but I have shared and others have tested it independently for great success.

The diagram below, is something that adds to the concept mix.



I simply don't have the time energy, or inclination to go over all the heavy stuff again and again because every time I post I get personal emails. Not that I don't want to help, but it keeps me away from doing things I enjoy that are getting harder to do.

[pjifl]

I always calibrated new powder lots by shooting perhaps 8 shots as 4 pairs with a variety of cartridges. This was done with 308, 6.5 Rogue, 284 and 7 RSAUM.
The load usually was increased in 1 grain intervals.

Then I would graph Velocity against powder. A trend line and or a glance at the graphs immediately gave me a starting load and the slope of the line gives how much V changes per grain of powder. Mostly this was 30 - 40 fls (never over 50) per grain of powder (not for 223) which is a useful figure to know when making loading changes. Saved a lot of powder and barrel life to know this and the graph made for much better accuracy without firing a lot of shots.

I NEVER saw changes as great as in the previous graph (David's last post) where 100 and even 200 f/s changes are happening with powder changes of often less than 1 grain.
And I was intrigued that my graphs of V against Powder were always close to linear. I know 1 grain increments could miss a node but even ladder tests that I did with much smaller powder increments usually showed a close to linear rise in VELOCITY.

I did see far more consistent results when I went over to a Magnetospeed which was very firmly attached to the barrel - far more firmly than the standard MS band.

I just wonder how the Vs were measured and whether there are some V errors involved. I have never been impressed with Optical Chronographs - some of the big name ones included.

In any case, it would be an ideal use for a LabRadar to verify or otherwise if Velocity changes in a similar experiment show similar or different variability.

Peter Smith

[DaveMc]

Thanks for joining back in David. There is a lot of stuff I can agree with there - My PHD was in population modelling so understanding population distributions is my field. I agree a lot of groups and shoots have larger outliers than a normal distribution. The less control we have over the variables (be they loading, components, atmospheric or other) the bigger the outliers will get. BUT just because populations aren't quite normal does not mean they don't behave in a similar way.

[williada]

Dave, I understand perfectly what you are saying and agree but my example is to take people looking at a image diagrammatically in terms of their knowledge base of understanding the wiggles of a graph who have not got an academic background and then transform them at 140 yards with the bigger sample size you mention which is statistically significant at a distance of minimum significance as all the people I have trained can attest to and I have repeated on my posts regularly. I am sick of repeating my self and going around in circles.

It is the same process you are engaging in by way of example now. I get that. But I don't get your need to explore this at the moment on one hand and on the other I do. While I do not have a Phd, my population studies were based on practical application of my undergraduate and post graduate tertiary studies at uni, say in crisis management leading a support team at the coal face in difficult family and individual situations. I understand population studies, psychology and motive and what role that plays in leadership and management too.

My understanding of people's behavior can find its roots coming through the college of hard knocks when as a young man my father said there is no money in farming as we lived in the bush and to which I have returned. Worked in an abattoir, timber mill, drove heavy equipment before during and after university mixing with a wide range of people. I have a real understanding of their sensitivities and exercise respect and tact and try not to talk down to people or use innuendo and selective examples taken out of context. Taught for a while but my best use of my understanding of people, processes and things comes from my initiative and successful negotiation with the Minister of Defence and his team which is on the record having flown to Canberra and followed up with the stakeholders to prevent the closure of rifle ranges in Victoria during the Howard era which set the model for other states. During my working life I have been a member of many high performing teams and in business, have always identified with my roots but chose the appropriate time and place for the exercise of value judgement in the simultaneous capacities I held with different responsibilities.

I understand the psychology of the opposition gaining a mental edge from knowing the plans of others in advance. I know what should be kept behind closed doors. I understand sources of conflict and a clash of value judgement that have to be nipped in the bud by appropriate procedures e.g selection in the case of high performing teams. It is the psychology and harmony that can determine results and explain the difference in performance at the elite level in sport or vocation. My experience in leadership is to inspire confidence and motivation and apply strategy. My experience in management is not to set direction and culture but to get the job done with technical application of knowledge of people, processes and things. There is is a big difference between the two roles. My training in economics, law and mathematics as a young man most of which I have forgotten now merely supplement my practical experience in the real world which seems to stick in the limited brain cells I have left.

The reason why I don't participate in the advanced discussion publicly is for fear of a conflict of interest for those to whom I gave commitment and an undertaking not to and a wider audience some of whom are technically literate. Its hard to pitch things at the right level at times. I like to keep things simple to develop confidence and the idea first and am reluctant post things that are private in nature without the consent of participants from which others can learn. I don't want to be seen as an old fool arguing over things of little relevance when there are more important things to do at my time of life. Hope you understand my thinking.

With regard to chronographs and I don't have a Lab Radar, I have always had to rely on the picture and group shape to tell me more when I drill deeper. Its a bit like a person with disability, they have to work out sometimes arduous alternative methods that work for them. Certainly new technology and tools can be more precise these days. But is all that necessary when you have discovered a new approach by looking at the pictures which many find easy to interpret if they are statistically relevant. Some methods can very accurate but not necessary for the exercise. David.

[DaveMc]

Why I am doing this at the moment is simply I have stayed out of the conversation over the last few years and Peter has posted some interesting graphs and questions at the start of this lead. Some that are worth discussing. "Just what is required to get compensation at different ranges?" and " How could we achieve it?". Our data agrees with the graphs he presented so the discussion needs to go towards the "How much and how do we achieve it."

In approaching this topic I only want to make sure there is a strong scientific basis for the discussion. We are living in a fortunate age with advancement of chronographs over the last few years and a bigger push by more people at 1000 yards.

I have withdrawn a few comments as they may have come across as disrespectful and I don't mean them that way. I will instead focus back on the original questions

[!Peter!]

Many thanks to both David's and Peter S for taking the time to share the information. I know those posts take a while to write.

For me, the purpose of running some numbers and playing with models such as Klobe's is to get a feel for the physics to deepen the understanding of what's going on when you're testing.

I've been digging and have found some more academic material which I won't post as it is very technical. I'm happy to share them on an individual basis though.

Unfortunately, I won't be in a position to do testing until May next year.

Thanks to all again for the time and information you've shared.

[AlanF]

DaveMc sent me this Excel spreadsheet and asked me to put it online : http://ozfclass.com/forums/xls/singletargetmodel.xls.

Press F9 (Calc) repeatedly to see the variation in size and shape of the resulting 10 shot groups (alter the red figures to suit your own circumstances).