G7

[brown dog]

Been thinking about use of the G7 BC, and the way it seems have been seized on by many as the panacea to ballistic prediction problems.

 

So, out of interest I modelled a few outcomes to see where it stands.

 

Bottom line up front:

 

• The critical piece is how accurate the BC in use is; manufacturer’s claimed BCs are usually ‘optimistic’; for example Lapua claims a G1 of 0.508 for its 155 Scenar. Using Bryan Litz’s (BL’s) measure G1 of 0.462 is much more accurate.
   
  Second to that is whether you use G1 or G7. In broad terms, for pointy spitzer boat tails, G1 normally seems ‘wander’ away from reality earlier than G7.
   
  And finally, in the case of the scenar, BL’s’s measured G7s are pretty much spot on to 700m; beyond that the G7 fit gets a little optimistic.
  

 

 

Below are a series of graphs each showing the data for a 155gr Scenar fired under the same conditions, but modelled using different methods.

 

The graphs show range against the scope elevation clicks required at that range.

 

GRAPH 1

 

This graph shows Lapua’s own data derived from Doppler firings. This is directly measured data and forms the baseline against which to judge the prediction methods.

(eg, at 1000m this Doppler trace tells us that 108 clicks of elevation are required).

 

 

GRAPH 2

 

Compares the Doppler baseline (ie ‘fact) against a G1 prediction made using Lapua’s published G1 BC of 0.508

(At 1000m it predicts that only 97 clicks are required – resulting in a miss 11 clicks low (ie 1.1 metres low))

 

 

GRAPH 3

 

Compares Doppler baseline against a G1 prediction using BL’s’s measured G1 (measured through firing) of 0.462 and his point mass ballistic solver.

(At 1000m it predicts that 105 clicks are required – resulting in a miss 3 clicks low (ie 30 centimetres low))

 

 

 

GRAPH 4

 

Compares Doppler baseline against a G7 prediction using BL’s’s measured G7 (measured through firing) of 0.236 and his point mass ballistic solver.

(At 1000m it predicts that 104 clicks are required – resulting in a miss 4 clicks low (ie 40 centimetres low))

 

 

GRAPH 5

 

Compares Doppler baseline against a G1 prediction using a ‘tuned’ pejsa based model and a BC similar (but different) toBL’s G1 BC.

(At 1000m it predicts that 108 clicks are required – resulting in a hit (ie spot on))

 

 

Is G7 everything?

 

Deductions:

 

It is evident that the biggest prediction improvement comes from using a ‘measured’ BC instead of the manufacturer’s claimed BC (regardless of whether it’s G1 or G7 (but see note below))

 

Use of a measured G7 is pretty much spot on out to 700 metres (using BL's Point Mass Solver, can't say if others are as good) ; after which it becomes ‘optimistic’.

 

A measured G1 can perform as accurately as a measured G7 (Note: I suspect BL has tuned his point mass solver to manipulate the G1 into a G7 approximation and then treats it as a G7 –hence the extremely close match between the two. I suspect that use of a different program, such as JBM would show a greater divergence between measured G7 and measured G1)

 

A ballistic prediction tuned to match fired data will beat both G1 and G7 measured BCs hands down

 

[Ronin]

In other words:

 

 

Actual field testing at known distances (100, 200, 300, 500, 700, 1000 yds / mtrs)

 

Measuring drop at these distances (or as far as you can go on your range), velocity, weight of bullet used, temperature, elevation , humididty, barometric pressure - ON THE DAY OF TESTING.

 

 

Will give a baseline model to work out an ACCURATE range card for any condition using Pejsa ballistic programme.

 

 

Which is the method employed when you do a card for anyone.

 

 

This ensures the card is accurate to YOUR rifle and bullet's behaviour.......

[brown dog]

In other words:

 

 

Actual field testing at known distances (100, 200, 300, 500, 700, 1000 yds / mtrs)

 

Measuring drop at these distances (or as far as you can go on your range), velocity, weight of bullet used, temperature, elevation , humididty, barometric pressure - ON THE DAY OF TESTING.

 

 

Will give a baseline model to work out an ACCURATE range card for any condition using Pejsa ballistic programme.

 

 

Which is the method employed when you do a card for anyone.

 

 

This ensures the card is accurate to YOUR rifle and bullet's behaviour.......

 

Spot on

 

But being a pedant, although I know this is what you meant, I'd say your rifle SYSTEM and bullet's behaviour - because we're calibrating what the scope's actually doing too.

 

Just thought I'd share the graphs; interesting to see the various predictions spelt out (well it was in my not-getting-out-enough mind ). BL's measured G7 and point mass solver is pretty amazing to 6-700m;

 

and anyone using G7 could almost adopt a G7 rule of thumb - for every 100m over 600m add one click

ie at 700m add 1 click to the G7 data, at 900m add 3 clicks etc.

 

 

But, of course, that'll be no use if they don't take account of pressure, temp, their actual scope click value etc too.

[Gandy]

Really intresting post, and whilst it took me a few reads really intresting information.

[Deserttech Europe]

Ronin and Brown Dog,

 

I am afraid I do not agree, whilst I agree with Brown Dogs graphs and conclusions 100% I totally do not agree with Ronin that to take your measured real world data is satisfactory in order to produce accurate charts. This form of data collection is far different from the Doppler measured real world data as used by Lapua.

 

Measuring all your real world environmental conditions whilst crucial, cannot take into considerations any vertical wind components and these would almost certainly have an impact on some of the measured results.

 

In fact there is another thread where there is a discussion about the accuracy comparison between real world data and predictions from accurate ballistic model.

 

Regards

 

 

DTA Europe

[brown dog]

This form of data collection is far different from the Doppler measured real world data as used by Lapua.

 

Measuring all your real world environmental conditions whilst crucial, cannot take into considerations any vertical wind components and these would almost certainly have an impact on some of the measured results.

 

 

Yes.....But think about it; doppler tracking a real bullet will always be measuring a bullet being affected by any number of vertical and other winds. It's how you handle the data afterwards. Regression analysis.

 

In terms of the field measurement available to us all (of carefully and skillfully collected data), data anomalies tend to stick out like a sore thumb; as such 'interpolation' is the word of the day

[Deserttech Europe]

Brown Dog , agreed, and I did not mean to state that Ronin would not have taken his data carefully. However in my experience I would trust the team that have gone to the effort of obtaining Doppler radar data to achieve a far more accurate result (after I would assume more extensive testing) than the usual real work data I see which is from a day in the field or Bisley with about 30 rounds fired.

 

 

DTA Europe

DTA Europe

[brown dog]

Wasn't reading it as a point on Ronin I actually think we're violently agreeing. The doppler directly and empirically measures the bullet itself - and regression analysis can extract the central 'theme' of that bullet's behaviour from the environment.

Field data has more 'noise' in it as it's got all the vagaries of the whole shooter system mixed in too.

That's good and bad.

Bad in being less 'pure' about the bullet itself;

but good in that it allows you to model the system - and it's the system we're interested in modelling to get a spot on firing solution - the best measured drag coefficient in the world will be no good if your scope's doing actually doing 0.12mRad clicks.

[TC]

Matt,

 

Interesting read!

 

What do you mean by by a "‘tuned’ pejsa based model and a BC similar (but different) toBL’s G1 BC." ?

 

The model itself is tuned (if so how?) or the BC is tweaked until the prediction matches collected field data at several points then extrapolate out?

 

Tony

[brown dog]

I can't reveal my secrets on the www

[TC]

I can't reveal my secrets on the www

 

[Davy]

• And finally, in the case of the scenar, BL's's measured G7s are pretty much spot on to 700m; beyond that the G7 fit gets a little optimistic.

 

In the case of Scenar's I would agree G7's are a tad optimistic in that you need to dial in more elevation than the theory would suggest as you move out, however this seems to be the norm with the G7 model in general with .308's and not just Lapua. However for smaller diameter projectiles such as the .224" and to a far lesser degree .243" the G7 fit becomes pessimistic in that it suggests more elevation than is actually required. We can very quickly attribute this to vagaries of elevation at longer distances, is your 1/8" MOA click really what it says? Does 40MOA of elevation on your 'scope really get you 40 inches of elevation at 100 yards, or a 'true' 41.88 inches or something different?

 

In an attempt to prove/disprove this theory I have tested the 80 grain SMK at 3500fps (OK a different manufacturer) and found that once you start moving it at over 1000m/sec (3280fps) things happen... By the time you get to 1200 yards it is shooting flatter in the order of 2MOA or 24" or just over 25" if you use true MOA instead of the inch approximation.

 

S&B, Nightforce BR and S&B all give varying results but all seem to show that little pills get flatter as they get out there...

 

Now given that you have probably crunched theoretical data for many loads and projectile sizes against user data do you see this yourself?

 

Great work BTW and apologies for going slightly off track with this post.

 

 

.

[brown dog]

David,

 

I wasn't suggesting that this is a Lapua issue; it's just that Lapua is the only manufaturer who makes Doppler data freely available for their bullets; therefore it's the only brand that you can benchmark 'truth' (ie their regression analysed doppler Cds) against any of the various approximations out there.

 

I'd recommend running some of their data at the velocities you think you're seeing something weird at and comparing it to your predicted data as I have above -that'll give you groundtruth.

 

I don't agree that scope click value causes divergence in the way that you say; a constant click error will have the same effect for any calibre, not optimistic for one and pessimistic for another.

 

...but I do agree that what a scope is actually doing (as opposed to what it claims to be doing) is a critical piece in the jigsaw of modelling a rifle system's performance.

[Davy]

I am with you on 'scope clicks, the error should not jump at greater elevation a result of the index ring on a 'scope as it should be cut equally therefore the error should be equal, the only variance would be the perceived error at distance which *should* be purely a component of the angular error.

 

Bottom line is if you are using a reasonable 'scope then the increments should be equal. I read an article on Nightforce scopes a while ago which actually gave a figure for the angular error on the BR 42X - apparently it is that well known, (Wish i could find it now)

 

The problem for me is I only see this error to any degree with .224" and only when it is moved at high MV's. Move an SMK80 at 3200fps and the error is evident, move it at 3500fps + and the error grows exponentially. The odd thing is calculated drops work very precisely until 900 yards are 'close enough' at 1000 but move beyond that and in practice it is like it has picked up a 100fps tail wind... Add to this it is my wife's rifle, she only shoots paper, has good drops for 200-1200 and suggesting a change to her 'perfect' load would proably be greeted with extreme violence all mean I cannot research this further.

 

Just curious to see if you have ever come across similar. I was talking to Vince about this last year and he was also of the opinion that this phenomena occurred, I need to catch up with him again and compare notes!

[brown dog]

Bottom line is if you are using a reasonable 'scope then the increments should be equal. I read an article on Nightforce scopes a while ago which actually gave a figure for the angular error on the BR 42X - apparently it is that well known, (Wish i could find it now)

 

I think Nightforce are generally operating on an MOA value of around 1.09" at 100yds, be interested to hear if that matches the value you've read.

[foxing2night]

Hi Gents,,

 

Very interesting read indeed,,

 

Keep it going??

 

Darrel

[Davy]

I think Nightforce are generally operating on an MOA value of around 1.09" at 100yds, be interested to hear if that matches the value you've read.

Checked and all I have is a spreadsheet that gives 1.076" MOA value for 100 yards and I would be inclined to challenge such a figure as it seems too precise.

 

Around 105% would give 'true' MOA. I currently run a Nightforce 42x BR on my 7mm and 20MOA was just over 21 inches at 100 yards so in that respect it was quite good. A quick check shows I still have my MOA set at .125 so it is at the true 1/8 MOA.

 

Urghhh... that opens a can of worms and it is only Monday!