Satterlee 10 Round Load Development

[Lemonaid]

Klem, adding to the mix now is the bullet jump testing from precisonrifleblog. https://precisionrifleblog.com/2020/...elopment-data/
Satterlee and friends now find the best bullet jump distance first, then do charge weight variations. Long story short is as they shoot up to 200 rounds a match, the lands would erode enough that they would be off their sweet spot before the match was over. So again to have a larger window of consistent loads they did a lot of testing. It's a good read, lots of data and graphs, I think it's up to four installments now.

[Klem]

Lemons,

Off their sweet spot in 200rds...Is that right? This sweet spot they are chasing is so fleeting that it is gone in a couple of shooting sessions. What are the chances we will even notice this in an AR15, made worse by using a bipod and cheap scope?

In F-Class we would be chasing the lands as the barrel throats wore. But, only firing 30rds a week it would take a year before it became an issue. By then people were buying new barrels anyway.

[Lemonaid]

A quote from the first section: https://precisionrifleblog.com/2020/...d-development/

[Harpoon1]

So you are saying Satterlee's goal is not to find the smallest groups but the most consistent groups over time?

https://youtu.be/O4K4l0KzRfg

[Klem]

Here's a work-up done yesterday with Satterlee embedded. 50 rounds fired at 100yd targets in ten increments of 1%.

All five rds fired per load fired at same point of aim (10 x 5rd groups).

It is 5 Satterlee's in a row, fired at targets.

Here they are graphed. You can see there are two to three 'accuracy' nodes per 10-shot ladder. Each Satterlee string produced different nodes which sinks the theory immediately. If however you combine all 13 nodes (bar graph, bottom right) you can average them out to three common nodes.

21.8 grains of H4895
22.7
23.0

Even the averaged nodes don't match the tightest group, 23.5gns. 23.5 is the best load I settled on a while ago so the results on target appear to be repeatable. There is no visual correlation between group size and elastic velocity ranges (bottom left graph).

On this occasion Satterlee did not work, five out of five times. I think the technique has more holes than Swiss cheese.



223 bolt gun
Bench rested
80SMK
100yd.
No wind
New Lapua brass
Lab Radar for velocity
On Target for group sizes.

[Lemonaid]

Klem, just out of curiosity, at what loads in your test did you get the best SD and ES?

[Klem]

Klem, just out of curiosity, at what loads in your test did you get the best SD and ES?

Lemon,

Here is a copy of the barrel data I keep in a digital diary.

• The first column of 15rounds was 15 warming/fouling rounds to settle the barrel. No MOA recorded as it was shooting steel.
• Note, the best group in the test is the same load as the fouling load. It is heartening that the previous work-up is repeatable.
• The powder 'AR2206H' is the same as 'H4895'. It's what it's called here.

Notice the Standard Deviation and extreme spread at the best group, 23.5gns. That suggests Satterlee works...and for a second I thought what a chump, it really does work. Hold on, looking across the SD's there's an even lower SD/spread, 22.3gns. Unfortunately, the group size for the best SD/spread is one of the largest. If I wasn't shooting at a targets and just going off velocities I might spend the rest of that barrel's life throwing large groups downrange not being any the wiser.

For full disclosure here are the groups, numbered from lightest to heaviest load;







.

[Lemonaid]

First klem let me acknowledge the great deal of work and brain power you have put into this, really great! I started the thread to gather information about the Satterlee method and how it works or dose not work for those that have tried it and the posts here have answered the call.
Let me now try to channel Satterlee to a small degree and take your data and play with it a bit.
In Satterlee's world he is more focused on hitting steel at long range (1000 yards plus) so the ES and SD are more important to him than absolute small groups at closer range. At some point in a long range trajectory a low ES load with average sized groups (at short range) will out preform a larger ES load that had a smaller group at short range. The variations in bullet velocity from a high ES load causing enough vertical variation to make a miss.
So he would see your node at 22.3 and see that it produced the best SD ES with your 5 round tests and the next step would be if he didn't like the group size would be to play with the seating depth to get that load to group as small as possible, then to try that load at long range (600 yards to 1000 yards) to verify that he can make hits with it.
So it might just be that if you used your superbly accurate 23.5 load for a competition on targets out to 600 yards you would score better than the 22.3 load but at 800 or more the 22.3 load would work better (imagine your data came from a 6.5 Creed).
This is all theory to me so far so I may be completely wrong! But the people that swear by it shoot a whole lot more than me and at a lot longer distance and are high in the rankings in the ELR game so I will give some credence that it can work.

[centerfire]

The Satterlee method isn't intended to find the tightest group but to identify velocity nodes in the least amount of rounds possible. Your group size is irrelevant. Your spreadsheets still invalidate the process but your commentary on the group size is more of a sidebar. The next step should be finding the optimal seating depth to tighten the group assuming you found a velocity node. Based on your spreadsheet I'd probably change primers or powder because the SD's are high and loads .3gr above and below don't suggest a velocity plateau.

[Arkhangel5]

As an aside to this topic. Question for Klem.

Have you ever gone higher in powder charge with H4985 and this weight class bullet?

H4895 is my preferred powder for the .223 in 600yd F-class matches.

SY

[Lemonaid]

That is an interesting hypothesis.

I've deleted the comment as I see where you are going with this...It is about bullet drop differences at long range caused by different velocities. We can still do a 'What-if' using the load data to see the difference between best SD and best 100yd group size.

I will run some numbers and amend this comment with the results soon

Thinking back to some of Satterlee's comments on the 6.5 guys site he may have chosen your 23.5 over the 22.3 load for the better velocity or even the super accuracy, but maybe tested both out at long range to pick the winner.

[Klem]

I have done a 'What if' at 1K yards using JBM's online ballistics calculator. It's not real like the 100yd test but it is consistent if you believe the MOA drops it spits out.



In the scenario I have given the Satterlee load (the best SD/Spread) the best chance of winning by assuming the following;

1. That the slowest round is at the bottom of the group, the 6 O'clock position.
2. That the fastest round is at the top of the group, the 12 O'Clock position.
3. That the actual Best Group at 100yds was a statistical anomaly (unusually tight) and have made it the size of the next best group 0.329MOA@100yds.

Assuming a cone of fire will never diminish as range increases I have multiplied the 100yd group sizes by 10 to simulate 1,000yds. Then added the max possible vertical bias that differences in bullet velocity would blow the group out by. This is the difference between fastest bullet at 12'Oclock and slowest bullet at 6'clock.

Even with biasing Satterlee with the best possible conditions and replacing that great group at 100 with the next best group, the Best Group still wins the day.

Sure, different velocities will expand the group at 1K with a vertical bias but looking for the lowest Standard Deviations/velocity spreads is not the best predictor of 1K group size when you have actual 100yd group sizes to inform your decision. If you accept JBM's drop predictions, and assume both groups and velocity spreads are repeatable we can now say with certainty that Satterlee does not work in all situations. Unfortunately, you will never know when Satterlee is correct unless you shoot all loads at 1K, which of course renders Satterlee moot as a predictor.

And the idea you can do all this with only 10rds across chrono is as useful as an ashtray on a motor bike.

[Klem]

The Satterlee method isn't intended to find the tightest group but to identify velocity nodes in the least amount of rounds possible. Your group size is irrelevant. Your spreadsheets still invalidate the process but your commentary on the group size is more of a sidebar. The next step should be finding the optimal seating depth to tighten the group assuming you found a velocity node. Based on your spreadsheet I'd probably change primers or powder because the SD's are high and loads .3gr above and below don't suggest a velocity plateau.

Centrefire,

Well, at least we agree that Satterlee does not work in all situations.

And if you want to compound the validation by suggesting it is invalid because the OAL is not being tested and lots of different powders are not being tested then again, how will you ever know that the lowest SD/velocity spread produces the best results at 1K unless you test all of these other permutations across all other loads to produce the tightest groups at 1K?

Are you saying that because my SD's were not low enough that this somehow confounds Satterlee, and that Satterlee only works with single digit SD's? Surely, as long as all variables across all loads are equal that shouldn't matter. Yes, I could now tinker with powder and primers to tighten the SD's but how much of doing this is enough to turn Satterlee into a good predictor of best groups at long range? You can tighten SD's by tinkering with primer/powders but how does that guarantee Satterlee is now a better predictor of long range accuracy than doing the same tinkering and seeing the effects on a target at 100?

Tightest groups is not a sidebar, this is what you are trying to achieve at 1K, to give your ability to drive this around the target the best chance of winning. Finding the tightest velocity nodes with the lowest round count at the muzzle is not the aim of the exercise, it is the mediating variable to finding the tightest groups at long range.

As shooters we could start a discipline where we shoot across chronographs into the butts without targets and the winner is the one with tightest velocity spread. But that's not what Satterlee is currently used for. It is used to predict results on a target. Results on targets are measured in group sizes and our ability to place those groups on an aiming mark; Precision and Accuracy.

[Mad Charlie]

"And the idea you can do all this with only 10rds across chrono is as useful as an ashtray on a motor bike."

Thanks Klem, this has been my suspicion ever since I heard of the Satterlee method and thanks for your work.

[Klem]

Arkhangel,

Not in 4895, but in the slower Varget. The only reason I am using H4895 and not Varget is it is more efficient in the shorter 23" barrel.

There were a few of us shooting 1K F-Class using the 80SMK in .223 30" barrels, loading 25.5/Varget which is a compressed load and over SAAMI at 64K lbs. 223 doesn't make 1K yards unless you overload it. It is OK if you have a strong action like the Rem700 or Panda or Barnard.

According to Quickload 24.9 - 25 of H4895 is equivalent pressure to what we were using in Varget for the 80SMK loaded long. 23.7 is QL SAAMI max for a long OAL 4895 load so I'd be careful but it is possible to go hotter than 23.5.