Well I have tried several more things.
Weighed all balls and removed outliers from 3 standard deviations.
Rattled the balls in a bottle with water and detergent to obliterate sprue lines.
Bought an endoscope and took pictures down the bore, but it wont focus on the rifling. Darn it!
Recovered patches from experiments.
All patches were measured with micrometer, generally 0.016" on the ratchet and .011"-0.013" compressed.
Tried Dutch Schultz dried lubed patches (20% soluble oil plus a drop of detergent in water, dried in sun beforehand);
tried beeswax/olive oil grease;
tried Ballistol 20% in water
1: duck fabric: recovered patches were shredded.
View attachment 24695
2: New 'ticking' looks like calico with ticking inked pattern - shredded
View attachment 24694
So I went back to my yellow hi-vis workshirt, which has much denser weave.
2 groups of patches: upper, soaked with 20% Ballistol, over polenta and a 55gn charge PPP black powder (Similar to FFF but narrower size range).
Lower group, same patch, lube and filler but 69gn PPP, increasing charge to get faster spin and I hoped more stability.
View attachment 24689
Groups from these shots were group 1 marked blue, group 2 marked orange. 50 m on standard pistol target, 8" dia black.
Please note that as I shot these 12 shots there was increasing tightness of fouling about 8" up the barrel.
The wilder shots were later in the strings. In the past a good wet patch or a wet patch above the rammer meant no increase in tightness over up to 45 shots.
I believe this thickening fouling comes from using PPP instead of PP, which did not give this problem but powder is hard to get here.
View attachment 24690
The groups, 8" off a benchrest at 50m using a familiar rifle, are appalling. I expect to see 4" or better if I do it right, as I do 1.5" with a .22.
View attachment 24689View attachment 24690
Finally I did the same 55+filler+Ballistol but cleaning between shots. There was heavy fouling removed each time, but now there was no feeling of constriction at 8" above the charge.
Patches are much better, but group size the same.
View attachment 24691
View attachment 24692
CONCLUSION: Nothing gave markedly better groups!