Plugging in your numbers but with the default weighting factor of 0.50, the calibrated output I come up with is very similar to yours (see attached file). It is not identical in terms of pressure, but pretty close.
The two things that caught my attention were your case volume (54.91 gr, which seems relatively low), and the amount by which you have to adjust Ba to match your actual velocity (~8.9% increase). The "low" case volume may simply be inherent to the brand of brass you're using, especially if it has noticeably thicker walls than Lapua. More importantly, increasing the case volume value in the program would effectively
lower pressure and require an even larger corrective increase in Ba to match the predicted and actual velocities. I have previously encountered similar QL predictions where I had to change Ba by more than 5% in order to get predicted and actual velocities to match.
It is not "ideal" to have to change Ba by that much, but when doing so, I have never had a problem when making subsequent changes to charge weight with the predicted velocity not matching the QL prediction. It cannot be ruled out that you have a batch of 4064 that is at the upper end of the burn rate range for that powder. Alternatively, there could be another reason that is not immediately clear. Was the temperature exactly 70 degrees when you measured velocity? What type of primer are you using? Another possibility is that a primer with significantly greater brisance can markedly increase velocity over a another brand of primer, even with the exact same charge weight of powder
Regardless, as long as predicted velocity matches measured velocity fairly closely over a range of charge weights, I work under the assumption that other predicted parameters such as fill ration, pressure, and barrel time are accurate within the limitations of the program. Your QL predictions fall well within the range at which I usually obtain optimized .308 loads in terms of pressure. That is to say, loads with predicted pressures in the range of 57-58K psi. With standard .308 brass, loads predicted to be in the 59.5-61.5K psi or so range, although still below SAAMI MAX, are generally hard on brass and may cause the primer pockets to become unusable after as little as 4-5 firings.
At this point, I would not be greatly concerned that your velocity is quite a bit faster than predicted by QL using only the default IMR 4064 powder input for temp and Ba. I have had plenty of Lots of various powders that gave velocities very close to those predicted using the preset Ba. But I have also had Lots that were in the neighborhood of 75-100 fps fast/slow. During load development, I generally aim for Optimized Barrel Time nodes whenever possible, using barrel times from Chris Long's OBT chart and those predicted by QL. In my hands, accuracy nodes on paper fall very close in terms of barrel time (as predicted by QL) to the values in Chris Long's table. Other factors, such as brass life, also seem to mirror the pressures predicted by QL quite well. As long as the velocities predicted by QL following "calibration" of the program for your specific setup match reasonably well, I see little reason not to accept the other outputs such as pressure and barrel time as being valid, or to go to great lengths by changing a number of other parameters in the program merely to improve the output by a negligible amount. QL outputs are, by definition, "estimates" or "predictions". They are certainly useful for expediting the load development process, but their true "accuracy" must always be questioned, and they can never replace the outputs of targets generated by actually testing various loads. In other words, the test targets should always have the final say. Once I am convinced that the target is telling me I have a good load, as long as the predictions in QL seem to match
reasonably well with actual measured data, I don't feel the need to adjust the program further. IMO - you're really just tweeking factors that are already down in the noise at that point.
View attachment 1079988
