Tell me if I’m observing what I’m observing:
I find over a session of calibration that the printer just gets worse and worse and worse. I’m thinking that a cold boot starts with certain values in firmware, but then settings over time (specifically M666 and Flow multiply) accumulate, until you have a printer that doesn’t really know what zero level is at each axis, and doesn’t really know what a good flow is, leading to a nozzle that doesn’t reach the bed, and gets caught on over extruded first layers, losing calibration.
Further, if the first print’s got good flow when you adjust things to 115% flow multiplcation, the next time it runs, it may, or may not be using that setting…and if you select 120%, then screw up and select 115%, you very well may be getting 115% of 120% of calculated extrusion…
ETA: Would it make sence that the M666 adjustment would be different in Kisslicer as it is in Slic3r? That wouldn’t make any sense if they were absolutes, but my behavior WOULD make sense if every print I’m attempting, it’s flailing about a little differently each time…,
I don’t get where your coming from.
Are you saying you update your Firmware every time you use your printer.
Or are you confusing your software settings with your Firmware settings.
I’ll power up the printer and have KISSlicer add X-1.2 Y-2.2 and it won’t quite stick…Z will be perfect and I’ll change to M666 X-1.4 Y-3.2 and it’ll do a reasonable first layer, but then something else won’t be right, cancel the print. Fix the tape, for whatever reason switch to Slic3r and to get it to work with M666 X-2.2 Y-4, and 120% on the extrude…
Go back to Kisslicer without changing it’s settings and it prints WAY differently than it did before Slic3r got involved, over extruding high on one part, low on another, catch on a high piece of plastic which throws off it’s calibration…I’ve had a number of episodes where I’ll look up and 90 minutes have passed and I haven’t had a single successful print. What I haven’t done is cold-boot the controller early on.
There’s just some behavior, which is hard to put my finger on, that appears to be accumulative error.
I was under the impression that M666 was a work-around kluge. The pots on my hall effect sensors can’t get it any closer and I haven’t gotten so fed up with it that I start loosening screws on the X and Y axes and retension the spectra.
