As the title states, I cannot get a successful automatic Flow Dynamics result.
Dry filament, white PLA+
Texture bed
Cleaned the lidar lens with a microfiber cloth
Completed full system calibration
Latest firmware
It’s been failing for the longest time but white on textured plate shouldn’t fail.
Also Auto Flow Rate Calibration works very well. It’s the Flow Dynamics that fail, no matter what color or type I use.
Have you tried on a smooth plate?
The flow rate calibration works perfectly on the same textured plate and filament but fails 100% of the time with the Flow Dynamics test.
I think the flow rate test varies the amount of flow while the nozzle travels at constant speed. The LIDAR measurement measures the line width, and result is the flow rate which matches the desired line width. The textured plate might cause some variation in the measurement, but it would average out.
The flow dynamics (pressure advance) test makes changes in the nozzle speed, increasing and decreasing the speed of the nozzle as it would approaching and leaving a corner. Moving in a straight line, this results in thick and thin “bumps” in the line. The Lidar evaluates the bumps and tries to adjust flow to match the changing nozzle speeds to produce a smooth line of consistent width. A textured plate complicates this evaluation as it adds random “bumps” to the line which cannot be distinguished from the intentional pressure advance changes.
Calibrate your filament manually, preferably on a smooth plate, save the results, and turn off the pre-print automatic calibration. You will save time and a little bit of filament on every print using that filament and your calibrations will be more accurate.
I agree with your explanation and with the conclusion that manual calibration, especially on a smooth plate, will produce more accurate and repeatable results. From a purely technical standpoint, that makes sense.
That said, my point isn’t that manual calibration is inferior, but that this is a built-in feature that should work reliably within its intended use cases. Sometimes the convenience of automatic calibration is exactly what I’m looking for, even if it’s slightly less precise than a fully manual process. If I’m swapping filaments frequently or doing quick prints, the time savings and reduced friction matter.
So while I accept that manual calibration is the “gold standard,” I still expect the automated system to be reasonably robust and usable without having to disable it entirely. Ideally, it would either handle textured plates better or clearly communicate its limitations so users can make an informed choice between accuracy and convenience.
In short: manual calibration for maximum accuracy, automatic calibration for speed and workflow, and both should have a valid place.
