I think Ed and my are of the opinion that any kind of auto leveling isn't going to help a bed that's nearly level because of stiction and other non linear effects.Bratag wrote:There was a discussion on the boards at one time where it compared auto levelling with auto mesh. Auto levelling simply handles tilt in the print bed but it makes the assumption that the bed is perfectly flat. Mesh levelling actually accounts for the print bed having deviations outside the tilt.
The current Marlin firmware we have doesn't handle mesh levelling, though I know Insta was looking into it and I had poked about with it myself. Perhaps we need revisit that.
The amount of force it takes to start something moving is almost always higher than to keep it moving. Stiction is that force to start something moving.
The other important concept is that your stepper is not infinitely strong. It has a torque rating, and that's how much torque it puts out across a full step. When you're micro stepping, it's better to think of the stepper as a spring attached to your shaft. With the other end of the spring attached to the commanded position. Until that spring stretches tight enough, the shaft isn't moving.
For a more concrete example, picture dragging a rock around with a rubber band. Move your hand a foot to the right, and the rock trails behind it because of the stretch in the rubber band. Move your hand to the left a foot and the rock doesn't move the whole foot because it's trailing behind in the other direction. For a small enough hand movement, the rock doesn't move at all. That's about what I think is happening with Ryan's fairly level bed. It doesn't matter during normal printing because you're always dropping the bed, or at least moving in larger increments.
Turning up the stepper current makes the spring tighter, which should help. I'm going to take a look at the stepper and driver when I get home and see how much higher it could stand to go.