More Simplify3D problems. Even though the printer is set to home the bed in

More Simplify3D problems. Even though the printer is set to home the bed in 3 places to detect the pitch of the bed, Simplify3D doesn’t take these measurements into account and when printing parts that span the bed, I have problems with the nozzle being too close to the bed in some areas and too far in others. I tried going back to Cura and that works completely fine so it can’t be the printer it has to be the slicer right? I’m also still sort of having a problem with the slicer thinking the print head is in the bottom left corner when the start gcode finishes no matter where the nozzle actually is, but I avoided that problem by moving the print head to the bottom left corner at the end of the start gcode. Any suggestions?

Are both start gcode identical?

@Evan_Nguyen Except for the portion that moves the print head to the bottom left corner, yes.

Just to clarify exactly what you said:

The same part was printed twice in exactly the same place on the bed, but the only difference was the slicer, Cura vs. Simplify 3D?

And yet you found that the Simplyfy3D sliced result has the nozzle too far from the bed in places and too close in others?

@Paul_Gross correct. I was trying to print multiple parts spannign the entire bed and they all have the same distance from the bed in Cura but in S3D the ones in the bottom right corner are squashed and the ones in the top left corner don’t stick to the bed. My bed is tilted about a third of a millimeter from corner to corner and Cura takes that into account.

@Aric_Norine is correct.

After the auto-level program runs, the printer will then do minor z-height adjustments during printing that are not part of the g-code it is executing.

It seems more likely to me that your level probe is not consistent. This happens to me occasionally too (about 1 failure every 50 prints), and I only use slic3r and I have an inductive level probe on a steel bed.

When I say failure, I mean that after the auto-level program runs the printer seems to still have no idea where the bed actually is, so I have to abort during the first layer.

There was a bug in older Marlin versions: if you home not all axes but specific ones the bed leveling data was lost.

@Adam_Steinmark Try my Marlin3DprinterTool. It contains a bed level feature that helps you to get a leveled buildingsurface. The effect of ABL/MBL is almost zero on a flat bed. If the bed is warped and impossible to get flat the Software can give you a visual picture of how and where the bed is warped.

What version of Marlin do you have?

The sequence of ABL is first a G28 and then G29. If you do a G28 AFTER G29 the ABL is lost.

This is working start of ABL
G28 G29 or
G28 X G28 Y G28 Z G29 or
G28 X G28 Y G29
( you can have the X and Y in reverse order)

This will erase ABL
G28 (zero or more axis) G29 G28 (Zero or more axis)

Well I need to move the print head to the bottom left corner at the end of the start gcode, how do you suggest I do that without homing first? I don’t know where the firmware is set to probe the bed at the last point but it’s offset from the bottom right corner slightly.

add this rows to Script/Start Script
G28 Y
G28 X
G28 Z
G29
M117 ABL done

and this to Script/End Script
G28 Y

Then each print will start with the right sequence

@Johnny_Linden So what would the full start script be?
G28 X0 Y0 Z0
G92 E0
G29
G28 Y
G28 X
G28 Z
G29
M117 ABL done

Simplify3D support helped fix the problem. My new start gcode is below.
G28 X0 Y0 Z0
G92 E0
G29
G1 X0 Y0 Z0.5

Why do you need to force the extruder to 0,0?

I explained this in my post. It thinks it’s at the origin as soon as the start gcode finishes executing and if it’s anywhere else, the printer tries to move the print head off the build plate. I don’t know why this happens but this is a workaround for the problem.