weird behavior voron tap & how to correct magnetic bed

[JBC]

I've just finished my voron 2.4 Build from a siboor kit last week and I've been testing with my machine. The parts i get are not flawless, but a lot better than i can get from my tuned (&somewhat modified) ender 3. The only weird thing, my z-offset changes when my chamber temperature goes up.

Things I've already changed:

- increased retraction after print (for less oozing)

- added a copper brush for automatic wiping before corner squaring and homing/bed level

- checked everything mechanically (everything is tightened).

When i print with TPU and pla for example, the z-offset is 0.68 . But for ABS+ (which i print at a chamber temperature of 40+), the z-offset is 0,58mm.

I know i did something wrong (i think) with the application of the magnetic sticker. See the bed mesh below. But a difference of 0,1mm with changing temperatures?

Does anybody know what could cause this difference in z-offset?

and for the bed mesh, is there some way to fix this? (btw this is a cold bed).

[mvdveer]

6 hours ago, JBC said:

When i print with TPU and pla for example, the z-offset is 0.68 . But for ABS+ (which i print at a chamber temperature of 40+), the z-offset is 0,58mm.

When printing PLA the bed temp is 60 normally whilst ABS and ASA would be 100-110. The heat expansion of machine will account for this difference. Also bear in mind that the gantry will react to the heat (gantry bowing, therefore it is advisable to both heat soak your bed and chamber before starting an ABS print.  This can explain the bow in your mesh.

6 hours ago, JBC said:

I know i did something wrong (i think) with the application of the magnetic sticker. See the bed mesh below.

Not at all - it is the aluminium plate expanding in the heat. It is why we don't tighten down the bed on all four screws in a 2.4 build. Normally would only tighten one screw and leave 3 loose to compensate for this expansion. There are plenty of articles about thermal expansion, and if you want to know more about the subject, just google "thermal expansion of aluminium"

Here is a video demonstrating the principle of thermal expansion in a nutshell without the complicated science behind it.

 

6 hours ago, JBC said:

Does anybody know what could cause this difference in z-offset?

Thermal Expansion

[JBC]

22 hours ago, mvdveer said:

When printing PLA the bed temp is 60 normally whilst ABS and ASA would be 100-110. The heat expansion of machine will account for this difference. Also bear in mind that the gantry will react to the heat (gantry bowing, therefore it is advisable to both heat soak your bed and chamber before starting an ABS print.  This can explain the bow in your mesh.

 

Hmm i think i asked my question a bit incorrectly, I know of thermal expansion and the effect on the build plate. When i bolted the plate i did leave 3 of the four bolts 'finger tight (very light torque with the hex wrench) so there was a tiny bit of give when trying to move the bed. But there are some strange places on the build plate, that i can't figure out. So here, the first picture is of the bed when the camber temperature is at 41 degrees after 40 minutes. and the second was it was still cold at 18 degrees (and cleaned, both above and beneath the plate and the sticker. the strangeness are the marked area's. That's what i meant with 'I possibly messed up the sticker'. so the question is more a sanity check.

 

 

And back to the Tap, if there is gantry bowing, it shouldn't affect the z-offset of the tap sensor, since it's nozzle probing (and if i do some very rough calculations even with the thermal expension of abs) it shouldn't be more than 0,05mm right? maybe I'm chasing ghosts with the final amount, but i do change from filament to filament, so i'm searching for a possible solution. 

Is the SET_GCODE_OFFSET addition to the material a good solution to tackle this 'problem'?

 

[cjkennedy72]

The front of the bed being higher might simply be the TAP. I just installed a Beacon clone (IDM) and that rise in my mesh is gone. When I had a TAP sensor, I could watch the gantry and notice a little twisting at the front during bed probing. From what I have read, it is really hard to fully eliminate. The action of pressing the head against the bed to lift the SB so the TAP triggers is fairly accurate, but all bed leveling solutions have some compromise. For me, the TAP system always produced a little twist on the X axis at the front of the bed, where Y is furthest from a cross brace. It happened with printed parts, aluminum parts, and carbon fiber parts.... a 2020 extrusion and a CNC'd X axis.

I believe that the mechanics of the TAP system and the thermal expansion issue explain 99% of what you are seeing.

Edited January 2 by cjkennedy72

[claudermilk]

Make absolutely certain everything--especially the carriage connections--are good and tight. If they are at all loose and allow wobbling it will negatively affect Tap. Ask me how I know...

[JBC]

6 hours ago, claudermilk said:

Make absolutely certain everything--especially the carriage connections--are good and tight. If they are at all loose and allow wobbling it will negatively affect Tap. Ask me how I know...

Thanks, i double checked everything when i build the machine, but an extra check is never bad.

 

22 hours ago, cjkennedy72 said:

The front of the bed being higher might simply be the TAP. I just installed a Beacon clone (IDM) and that rise in my mesh is gone. When I had a TAP sensor, I could watch the gantry and notice a little twisting at the front during bed probing. From what I have read, it is really hard to fully eliminate. The action of pressing the head against the bed to lift the SB so the TAP triggers is fairly accurate, but all bed leveling solutions have some compromise. For me, the TAP system always produced a little twist on the X axis at the front of the bed, where Y is furthest from a cross brace. It happened with printed parts, aluminum parts, and carbon fiber parts.... a 2020 extrusion and a CNC'd X axis.

I believe that the mechanics of the TAP system and the thermal expansion issue explain 99% of what you are seeing.

and good to know! Maybe that's something for the future. Thanks for sharing your experiences!