exxos blog - random goings on

[exxos]

What a surprise

IMG_0264.JPG (56.27 KiB) Viewed 2591 times

I used the "baby step" option to move the Z-offset to get it level with the bed etc so its on -0.20. But it was like 1mm above the bed

I homed it like 20 times, and it was always consistent at that, so I do not think my homing switch is to blame. However I had to move the offset to +0.20 to get it correct height. Then I homed again and had to put it back to what it was originally -0.20

So I think what is happening is that when I turn the printer off and on again, the display shows the correct offsets, but is not actually applying it to the motors. So when I manually adjust it and then rehome, it then reads the value on the screen. So I guess if I simply moved it just 0.01mm and back again and rehome is that it would then home correctly.

I will see if I can do this with Gcodes instead. I think there is a bug in the firmware somewhere

[sporniket]

At some point, I would not be surprised that you end up replacing the microcontroller by a ST through the parallel port, and writing a STOS program to drive the whole thing...

[Badwolf]

So I think what is happening is that when I turn the printer off and on again, the display shows the correct offsets, but is not actually applying it to the motors. So when I manually adjust it and then rehome, it then reads the value on the screen. So I guess if I simply moved it just 0.01mm and back again and rehome is that it would then home correctly.

Probably asking the obvious, but are all the temperatures (hotend and bed) consistent throughout?

BW

[exxos]

At some point, I would not be surprised that you end up replacing the microcontroller by a ST through the parallel port, and writing a STOS program to drive the whole thing...

That's a great idea! I'm starting to think I need to write my own firmware to drive these things! Imagine that, a Atari ST printing out parts for future Atari ST's

[exxos]

Probably asking the obvious, but are all the temperatures (hotend and bed) consistent throughout?

Yep. I tried a test print this morning and it failed. Then I left the temperatures on for half an hour why went and had lunch. Then came back and repeated the test and was exactly the same. It does look like the bed not perfectly flat because of the magnetic stuff.. But it only seems to be a problem towards the edges of the bed. I am really thinking of getting some glass cut and using that because I am not convinced that all this metal is not warping all over the place.

[Badwolf]

Yep. I tried a test print this morning and it failed. Then I left the temperatures on for half an hour why went and had lunch. Then came back and repeated the test and was exactly the same. It does look like the bed not perfectly flat because of the magnetic stuff.. But it only seems to be a problem towards the edges of the bed. I am really thinking of getting some glass cut and using that because I am not convinced that all this metal is not warping all over the place.

Glass isn't a silver (or transparent) bullet. At the levels of precision you're dealing with here, glass will bow too.

If it's varying heights across the bed, perhaps up the mesh levelling points? It's slower and boring but it can do wonders.

I started with four points IIRC (a 2x2 grid) and it worked terribly. It turned out there was a peak in the bed down the centre running front to back. Possibly a heating bow, but that's not really the axis upon which it's secured, so I dunno.

Anyway, long story short, changing to 6 points (3x2) improved things greatly as now it's actually taking a measurement at both edges and the 'high' ridge.

BW

[sporniket]

I was wondering, at least for small pieces like the square, whether UV 3D printer would yield more consistent results ?

[exxos]

If it's varying heights across the bed, perhaps up the mesh levelling points? It's slower and boring but it can do wonders.

The ABL actually seems to make things worse. I really do not know why. It seems to think the bed is leaning across the X-axis for some reason, when its not.But again I think it does not recover values correctly on power up. I don't really have a problem with doing the bed levelling if it actually improved things. But every time I turn it on, it ends up making things considerably worse Another problem is there is no manual for this printer either. So I pretty much have to guess what everything is doing, but ultimately gave up in the end and just turned it off.

I don't know if you ( or anyone for that matter) knows how to read the Z-probe via Gcodes ? it is a magnetic one, so I assume it will change value based on how close it is to the bed.This way I could see with the probe if the bed is walking over time or something.

I really would like to use this for my "z-end stop" rather than my hall effect switch, but no idea how to set that up either. I have half a feeling @PhilC Might have done something like that already but cannot remember ?

[exxos]

I was wondering, at least for small pieces like the square, whether UV 3D printer would yield more consistent results ?

The ones which have the "bath of liquid" ? I did look into that, but seemed a lot of messing about and cleaning up afterwards.

I think the bulk of my problems are some of the calibration settings on the printer are not actually stored or loaded correctly. A lot of them I am sending Gcodes already in my start-up script.

[PhilC]

@exxos I use a bltouch which most firmware supports as standard, then do an auto bed level before each print by putting the gcode G29 in the startup code in Cura.