G30 and G29 show different results

TC

Hi I am trying Mesh compensation for the first time. I am using a Precision Piezo inside my hotend to home and probe the bed and it es extremly precise. First I home the Z axis using a G30 command and then probe the bed at 9 points with G29. The center point is identical to the one used for G30 but when looking at the result the complete hightmap has an offset of about +150 microns even the center point. I am really confused how that is possible since the wayy of probing should be exactly the same. The trigger height is set to 0.0 since the piezo is really precise but this should effect both so should not matter for this problem. Does anyone have an ideo what is going on here?

lg TC

fcwilt

@TC

Have you done this multiple times to see if it varies?

And you are sure the G30 point is the same as the center G29 point?

Frederick

TC

@fcwilt I probed it several times manually with G30 and it allways showed an error of +-0.005. Nothing compared to 0.150...

fcwilt

@TC said in G30 and G29 show different results:

@fcwilt I probed it several times manually with G30 and it allways showed an error of +-0.005. Nothing compared to 0.150...

Sorry, I meant have you done both the G30 and the G29 multiple times together.

Frederick

TC

@fcwilt Yes its allways the same. G29 has a constant offset

fcwilt

Thanks.

Well at this point we need to see these files:

CONFIG.G HOMEALL.G HOMEX.G HOMEY.G HOMEZ.G BED.G and a screen shot of the displayed height map.

Frederick

TC

@fcwilt config.g homex.g homey.g homez.g bed.g

TC

@TC Ignore the commented stuff
I played around a lot

fcwilt

Thanks.

Looking at them now.

Just FYI in the future you may want to try using the </> tag which allows pasting g-code into a scrolling region of it's own.

See below where I pasted your bed.g file:

; Auto calibration routine for an example delta printers
; Before running this, you should have set up your zprobe Z offset to suit your build, in the G31 command in config.g.
;you should use https://configurator.reprapfirmware.org/ to setup a correct bed.g file for your printer, what is below is just an example (and its commented out)


G1 Z3.0 

G31 P{sensors.probes[0].value[0] + 30} Z0.0
G29 S2

G29

Frederick

TC

@fcwilt Ah thanks for the hint

fcwilt

Hi,

Well that was fun - those files are a bit cluttered with unused/unneeded commands.

Anyway what is the intent of the the command below?

G31 P{sensors.probes[0].value[0] + 30}

Aside from that as near as I can determine you are running G30 at X150 Y130 but the center point for G29 is X152.5 Y130.

It's possible those points are not at the same Z height but a difference of 0.15 doesn't seem likely.

Please take screen shot of the height map.

Frederick

TC

@fcwilt I know I have to tidy them up a bit
The purpose of this line is to have a reliable trigger value in case it chanches over time. I found this method here in the forum and it works perfect.

Regarding the probe point you are right but thats definitly not the reason. The height map is allmost plane just all red... I can post a pciture tomorrow

Phaedrux

Sinec your config files are rather convoluted I would suggest getting a very basic stock config set from the web configurator with the minimal adjustments needed to match your hardware and then checking again if you get the same issue. Chances are there is something in the configuration that is causing a problem, but there's only one way to be sure.

Also, please post the results of M122, and M98 P"config.g"

fcwilt

@TC said in G30 and G29 show different results:

The purpose of this line is to have a reliable trigger value in case it chanches over time. I found this method here in the forum and it works perfect.

And what data is contained in sensors.probes[0].value[0]?

Thanks.

Frederick

TC

@fcwilt its the analog input value provided by the untouched piezo sensor

lord binky

My first guess is that if you're seeing an constant offset across the board, that'd be the discrepancy from programmed Z height, likely related to some missed steps traveling fast / accumulated after homing to the first probe point. I would reduce the travel speed in M558 to something fairly low to and see if it clears up. If it does, then you can play around with adjusting acceleration/ jerk settings to make it go away. If it doesn't, try lowering the feed rate too. Hopefully you'll find something with that.