Autolevel sequence taps screw on Taz 6

General Software
1

I’m hoping to get some help here as Lulzbot support was not able to help me.

I am on the latest released version of Taz 6 firmware for single extruder. 1.1.5.71

During the G29 auto bed level sequence, the back-right tap (3rd point) is tapping just on the edge of the screw head. This screw head is about .4mm higher than the metal disc itself (yes - I’ve tightened it all the way down… its just a slightly taller screw head).

What I want to do is just move the tap point so it hits the disc like the other 3 points do.

The current 3rd tap point is X: 288 Y: 289

I would like to simply move it (minus) 1-2mm in each direction X:287 Y:287 so that it hits the disc instead.

Are there any straight forward steps to modifying the firmware? Everything I have found is so disjointed between Lulzbot, GitHub, Marlin, etc…
Any help appreciated…
Thanks,
TM

2

SOLVED…

A friend from another forum suggested just modifying the G29 in the start script.
G29 V3 B287 R287

cheers,
TM

3

You will need either the Arduino IDE or Microsoft’s Visual Studio Community with the Arduino for Visual Studio. I use both. Next you will need GIT for Windows. Last you need to clone the LulzBot GIT Marlin repository. Start at https://code.alephobjects.com/diffusion/MARLIN/.

I would send an email to support@lulzbot.com asking them for instructions (branch, tag, etc.) to fetch the source code for the version of Marlin you are currently using, probably 1.1.5.71. See https://code.alephobjects.com/T1598 for some hints on getting the correct version.

You have to edit one file to select the toolhead and printer and then build using the IDE. I like to look at the code using Visual Studio because it has highlighting so you can see what code is actually being used based on the #defines in the .h files. I’ve been successful generating a .hex file using the Arduino IDE so I stick with that. The only other hint is that you want the .hex file without the bootloader.

4

Based on http://marlinfw.org/meta/gcode/, I’m not sure what that G29 is doing (OK, I understand V3 is verbosity).

5

You could loosen up the Y frame mounts on the main frame and slide the assembly a couple mm to ensure the nozzle taps the washer. There is generally enough room on the z switch and wiper to accommodate a small adjustment. Make sure the Y frame is square with the main frame before you retighten the mounts.

6

@ b-morgan,

F Set the Front limit of the probing grid
B Set the Back limit of the probing grid
L Set the Left limit of the probing grid
R Set the Right limit of the probing grid

I left out the front and left limit so it would use the firmware default. Thus changing the Back and Right so I could get it to stay off of the 3rd point screw head.

7

My bad! I didn’t notice there were 3 entries for G29!

8

A stock machine should not hit the screws with stock firmware values. If it does… Using the F/B parms for G29 may provide a workaround, but the root cause is something wrong with the geometry of the machine. I’ll second the advice of @nopick above: Make sure the frame is square and everything is assembled correctly, as the root problem may cause other issues besides incorrect probe position.

9

Ok… if two of you think that, then I’ve got to go do my due diligence… (insert OCD eye twitch). I’ll report back after I check the framework and such.

10

Also, they do have a known issue with some units coming out of square due to loose fasteners.

https://www.lulzbot.com/sites/default/files/TAZ-6_Square_Frame_service_bulletin.pdf

11

Well I took the Y assembly off checked it for squarness and level - All good.
Checked the Main Frame Assembly for squareness and level… all good too.

Put it all back together and still taps on the back right screw during auto level. Here are the readings.
READ: Bed X: -9.000 Y: -9.000 Z: 1.179
READ: Bed X: 288.000 Y: -9.000 Z: 1.111
READ: Bed X: 288.000 Y: 289.000 Z: 1.374 <---- taps down on screw head slightly higher than the disc itself.
READ: Bed X: -9.000 Y: 289.000 Z: 1.161

weird…

12

Sounds like the chamfer of the washer is not deep enough.

Where are you hitting the wiper pad?
Middle = fix washer chamfer
Right side = move Y table frame mounts a little to the right.

13

I’d switch the washer and screw with another corner.

If the problem moves with the washer and screw, send an email to support and ask them to send you a new washer (or screw, you can figure out which is out of tolerance). If the problem stays in that corner, then figure out what’s wrong and email support asking for that part. If your printer is still under warranty, they should fix it for free. If not, they should fix it anyway but a washer shouldn’t cost that much (and the other corner part is 3D printed).

Be prepared to take a picture of the bad part (or just send it before they ask for it). Side by side with the good one since you’ll have two off at one point!

14

My machine is square and level to an insane level of accuracy. I’m convinced that someone fat-fingered the 3rd tap point in the firmware. If you look at the X & Y you can see that they are all either 9 or 288… except the Y on point 3… Its 289 (see above). I’ve changed mine to 288 and it misses the screw head. So i’m going with that.

Thanks all.

15

Really not sure what’s going on, but I don’t think it is a firmware issue. The firmware doesn’t have separate Y values for the two rear locations – there is only one “rear” location value and it is used for both rear probe points (X=-9 and X=288). If everything is squared (frame, bed, X carriage) and belts/pulleys are tight, then position 289 should be in the same relative position on both Left and Right washers (and shouldn’t be hitting the screw on either).

Here are the relevant declarations:

#elif defined(LULZBOT_USE_AUTOLEVELING) && defined(LULZBOT_TAZ_BED)
    #define LULZBOT_LEFT_PROBE_BED_POSITION       -9
    #define LULZBOT_RIGHT_PROBE_BED_POSITION     288
    #define LULZBOT_BACK_PROBE_BED_POSITION      289
    #define LULZBOT_FRONT_PROBE_BED_POSITION      -9
#endif

But whatever is going on… If it prints ok (aside from having to use the position overrides for G29 in start script), then it may not be worth trying to chase down that last mm. :slight_smile: