XT backlash calibration wipe off side?

I don’t recall messing with touchscreen stuff. I assumed it just worked based on the display setting. It may be running the same code from the last official firmware I had installed.

Update: The “About Printer” says 2.0.9.0.14, while I’m pretty sure I was going off a 2.1 nightly build. It was done on a different computer, but the timestamp on the compiled firmware file was mid-May 2023.

Thank you for pointing out the NOZZLE_TO_PROBE_OFFSET, It was indeed a typo and will be updated in the 2.1.3.0.29 build.

Great!

Any ideas on the wiping issue?

Based on the video, there are a couple things that are not functioning properly.

The first is that the nozzle is wiping to the right by close to 2-3mm. One Cause of this is the bed mounts shifting during shipping or assembly of the bed. The other cause that comes to my mind would be if the ninja flex bumper is not fully seated on the left side of the bottom X axis rod. This would cause it to home to the wrong location.

The other thing that is strange is that it seems to be wiping too far forward, this is an issue I rarel;y ever see which makes me question the homing process. Can you make sure there is nothing interfering with the homing of the Y axis.

Is the wiping consistently wiping in the same wrong location or is is sporadic with where it wipes?

It only started doing this once the BL Touch was added

Same spot each time.

Also, can we also get 5x5 mesh back?

Looking though the code I no longer see the correction for BLTouch adapter plate in the wipe. I had a check put in place to correct the Y distance with the adapter but it must not have been pushed up. I will also correct that issue withy the next build of firmware.

The bed mesh was set back to a 3x3 to speed up the probing time when we went to the next version of mag plates which are a machined surface and a 5x5 had very little benefit. I can see where an option of doing a quick mesh vs detailed mesh might be a compromise to both situations. This could then be an option in cura to speed up the startup time if you so wish to or have longer but more detailed mesh.

Thanks for investigating

And the compromise on the mesh. More points are better. You should see how long my mesh takes on the Modix 120X (24 minutes).

The more convenience you add into the user interface, the better. I’ll take pictures of the areas of concern you highlighted when I get home. It’s been non stop meetings today. I still have to get to my emails !

See below for my Taz Pro XT 11x11 mesh with the Octograb build plate.

How much of this is the surface itself or the toolhead drooping in the middle I can’t tell. Not sure how 12mm rods could droop this much though.

When will these changes be applied to the latest firmware?

I’m wrapping up my YouTube review on the machine and certainly, these helpful changes will definitely be a positive for the XT.

I’m coming close to a deadline and I know Lukzbot would really like me to send the machine back to them.

Getting those irksome issues resolved and then some test prints after would be very useful.

Judging from this, they get a new release out every 1-2 weeks.

1 Like

I can get you a prerelease firmware to make sure that it solves your issue. I am compiling this morning and will let you know when it is ready to test.

Could you?

Or shoot me a PM here.

It’s kind of ironic they ask for the machine back at just the same point some of the big issues might be resolved. (I’ve been asking for months about the BL Touch mesh issue and off bed wiping for quite some time.)

Adding a quick vs detailed option to the bed mesh is a little more extensive than I originally thought. I have to do more testing before I release an experimental firmware build. These changes will likely end up being implemented in the next cura build.

In the future, a great way to submit issues or odd behaviors like this would be by submitting a bug report through our github issues. If it is something that you see consistently, the sooner the better so we can look into it as quickly as possible.

Hi,

Well I created a support request in July of 2023 asking for assistance on installation and usage of the BL Touch. Unfortunately Lulzbot never responded to this ticket.

If you check my posts, I’ve asked repeatedly about the BL Touch support in Cura LE, the mesh pattern and other questions in the forum. I mean, where else can a customer ask?

And the answer so far from the community have been sympathetic and some advanced users offering ideas to modify the start gcode , namely with the wipe issues. Another XT owner has vented a lot of frustration with his machine and requiring extensive tweaking to get his machine reliable , not to mention the QC issues he has pointed out in this thread. (I don’t know if the machine I have has similar problems)

So telling me I have to go to GitHub is not really a customer friendly answer. Support failed to reply, I went to the forum here and I just don’t know who else to ask? Advanced users could certainly drill down further.

But let’s be honest here. This machine is designed for print farms , schools, prosumers and others to reliably mass produce prints (large!). If a customer encounters issues, they shouldn’t have to wonder when someone at the company will address the problems. At $6500, this is a premium machine. A customer shouldn’t have to watch his investment sit idle while awaiting a fix.

I think this machine has great potential. Unfortunately the software (Cura LE) is several version behind the newest by well over a year and the firmware , as you have pointed out, certainly needs to fixes.

LulzBot wants this machine back very soon. I understand. But I’d sure like to see this XT work to its max potential. And crank out a few great prints before I have to ship it back.

I had a great chat with John (CEO) about my questions and concerns.
Sometimes a telephone call does wonders since text/email/forum posts sometimes come off bland or without the intended emotions (humor and such!)

I wasn’t able to get any fixes or updates to resolve the problems I’ve detailed and it was time for the printer to head home. I packed it all up and dropped it off at the UPS Store today.

So I’m down to one Lulzbot machine, the Taz 6, for now :slight_smile:

Were you able to print up anything of significant height successfully? I don’t recall any posts about them.

I asked about this before I bought it, and to nobody’s surprise they were glowing about its abilities:

"Thank you for reaching out to us. I can confirm that the Pro XT’s stability is still exceptional despite its taller size. Our development team worked very diligently to provide a larger format printer without sacrificing the quality people have come to expect from LulzBot. "

Fine, I said. But, it’s not that… great. The Taz Pro platform is pretty wobbly along the y axis due to the larger footprint along the x axis. Imagine it but twice as tall in the z direction. You can mitigate it of course by significantly lowering the machine acceleration and feedrates. And by lower, I mean to almost comical slow. But you really only need to do that around halfway up - but I don’t really know how to do that with Prusaslicer. But anyway, yes, it prints… OK. But not exceptionally. And for the price I could have purchased a Raise3D Pro3 or Modix printer of similar size.

I only print things on it that use larger nozzles, so it really can’t go “fast” anyway.

I am going to heavily modify it, if for no other reason, I have way too many tool heads and enjoy the ability to swap them out quickly. This is the few remaining graces I have with keeping these things around.

Large prints - yes, I did some but it was a good deal of struggle with Cura 3.6 a year ago and the 4 version very much a beta back then. My prints were using the latest version of Cura, especially for tree supports. You helped provide some potential Start Gcode workarounds to use the latest version of Cura. I didn’t try that, trying very much to stick to Lulzbot solutions like most customers would. (Remember I’m reviewing this machine so I’m avoiding doing advanced things most consumers wouldn’t dare do)

Here’s a print of Threepio’s (C-3PO) thigh piece. It took two tries since the it got wobbly and failed up high on the first. It was a 3 day print and I think an enclosure containing the heat would’ve made the difference

So - it can do some good big prints. Adding the BL Touch still gave good prints, but any time I did the backlash calibration, off the pad it would go.

I’m still very much hoping there’s some way the LulzBot machines can be part of latest builds of Cura. I get they have their own very unique gcode start with pre heating and such. I don’t think that’s entirely necessary anymore - maybe in 2014 :slightly_smiling_face:

The pre-heating for wipes is relevant again with load cells on the nozzle being the new cool thing. Also, pre-heating the next hotend should, again, be important with PrusaXL supporting multiple toolheads, TriDex being a thing as well. They were ahead of their time, then behind the times, now they’re about right again… Lulzbot is a broken analog clock? what’s next, coreXY gets dropped for cartesian again?

You’ve just described the direction Bambu Labs has been going in for the last 6 months.