XY plane not parallel to working surface/calibration issue when drawing

[hoodel]

Rather than the current 1-point 'B' calibration, I'd like to see a 9-point calibration so that the xy-plane can be properly flattened. My arm leans a bit to the right, so I put 2 or three sheets of paper under the right side. With a 9-point calibration we could fix the z as a function of R problem, and also correct for tilt.

[tony]

You mentioned the focus, does it mean height changed several millmeters during the engraving? Since before release, we tested it in our assembly line and it works well. Would you please share me your details about the calibration steps? That would help us to find the issue, thanks

[serdan]

here you go: http://mezhov.ru/v/V70904-233222.mp4

[serdan]

Hi @tony, any update please?

[tony]

@serdan @virtualritz @hoodel Thanks for everyone's feedback, our engineer are working on the multi points calibration right now, they have just finished working on the Grove modules. I am sure this function will be added later on.

BTW, I also find out a potential cause of the horizontal error, please check the parts between lower arm and upper arm like the picture below.

The parts in red circle was bended a bit, although it is very rare but it might cause the horizontal error. The damage was caused during the delivery, and please make sure your robot is fine.
Thanks

[hoodel]

If there is a bend in the part at the elbow, as you describe, it is too slight for me to see it. Another possibility for a bent part might be at the wrist (blury, up and to the left in the photo). I can't tell from looking at that one either. All of the screws seem to be tight.

[serdan]

Thanks. Will be waiting for the new procedure.
You can always contact me to test it on problematic arm before the release.

Don't think the elbow is bended on my arm. Looks solid, straight, and properly assembled.

[tony]

Maybe you can let me have a try, thanks for your help in advance And I am glad to see the part you mention at the wrist, thanks a lot

[tony]

Hi serdan, thanks for your understanding. Would you please show me the picture of the joint, I want to make sure it's ok.
BTW, here is a picture from my colleague. The elbow is completely destroyed.

[serdan]

Thanks for this opportunity to test how sharp is my new camera.

[serdan]

It has been 2 weeks from last reply. I'm very glad you guys have a well deserved vacation soon.

Any chance to provide fix for uArm calibration? It can't print, can't engrave, can't cut from the day of delivery. The basic calibration is not resolved for 2 month already.

[tony]

Just send you the message to confirm your order number, would solve it after the vocation, thanks

[virtualritz]

Guys, any update on the three point calibration?

.mm

[Semertl]

Are there any updates? I have the same issue. THX

[piddhd]

I’m also interested in this calibration solution - sounds like a great idea!

[bloudon]

Hi there, I have also had this exact same issue as reported here from day 1. I have not progressed from drawing to laser engraving or 3D printing as the results are so bad it seems pointless to try.

I am running uArm studio 1.1.15-a1 with uArm firmware 3.2.0

I have tried running the blockly script you suggest on another thread
This seems to work OK but does not pick up on the XY plane misalignment, here is a video of this running;

Trying a drawing in uArm studio shows the pen lifting off issue;

I align the pen with a block under the universal gripper to ensure that the pen is touching the paper;

Here is a video of this running;

As you can see in the video, the pen leaves the page quickly. I have also had runs where the pen digs into the page so much the base starts to rotate!

I have used the readout from the “move to” component in blockly to read off test coordinates on the drawing surface. I have then input these to Rhino3D to look at the best fit surface through these points. This shows a significant angle in two directions relative to the base.

I have tried the calibration routine recommended issuing commands M2019 and M2401 B using the Arduino IDE, but this has not improved the situation at all.

To the naked eye the arm seems mechanically sound with no obvious bent components.

I have never attempted 3D printing or laser engraving with the arm as I doubt it would work given the issues I have just with drawing with the pen. What can you recommend, do I just have a dud arm?

[bloudon]

my videos and photos were blocked apart from one as I am a new forum user

[piddhd]

Hi,

i’m also still waiting for a solution to calibrate the uArm in a way that works. I have not been using the robot for a long time because it is impossible to make good 3d-printing.
It would be nice to get any comment from the uArm staff concerning this issue.

[virtualritz]

The uArm is another failed Kickstarter for me. Great hardware, software has issues that are still not solved after many months. The issue is also not random. I ordered two uArms and they both have exactly the same issue with the XY plane. uArm is only useful for learning and picking tasks. 3D printing, laser engraving or drawing: forget it.

I have used my uArms barely at all because of this.

I am also kinda fed up with uFactory sitting this issue out. I’m about to publish a blog post about this.

[bloudon]

I have created a fuller post with all videos and images on the facebook group, I appreciate you are all on holiday, I hope you get the chance to respond fully after the break.

Edit… fb post:
https://www.facebook.com/groups/1653721141585397/permalink/1929764883981020/?comment_id=1931622497128592&reply_comment_id=1931762843781224&notif_id=1518964324816105&notif_t=group_comment_reply&ref=notif

hope this works…