XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners
 
Notifications
Clear all

XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners  

  RSS
ty.b
 ty.b
(@ty-b)
Eminent Member
XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

My apologies for the long title, but wanted to keep it descriptive

I have gone through the initial self-test, and everything seems to be working. Upon attempting the XYZ calibration, my printer seems to recognize the 4 corner spots (In fact, it seems to be seeing the front and back edge spots), but is clearly missing the spots in the middle (The extruder is crashing into the bed surface).

Additionally, when I auto-home the system, it appears to be -6mm on the x-axis away from the spot (The PINDA probe is clearly outside of the circle). This is really strange to me. The PINDA probe appears to be too sensitive in some locations (Reading the spot from 6mm away), and not sensitive enough in other locations (allowing the extruder to attempt to move through the bed).

Finally, on the last XYZ calibration I did to observe and write this request, the nozzle still crashed into the bed, but no error messages were displayed.

Posted : 24/03/2017 2:14 pm
ty.b
 ty.b
(@ty-b)
Eminent Member
Topic starter answered:
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

I talked to tech support this morning, and they asked that I post a video. It's the XYZ calibration process, the problem happens around the 20:41 minute mark (The video is huge, as my phone won't capture at less than 720).

I apologize in advance for the rambling and shaky hands.

Edit: Previous video did not upload

Posted : 24/03/2017 3:58 pm
richard.l
(@richard-l)
Member Moderator
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

I talked to tech support this morning, and they asked that I post a video. It's the XYZ calibration process, the problem happens around the 20 minute mark (The video is huge, as my phone won't capture at less than 720).

I apologize in advance for the rambling and shaky hands.

Video is unavailable.

Posted : 24/03/2017 5:01 pm
ty.b
 ty.b
(@ty-b)
Eminent Member
Topic starter answered:
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

Amended the original post. Video should be available now. Thanks for looking

Posted : 24/03/2017 6:04 pm
gabriele.s2
(@gabriele-s2)
Estimable Member
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

The video is private... You may want to set it to unlisted instead.

- Gab

Posted : 25/03/2017 2:57 am
ty.b
 ty.b
(@ty-b)
Eminent Member
Topic starter answered:
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

Fixed. Video is now public.
I've had more issues getting this simple video uploaded than anything else this week, and it's been a rough week.

Thanks for looking, folks. I really appreciate it.

Posted : 25/03/2017 3:26 pm
Mike B.
(@mike-b)
Active Member
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

Your bed must be going back too far before triggering the home switch. If you have the Y belt holder with the slotted holes, it doesn't look like it was designed correctly. I went to print this part thinking I could use it to fix a problem I thought I was having but the part doesn't look correct. Wall is missing at one end.

Here is an interesting blog on the PINDA probe: http://www.prusaprinters.org/first-printer-to-automatically-correct-geometry-in-all-axes/

Posted : 25/03/2017 5:06 pm
richard.l
(@richard-l)
Member Moderator
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

It would actually be helpful to remove the paper after it detects the first four points. It it really hard to tell exactly where it is probing in relation to the bed.

Posted : 26/03/2017 3:05 am
gabriele.s2
(@gabriele-s2)
Estimable Member
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

Just looking at the start of the video - the probe is way too far from the first point. It's probably triggering because of the heating copper lines in the bed.

I would say start from fixing this - the probe should land more or less on the circle (it doesn't need to be perfect, the calibration will take care of a small offset, but the closer you can get it to be centered the better. You'll basically want to move the Y belt holder part so that it hits the end stop earlier.

Quite a few people have reported issues with this part, I've got the older version of it and I'm not sure why they changed it. Worst case, shim it so that it triggers the end stop correctly then print it and replace it.

- Gab

Posted : 26/03/2017 6:10 am
ty.b
 ty.b
(@ty-b)
Eminent Member
Topic starter answered:
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

Just got home.

Tomorrow I'm going to try to put a M2 screw in place to trigger the endstop. With a screw, I can adjust it, and hopefully get it in the right spot with minimal adjustment.

I'm surprised I can't A) adjust the location within the firmware, B) the auto-calibration doesn't look "further" for the correct calibration point. It seems like it scans a significant area when it is calibrating the 9 points.

I'll post up about my progress tomorrow. Thanks for the suggestions.

Posted : 27/03/2017 2:53 am
ty.b
 ty.b
(@ty-b)
Eminent Member
Topic starter answered:
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

I drilled a 2.5mm hole in the surface that contacts the Y-axis-endstop, and put an M3 screw in there (with two nuts as backers) to contact the endstop. When running the XYZ calibration, the system seems to "see" all of the endpoints, but ultimately the final report is:
"XYZ Calibration compromised. Left front calibration point not reachable."

I just advanced past that message, and got another one regarding first layer height. I'm re-running the calibration, because I swear that the PINDA probe is "seeing" the first spot (left front), and is detecting it over the partial circle.

It is detecting the front left spot, and appears to be right over the calibration circle.

Going through the troubleshooting guide, it suggests that the frame is skewed. I'm rerunning the test now, but the left side rear frame block and the frame, is now around 96-97mm, and I can't really "slide" it anymore without taking a wrench to the nuts.

Any additional thoughts?

Posted : 27/03/2017 9:56 pm
ty.b
 ty.b
(@ty-b)
Eminent Member
Topic starter answered:
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

Most recent test:
XYZ calibration failed. Front calibration points not reachable

Followed by:
XYZ calibration all right. X/Y axes are slightly skewed. Good job!

I'm suffering from mixed feelings... It seems to me that the system is seeing the front calibration points.

I'm going to try the next step.

Posted : 27/03/2017 10:44 pm
vitor.j
(@vitor-j)
Trusted Member
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

Sometimes it's a pain to get this printer set up 😐
I also got that error. I tried so many things and had so many problems that I cannot honestly say what fixed it.

Some things I remember:
When you get the X carriage to the top (when the LCD instructs it), check that both screws are touching the top mounts. Sometimes, one of the screws touches the top and rebounds. In this situation, the X carriage will be lower on that side.
Check the pinda probe. The bottom nut likes to slide down. I used hot glue to fix it in place. The new MK2S kit has a different mount.
Check the selftest function. It allowed to me notice that the belt was not tight enough.
Check that the bed and the hotend assembly slide easily (disable motors in the menu), with little force. Clean the smooth rods and check the bearings. I also used oil to make the sliding easier. It also makes them quieter.
The the bed belt alignment.
When you have everything setup, try not to move the printer, or at least tie it down to a board, and move it instead. If you don't, you risk messing with the aligment.

Don't give up. I've been there more than once. Today I failed another print due to "operator error" :mrgreen: But the part that printed ok is awesome hahaha

Posted : 28/03/2017 2:34 am
ty.b
 ty.b
(@ty-b)
Eminent Member
Topic starter answered:
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

Good tips (all of them), but unfortunately I've either done them, or checked them. I've had some experience in the past.

Last night, I went ahead with a test print (Treefrog, it's a tradition at this point), despite the printer stating it can't find the front calibration points. The print was finished this morning, and it's gorgeous. This is a small (1mm x 1mm) underprint at the crown of the head, but a portion of the back is insanely smooth.

I don't know what the deal is with that error message. The printer is going through a pre-print calibration without raising any errors. I'm almost tempted to rollback the firmware, but this system seems to be incredibly dialed in.

Posted : 28/03/2017 9:04 pm
ty.b
 ty.b
(@ty-b)
Eminent Member
Topic starter answered:
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

More updates!

Ok, the printer was performing almost flawlessly (When I did a manual XYZ Calibration, it would report that the front edge was not recognized, but when I attempted to print something, it went flawlessly).

I printed the geared bearing included on the SD card, and it's gorgeous. It's so nice that I had a couple of requests to print one for folks. I came home, pulled the treefrog off the bed, cleaned off the perimeter and starter feed, and set the printer to print another bearing. When I checked on it sometime later, it had printed about 4mm worth of layers, and seemed like it had extracted one of the gears when it stopped. No tangled mess of the spaghetti monster, no error messages. It's like the printer hiccuped and decided to stop (I've never had that happen with any of my printers).

I attempted to print again, and the system stopped after it tried to read the front right calibration point, stating it could not see the calibration point.

I ran through a self-check, which passed.

I ran through another XYZ calibration, and the system crashed the extruder into the middle right calibration location (the PINDA probe apparently did not read the print bed). I took a break at this point, and stepped back.

My latest attempt has been to run the self-check, which is now failing on the Z endstop (PINDA probe). Where my PINDA probe was sensing the tip of my screwdriver (Same one that I've used for other tests) around 2mm before, it is now sensing it at less than a millimeter, and only at a specific point on the probe.

What does it look like when a PINDA probe fails (I already have another one on the way from my earlier problems).
Has anyone else had a PINDA probe fail in less than a week? (essentially out of the box)
Is there a way to extend the life of these things? (Beyond keeping them from crashing in to things... I've never had the probe lower than the extruder tip).

Thanks folks!

Posted : 05/04/2017 9:16 pm
vitor.j
(@vitor-j)
Trusted Member
Re: XYZ Calibration Problem, 6mm away from calibration spots, but PINDA still seeing corners

Any more updates?

Posted : 12/04/2017 1:22 pm
Share: