Notifications
Clear all

Looking for help  

Stránka 1 / 2
  RSS
lee.v
(@lee-v)
Active Member
Looking for help

Took ULTRA care building the Prusa i3 MK2, had no problems at all. I've measured, remeasured and remeasured everything many times and the machine is perfectly square. XY is perfectly square. Z height is exact all around. I can't find a single flaw with it. Self test OK, Calibrate Z OK. All firmware was updated when I got the machine. When running the XYZ calibration, everything seems to run really smooth. THe probe hits the center of the 1st and last dot exactly. But when it returns home the probe is to the left of the circle on the bead. (Is that ok?) Then the error message:XYZ calibration compromised. Left front calibration point not reachable. I've measured the distance of the M10 nuts and they are EXACTLY the same. Is there a distance that the double M10 nuts need to be from the orange feet for example as a reference, or if you can tell me how far those nuts should be from a certain point for me to use as a reference. I kept the 100mm distance from the correct nut/washer as in the manual.

Then clicking the orange dial again, I get a second message: X/Y skewed severely. Skew will be corrected automatically. I don't know how that is possible. Everything moves freely and everything is alligned perfectly. Could the PROBE be in the wrong place. I have the clear cover of the probe end lining up with the tip of the orange holder where the cable goes. The build went perfect but this is causing trouble and I just can't figure it out. Thanks

Napsal : 10/10/2016 8:37 pm
David T.
(@david-t)
Noble Member
Re: Looking for help

Are X and Y belt pulleys tightened well on motor shaft? What is height of the probe relative to the nozzle = vertical distance of nozzle tip and probe tip?

Napsal : 10/10/2016 8:38 pm
lee.v
(@lee-v)
Active Member
Topic starter answered:
Re: Looking for help

X and Y belts are nice and tight, perfectly straight, nice ping to them. The PINDA probe is about 2 mm higher than the nozzle. I just tried the V2 calibration code print from the SD card and the PLA filament came out but did not stick to the bed. It was very nice and fine but dragged around the print area. I have something that is not quite right yet.

Napsal : 10/10/2016 9:06 pm
David T.
(@david-t)
Noble Member
Re: Looking for help

I don't mean belts, but belt pulleys. http://manual.prusa3d.com/Guide/3.+X-axis+assembly/105 Step 13, "Tighten up the pulley".
The same for Y axis.
If the pulley(s) got loose, calibration will most likely get unexpected results.

Also try to lower the probe a bit (about 0,5mm) and try again.

Napsal : 10/10/2016 10:06 pm
lee.v
(@lee-v)
Active Member
Topic starter answered:
Re: Looking for help

Ok, I'll check both those things and see how that works. Thanks.

Napsal : 10/10/2016 10:17 pm
lee.v
(@lee-v)
Active Member
Topic starter answered:
Re: Looking for help

All pulleys are good too, tight, belt moves well over them. I tried the PINDA probe higher, then the nozzle hit the bed, lowered it, and things seem fine. The manual shows the tip of the clear cover of the probe in line with the orange clip on the extruder assembly where the cables are held. Mine is now lower than that. I would say the black tip of the pinda probe is exactly the same height as the extruder nozzle. All tests perform perfectly except the XYZ calibration. Gives me the same two error messages.

XYZ calibration compromised. Left front calibration point not reachable.
X/Y skewed severely. Skew will be corrected automatically

Totally lost as to what to do now, the unit seems perfectly built. Measures perfectly square all over, heights are exactly, pulleys good, everything moves smoothly across x, back and forth y and up and down z. just seems to fail the xyz calibration. Have no idea what to do next. My firmware that came with the machine is 3.0.8 if that has anything to do with it. The only thing that doesn't seem perfectly right is after performing the test the probe to me should go back on the center of the first position circle, it is off to the side......unless it is supposed to do that. It starts well in every test and goes a little beside the circle in the other tests and the LCD says ok, but this xyz fails. The z alone passes. Do I need xyz to work perfectly?

Napsal : 10/10/2016 10:51 pm
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: Looking for help

Lee

After calibration, the nozzle returns to 0,0 and not the first probe point.

The probe needs to be about 0.75mm higher than the nozzle.

Sounds as though your front left is too far forward; you need to move it back by about 1mm.

Peter

EDIT: did you check the Y frame diagonals?

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Napsal : 10/10/2016 11:27 pm
lee.v
(@lee-v)
Active Member
Topic starter answered:
Re: Looking for help

Sorry if I sound ignorant. I turned the machine on its side. Using the 4 orange bases with the cushioning on them. One side diagonal is 38.4 mm, the other is 38.2 mm. If that makes a difference, I just don't know what to adjust to straighten it.

When your comment says: Sounds as though your front left is too far forward; you need to move it back by about 1mm:

What would I adjust to do that? Sorry, I am just not sure what to move to make these changes. Appreciate the help though.

Napsal : 10/10/2016 11:36 pm
lee.v
(@lee-v)
Active Member
Topic starter answered:
Re: Looking for help

Ok, I loosened and re tightened some of the nuts on the Y frame and it is corner to corner 38.2 mm from ends of the orange feet with the soft pads, that seems perfect now. Not sure if the 2 mm would have made a difference. Now I am not sure how to move the "left back 1 mm" unless what I did solved that problem. I'll try again.

Napsal : 10/10/2016 11:44 pm
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: Looking for help

Lee

Apologies, I should have been clearer, but I had to go...

You need to loosen the large nuts holding the Y frame to the Z frame on the left side only - loosen the nut at the front, tighten the rear nut by half a turn or so and then tighten the front nut to grip the Z Frame. This should move the left side of the bed further back and allow the calibration to find the front left calibration point.

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Napsal : 11/10/2016 9:32 am
lee.v
(@lee-v)
Active Member
Topic starter answered:
Re: Looking for help

Thanks for the reply. I have also been emailing the company I bought this from in Canada and getting suggestions. He told me to check that no cables are hitting the bed , the motor isn't either. That works fine. he told me about the nut adjustment as well. I did all that very carefully. The X is square, the Y is square, the Z is exactly the same height, the diagonal of the base is perfect. The machine is perfectly squared up and leveled. The probe is about 1 mm above the extruder and right in the center of the circle in position 1. Everything works in calibration except the xyz calibration. I keep getting the same 2 error messages:

XYZ calibration compromised. Left front calibration point not reachable.
X/Y skewed severely. Skew will be corrected automatically.

There is just no possibly way the x/y is severly skewed, everything is squared up perfectly. He told me he would contact Prusa directly and try to get me an answer. The machine simply won't calibrate.

Napsal : 12/10/2016 3:50 am
Nigel
(@nigel)
Honorable Member
Re: Looking for help

@pjr only 1mm out affects the automatic alignment? I thought the new firmware was able to cope with a skewed bed and correct it? I agree checking bed frame diagonals is a place to start.

Nigel
Life is keeping interested and excited by knowledge and new things.

Napsal : 12/10/2016 4:27 am
lee.v
(@lee-v)
Active Member
Topic starter answered:
Re: Looking for help

Yeah, everything is squared up perfectly. I've seen videos of this thing built so badly that it is beyond belief and the software is supposed to still be able to build correctly. mine is 99% perfectly square and it won't calibrate at all. I'm lost on this one. Maybe I got a dud.

Napsal : 12/10/2016 5:09 am
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: Looking for help

Lee

You could try lowering the probe by 0.25mm.

And a brief explanation for Nigel...

The front 3 probes are positioned such that the Y axis has to travel beyond the origin to reach them (ie negative Y). If the Y frame is not installed in the correct position the Y end stop will trigger before the calibration is able to see the downward slope of the probe sensitivity. As the firmware is suggesting a severe skew and also front left point cannot be reached, but that Lee is confident everything is square, only a very small adjustment is indicated on the front left side and that may well have resolved both issues (please also note that Lee has confirmed that his supplier also suggested the same as me).

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Napsal : 12/10/2016 10:00 am
MrMik
(@mrmik)
Honorable Member
Re: Looking for help

Maybe a faulty print bed?

What are these calibration points made of? Magnets I assume? Maybe the front left one is missing?

Just guessing here.

Napsal : 12/10/2016 2:05 pm
David T.
(@david-t)
Noble Member
Re: Looking for help

Calibration points are only small copper circles printed on PCB together with heating pattern, that's all. No magnets, no metal inserts, nothing. I know it, I discussed it with Vojtěch Bubník (PR developer). There is no possibility of one calibration point missing or misplaced. 🙂

Napsal : 12/10/2016 2:37 pm
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: Looking for help

All explained in this blog post: http://prusaprinters.org/first-printer-to-automatically-correct-geometry-in-all-axes/

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Napsal : 12/10/2016 3:32 pm
lee.v
(@lee-v)
Active Member
Topic starter answered:
Re: Looking for help

Thank you to everyone for the replies and help. I appreciate it.
Hi,
I checked all the connections on the board very carefully again and they are all in the right place. I lowered the probe half a turn, ran the xyz calibration and got the same two error messages. So I lowered it another half turn, ran the xyz calibration again and finally some good news. The first error message went away. Now I am left with the error message:
X/Y skewed severely. Skew will be corrected automatically.

Please advise of what the next step should be to have the unit function properly.

Napsal : 13/10/2016 4:10 am
Nigel
(@nigel)
Honorable Member
Re: Looking for help

Thanks for your reply pjr. I learnt something. My comment was genuine. I am trying to understand and learn. My new mantra.

Nigel
Life is keeping interested and excited by knowledge and new things.

Napsal : 13/10/2016 4:21 am
Omikron
(@omikron)
Estimable Member
Re: Looking for help

Thank you to everyone for the replies and help. I appreciate it.
Hi,
I checked all the connections on the board very carefully again and they are all in the right place. I lowered the probe half a turn, ran the xyz calibration and got the same two error messages. So I lowered it another half turn, ran the xyz calibration again and finally some good news. The first error message went away. Now I am left with the error message:
X/Y skewed severely. Skew will be corrected automatically.

Please advise of what the next step should be to have the unit function properly.

Next step is to print V2_Calibration and do live Z adjustment! Looks like you got your probe close enough for it to find the points and auto-correct the skew!

Napsal : 13/10/2016 8:46 am
Stránka 1 / 2
Share: