Notifications
Clear all

XYZ Length Failures  

  RSS
Orochium
(@orochium)
Active Member
XYZ Length Failures

Hello, I am new here, but thought I would bite the bullet and reach out for assistance.

Before anything else, I want to make absolutely clear that building this MK3S has been a side project. Because I didn't have money to buy a kit directly from Prusa, or a pre-assembled one, I've been buying it one part at a time for over a year. The parts are spec'd, to the best of my knowledge, matching the original equipment, down to the Einsy Rambo and LDO 019-006-025 steppers, and 16T GT2 pulleys. The only exception is an upgraded PSU with a break-out power-panic module. 

Basically, the title of this says it all. I currently cannot get past the Self-Test as it fails immediately on the X-Axis length. I've already done the business of checking width, moving zip ties out of the way, and whatnot. The carriage will successfully go to the ends, and back again, and moves freely.

Currently, if I skip out of the self-test, and go to "Move Axis" in the menu, I can get to the following values manually, before running out of space:

X Axis: 201

Y Axis 168

Z Axis 210

Otherwise, everything works. PINDA Probe, temperatures sensors, PID calibrations, Bed and nozzle heating, all fans, etc.

It's a head scratcher, so I'm looking for input. Would pulley and idler diameters effect it? Even going over the manual, I believe the belts to be tightened and mounted/tensioned right. Everything else runs quietly, and smoothly off the back of the Einsy Rambo trinamic drivers. 

 

Penny for your thoughts?

Posted : 14/06/2020 9:44 am
Orochium
(@orochium)
Active Member
Topic starter answered:
RE: XYZ Length Failures

Possibly, at least partially, answering my own question.

Printing handbook on pg. 75 says values for belts shouldn't be under 240 nor above 300.

Currently X-axis belt status is 220. So that would reason it's nowhere's near tight enough. Gonna look at that then in the morning.

Posted : 14/06/2020 10:06 am
JoanTabb
(@joantabb)
Veteran Member Moderator
RE: XYZ Length Failures

I think you will find that 220 is too tight. 

have you lubed the linear bearings? 
If not now might be a good time

regards Joan

I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK

Posted : 14/06/2020 10:45 am
Orochium
(@orochium)
Active Member
Topic starter answered:
RE: XYZ Length Failures

@joantabb

Hey Joan,

I'll admit I'm a little confused by that. The handbook says belt status should be between 240 and 300. If 220 is too tight, what do the 240 to 300 figure represent? 

I have spare gt2 timing belt so it's not like I can't play with lengths either way.

 

Posted : 14/06/2020 9:19 pm
Orochium
(@orochium)
Active Member
Topic starter answered:
RE: XYZ Length Failures

OH. Gotcha, flipping through the handbook and researching, the numbers are inversely proportional. So 220 is "pretty tight" while something like 255 or above is more the correct value. Hmm. Might be worth it then to re-cut the timing belts and try again. 

Posted : 14/06/2020 9:59 pm
JoanTabb
(@joantabb)
Veteran Member Moderator
RE: XYZ Length Failures

Hi Dracas, 

there are a number of odd things in the Prusa Printer... 

Belt tension, 
Pinda Setting and
Live Z, are probably the most confusing.

Glad you sorted it. 

Joan

I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK

Posted : 14/06/2020 10:23 pm
Orochium
(@orochium)
Active Member
Topic starter answered:
RE: XYZ Length Failures

@joantabb

Well, I completely removed, and recut the belt, and followed the directions, this time trying to leave it looser and...I guess I effed that up again somehow because it's still failing axis length test, and showing belt as 224. I guess I need to make it even looser than that...

Would inside-diameter of the Idler pulleys matter? Is there any way to get a measurement of that, I might be able to print a different housing if the idler I found and am using is too small. Otherwise, recutting the belt again I guess...

 

Posted : 15/06/2020 12:01 am
JoanTabb
(@joantabb)
Veteran Member Moderator
RE: XYZ Length Failures

Did you lubricate the linear bearings? 

I believe the belt tension value is a misnomer,  I believe it is actually a reference tot he resistance to movement of the whole mechanism. 

It's really easy to smear grease on the smooth rods and see if that makes a difference! 

the idler pulley has one bearing, 

I have had these bearings fail...  it shouldn't break on a new printer but you never know, 

the bearing should be central in the pulley, if not it can cause friction

the bearing retention screw can be too tight, causing friction, and the
motor pulley can be too close  to the plastics
the back of the X axis and the ubolts on the y axis, can be too tight... 
and the problem could be a mixture of the above

Best wishes, Joan

I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK

Posted : 15/06/2020 12:53 am
Orochium
(@orochium)
Active Member
Topic starter answered:
RE: XYZ Length Failures

@joantabb

Apologies, you did ask that earlier - and yes, I greased it earlier on. Idler pulley seems to be spinny freely, and nothing is tightened down what I would call tight. When the belts not on, both the gantry and the print head glides freely and without resistance.

Even loosening the belt to the point it's almost skipping teeth I still get that Belt test value of 230. I'm really starting the pulley diameter is what's throwing it off here. These are the pulleys I used:

https://www.amazon.com/gp/product/B07LGTSYQ5/ref=ppx_yo_dt_b_asin_title_o03_s00?ie=UTF8&psc=1

 

Everything else up there, I've tried and have been messing around with, to no avail...

Also went back and reconfirmed the steppers.

X,Y, and Dual Z are all LDO 019-012-015 steppers

Extruder is LDO 019-006-025

 Man, I'm at a loss here

Posted : 15/06/2020 1:18 am
Orochium
(@orochium)
Active Member
Topic starter answered:
RE: XYZ Length Failures

Sorry Joan, I appreciate your help, but looking closer at the stepper motors model numbers, they're different than in the Prusa kit after all, at least looking on their store. Store shows LDO 017 007 006 for the extruder and LDO 017 007 005

Which means I need completely different values for the firmware. Is there a user compilable for the MK3S firmware somewhere or will I have to build from a Marlin scratch?

Posted : 15/06/2020 1:47 am
Orochium
(@orochium)
Active Member
Topic starter answered:
RE: XYZ Length Failures

So just for follow up information...

I decided to try and print a calibration cube, since everything else works; and I think I've figured out why it thinks the X-axis length is wrong. The firmware thinks the printing area is bigger than it is. 

When trying to use the PINDA sensor to bed level, once it got to the far right side, the sensor completely overshot the printing area by about 1-2 cm before crashing down towards the Z-axis stepper and I had to kill power.

Even in terms of 1/16 microstepping, the steppers are 1.8 degree, and are zaribo 220 steppers, they supposedly match the original Prusa drivers but what on earth is going on here? the gantry and smooth rod lengths are correct, so is the frame and MK52 bed

So I guess I'm still fishing for info. This is the latest MK3S firmware; while I keep researching, I guess - trying to see if anyone else has experienced this. If you need me to crank out a video or something of what it's doing I can do that.

Posted : 15/06/2020 6:35 am
Orochium
(@orochium)
Active Member
Topic starter answered:
RE: XYZ Length Failures

Sorry for the stream of consciousness here.

Found the problem.

I originally ordered 16T gt2 pulleys for the steppers with a 5mm bore. What I got were 20T gt2 pulleys of the correct bore and I just discovered that pulling them off and counting teeth.

That accounts for the 1.25x movement factor so will shelf this until the replacement pulleys get here. Thanks for helping out, too, Joan!

This post was modified 4 years ago by Orochium
Posted : 15/06/2020 7:05 am
Orochium
(@orochium)
Active Member
Topic starter answered:
RE: XYZ Length Failures

This will be my last update on this thread, provided for early troubleshooting tips if you run into troubles like I did.

Symptom: X-axis and/or Y-axis length failure, despite proper lubrication and setup of assembly and pre-flight checks.

Checklist:

  • Check X-axis pulley, ensure it is 16T (roughly 0.350" or 9.5mm) and NOT a 20T (roughly 0.475" or 12mm diameter)
  • Check Y-axis pulley, ensure it is 16T (roughly 0.350" or 9.5mm) and NOT a 20T (roughly 0.475" or 12mm diameter)
  • Loosen belts to above 250 under Support --> Belt Status
  • Run "Auto Home" before beginning Self-Test and Calibration respectively

Idiosyncrasies: 

  • Modifying the 3.9.0-3421 firmware to accept the 20T pulley by modifying the steps-per-mm setting in configuration_prusa.h to (80, 80, 3200/4,280) did not correct the self-test failure, as the head continued to overshoot. I had to use Pronterface to send M92 X80, followed by M92 Y80, followed by M500 to store those settings in the EEPROM for the steps-per-mm change to take effect
  • Using the M92 X80 + M92 Y80 trick to fix the ranging issue would allow the Self-test and manual adjustment to pass, but would fail calibration as it would again over-range. Likely firmware would need to be set to 80,80, 3200/4, 280 as well as  sending the M92 X80 + M92 Y80 + M500 command to set it in EEPROM. Review settings with M503
Posted : 20/06/2020 3:59 am
Share: