Cannot Get Perfect Calibration

[Herrminator117]

Hello all,

So I've had my RM2 for about 5 days now and have 6-7 "successful" ABS prints under my belt. I say "successful" because I have one issue with every single one. I noticed my corners lifting on my very first print and immediately thought, oh my bed needs to be hotter, or the filament hotter, or blah blah blah. I've tried all the fixes. But I finally figured out what it is last night. All of my lifting corners are on the positive X side (right side) so I used my Bed Center macro and went to manual control with X+/- and found that my beds tilted (or calibration is off). As in when I put the nozzle on the positive X side it begins to get farther and farther from the build plate. And on the negative X side I get maybe 20mm to the left before I'm touching glass.

I recalibrated 3 times, I even tried using the only EEPROM values (which funny enough changing my endtop steps didn't change a darn thing) i have no idea what to try next or what the issue could be. Maybe my towers aren't square? Either way I can't print anything with a large build area until the is fixed unless I want a janky print.

I apologize if this is covered somewhere else, but for the life of me I couldn't find anything.
Thanks for any help you guys can give me!

[KAS]

This would be the first stop: http://forum.seemecnc.com/viewtopic.php?f=36&t=7361

Most likely it is a mechanical calibration issue, either the towers are not 90 degrees to the build table or it's possible that the tower lean left/right is not square. I don't believe that one is covered in the build manual, and a little bit harder to adjust with basic hand tools. Something like the Wixey WR300 Digital Angle Gauge comes in real handy.

[Herrminator117]

Thank you for that post! So much information!

I have to disassemble anyways to pu tin my dampeners, maybe I can take off the bed and use a carpenters square to see if the towers are straight. Ill report back after I work up the motivation to take it apart...

[KAS]

Hopefully you see this before you dismantle. You can remove the steppers and installed the dampers without destruction of the printer. Just takes creative screw driver angles, and the holes already in the mounts. You can reach the pulley set screw from this angle ( image below).


Good luck!


[img]http://i482.photobucket.com/albums/rr189/Onuxis/20150302_201501.jpg[/img]

[Herrminator117]

Hopefully you see this before you dismantle.

Thanks for that! It made my life a whole lot easier!

Sorry for the delay I just went back to school from break so Ive been trying to get everything set back up! Thankfully though when I put in the dampeners I had gone ahead and loosened all the towers and took a carpenters sqaure and straightend everything up. I also went through and made sure I had even pressure on each of the tower belts. And low and behold its flat as can be! Calibrated no problem and printed off something with outta hitch! Thanks again for all the help!

Also as a side note, these dampeners are a god send! I literaly sit within 3 feet of my printer so I needed it as quiet as possible! Now I just need to find a way to silence the extruder motor!

[Herrminator117]

Okay I was very much wrong, I am still having issues. I squared all the towers recalibrated and then I ran the circle test (this one: http://www.thingiverse.com/thing:50505/#instructions)

And I ended up with this:

When looking at the outer circle:
Quadrant 1 and 2 (top right and left) were off the bed with zero adhesion as you can see.
Quadrant 3 (bottom left) was a little high but adheared
Quadrant 4 (bottom right) was too close to the bed

Looking at the inner circle though was pretty much great! The center dot was a little close to the deck but not too bad.

Looking around on the forum I cant seem to pin point the issue. And its messed up every print Ive tried for the past week (for obvious reasons) even after constant calibration. Any ideas? thank you all again

[KAS]

It's honestly still going to come down to mechanical calibration somewhere. Couple things I ran into when building mine. First, none of my framing carpenter squares were not 90 degree square like I expected. You can test this a few ways but draw a sharp line pencil (x-acto line is better) on a piece of paper, flip the square 180 and draw another line beside it. Measure the distance with a set of digital calipers at the base and top to get an idea of squareness.

Next was the left/right tilt of the tower that the manual didn't cover. I ended up purchasing a Wixey WR300 Digital Angle Gauge from Amazon which solved this issue and the one above. It's way easier to zero the gauge to the bed and align each tower while it's clamped to the towers. I'm not sure how you can check/correct without having a digital gauge.

Also, used the depth gauge of the digital caliper to set the heated bed height at each screw to the same distance from the melamine base. This just lowers the chance of any waviness from over/under torquing the screws.

[teoman]

What is the technique for adjusting the towers?

Undo the screws and press push pull curse it in to place or is a more algorithmic shifting and rotating of the entire top assembly in order?

[KAS]

Nope, you had it right the first time. Loosen the top and bottom screws from the towers.
Place the digital gauge parallel to the axis you want to square and zero it out.
Place the gauge on the tower somewhere up high and clamp it.

Curse and get pissed as you perform this same routine over and over while trying not to mess up the 90 degree square in the other direction. Tighten one screw then check the front/back of the tower by re-zeroing the gauge parallel to that axis. Hopefully when it's all said and done, everything will be tight and 90 front/back and left/right to the build plate.

Crappy lighting in this room,

[Herrminator117]

This sounds like a party... But you gotta do what you gotta do! Thanks for explaining that for me!
As soon as I get my scale and have some free time Ill report back if all is well with the world!

Also what is that niffty looking plate on top of the glass on your bed?

[KAS]

Failed experiment for now.....

http://forum.seemecnc.com/viewtopic.php?f=59&t=8071

[Herrminator117]

Alright well it wasnt my towers, I got the digital angle gauge (same as you, good choice, I like it a lot!) But before I even tried to tweak them my max deviation was 1 degree on only one tower. I dont think that is enough to cause this many issues. Also checked my bed height with a caliper compared to the base and ever where I checked was within .2 mm of one another.

So at this point it is either the software or the arm assemblies. Im going to go back through the manual and see if I just happen to miss anything (I was using an older manual at first, but switched to the newer one after I found it)

With the arms, do you think it could have anything to do with the cheapskates? I did have a lot of issues with the Z skate, either too loose and had a bit of wiggle and didnt bind, or it was too tight and bound up.

Thanks for bearing with me guys and helping out, I appreciate it!

[KAS]

My hand calibration with the skates was to adjust the eccentric nut just tight enough that I could spin the wheel and move the carriage up/down. But just a fraction looser and the wheel would just spin in my fingers. I did that for all 6 eccentric nuts.

Hope that's understandable.


1 degree doesn't seem like a lot but it starts to compound when X is closer to Z but further from Y. ( or whatever your towers measured) That's why in your circle print, you were able to print between X and Y but not between X/Z or Z/Y. The Effector starts lifting at odd angles, and in printing between the towers has been a huge headache for many people.

This post might give more insight to what's going on:
http://forum.seemecnc.com/viewtopic.php?f=37&t=4868



Also wanted to add, after you've been printing for a week or so it's best to go back and retension the belts and check all the screws and nuts. Things like the end stops and steppers, from screws under the base can work loose and cause issues when trying to calibrate.

[Herrminator117]

Thank you for posting that thread! A whole poop ton of emotions in there, lemme tell ya!

But one thing that stuck out to me, was when someone had said that they were getting much better prints when they had left their printer hot for 4.5 days. And everyone kinda agreed noting some "thermal equilibrium". And in the past few days when I have attempted to print things, I turn it on preheat and as soon as it hits temps I try to print, and I cant get ANYTHING to stick. Not in the center, not in the outside, nothing. So today I turned on the printer and preheated then left for class. When I came back I tried to print and low and behold its sticking. Now I doubt that it fixes my outter edge adhesion as for right now I am printing inside the happy triangle as I deem it.

Being in college with engineering, Air Force ROTC and all my other college-y duties, I have not had the time to tinker around. But I think when I get some Ill go through, shim the thing as Jim did, make sure my towers are as straight as can be and then see where I sit.

Thank you all for your time and support! I know we all have our busy times, so your time is always appreciated!

[hubrigant]

I'm about to go through and try to resquare the towers, and this thread's gotten me thinking about whether there's a better way. My current thought is to combine carefully machined holders for some cheap laser diodes, one per tower. These point to a set of three mirrors mounted to the effector plate, all of which are arranged based on the arms being all the way to the top.

My thought is that, if the holders are square to the towers and the mirrors are positioned right, once the three towers are squared, all three laser dots will be centered on the build plate. If the diodes are three different colors (red, green, blue?), it'd be easy to see which direction to push in order to square everything up, right?

[Herrminator117]

I think if absolutely everything was perfect it would work, but I think in reality there are such small tolerances, whether it is in the laser cut Melamine or the effector plate being ever so slightly tilted it was cause you massive ammounts of headaches. Maybe you could build the mirror plate so that it sits centered on the bed with the mirrors raised off the bed (like a tall, skinny pyramid stand) and then the diodes on the towers. That takes out the variable of the effector plate.