Titled prints

[LASERMAN]

Greetings!

I have been running the crap out of my V1 Max for a few months now. If I'm home, it's printing something. I had a heater core go out, and in the course of trouble-shooting I replaced the Rambo board. I am now using the original board again. After I got the new heater core in, the machine now prints with a distinct "tilt" towards the right. I have the TrickLaser carbon arms, and all the mechanical stuff looks fine (belt tension, cheapskate adjustment, etc). I am out of ideas as to what could be the issue. The only other thing I can think of doing is to replace the stepper motors.

Any ideas?

[Hansen]

I had a similar problem with my V2. Did you check if the towers are right angled to the print bed? I had to loosen the tower screws, move the towers a little until they were all perpendicular to the print bed, and then carefully tighten the screws again.

[Polygonhell]

Assuming it's actually level within you home, i.e. Check your endstop calibration.
The usual culprits are,
1. Loose grub screw on the stepper pulleys
2. Tight or binding UJoint
3. Tight or poorly adjusted cheap skate
4. Damaged belt, check the teeth on the inside of the belt about half way up when homed
5. Excessive belt tension
6. Too little current to the steppers
7. Too much current to the steppers

My first guess knowing nothing else would be mechanical rather than stepper current, there are too many people with identical setups running the same settings. I think the easiest ones to miss are belt damage, which is usually caused by the head hitting the bed at some point, and excessive tension, there is a point where if the belt is too tight the stepper bearings bind, which will cause issues, and isn't very good for the steppers.

[LASERMAN]

Thanks for the replies!

I checked the alignment of the towers. I have a digital inclinometer - they are all within .4 degrees of each other and the bed.
I checked for belt "dust" underneath the motors - I'l have to actually look at them I guess.
I re-adjusted the cheap-skates - they were a little loose, but not terribly so.
I'l check the other stuff you mentioned.

I have had this thing running great for awhile now. I pop a fuse, replace the heater core, and all of a sudden I get this issue.

[LASERMAN]

[quote="Polygonhell"]Assuming it's actually level within you home, i.e. Check your endstop calibration.
The usual culprits are,
1. Loose grub screw on the stepper pulleys screws are tight
2. Tight or binding UJoint switched to the Tricklaser carbon fiber set-up - no play here!
3. Tight or poorly adjusted cheap skate RE-adjusted them - they weren't off by much...
4. Damaged belt, check the teeth on the inside of the belt about half way up when homed Inspected belts - they look good
5. Excessive belt tension re-adjusted them a little looser although they worked fine before
6. Too little current to the steppers They worked alright for quite awhile
7. Too much current to the steppers They worked alright for quite awhile

I also disassembled the effector platform, since I saw signs of excessive heat on my light shield.....no visible problems there either.
I am running out of things to check. I did notice that the prints seem to start out OK (first 2mm) then seemed to "step" to one side, print OK for a mm or so, then "step" again. Always to the same side. I did a "spiral vase" print and I did not see this.

Right now a motor swap is the only thing I can think of to do.

[dtgriscom]

Sounds like someone had a similar problem a couple of weeks ago.

Does this match what you're seeing? (The other thread's OP disappeared without further comment.)


Dan

[LASERMAN]

Hey Dan!

I read through that thread - kinda sounds like my issue. Mine seems to be in the X tower. I looked in the config, and I only see 1 place for steps\mm - I don't see a separate value for each tower.

[dtgriscom]

Things to check:

- Try using Repetier-Host to move the head some fixed distance up or down. Do all three CheapSkates move the same distance?

- Were you able to calibrate such that "Z=0" was at the same height anywhere on the bed?

- Once you calibrated, are the three end stop screws in similar positions (e.g. one isn't a half inch longer than the other two)?


Dan

[LASERMAN]

I'll get out my calipers and measure the movement.

I can get Z=0 all round the bed, and the screws are very close to the same height.

I'm getting about 2.5mm drift for every 37mm height.

[dizzy1]

I am having a very similar problem. All my print for the last week have been tilted like the linked pictures. The cube's tilt is reproducible. I have 7 of them, 3 that interlock with each other.

1. Loose grub screw on the stepper pulleys screws are tight
2. Tight or binding UJoint checked all 3 times
3. Tight or poorly adjusted cheap skate all have been adjusted to be both tighter and looser and it doesn't change
4. Damaged belt, check the teeth on the inside of the belt about half way up when homed belts look like new apart for a little discoloration
5. Excessive belt tension belts are tensioned according to the assembly guide
6. Too little current to the steppers current is at the default 175 and has not been changed
7. Too much current to the steppers current is at the default 175 and has not been changed

I do have a second hotend but no motor connected for it. I had a print run with these firmware settings about 2 weeks ago and it ran fine. The bed is level and the first layer go down fine. The endstop screws are within a couple millimeters of each other. Mine also seems to be roughly tilted from the x tower but the belt is about the same tension as the other ones and the ujoints move freely. I did have one day that I had to replace the fuse for the bed but that's all I've done concerning the printer. I have also tried printing with gcode from kisslicer, slic3r and simplify3d.

[img]http://freeter.me/IMG_20140710_004504.jpg[/img]
[img]http://freeter.me/IMG_20140710_004517.jpg[/img]

[LASERMAN]

I also had to replace a fuse - the F3 one. That is when my issue started.

I am now switching the motor (and end stops) on the Rambo board to see if the issue "rotates".

[dtgriscom]

Those pictures really helped: thanks.

The tilt isn't continuous. It looks like there was a some per-layer-shift early on, and then it increased, but it eventually got better (and it may even have been gone in the last 30% of the print).

I think what's happening is, once in a blue moon, and only when going up, one of your tower steppers is losing a step. It's the tower that the lean is AWAY from.

I bet if you printed three calibration cubes, widely spaced, one near each tower, plus one in the middle, you'll find that one tower's cube will have squished layers, and it will be the one that the middle cube is leaning away from. Or, if you could stop a print after some lean without re-homing (don't know how you'd do this), and then used Repetier-Host to go to Z=0 near each tower, you'd find that the head hit the bed near that tower.

I don't know much about tuning stepper motors, but my guess is a new tuning script would help in this case. It would move the head to Z=0 next to a tower, do rapid Z movements (as fast as you'd get while printing), and then return to Z=0 at that tower. The problem would appear if the head started hitting the glass. You'd then play with the stepper variables. I'll bet that a small change in the wrong direction of one of the variables would make the problem a LOT worse; the solution would be to adjust that variable in the other way.

Would one of the pros step in and comment?


Dan

[LASERMAN]

Those pics aren't mine -someone else is having similar issues.

I couldn't finish my rotation test - sometimes my job gets in the way of my obsessions
After I get home I'll try your suggestion - print one at each tower and one in the center.
I may swap the stepper motors around if the above two tests aren't conclusive.
I was thinking of going to 0.9 deg steppers anyways - do I need to have the extruder stepper the same as the tower ones?

[dtgriscom]

I may swap the stepper motors around if the above two tests aren't conclusive.

The problem may or may not follow the stepper. But, my guess is that tuning the stepper control variables is what you'll need to do in the end. You might want to wait until someone with more stepper motor experience than me weighs in; there might be a simple solution that just involves some EEPROM changes.

And, with the effect being as tenuous as it is, it could go away at any moment. If it does, though, you'll never know when it might show up again. So, be careful not to accidentally make it go away.

I was thinking of going to 0.9 deg steppers anyways

My gut is you should figure out this issue before you introduce another variable.

do I need to have the extruder stepper the same as the tower ones?

I don't know, but I'll guess not.


(Lots of guessing,
Dan)

[Eaglezsoar]

The stepper motor on the cold extruder (EZStruder) does not have to be the same as the axis steppers.
What is important is the current and the steps per mm are set up properly which they should be if you used
the firmware that came on the Rambo or upgraded using a version of Repetier that was on the Seemecnc site.

[dtgriscom]

Another idea (since I'm not the one who does the work): add a small weight to each of the CheapSkates and see if the problem gets worse.

[LASERMAN]

Another idea (since I'm not the one who does the work): add a small weight to each of the CheapSkates and see if the problem gets worse.

I like that kind of test - non-invasive and easy to implement!

I'll try it tonight.

[LASERMAN]

UPDATE: I didn't get much testing done last night, but I did increase the stepper current. I saw the towers were set at 175 and the extruder was set to 200. Since they are the same stepper, I figured it would be safe to increase the towers to 200.

The test cube looks much better. It still has the slightest bit of "drift". I didn't get a chance to run anything more complicated.
I hope I can get through this issue this weekend.

I noticed the steppers are rated at 2.5A. How high can I safely go in the configuration.h? Is there a chart somewhere to link the number they use with an actual current value? I read somewhere that 185 is about 1A...

...And where can I get replacement motors - SeeMeCNC seems to be out of stock.

[LASERMAN]

UPDATE: further testing...

I swapped the motors between the X and Y towers, since the problem seems to be tilted away from the X tower.

No change.

I did a complete re-alignment of all towers. They were not really off, but now they are perfect.

No change.

I did notice some play in the TrickLaser arms - They attach to a threaded rod, and the rod itself had considerable play on one side. Another side had a loose nut that allowed the bearing to slip. I tightened everything up.

Still no change.

Anything more complicated than a test cube shows this drift. Increasing the speed makes it worse.

After running successfully for a few months, this is getting very frustrating.

[Eaglezsoar]

UPDATE: further testing...

I swapped the motors between the X and Y towers, since the problem seems to be tilted away from the X tower.

No change.

I did a complete re-alignment of all towers. They were not really off, but now they are perfect.

No change.

I did notice some play in the TrickLaser arms - They attach to a threaded rod, and the rod itself had considerable play on one side. Another side had a loose nut that allowed the bearing to slip. I tightened everything up.

Still no change.

Anything more complicated than a test cube shows this drift. Increasing the speed makes it worse.

After running successfully for a few months, this is getting very frustrating.

Do you have the rubber bands in place on the Trick Laser arms?

[LASERMAN]

UPDATE: further testing...

I swapped the motors between the X and Y towers, since the problem seems to be tilted away from the X tower.

No change.

I did a complete re-alignment of all towers. They were not really off, but now they are perfect.

No change.

I did notice some play in the TrickLaser arms - They attach to a threaded rod, and the rod itself had considerable play on one side. Another side had a loose nut that allowed the bearing to slip. I tightened everything up.

Still no change.

Anything more complicated than a test cube shows this drift. Increasing the speed makes it worse.

After running successfully for a few months, this is getting very frustrating.

Do you have the rubber bands in place on the Trick Laser arms?

Yes.

[Polygonhell]

If it's a continuous tilt, it won't be slop anywhere, it has to be either missed steps or something slipping. Probably the former, it could be an overly tight Cheapskate, or belt.
The best bet for diagnosing these things is to move things by hand both with and without the belt in place, given it moves consistently one way, you're looking for something that feels different on one of the towers.
Another thing you can try is resting your hand on the stepper while printing, you can normally feel a stepper miss.

What are your accelleration and Jerk settings for the axis?

[LASERMAN]

All acceleration and jerk settings are set to the "stock" settings...Nothing has been changed since the last normal operation.

I guess I can swap the belts and cheapskates. I do a lot of trouble-shooting in my job - usually swapping is the easiest. When the problem moves, you know you've got it.

I took apart a laser system for cataract surgery today - and put it all back together too. And THIS is what stumps me....

[LASERMAN]

UPDATE: Still tilted....

I changed out the X belt, since the tilting seems to be away from the X tower. No change.

I'm going to recap so I can get it straight in my head.

I have a tilting in all my prints. The tilt is away from the X tower. It gets worse at higher speeds.

I swapped the plug locations on the Rambo. If the card was the issue, the problem should have "rotated" on the bed. Nope.

I swapped the motors between the X and Y towers. This also should have caused the tilt to rotate. Nope.

I re-aligned all towers. No change.

I re-adjusted all cheapskates. No change.

I took a little slop out of my linkages (I have trickLaser arms and rubber bands). No change.

I can print a test cube at normal speed, and the tilt is barely there. When I try to print anything complicated, the tilt is more apparent. When I try to print at faster speeds, the tilt is more apparent.

In all my adjusting\swapping\testing, the tilt is always away from the X tower.



I

[dtgriscom]

The tilt is away from the X tower. It gets worse at higher speeds.

I really, really think you're losing steps on the X tower. The "why" is the question, because you seem to have addressed all the obvious causes.

Have you tried another version of the printer firmware?


Dan