1. Got a question or need help troubleshooting? Post to the troubleshooting forum or Search the forums!

Z Axis Artifacts - Known Issues and Fixes

Discussion in 'Mods and Upgrades' started by Ziggy, Jun 1, 2014.

  1. Ziggy

    Ziggy Moderator
    Staff Member

    Joined:
    Feb 20, 2013
    Messages:
    707
    Likes Received:
    530
    No. I don't think anyone is saying that. But there are lots of opportunities to improve the basic Robo.
     
  2. Peter Krska

    Peter Krska Active Member

    Joined:
    Nov 30, 2013
    Messages:
    766
    Likes Received:
    229
    Just to chime in here late, if your going to replace something, then you're gonna run into configuration calibration issues again since the Robo3d printer is based on the parts it came with.

    But if you're going to replace something, do one thing at a time and work on it: software and hardware configuration.

    There is no use in replacing everything and expect to have it work. It's unproven.

    Take the advice of those that have successfully replace their parts and have great success. Such as the extruder and bed rails as well as to the metric rods.

    You might wanna share your STL file and see if anyone else has that ribbing Issue as you describe. There are 3D program that cam further smooth the parts, adding to the file size and resolution. Takes longer to slice but worth the wait.

    I've seen some oft prints come out very smooth and i think it was because of the resolution it was saved as.
    ImageUploadedByTapatalk1417141878.847032.jpg ImageUploadedByTapatalk1417142013.734338.jpg

    The smoothness also depends on the type of filament you use. Some seem to print really nicely and others are a bit ruff.
     
  3. AxisLab

    AxisLab Well-Known Member

    Joined:
    Dec 28, 2013
    Messages:
    322
    Likes Received:
    269
    Who was that post directed to Krska?
     
  4. Kristian von Bengtson

    Joined:
    Nov 27, 2014
    Messages:
    4
    Likes Received:
    0
    Hey. well you are right.. Thanks for the list of steps for improvement. I have just downloaded your document and will try working it out. Thanks a lot! I should do some before and after pics. got some serious wobbling going on. Are the M8 rods really supposed to move like crazy to any side (in the top end) when used by the z-stepper?
     
  5. SteveC

    SteveC Well-Known Member

    Joined:
    Jun 12, 2013
    Messages:
    894
    Likes Received:
    316
    No one said that. Axis-lab is having a problem with his specific machine.

    But if you have Z-ribbing you can eliminate some of it by adding a ball bearing in the couplers as outlined in Ziggy's document.

    (Edit: Sorry, somehow did not see the other replies to this.)
     
  6. nickster

    nickster Member

    Joined:
    Oct 7, 2013
    Messages:
    82
    Likes Received:
    52
    @Kristian - Assume you mean wobble in the ends of the lead screws and not the M8 smooth rods. You should be able to remove the entire Z carriage assembly and run the Z steppers with the Z screws unsupported. If there is any significant wobble, the show is over. Problem is the lateral forces on the Z nut cause a vertical displacement. Also the Z nuts need to be supported on a perfectly flat surface or held snugly in the plastic housing (not as good). The plastic injected parts that touch the Z nuts have some draft. Put a dial indicator on the bed to the hot end and wiggle the Z screws side to side. In my first adventure of using ball bearings, I totally eliminated any Z screw displacement at the stepper, and the ball was also locked in place by the cone I cut into the screw. This had issues though because the free ends of the screws picked up a huge amount of wobble because the metal Z coupler does not expand uniformly. It tilts to one side as the coupler expands to preload the ball. Ziggy's set up did not use the cone in the Z screw, but still that end of the screw needs to be absolutely flat. Probably want just enough preload to make contact and no more. My final fix which I am still using today, has a secondary offset adjustment to allow the Z screws to be centered so the ends don't wobble more than a few mm in free space without the carriage in place. PITA to adjust. Any remaining wobble puts negligible lateral forces on the Z nut. Also my Z nut forms a ball and socket under the carriage. So there is a ball and socket at the stepper and a ball and socket at the carriage.

    http://forums.robo3dprinter.com/index.php?threads/precision-z-coupler-upgrade.1471/#post-10294

    See follow up post #4 in the thread.

    Also apply some science. Measure the repeatability of your Z with a dial indicator and if you wiggle stuff around (really scientific) and see a change in the Z, you're not going to get the best possible print. When I first started "fixing stuff" I had over 0.3mm of Z displacement from wiggling stuff. Now (after the saga) it is rock solid.

    BTW Metric vs English Z screws are an optical illusion of a badly controlled "experiment". Thread pitch makes -ZERO- difference once you get rid of all the slop and wiggle.

    @Peter - Some thing I just discovered on surface quality at least with ABS is to crank the temp up and leave the cooling fans always on at some minimum setting (Slic3r knobs). Idea is to make sure the previous layer has dropped below the glass temp of the filament before the next layer gets applied. Even if layer times are like 90 seconds, keeping that small amount of air moving over the surface creates a crisp underlayer with no deflection as the new filament is being applied. I was tending to turn the fans off with layer times greater than 50 seconds, but slicer has a setting that allows the fans to stay on always at a min setting.

    -Nick
     
    2 people like this.
  7. 1nxtmonster

    1nxtmonster Active Member

    Joined:
    Oct 27, 2014
    Messages:
    118
    Likes Received:
    77
    Thanks for the tips, I changed my layer height to .09878 instead of .1 and it helped greatly on a 15mm calibration cube. Not sure if the picture really shows it, but the sides are super smooth on the right cube, which used the new layer height.



    20141209_181515.jpg
     
    2 people like this.
  8. Quest

    Quest Member

    Joined:
    Jan 7, 2015
    Messages:
    59
    Likes Received:
    9
    Humm.
    The R1 I have that shipped early December 2014 with bottom cover has 7.94mm on the X&Z rods but the Y rods are 8mm.
    When I replaced the z supports I did see some play on both the x and z.
    I guess I need to look into replacing these rods with 8mm ones.
     
  9. 1nxtmonster

    1nxtmonster Active Member

    Joined:
    Oct 27, 2014
    Messages:
    118
    Likes Received:
    77
    I did the firmware modification with the enabling of the Z axis steppers. Here is the result (15mm cube next to a quarter for scale):

    20150209_190930.jpg
    There still is some z artifacts however. They are very close together, and do not match up with the threads on the rod like all of my previous ribbing. Perhaps it's an extrusion issue like mark tomlinson mentioned previously?

    Update:

    I turned down the extrusion rate by a few percent and reprinted the cube. It turned out much better, almost perfect:

    20150209_204659-2.jpg

    So thanks to everyone for taking the time to write up very detailed solutions to the Z ribbing plague.
     

    Attached Files:

    #69 1nxtmonster, Feb 9, 2015
    Last edited by a moderator: Feb 9, 2015
    Jimmy Husain likes this.
  10. Ziggy

    Ziggy Moderator
    Staff Member

    Joined:
    Feb 20, 2013
    Messages:
    707
    Likes Received:
    530
    I could imagine the Z ribbing caused by some over extrusion is happening because of the infill layering/structure inside the cube.

    Otherwise there is no real reason for the Z ribbing to have repetitive pattern.

    What layer, primeter and infill settings were you using when the Z ribbing was noticeable?
     
  11. tonycstech

    tonycstech Active Member

    Joined:
    Dec 16, 2013
    Messages:
    606
    Likes Received:
    196
    Think about it.
    How would over extrusion cause ribbing ?
    It just makes it more obvious to the eye because extra material flows far out of there it should have been making the robbing more obvious.
    Didnt someone already conclude that its an issue with Z rods by looking at the pattern matching up the threads on those rods ?
    Its impossible for extruder to change extrusion rate every 1 mm or so, it just doesnt make sense.

    When your car needs wheel alignment, dont blame license plate. (sorry, thats how my mind operate)
    I replaced rods with metric and problem instantly went away 95%
    5% ribbing still exist but i am more then confident that if i go tweak the steps, it will go away completely.
    I confirmed that by doing everything i could per Ziggy's (i believe) long detailed post way back except for one thing. Every single step proved to by useless.
    If it worked for someone then am assuming either we all or that someone got a rod that was way of the specs.

    Homedepot has those rods for like $3-$5 each.
    No need to take extruder carriage off the printer to replace.
     
  12. 1nxtmonster

    1nxtmonster Active Member

    Joined:
    Oct 27, 2014
    Messages:
    118
    Likes Received:
    77
    But the ribbing that I'm seeing does not match up with the thread pitch. Its' much too fine.

    Screen Shot 2015-02-09 at 10.01.18 PM.png Screen Shot 2015-02-09 at 10.01.24 PM.png
     
  13. tonycstech

    tonycstech Active Member

    Joined:
    Dec 16, 2013
    Messages:
    606
    Likes Received:
    196
    print with standard settings 0.3mm height.
    I also suggest you to try either craftware or cura. They have less settings to tweak that you should not mess with at all unless your filament diameter is way off the chart.
     
  14. Robert Choban

    Robert Choban Active Member

    Joined:
    Aug 22, 2014
    Messages:
    279
    Likes Received:
    27
    Ziggy

    I have read through this post and your document Z Ribbing Issues. I have included some files to show my problem with what I think is Z ribbing I have. Before I go ahead and make your suggested changes I thought I would ask if you still standby your changes and if you have anything you would change or add. If you are still OK with your 5 changes I have some questions before I make them that I hope you could answer.
    1. Firmware Z disable configure change. Have you heard of any problems with excessive heat on the stepper drivers or the motors. Do you suggest turning the stepper drivers or just leave them alone. Did Robo every comment on why they never make this change or are they making it on newly shipped Robo's.
    2. Making sure the threaded rod/nut stiction and friction is as low as possible. Any suggestions on what type of lubrication to use. I have attached a picture of the lubrication I am using, any comments about how good or bad it is would be appreciated. Where do you find good quality nylon nuts that would replace the metal nuts that sit in the X carriage nut traps. Do you know if the metal nuts are glued into the nut traps. Do I need to replace the nut traps.
    3. Setting the motor shaft in the coupler so it does not slip I'm hoping I don't have to do this.
    4. Stopping any height variation due to the coupler geometry (Z Ball Coupler) If I decide to remove the threaded rob, is there an easy way to do this without removing the motor and the x carriage assembly. Could I loosen up the hex screw that holds the thread rod in place, then start screwing it up and out of the coupler. It would still be going up through the nut in the x carriage. I can then check the flatness of the threaded rod assuming it is flat. The next question is how to get the ball bearing centered on the motor shaft and the threaded rod, without disassembling the coupler from the motor shaft, can I drop the ball bearing down the top hole of the coupler and then screw the threaded rod down on top of the ball bearing, will the bearing be centered??
    5. Seating the threaded rod properly in the coupler. I have a 8mm threaded rod, so I hoping this part goes smoothly with some loctite on the threads

    If what I am talking about above will not work, will I have to remove the entire X carriage from the threaded and smooth rods to do some of the above recommendations???
     

    Attached Files:

  15. Ziggy

    Ziggy Moderator
    Staff Member

    Joined:
    Feb 20, 2013
    Messages:
    707
    Likes Received:
    530
    Robert,

    I can see from your pics you have a classic Z ribbing problem. I used to have prints that bad.

    Yes. I standby all these improvements. Only thing that has changed is I am even more convinced that squaring off the rod ends and the ball bearing in the coupler are essential to minimising the Z height variation. I have not had to fiddle with my Z axis mechanics since I made the changes many months ago. I have not even had to apply more lube to the rods/nuts - they are still quiet.

    I would also add that there is some evidence that over extrusion can also cause some degree of Z Ribbing. I haven't seen this myself but I believe it is possible.


    I have had no issues with heat or motors stalling. Since you already have 8mm rods, I don't think you will need to tune the driver current. I have no clue and was never given any reason why Robo have not made the change. It is an obvious problem with a simple solution. There are other examples of simple firmware improvements which Robo have not done - go figure.

    I used 3 in 1 Brand Dry PTFE lube as shown on the Z Fixes write up. If you replace the metal nuts with nylon I suggest you do NOT use white lithium grease. Clean the rods and nylon nuts as best you can and use dry PTFE lubrication. You will find Dry PTFE lube is the best for the metal rod/nylon combination. (White litium grease is more suited to metal nuts and metal rods)

    I don't know if the new Robos have the nuts glued in - but I can't see why they would do that. I found 8mm nylon nuts on eBay - quality was ok. Should not need to touch the nut traps as the 8mm nuts are standard size across the flats.


    I can see you are reluctant to disassemble the printer to do this. You probably won't need to pull the motors out and do all of this. As I mentioned earlier the rod ends being dead flat and the ball bearing in the coupler are the most important. If you still have some residule problems then disassembling and doing all the fixes may be necessary.

    It's actually not that difficult to remove the Robo cover and lift the X carriage up so you can remove the threaded rods. Just takes patience (and making sure you protect the glass bed). Then everything becomes much simpler.

    There is a possibility your Z Motors do not have detent in the shaft where the ball bearing can sit. In this case you can add a small printed bearing seat (I published the stl file for this in the thread)

    8mm rods should sit properly in the coupler with no real need to change anything - UNLESS you can see there is some issue with the coupler which causes the threaded rod to wobble excessively when it turns. Excessively means more than about 5mm at the top.

    Maybe, and as I said, it is not that difficult. But just try the flat rod ends and ball bearing first. Firmware change to stop the motor disabling will also help. That may be enough to solve your Z Ribbing problem.

    I would also suggest as a precaution that you check for over extrusion. Check your filament diameter is correct. Nozzle size is correctly set. Extrusion percentage is close to 100%. Extruder steps/mm is correct (which is very unlikely to be wrong).

    If you need some more help just PM me.
     
    #75 Ziggy, Mar 26, 2015
    Last edited: Mar 26, 2015
  16. Robert Choban

    Robert Choban Active Member

    Joined:
    Aug 22, 2014
    Messages:
    279
    Likes Received:
    27
    Ziggy
    thanks for getting back to me. I don't like the idea of having to take the Z axis motor coupler apart, or completely removing the X carriage, but I also don't like Z ribbing on my parts. So I will probably be removing the X axis carriage and do everything according to your document. Thanks again.

    I have additional question about extruder calculation, I have read a number of post and have review tom's guide on extruder calculation. The one thing I'm not sure about is where you make your first mark, where should that mark be when you start the extruder, should it be right at the hobble bolt, after the hobble bolt, on top where you can see it (see attached picture). I know you measure off 150 mm, then run 100 mm, subtract 50mm, then check to see how much you actually extruded. But again where is the first mark suppose to be when you start to extrude the 150mm of filament.
     

    Attached Files:

  17. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    It is not important as long as you have a reference point to know how far it moved. Use whatever reference works for you, you just want to be able to confirm how much filament was actually consumed/fed.
     
  18. John Huggins

    John Huggins Member

    Joined:
    Aug 22, 2015
    Messages:
    31
    Likes Received:
    6
    Is this report available anywhere else?. Thanks
     

Share This Page