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

Solved R1+ Adhesion for PETG

Discussion in 'Troubleshooting' started by Eric Viglotti, Mar 12, 2018.

  1. Eric Viglotti

    Eric Viglotti Member

    Joined:
    Jan 18, 2017
    Messages:
    116
    Likes Received:
    22
    Hello,

    I am struggling to print something in PETG on my R1+. Per the attached, I am just getting terrible adhesion so the perimeter is lifting off the bed and no longer forming a curve. Here are the specifics:

    - PETG filament
    - Simplify3D slicer
    - First layer is 90% height, 100% width, 30% speed
    - Temp is 240 extruder, 80 bed.
    - Gizmo dorks PEI sheet on the R1+ glass bed
    - Glue stick on top of that because I know that PETG can be brutal to remove, so a great suggestion was to use something like Glue Stick as the release layer

    My problem is that the Z-offset on the R1+ control panel set at 0.65 which appears to be a bit too high based on these pictures, but as soon as I move to down to 0.60, the nozzle rubs up against the print because it's too low. That's only a 0.05 difference. I could try 0.63 or something, but I think this is trying to be too exacting and I think I have to try something a bit more reliable and foolproof.

    I have BuildTak on my R2 v2 bed that I just started using which seems great, but I'm very concerned about not being able to get PETG off at all if I do that.

    What are some options?

    - Rip off the PEI and go glue stick on the glass bed, would that help?
    - Rip off the PEI and go BuildTak and hope for the best?
    - Buildtak and glue stick?
    - Buildtak on an aluminum sheet binder clipped to the bed so I can pull it off and twist it to get the PETG parts off?

    I feel like right now my margin of error is terribly thin that even small tweaks that I have tried won't stay reliable.

    Any thoughts would be awesome!
     

    Attached Files:

  2. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
  3. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    However, that said, it would not normally impact the ability of a subsequent layer to come back over the top. If that is the issue then something else is going on (which is why I asked if you had done anything with the firmware)
     
  4. Eric Viglotti

    Eric Viglotti Member

    Joined:
    Jan 18, 2017
    Messages:
    116
    Likes Received:
    22
    Mark,

    1. Thanks as always. I have never done anything with the firmware since the day I got it. How do I tell what version it is? I have the little LCD screen thing and it says ROBOR1PLUSV2 ready if that means anything.
    2. The big challenge I have is that my print goes from WAY too low to way too high in a matter of 0.05. Someone, perhaps you, said the answer is to have the z-offset set in the GCODE in S3D instead of the printer, would that matter? I just always felt that doing that is a bit of a pain because if I have to fiddle with z-offsets (which I seem to always be doing), I have to scrap a bunch of gcode files and re-generate them all.

    Thanks, and I know this is a tough challenge as the first layer is the holy grail of 3d printing, especially on the older R1+.

    Thanks.
     
  5. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338

    Cool, that is fine for an R1+. Just wanted to be sure. :)

    Your Z offset should ONLY be done in the startup GCode block using the M565 command -- anything else will be problematic. You can't define it on the LCD or even as a global Z offset, both of those are incorrect it is not a global Z offset it is an AUTOLEVELING Z offset (and that is what the M565 does in v1.0 of Marlin)
     
  6. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    If you had swapped to a different version of Marlin then there would possibly be different commands to use, but for the stock firmware ... use the M565
     
  7. Eric Viglotti

    Eric Viglotti Member

    Joined:
    Jan 18, 2017
    Messages:
    116
    Likes Received:
    22
    Perfect. Ok, I guess I will stop fiddling with the Control > Z-Offset and "Store memory" command in the display. I thought that to be handy, but I need to heed advice previously given and now given again :) and stick with the M565. I assume I just add this block to the startup gcode in S3D? (I will start with a really high z-offset and fiddle from there once I see my correct starting point):

    M565 X0 Y0 Z-0.65

    (assuming 0.65 is what I want as the z-offset)

    http://reprap.org/wiki/G-code#M565:_Set_Z_probe_offset

    thanks.
     
  8. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Yes, add it before the G29 and after the last G28 (because every G28 will reinitialize it).
     
  9. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    In later versions of Marlin (1.1.x) replacements for ABL (what you have now) like UBL or MESH do not need to do the sampling/testing every time the print starts and they do store their results in the EEPROM. That is a different story :)

    Honestly I don't suggest upgrading to 1.1.x unless there is a specific issue you know it will solve. I prefer to get it printing and leave it.
     
  10. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    G28; home all axis
    M565 Z-.65; set the offest for autolevel
    G29; autolevel dance

    thats my start up script for my R1+ machines before I moved to a different style firmware. Once your offset is dialed in (mine was not the .65, just using your example) it wont need changed unless you physically change the nozzle, and even then most times its still "ok". If you find a sweet spot and it "moves" it not moving the level switches likely are and you'll want to tighten them up and reseat if needed.

    I run a few R1+ machines and back when stock firmware each ran its own offset and didn't change. Hang in there- this is the hard part.
     
  11. Eric Viglotti

    Eric Viglotti Member

    Joined:
    Jan 18, 2017
    Messages:
    116
    Likes Received:
    22
    Thank you @Geof and @mark tomlinson, you guys are awesome as always. Good news, I was able to get the M565 working perfectly on my 2nd try and hopefully it should be solid from here. I also realized I was forgetting to clean my PEI sheet with alcohol between prints and hadn't in a very long time. I was thinking that wouldn't matter with the glue stick, but the glue stick did seem "tackier" once I had done that. And thank you @Geof for mentioning the potential issue regarding if you have to replace or pull out the hot end, that was going to be my next question!
     
    Geof and mark tomlinson like this.
  12. Eric Viglotti

    Eric Viglotti Member

    Joined:
    Jan 18, 2017
    Messages:
    116
    Likes Received:
    22
    By the way, I tried to run this same exact print and this time it looks fine from the nozzle distance to the bed as it's printing but it isn't creating a full first layer, there are gaps between the passes and it's not sticking. I may tinker with having S3D do a 200% first layer height to see if I can dump out more plastic and see if that helps.

    But I also wonder, since I'm using glue stick, is there any rule of thumb on how much thickness +/- this adds to the bed each time and maybe I'm not putting on the same thickness each time? Just seems rough that it needs to be this perfect, but maybe...

    Thanks!
     
  13. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    If you just want to dump more plastic on the first layer use layer width not layer height in S3D
     
  14. Eric Viglotti

    Eric Viglotti Member

    Joined:
    Jan 18, 2017
    Messages:
    116
    Likes Received:
    22
    Thanks @WheresWaldo. I know, that's what I would think too, and maybe it's operating like it's supposed to, but if you look at the screenshot from S3D and the 2 pictures when I put it at 200% first layer width, there are just big gaps because it is extruding (or thinking it is extruding) really wide passes. And FYI, I stared at the nozzle and boy the height looks absolutely perfect. I actually heard something similar from someone at BuildTak who mentioned to me: " I, too, have never noticed a difference when changing S3D’s first layer WIDTH, but increasing first layer HEIGHT definitely modifies the first layer."

    Anyway, I'm running this print now and I'll see how it does with adhesion and then figure out if these gaps are going to be ok in the final print or not as it's definitely not going to be a nice smooth bottom layer :(

    Thanks.
     

    Attached Files:

  15. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    I can say with 100% certainty that changing first layer height actually changes the first layer height, if for instance you are using 0.20 mm layer height but your S3D First Layer Settings has a First Layer Height of 150%, the result is a first layer that is 0.30 mm. First Layer Width will force more plastic to flow without affecting layer height. But since we are dealing with very small numbers you would need a larger percentage change to actually see an affect, < 10% and you likely will not notice any difference. S3D uses a layer width of 0.48 mm when set to Auto for a 0.40 mm nozzle, A 1% change in width would only represent a 0.0048 mm difference in width. If you do not want to modify the height of layer one, but you want to assure first layer adhesion at the called for layer height, you will likely need about 120% or more First Layer Width setting.

    I never rely on the S3D preview as I have never found it to be 100% accurate. Inspect the actual 3D print of a small test object to make your adjustments visually.
     
  16. Eric Viglotti

    Eric Viglotti Member

    Joined:
    Jan 18, 2017
    Messages:
    116
    Likes Received:
    22
    Thanks so much. So yes, I am using 0.2mm layer height. I went big and went all the way to 200% width in that example I sent and indeed the print was unusable as the print had way too many gaps. What's perplexing to me is that it looks like it's functioning as S3D intended as the print looks exactly the same as the screenshot, i.e. the screenshot told me it was going to print really thick with all of these gaps. So I'm not sure why anyone would want this type of look and I asked S3D and I didn't get a real conclusive response.

    So basically it just seems like every time I adjust the first layer width, it doesn't print plastic any closer together (which is what I want), it just prints thicker stuff that is further away from my first pass leaving all of these gaps shown in the screenshot and print and I'm no further ahead. Is there another way to have it take the normal pass of the extruder along the original route but have it dump out more plastic in the process so that it "flows" out the sides making one pass more likely to touch the other passes?
     
  17. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    You also need to change the extrusion multiplier, not a percentage, it has a scale of 1 so 1 = 100%. Extrusion Multiplier is the only setting that affect how much flow there is directly. And if the gaps are between layer and infill you need to change Outline Overlap in the Infill section.

    One more point, the actual layer height should be in a range between 25% - 75% of the nozzle diameter. So if you have a 0.20 layer and you use 200% as the first layer height, then your first layer will be 0.40 mm which is outside the generally accepted range.
     
  18. Eric Viglotti

    Eric Viglotti Member

    Joined:
    Jan 18, 2017
    Messages:
    116
    Likes Received:
    22
    Ultimately though all of these problems are only on the first layer, so we're talking about adding two processes, one for the first layer and one for everything else? Bit of a pain but I'm game to fiddle to make this work so long as there isn't another easier way :)
     
  19. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Actually you shouldn't need to do that. One single process should handle the issues you are having. Maybe it is just that layer 1 is too tall like I added to my above statement.

    @Geof won't like this, but S3D isn't perfect, there are a lot of dumb things that it does when working with profiles, and there just isn't enough granularity when dealing with layer one. But that said, I use it because it does what I need it to. I would love a combination of Slic3r and S3D then I think it might be nearly perfect.
     
  20. Eric Viglotti

    Eric Viglotti Member

    Joined:
    Jan 18, 2017
    Messages:
    116
    Likes Received:
    22
    Perfect, all makes sense, thank you. I'll give it a try.
     

Share This Page