Close



Page 1 of 5 123 ... LastLast
Results 1 to 10 of 42
  1. #1

    so what causes this?

    so i am printing a kossel mini set for a friend. I am having some artifact on the base pieces around the lettering and i am not sure what is causing it.

    here's a pic


    the left hand side of the model is the base.

  2. #2
    Engineer-in-Training TopJimmyCooks's Avatar
    Join Date
    Jul 2014
    Posts
    204
    Usually called ringing. it happens right after sharp turns like the 90 degree turn from the letter recess out to the face of the part. Due to the Prusa's design not having a huge amount of rigidity in the Y direction. Lower your acceleration rates to help with this. printbus did a great write up of the several places you can set and change acceleration values. 2x in firmware and again in the slicing program IIRC. Also, you can make a small improvement by orienting the parts so the face with the worst ringing issues/best appearance needed is parallel to the X axis.

  3. #3
    Staff Engineer printbus's Avatar
    Join Date
    May 2014
    Location
    Highlands Ranch, Colorado USA
    Posts
    1,437
    Add printbus on Thingiverse
    TJC beat me to it. Sniffle, clarify how the X & Y axes are oriented when you print. When I see ringing, it's usually on the side parallel to Y. Rereading TJC's post, I think he's saying the same thing. It's ringing because it fades away. Also notice how the interval between the ringing pattern increases - this is because the printer is accelerating away from where the ringing started. As it speeds up, the ripples will be farther apart.

    Fine and dandy. I personally think there are issues in Marlin move planning.

    A counter argument to ringing in this case is that we're not at a 90 turn. At least not one where the axis was running full speed before decelerating to the corner. My 90 degree corners come out pretty clean. Where I tend to always see ringing, and sometimes substantially, is on shallow artifacts like your lettering or on holes in relatively thin sidewalls. In the case of your lettering, the print is moving along, stops at the recess created by a letter, shifts slightly perpendicular to the direction it was going, and then reverses direction. At least that's what I've watched my printer do. So, the ringing is supposed to be caused by that little bit it shifts for the depth of the letter recess? That's hard to buy.

    Same thing for holes on a thin sidewall. Just yesterday I printed a small project box with a large hole on a 1.75mm thick sidewall, and got substantial ringing on one side of the hole. The x carriage built up a lot of momentum crossing that 1.75mm sidewall? The 90 degree corners came out far cleaner than the hole. The inconsistency is what causes me to question what Marlin is doing. EDIT: It's almost like the X motor is being twitched during that turn-around in the Y axis - like Marlin is not starting the X movement out slow. The jerk setting should influence this, but again, 90 degree corners come out fine. I really wish my oscilloscope had storage capability so that I could monitor what the driver signals look like for the X and Y motor where I see this ringing occurring. If nothing else, I may rotate that object and reprint it just to see what kind of ringing I get on the hole when it is parallel to X. With the aluminum v-rails on the i3v, I have a hard time understanding how the Y-axis is any less laterally rigid than the X-axis.

    If you want to experiment, the settings involved would be DEFAULT_XYJERK, DEFAULT_MAX_ACCELERATION for the axis perpendicular to the one where you see the ringing, and DEFAULT_ACCELERATION. Details on how these interact are discussed in thread Marlin Motion Related Configuration.h Settings for MakerFarm i3v. It could also be argued that your belts are too loose. Others might argue too tight. Carriage wheels could be adjusted too loose.

    Hoping to hear other opinions on this.
    Last edited by printbus; 01-20-2015 at 04:13 PM.

  4. #4
    Engineer-in-Training TopJimmyCooks's Avatar
    Join Date
    Jul 2014
    Posts
    204
    Rigidity: motion in the x axis is resisted by the primary square frame of composite wood in plane, which is super stiff in that direction. In effect the stops and starts have something firm to push against. In the Y direction the stops and starts are resisted by a very small piece of wood in the bendy direction especially at the front piece where the y idler attaches, hence more oscillation/ringing.

    If I'm right ringing might be reduced a good bit by stiffening that y idler mounting piece. I need to test that. I bet it's less of an issue with the original linear bearing I3 version because that piece was doubled up in that design. Anyone who had both care to comment? or are there too many other variables to compare? From what I've read acceleration/jerk/move planning is a bigger factor, though.

    I also read somewhere that the more the machine is allowed to move overall during printing the less this happens. An argument for rubber feet.

  5. #5
    Staff Engineer printbus's Avatar
    Join Date
    May 2014
    Location
    Highlands Ranch, Colorado USA
    Posts
    1,437
    Add printbus on Thingiverse
    Quote Originally Posted by TopJimmyCooks View Post
    Rigidity: motion in the x axis is resisted by the primary square frame of composite wood in plane, which is super stiff in that direction. In effect the stops and starts have something firm to push against. In the Y direction the stops and starts are resisted by a very small piece of wood in the bendy direction especially at the front piece where the y idler attaches, hence more oscillation/ringing.

    If I'm right ringing might be reduced a good bit by stiffening that y idler mounting piece. I need to test that. I bet it's less of an issue with the original linear bearing I3 version because that piece was doubled up in that design. Anyone who had both care to comment? or are there too many other variables to compare? From what I've read acceleration/jerk/move planning is a bigger factor, though.

    I also read somewhere that the more the machine is allowed to move overall during printing the less this happens. An argument for rubber feet.
    I think we're miscommunicating. In my case the ringing is observed on the print sides parallel to the Y-axis. In other words, the print sides that go from Y+ to Y-. Since the ringing is observed on that axis, the ringing has to be lateral movement on the X-axis. Right? Or maybe not... I guess I can kind of see that if the nozzle is moving in the Y direction and the frame is also ringing inline with the Y-axis, the net effect would be the nozzle accelerating and decelerating with respect to the print, and that could cause an artifact to show up. TJC, is that your theory? Interesting possibility I'll have to ponder.

    I long ago added 1/2-inch wood reinforcement plates to both Y end plates. The 10 and 12 inch printers likely need something like that even more. As of my last maintenance cycle, I've got the belts tighter than ever or I'd focus on that. I've got the Sorbothane feet installed.

    I picked a bad day to do some power rewiring. Might have to kluge things back together to get some test prints underway...
    Last edited by printbus; 01-20-2015 at 04:48 PM.

  6. #6
    Makes me start thinking about some type of triangular frame replacement to print to replace the stock wood.

  7. #7
    Engineer
    Join Date
    Jul 2014
    Location
    Eastern Colorado
    Posts
    536
    Quote Originally Posted by printbus View Post
    I think we're miscommunicating. In my case the ringing is observed on the print sides parallel to the Y-axis. In other words, the print sides that go from Y+ to Y-. Since the ringing is observed on that axis, the ringing has to be lateral movement on the X-axis. Right?
    That is my understanding. I would check the tension of your X belt, and anywhere on the X axis that may be a little springy. As I understand ringing, it primarily happens where the nozzle is undergoing a sharp change of direction, and something springy is letting it "bounce" a little bit.

    With the aluminum v-rails on the i3v, I have a hard time understanding how the Y-axis is any less laterally rigid than the X-axis.
    The X belt is mounted parallel to the wood plates of the X axis. The Y belt is mounted perpendicularly, in between the rails, to single-thickness wood. Didn't you have a modification in your build thread about putting another piece of wood in front because it was too bendy?
    Last edited by AbuMaia; 01-20-2015 at 06:58 PM.

  8. #8
    So to eliminate bounce it wounfs like we need to repl a ce the belts with micro chain...

  9. #9
    Staff Engineer printbus's Avatar
    Join Date
    May 2014
    Location
    Highlands Ranch, Colorado USA
    Posts
    1,437
    Add printbus on Thingiverse
    Quote Originally Posted by sniffle View Post
    So to eliminate bounce it wounfs like we need to repl a ce the belts with micro chain...
    Not necessarily. The belts are supposedly pretty good at maintaining dimension. They're designed for that. But getting the right tension seems to be tricky, sort of like bed leveling. (lol)

  10. #10
    Engineer
    Join Date
    Jul 2014
    Location
    Eastern Colorado
    Posts
    536
    I found it easy to tension the Y belt, I just used a spring clamp to pull the Y motor to the rear, then tightened the motor mount bolts. The X belt was harder to tension, as there's not much room to get a good pull on the X motor. I've been thinking up a redesign of the X belt idler so one can turn a bolt to adjust the X belt tension.

Page 1 of 5 123 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •