9
$\begingroup$

I'm having issues with ripples on the first layer of big flat prints. The initial corner of a big flat print is fine, but then ripples begin to form as shown in the screenshot.

I'm just a newbie, so I was thinking they might have something to do with heat or contraction or something. Normally, I use the default and print with no turbofan on the first layer. When I tried adding fan 20% or 50%, nothing much changed (slight differences in the ripple pattern and area, but that pattern varies anyway).

I also wonder if one strip gets bent, then maybe the rest just follow the bends. As far as I know, my heating plate is working fine, has no serious hot spots, and I'm using a high-quality PLA+ filament. I also tried adjusting the print temperature from 205-220 (the range on the box is 205-230). Nothing seemed to help. I am running a default first layer thickness of 0.3 mm because that is supposed to help adhesion (and adhesion is fine).

The ripples look worse than they feel. They feel fairly flat, only slightly rippled, even though they look terrible! (And I don't know what that weird row with blobs is in the top left of the picture. That only happened once; almost like junk was in the nozzle or the feed gears slipped or something).

I'm running a Qidi Xpro machine, Sunlu PLA+ (wonderful) filaments, bed 50 C, print temp 205-215, print speed 30-40 mm/s on the first layer, and first layer thickness 0.3 mm (normal layer thickness is 0.2 mm). This machine has a direct drive with gears immediately above the nozzle.

Does anyone know why this rippling effect occurs, and what I might to do to correct it? Thanks

UPDATE: I'm adding this info here to respond to several comments concerning bed leveling, etc. (Thank you to those who made comments!)

1) I'm sure that the bed is as level as I can make it because I always go through the cycle twice).

2) Regarding clearance, if anything I worry that my clearance is too small since there is a fair amount of drag on my leveling card under the nozzle. So, there is definitely drag on all three level points, about midrange between the lightest drag and the heaviest drag that makes me think I'm filing off part of the nozzle.

3) I do have two nozzles, so I suppose the problem could show up on one but not the other if the nozzles were screwed into the block to give different heights. But the ripple shows up on both nozzles, always in the middle of the build plate, always in the middle of a big flat print. Corners don't usually show ripple effects. I don't want to believe that my build plate dips in the middle on my new machine, either ... :-) Adhesion is fine on small prints in the middle of the plate.

Here is a picture of the bottom of the piece. A careful examination shows an oscillation in the squished filament segments on a filament thread. Almost like the extruder was oscillating vertically in the z-axis at that frequency, or perhaps the filament squishyness was oscillating at that frequency. Looks almost like a weave pattern, since the squished parts alternate position on alternating lines.

It's worth saying again that the piece feels pretty smooth on both the top and bottom sides, even though it looks awful. I don't know what to make of that.

Ripples on the top side towards the middle of the plate

$\endgroup$
  • $\begingroup$ I'm wondering if you are getting enough "squish" on the first layer ... this would have to do with bed leveling and how close the nozzle is to the bed for the first layer. I cannot see your image at work, but your description got me thinking along these lines. $\endgroup$ – Pᴀᴜʟsᴛᴇʀ2 Oct 22 '18 at 16:17
  • $\begingroup$ If you screw in a new nozzle you need to re-level. I recently changed nozzles and found a difference of a few hundreds of a millimeter, just enough to get a really too thin first layer, which I easily changed by changing the Z offset with M851 and a slightly less negative Z value (as I am using an ABL sensor). $\endgroup$ – 0scar Oct 23 '18 at 22:49
  • $\begingroup$ You could take your 'problem solved' part (which is I think a more complete answer than the suggestions) and post it as an answer yourself. $\endgroup$ – Sean Houlihane Oct 24 '18 at 18:39
  • $\begingroup$ Basically the filament flow is too large for the initial layer height you had, reducing that would have had an instant effect. Your solution is equivalent, by leveling with less drag the height increases for that unchanged amount of flow. An additional Z offset would have given the same result. $\endgroup$ – 0scar Oct 24 '18 at 21:11
7
$\begingroup$

First layer rippling is usually caused by a too low of a first layer height (for the amount of extruded filament).

Are you sure that:

  1. Your bed is leveled as good as possible, and
  2. the initial height between the nozzle and the bed is correct when Z=0 (A4 paper thickness, when moved should be giving some drag), and
  3. the bed is flat. (This is most probably your actual problem!)

To minimize the effects, you could try to:

  • increase the first layer height, or
  • set an additional Z offset in the slicer, or
  • reduce the filament flow for the first layer, or
  • install an automatic bed leveling sensor, or
  • perform a manual bed levelling mesh procedure (if you have Marlin Firmware).

This usually helps fighting these ripples.

$\endgroup$
  • $\begingroup$ please see my updates in the original post. Yes, I have leveled as best I can and there is medium clearance drag. I want to say that "long lines" are associated with ripples, but the long lines on the outer walls never ripple. (But, they aren't laid down side by side like a flat layer is, either.) $\endgroup$ – Kevin Oct 22 '18 at 19:53
  • 2
    $\begingroup$ Yes, this is a nozzle height issue with a little "over extrusion" due a high temperature and low speed. I was tired of seeing this on an uneven surface until I could set the right offset with the auto level sensor. $\endgroup$ – Fernando Baltazar Oct 23 '18 at 4:54
  • $\begingroup$ @FernandoBaltazar indeed, it looks like the build plate is not completely flat, thanks for the addition! Haven't we all ran into such problems at one time? :) $\endgroup$ – 0scar Oct 23 '18 at 19:17
  • $\begingroup$ Yes, the aluminum plate has a little bow like 0.3mm, but I fixed using some glass cover, I was refusing to use them but is the best method for adhesion; on the change I notice that ripples or over wrapping lines but this is due an error on the Z offset between 0.04 and 0.08 mm; now prints smoothly. I Had an extremely training for debugging 3D printers with my Own printer. $\endgroup$ – Fernando Baltazar Oct 23 '18 at 22:16
2
$\begingroup$

The main problem is solved (first layer thickness vs leveled nozzle height).

The following image shows the problem. I was running with a default 0.3 mm first layer (the tooltip setting says a slightly thicker layer helps with adhesion). The build plate was correctly leveled with "midrange" friction on the leveling card at the leveling points.

Problem cause: The midrange leveling height put the nozzle too close to the plate and caused rippling. The first layer thickness was set to 0.3 mm and the thickness of the leveling card was 0.25 mm.

The following image illustrates the problem (and one of the solutions). The bottom right of the image shows rippling. Not knowing what else to do with the "too close" or "too far" or "unlevel" tips in the comments, I just manually lowered the build plate knobs 1/4 turn while the print was in progress. The print began in the lower right. You can see the smooth area where I manually lowered the build plate. Then, to be sure, I raised the plate by restoring the 1/4 turn on the knobs. The rippling returned.

Ripples disappear when build plate is lowered 1/4 turn on the knobs

To further explore the 0-90 degree suggestion provided by profesor79, I changed the slicer degree settings to 0-90 degrees and set the first layer thickness to 0.2 mm, which was equivalent to lowering the build plate knobs by 1/4 turn. I kept the same "midrange" friction settings when leveling. The result was a first-layer print with no rippling.

No ripples with 0-90 degree settings and new build plate distance

Closing Thoughts

From this experience, I think:

  1. 0.05 mm difference between a thickness of 0.3 mm on the first layer and a leveling-card nozzle height of 0.25 mm makes a rippling difference.

  2. Using mid-range friction vs light friction on the leveling card also makes a difference. You don't need much of a height difference to reach 0.05 mm. Maybe even less is required to cause a ripple.

  3. When printing with a first layer thickness of 0.2 mm, tolerances were tight and I discovered a spot on my build plate that had no adhesion because of a buildup of old adhesive. It left a 1/2-inch hole in the 0.2 mm-thick first layer. I also noticed just a hint of ripple in another place on the build plate, which (I think) indicates a tiny magnetic build plate thickness or warp issue of some kind. Hardly noticeable.

  4. I think I will go forward with a 0.3 mm layer thickness to "absorb" minor flatness inconsistencies in my plate. (I have a glass plate but I have never used it because the magnetic plate is vastly more convenient.) But, to compensate for rippling effects, I will also use a "very light" friction amount when leveling the plate to ensure that the nozzle doesn't get too close to the plate on the first layer.

  5. I found that manually adjusting the build plate height during a solid first layer print was a wonderful way to detect, see, and explore all the relationships between plate leveling, plate flatness, first layer thickness, and friction adjustments on the nozzle. It's very easy to immediately see, understand, and adjust all the related settings to get the best print possible from the machine.


Thank you again to everyone who contributed ideas to understanding the problem. It's hard to pick any particular answer because the solution involved multiple ideas, so I have added my own answer to share.

$\endgroup$
  • $\begingroup$ I fail to understand why an attempt to investigate and justify this has been down voted. It is a sad reflection on the community here. $\endgroup$ – Sean Houlihane Mar 8 at 14:25
1
$\begingroup$
  1. The first that I have in mind was connected with an acceleration, so you could play with it (set to half the current value and see the results)

  2. The other source of that could be drive belt that is fiddling a little bit on the motor and idler shaft (visual check for any play on the motor/shaft)

  3. Next one could be connected with some obstructions in the filament path (as this is direct drive, the Bowden tube could add an extra load if it was bent or spool is blocked)

  4. As this is coreXY type printer could you set in slicer filing angles to 0deg and 90deg? That will force booth motors to run in the same time and eliminate a not holding torque on the other motor (or from other hand please check if the other motor gets some play when the head is going diagonally)

$\endgroup$
  • 1
    $\begingroup$ The print speed on the first layer is already slowed down to 30 mm/s (down from 40 or sometimes 50). I will try to slow it more to see what happens. Belts are tight (and it's a new machine too). Prints look good, so slack seems unlikely. Ripples show up on both nozzles and both spools on this machine, so a single-source common element to do with the print bed or settings seems more probable than one of the filament-extruder paths, I'm thinking. $\endgroup$ – Kevin Oct 22 '18 at 19:57
  • $\begingroup$ @Kevin, that sounds right. Interesting case. As this is coreXY type printer could you set in slicer filing angles to 0deg and 90deg? That will force booth motors to run in the same time and eliminate a not holding torque on the other motor (or from other hand please check if the other motor gets some play when the head is going diagonally). $\endgroup$ – profesor79 Oct 22 '18 at 20:24
  • $\begingroup$ thank you for your suggestion - I learned something new about the ability to control first-layer print angles. I will print a test and report back. Please see the bottom image that I posted up above, with the interesting weave pattern that shows up between successive lines. $\endgroup$ – Kevin Oct 22 '18 at 20:59
  • $\begingroup$ @Kevin that looks like lack of adhesion - so probably 0scar's answer is closer to the real problem. $\endgroup$ – profesor79 Oct 22 '18 at 21:01
  • $\begingroup$ What exactly makes you think "adhesion problem?" I guess that means my build plate has a ripple-pattern lack-of-adhesion problem? If so, then I suppose the ripple pattern should show up in the same place all the time. Now I wonder why adhesion would vary in a ripple pattern, and how a ripple pattern would get there... I have the ability to swap my magnetic blue build plate with another plate, which should absolutely have a different adhesion/wear pattern. I will test it out and report back. $\endgroup$ – Kevin Oct 22 '18 at 21:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.