banding/artifacts on the surface of layers with holes

Having a problem? Post it here and someone will be along shortly to help
Post Reply
klajo
Noob
Posts: 3
Joined: Fri Jan 01, 2016 4:28 pm

banding/artifacts on the surface of layers with holes

Post by klajo »

Hi all,

I have a Rostock Max V2 which is working well, except I see some banding/artifacts on the surface of layers with holes. I was asked to print a bunch of STLs and the effect is especially visible on some of them since there are so many nooks and crannies. I've printed smoother things like helmets without any such artifacts. The effect is somewhat exaggerated by the lighting, but still clearly visible in regular daylight.

I'm attaching a number of pictures as well as two screenshots from cura.

The STLs were sliced in cura and I decreased the outer shell speed to 10 mm/s (I normally print those at 20 mm/s but decreased to see if that helped, it didn't). Resulting gcode was printed through octoprint. Looks like I'm using Repetier 0.91S

Code: Select all

Send: N1 M115*39
Recv: ok 1
Recv: FIRMWARE_NAME:Repetier_091S FIRMWARE_URL:https://github.com/seemecnc/Repetier-091-ROSTOCKMAX PROTOCOL_VERSION:1.0 MACHINE_TYPE:ROSTOCK MAX EXTRUDER_COUNT:1 REPETIER_PROTOCOL:2
  • Areas where there no holes on the same horizontal layers print nicely, see for example the top section of the 4th picture
  • Take the hex-shaped hole to the right in the 2nd picture: the area to the right of that hole is wider all around to the other side, it's not like it's shifted
  • It doesn't seem to be related to which height the hole is at.
  • If it is a mechanical problem, I'd expect a more random result and/or shifted layers. These are... wider/narrower somehow.
  • Is this software/firmware related? Something like a rounding/calculation thing?
I'd be happy for any suggestions. I've tried searching around for solutions but haven't been successful so far.
IMG_20161211_215320.jpg
IMG_20161211_215334.jpg
jonaskuehling_gregs-wade-v3_groovemount.png
IMG_20161212_080052.jpg
carriage_for_traxxas.png
Thanks!

/Klas
Doug68
Printmaster!
Posts: 61
Joined: Sun Nov 15, 2015 11:42 pm

Re: banding/artifacts on the surface of layers with holes

Post by Doug68 »

I've no experience with CURA, but...
But with a largely parallel part, if you can look at the Gcode the X and Y's should be the same as the Z increases with each layer.
So for example get the code for a 50 x 50 x 100 column, see what that looks like and then cut a hole in the column and redo the code and examine again.

My guess is though its not the code but something else instead, at least you'll know though that the machine is being asked to go to the right place even if its not getting there.
Xenocrates
ULTIMATE 3D JEDI
Posts: 1561
Joined: Wed Sep 23, 2015 2:55 pm

Re: banding/artifacts on the surface of layers with holes

Post by Xenocrates »

I've seen similar banding for 3 different reasons.

1st, the slicer is compensating to the wrong side when there are holes (Check the layer view, see if it shows up there. If so switch slicers or play with parameters. Rare).

2nd, the joints in the printer have some slop. If you have metal axles or laser cut carriages on your V2, you may want to upgrade to the latest carriages and arms to try and eliminate this. If you have the metal axles AND metal joints (which I believe was a V1 only thing), I would definitely upgrade.

3rd, the bed is bouncing around in temperature, causing it to flex. Since you have Octo-print, graph the bed temperature and look for swings and pulses. Some later V2's have a central hole that was never supposed to be cut, and when the cooling fan on the power supply turns on, it will cause the thermistor on the bed to drop in temperature, sending power to the bed that it really doesn't need, or the PID loop could be poorly tuned for the current conditions.
Machines:
Rostock Max V2, Duet .8.5, PT100 enabled E3D V6 and volcano, Raymond style enclosure
Automation Technology 60W laser cutter/engraver
1m X-carve router

Sic Transit Gloria Mundi
01-10011-11111100001
IMBoring25
Printmaster!
Posts: 616
Joined: Wed Mar 18, 2015 1:11 am

Re: banding/artifacts on the surface of layers with holes

Post by IMBoring25 »

I'd add another. If there are overextrusion issues or the retraction and coast settings are particularly bad, they'll show up worse where the infill percentage of the total volume is higher including small cross-sectional areas where the slicer might go to 100% infill (i.e. where the region is interrupted by a hole).

I see the bed swings mentioned a lot, but:

1) That would show up as a repeating cyclical pattern.

2) I've been running both my beds bang-bang since the SSR upgrades and I've never seen it.
ccavanaugh
Printmaster!
Posts: 298
Joined: Wed Dec 12, 2012 5:03 pm
Location: Indiana

Re: banding/artifacts on the surface of layers with holes

Post by ccavanaugh »

I've seen the same issue on my prints... always shows up with round holes. Mine are nowhere as evident as yours.

I've tried Slic3r, Cura, KISS, and Simplify 3D. The banding is less noticeable with Cura while using as close as possible settings for each slicer.

I need to test some more, but I suspect the travel direction may be reversed when dealing with the holes, and maybe a less than round nozzle could be the root cause.... just a theory until I play some more. Time to use a gcode simulator with the different slicers.
nebbian
Printmaster!
Posts: 175
Joined: Wed Sep 23, 2015 12:31 am

Re: banding/artifacts on the surface of layers with holes

Post by nebbian »

This happened to me when I had:
a) A long bowden tube
b) Variations in extrusion width*speed (caused by infill changes)

What happens is that when you push hard on the filament (when going fast, or using large extrusion widths), and then push lightly (going slow, and/or using thinner extrusion widths), that stored pressure in the bowden tube pushes a fatter extrudate out. Hence you get the banding.

It took me about 6 months of experimentation before I could tune it down to a level so that it's almost unnoticeable. Using a flying extruder will lower this effect by around 70%.

For now try setting all your extrusion widths to be the same width, and all your speeds to be the same. This will help a lot (at least it did for me).
klajo
Noob
Posts: 3
Joined: Fri Jan 01, 2016 4:28 pm

Re: banding/artifacts on the surface of layers with holes

Post by klajo »

Thanks a lot for all the ideas and suggestions!

changes in coords visible in gcode: I looked at the gcode using gcode.ws and it looks ok (no shift in gcode apparent in 2d view)

arms/carriages: I have the injection molded cheapskates and arms. The printer is from Jan 2016

varying bed temp: I noticed the problem with the hole below the bed during first runs. The firmware/software never (or took forever) thought the bed had reached its target temperature so the print never started. I put some tape over that hole during assembly. I'm thinking the effect would be less connected to the presence of holes if this was the problem?

travel direction may be reversed when dealing with the holes: According to gcode.ws, it always moves counter-clockwise both on layers with and without holes :?

I'll play around with retraction and speeds on the machine and get back to you.

Attaching two screenshots of the area to the right in the second picture for reference.
A layer below the hole:
lower-layer.png
A layer at the hole:
hole-layer.png
hole-layer.png (54.75 KiB) Viewed 5644 times
Thanks again!

/Klas
klajo
Noob
Posts: 3
Joined: Fri Jan 01, 2016 4:28 pm

Re: banding/artifacts on the surface of layers with holes

Post by klajo »

I've now done some more testing, using a few versions of speeds and retraction settings. It looks like the problem was a difference in speeds between inner/outer layers.

I've normally kept my values at these:
  • travel speed at 150 mm/s
  • outer layer speed at 20 mm/s
  • other layers at 40 mm/s
I've seen slight banding around holes there before, but haven't printed that many objects with so many holes in them.

For the versions I've shown earlier I thought I'd be clever and get a nicer result bu reducing the print speed of the outer layer. So I set them to:
  • travel speed at 150 mm/s
  • outer layer speed at 10 mm/s
  • other layers at 40 mm/s
... thereby evidently increasing the effect. These last pictures were printed at the following values. The new prints are closest to the camera. The new ones look a lot better (and I'm not being kind with the lighting in these pictures), although there's still some slight banding. I'll experiment some more and increase the speeds a bit to reduce the print time.
  • travel speed at 40 mm/s
  • outer layer speed at 20 mm/s
  • other layers at 20 mm/s
IMG_20161215_224539.jpg
Thanks everyone!

/Klas
Post Reply

Return to “Troubleshooting”