Spacing between shells

Hi all.

Perhaps someone knows what's happening here on attached photo.
Printed in PLA on a delta with 0.4mm nozzle, 0.2 layer height @ 205*C hotend and 65* bed, Marlin 2.
Slic3r'd with 3 perimeters, 3 bottom and 3 top layers, 30% infill, and 75% infill/perimeters overlap (upped from default of 55%).

As far as top layer goes - the dude's arms have 100% infill and look good enough i guess, his legs not so much but probably just not enough infill?

The uneven gaps between perimeters and infill, and between actual vertical shells are a mystery to me though. Tried upping the overlap but it had no discernible effect and as far as i can tell I'm not underextruding.

Regards,
Peter

I notice a couple things. First, the line widths are narrower at the bottom of the photo than at the top, suggesting that the nozzle is further from the bed at the bottom. Maybe the bed or X axis is tilted a bit.
Second, the lines are grouped in pairs at the bottom of the photo suggesting backlash in the mechanism that positions the extruder and/or bed.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

The bed is flat and perpendicular to towers, will take a closer look at the pulleys, check for backlash.

Thanks, and a cool blog, the_digital_dentist

Quote
carro
Hi all.

Perhaps someone knows what's happening here on attached photo.
Printed in PLA on a delta with 0.4mm nozzle, 0.2 layer height @ 205*C hotend and 65* bed, Marlin 2.
Slic3r'd with 3 perimeters, 3 bottom and 3 top layers, 30% infill, and 75% infill/perimeters overlap (upped from default of 55%).

As far as top layer goes - the dude's arms have 100% infill and look good enough i guess, his legs not so much but probably just not enough infill?

The uneven gaps between perimeters and infill, and between actual vertical shells are a mystery to me though. Tried upping the overlap but it had no discernible effect and as far as i can tell I'm not underextruding.

Regards,
Peter

9dYmBSQ.jpg

To me it looks like you are underextruding or have an axis that isn't calibrated.
When you see lines not even close to overlapping (leaving gaps) or walls that do not touch, you may not be putting enough plastic out to fully fill your lines. Sometimes this comes because you are going too fast for the temperature you are using. Listen for "cracking" or skipping noises from your extruder. Check to see if your extruder is sending 20mm of filament through when you tell it to put 20mm out.

Problems can also come from your X or Y axis not being tuned. An easy test of this is to print one of the many 20mm cubes from Thingiverse and measure them. Tweak your steps/unit settings in your firmware to get as close as possible to expected results. You have different parts of the print with different gaps between the lines. It is unlikely that this is a bed level issue because after the first layer, the layers then be level. However, on a delta printer "level" has a whole different meaning than with the "bed-slinger" printers. If your delta configuration settings are off, your end-effector will either make your bed concave (higher on the edges than in the center) or convex (the other way around). You will need to correct DELTA_DIAGONAL_ROD which affects X and Y accuracy and/or DELTA_RADIUS to "flatten" that movement out. This is always a touchy process.

DLC

---

Kits: Folgertech Kossel 2020 upgraded E3Dv6, Anet A8 upgraded E3Dv6, Tevo Tarantula enhanced parts and dual-head, TronXY X5SA Pro(E3DHemera).
Scratch: Large bed Cartesian, exchangeable heads, Linear slide Delta, Maker-Beam XL Micro Delta, 220x220CoreXY.

Don't calibrate with 20mm cubes. They are too small to calibrate anything because extrusion errors can make up a relatively large portion of the dimensions. Print the largest object you can reliably measure. The diagonals of a 100 mm cube/square can be measured with a common 6"/150mm caliper.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
the_digital_dentist
Don't calibrate with 20mm cubes. They are too small to calibrate anything because extrusion errors can make up a relatively large portion of the dimensions. Print the largest object you can reliably measure. The diagonals of a 100 mm cube/square can be measured with a common 6"/150mm caliper.

While I agree in principle, I respectfully disagree for reasons of time and material used.
A 20mm cube won't get you high resolution measurements, but it will tell you if you are way off. You can then tune in and when reasonably close, then print the bigger object. The 20mm cube will take you a half hour, a 100mm cube could take you hours, depending upon your printer and settings.

Everything is a trade off,
IMO,
DLC

---

Kits: Folgertech Kossel 2020 upgraded E3Dv6, Anet A8 upgraded E3Dv6, Tevo Tarantula enhanced parts and dual-head, TronXY X5SA Pro(E3DHemera).
Scratch: Large bed Cartesian, exchangeable heads, Linear slide Delta, Maker-Beam XL Micro Delta, 220x220CoreXY.

I agree with dlc60 for using smaller test parts first. I use even smaller tiles of 20x20x2mm just to get into the ballpark and later (if necessary) use larger test parts or actual objects I want to print. That saves a ton of filament and time, especially if the printer is of poor stability and repeatability is low (a lot of backlash, springiness in parts, compliance).

Also, slic3r default perimeter/infill overlap ratio, to my knowledge is 25% (used to be 15% in the past), 75% would cause a lot of problems once you've dialled your printer in properly. I'd first up the extrusion multiplier (or reduce the filament diameter). Once you've got your multiplier dialled in, only change the filament diameter based on actual measurements.

just grab the (cold!) nozzle and wiggle it with the motors powered on. Any motion you feel will result in backlash unless you feel considerable resistance. Try to eliminate any compliance.

Don't base your extrusion multiplier off of the first layer. The first layer may be too low. So increase the multiplier and up z the distance from the bed a tenth of a mm if it seems to over extrude/squish out too much.
After a couple of layers the print equalizes out and getting the correct extrusion multiplier becomes more reliable. I usually up the extrusion feed rate during the print to get to the correct value quicker. Then adjust the multiplier accordingly in slic3r.

Also make sure the velocity isn't too high so that the extruder can keep up. 30mm/s is a safe starting value. Increase it once you've dialled it in and find out what the limits are.

---

--
Kind regards
Imqqmi

NFAN CoreXY printer:
[reprap.org]

Calibrating a broken mechanism isn't calibration. If there's backlash, the mechanism is either broken or poorly designed and built and should be fixed before you even consider calibration. Calibrating multiple times saves neither time nor materials. To paraphrase the dude, "that's just like my opinion, man."

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

"Smokey, my friend, you're entering a world of pain. ...."
On a well made cartesian you dont calibrate xyz you compute from the pitch, gear/pulley ratio ...

Edited 1 time(s). Last edit at 06/01/2019 06:32AM by MKSA.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Well, it depends...

---

--
Kind regards
Imqqmi

NFAN CoreXY printer:
[reprap.org]