Hey all,

I'm still trying to get my delta 3D printer working well but my first layers are a bit strange. I do have raft on but the edges are a bit odd... have a look at this pic:


High res version here: goo.gl/WEl4Nw

You can notice that there's a bit of excess plastic but it's only on the edges furthest away from the camera. I'm not sure what causes this but if I try to print something smaller, or there's a part with a small footprint, these blobs can lift the whole layer off the print bed when the nozzle passes near them and grips.

Now I know that my bed isn't terribly level but the autocalibration thinks it is. You can see in the top left hand corner of that print that it's so close to the print bed on the skirt that no plastic gets put down. I've only use autocalibration; haven't done any manual tuning of my own.

Detailed info: It's a Cherry Pi IIIS design that's slightly taller. It's got a print diameter of around 220mm with the bed being 2500mm. This print was using Flashforge 1.75mm PLA at 210°C on a bed heated to 60°C. I use glue stick to get better surface adhesion. Layer height is 0.2mm printing at around 25mm/s I think for the first layer.

Ignore that blobby thing in the middle (that's just where I killed the print).

Edited 2 time(s). Last edit at 08/07/2015 08:19PM by pugzor.

Lifting edges can be a result of hotend temperature to high. Sometimes reducing 5°C does miracles to a print.
But wouldn´t that occur at the front, too?

I´m not familiar with all the delta parameters, you need to dial in correctly to make the nozzle stay parallel to the bed. Maybe you ask this question in the delta forum?
-Olaf

Edited 1 time(s). Last edit at 08/08/2015 03:14AM by o_lampe.

Looks like you need a better auto calibration algorithm in the firmware. I implemented one a few months ago. It lets you define up to 16 bed probe points, then does a least squares fit of 3, 4, 6 or all 7 delta parameters to minimise the sum of the squares of the height errors at the probe points. It needs only a single iteration, or 2 or 3 if your parameters are a long way off to start with. I recalibrate every time I turn the machine on, as it takes under 30 seconds and saves me setting the Z=0 height separately.

Other developers have stated their intention to port the algorithm to Marlin and Smoothieware, but it hasn't happened yet. So for now you have to use Duet electronics running RepRapFirmware to use it. Follow the link in my signature to blog for more.

You may be able to get your existing auto calibration to work better by using different bed probe points.

One other thing to check is that your effector is not tilting as it moves in the XY plane. If it does and you are using a Z probe that is displaced from the nozzle, the tilt changes their relative heights, causing probing errors.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Looks more like over extrusion to me.

Have you calibrated your e-steps? measure how much filament comes out of the extruder (not the hot end) or how much goes into the extruder when you press 10mm or 50mm...

Fancy auto levelling aside, you might want to practice at doing it manually. It brings you closer to the soul of the machine