Perfect Pyramid

[Terrence @terrence My Robox is a Green Kickstarter Limited Edition]

@gid: Are you using a G-Code visualiser of sort to show those images, and if so, how do you get hold of the g-code output of a model, to send to a g-code visualizer?

[gid @gid My Robox is a Green Kickstarter Beta]

@terrence: Yes, I used Repetier (I think), but also have Pleasant3D installed. When AutoMaker starts a print job, it creates a folder for that job (currently with a randomised filename) in the PrintJobs folder. Inside that folder there will be a number of files including 8ac75ba785b65.gcode (the original gcode from Slic3r) and 8ac75ba785b65_robox.gcode (the post-processed result actually sent to Robox)

Tom Gidden -- Bristol, UK

[gid @gid My Robox is a Green Kickstarter Beta]

I just tried a pyramid with “Go slow if layer time < 15 secs” (default) but with the setting below, “Min print speed” set to 2mm/s. The apex of the pyramid was significantly less blobby.

Tom Gidden -- Bristol, UK

[Ian @datman]

Has anyone actually watch what happens at the tip? I printed something which is 2 piece. Both peices have pins (about 2mm dia) on the surface at different levels, on one peice it was ok, on the other blobby. I do wonder whether the fill nozzle is being used when it might be better to just used the small nozzle at that stage. I will print another and watch closely.

[gid @gid My Robox is a Green Kickstarter Beta]

Well, with the pyramids you can clearly see it wobbling around: because it progresses through the smaller layers so quickly, a few successive layers are still very soft at the same time. The motion of the head with the viscosity of the filament being laid down distorts the still-soft layers below. That’s why printing them slower results in less blobbage: it gives the lower layers more time to cool before they’re overlaid.

Tom Gidden -- Bristol, UK

[Terrence @terrence My Robox is a Green Kickstarter Limited Edition]

@gid: Thanks for the g-code/printjobs info, that’s a great help.

[gid @gid My Robox is a Green Kickstarter Beta]

I just did a pyramid in Taulman Bridge (Nylon), using @BHudson‘s profiles.

As it’s translucent, the inside structure is partially visible. At the ridge point there’s definitely an internal change: it becomes clearer, due to the fewer internal air/plastic interfaces. Interestingly, there’s another change lower down that I hadn’t noticed before.

The apex is even more blobby, unsurprisingly due to the gooey-er nature of Nylon. Lift-off from the bed is greater than ABS as shrinkage is also greater. Other than the shrinkage liftoff, bed adhesion was great.

If I can get used to the shrinkage issue, I think Bridge might be something I’ll use quite a bit of.

Attachments:

You must be logged in to view attached files.

Tom Gidden -- Bristol, UK

[dave @dave]

Did anyone get to print a perfect pyramid without the apex blob yet? How do the advanced settings with slowing down layers work? At the start of my print I could see the underlaying layers flex with the print head for the next layer and my Walter White has a dent in his head and a broken nose 🙁

 

Attachments:

You must be logged in to view attached files.

[BHudson @bhudson I have several Robox units Kickstarter and Commercial Arizona, USA]

I think that the clue for the fine details on all of these prints will be to slow the head movement by either running multiple - 6 or more - of the same part at the same time to give the plastic time to cool. Once I get my machine working again, maybe tonight, I will see what I can turn out.

As far as the Bridge warping, yes, that is an issue. It is an issue with any nylon, but Bridge is FAR better than the 618 I tried first. I print everything in Bridge at 102% scaling and with full brims/skirts. You may want to add lilypads to the corners for anything long and thin. My pyramids with 618 came out very well, though, I still use one of them to knock print residue off the PEI. I had to literally pry them off the bed with a scraper blade.

I operate two Betas and one Production Robox.
See my 3D Hub site at https://www.3dhubs.com/phoenix/hubs/ben

[Terrence @terrence My Robox is a Green Kickstarter Limited Edition]

After having some success with the retaining clip, I have revisited the pyramid.

Firstly, I tried to generate some cooling time by inserting some g-code to move the head up and then back down for the top layers, but I appears the Robox firmware ignored them. I guess it figured the move was redundant, as nothing was extruded.

So instead, I have done a print using no fill, and 4 perimeters. I did not put a base on it, so as to inspect the inside, which looks pretty decent.
On of the outside edges does have a horrible seam / gap, and there is visible hot swirl on the top 10 or so layers.

Attachments:

You must be logged in to view attached files.

[gid @gid My Robox is a Green Kickstarter Beta]

@terrence : When you added the gcode, did you add it to the normal “.gcode” file or the post-processed “_robox.gcode” file? I haven’t spent much time tweaking GCode so I’m not sure how AutoMaker handles it.

If it is the firmware ignoring the GCode (as opposed to AutoMaker editing it out) maybe if the inserted code dropped the head, extruded a little, then returned, it would have more effect?

I’ve noticed the bottom edge of your pyramid’s a little bitty… I’m not sure your calibration’s good. Can I suggest that you try to print a pyramid at 1000% (or other rectangular-base object at very big scale) and stop it after the first layer and take/upload a photo? Also, the same thing with the first layer set to the other nozzle too.

Tom Gidden -- Bristol, UK

[Terrence @terrence My Robox is a Green Kickstarter Limited Edition]

@gid: I edited the Robox g-code file, and then did a “Transfer and execute GCode file: action. My changes looked like this:

G1 Z24 F18000 ; move to next layer THH
G1 Z19.2 F18000 ; move to next layer (95)

etc

…simply moving up a few mm’s before the next perimeter.

On the calibration, I recently did another Nozzle height calibration, and I think perhaps it was a little too loose perhaps. I noticed this when printing the retaining clip. It looked like the first layer was a bit iffy.

I’ll do a 1000% test and see what happens. BTW, I’m only printing with PLA.

[Author]

Posts