Separate names with a comma.
Discussion in 'General Questions' started by John Wohlfeil, Nov 14, 2017.
I have successfully printed 4-Color objects on Robo R2 using Mosaic Palette+.
Pictures or it didn't happen
I have been back and forth and hit a number of road blocks along the way.
Both Robo and Mosaic have been very helpful and very responsive.
Most of my challenges were, I THINK, cause because of the printer set up in Cura. Simplify 3D had some default printer stuff that worked for me, while I do prefer using Cura, I used Simplify 3D for slicing my successful multiple color files for printing. There is a LONG VERSION of this, if anyone is having trouble, ask for more details.
I experienced several failed prints, and cancelled prints, different set up challenges, temperatures issues, tuning and fine tuning the Palette+, while trying to get this figured out and printing.
Finally, I got the Mosaic Keychain to print:
But not before this happened:
And This happened:
FINALLY, I successfully printed a few different Mosaic Logo Keychain files in different color combinations, then finally got this snowman to print:
Honestly, it sounds like the problem could be solved in the slicer. If it were me, I'd handle the color change on the infill and forecast so that you're sure that it's taken place before you then print the "for show" side of the part for that layer.
Good luck with that.
One of these days, I'll get that three-extruders-one-head thing and work that up.
For large parts yeah, but for small ones there often by far isn't enough infill on a part to purge the required amount and a place is needed to put the extra.
Take the snowman where there are 3 colors on the same layer, need to extrude probably 60cm of material on a 3? cm circle - not gonna happen.
That said the Palette+ comes with most likely conservative transition settings that can probably be optimized quite a bit. Not sure if it starts the purging on the infill or not, will see when I ever receive mine...
Its my understanding it all happens in the purge block but you can change the purge block dimensions manually once you know how much it takes to move color A to color B and so on. Frankly waste is the least of the concern IMO when it comes to multiple extrusion (2, 3, 5 100 whatever lol) as long as the print is correct do you REALLY care if it wasted a few bucks in material? (still cheaper than running bronzefill )
And your point is? (clutching my spools of bronzefill and giving you the evil eye)
LOL! I keep about 5 spools in inventory, just using as an example No hate buddy. .
The lady bought 3 spools from Matt black friday lol. those arnt mine no touchy.
That's what I meant, while not giving control of the purge block itself the Palette software allows you to spec the length to purge on a transition and which percentage should be before and after the weld (i.e. what affects what the purge block must be), so once you've got things tuned right you can likely reduce this significantly from the default values.
Not caring about the material, but more about time and a shorter/simpler print naturally has less risk of failing.
Well, I getcha but... look at the signal-to-noise ratio (snowman-to-purge). It just feels like we're going to look back at this style of printing five years from now and laugh.
I think I'd rather print one "for-show" snowman and two "anything-goes" snowmen next to it and to use those for purge stacks. At least you get one good-looking snowman and two artsy/colorful ones that a kid (or a modern art teacher) might appreciate. And okay, you do transitions on the infill of the anything-goes snowmen. Under this ratio, it might accidentally produce three good-looking snowmen.
I think it's misleading, the purge stack is large and looks solid but likely quite light/empty.
Think about it, when rectangular like this software produces its size needs to be determined by the one layer that requires the largest amount of purged material. Other layers with no purge might have 5-10% infill just to support what will come on top of them.
Not sure, it seems to just be one compromise like many other things are. There already are many multicolor solutions, they all have significant drawbacks. This one avoids most of them and apparently produces good quality output, the extra waste is just the price to pay for it which IMO is pretty small in comparison to the gains (and is why I've ordered one, first solution that seems "good enough" to me.
This. Read this.
You might think multiple extruders sorts it, until you actually try to use them. None of the solutions at this stage of the technology are perfect, some just stink less than others. I always try to either:
1) design around needing multiple colors (multiple parts in their own colors for example)
2) Use paint
Occam's razor folks...
Yay my palette+ has finally shipped and is due to arrive on Monday!
Boo as I'll be right in the finish line of a project and leaving Tuesday morning for 2 days and likely won't get to try it until I'm back
Grr freaking DHL, they again pulled the "out for delivery so you stay available doing nothing to be sure not to miss it, but don't deliver and postpone to the next day in the end"...
So unless it comes tomorrow early before I leave it won't get to me before Thursday. Obviously won't be able to look at it before then anyway even if it arrives tomorrow morning.
Guess that leaves me time to print a few spool holders.
No time to print more than one spool holder because printer busy with other stuff... but it arrived quite early and I just had time to do the setup and calibration print uneventfully.
OK, got my first successful print. But that's out of 20+ attempts...
0 actual printing issues unlike @John Wohlfeil nor splicing process issues but total unreliability in splicing lengths/alignment.
Every single print (except this one) gets a color shift within the first 10 layers, which usually rights itself but the print is still ruined by 1 or 2 layers being off.
Calibration done several times with similar results.
IMO the issue is filament discrepancy, which is a thing (all spools I used are the same brand) that just has to be accounted for to gain any kind of process reliability in a system where errors cumulate, but currently isn't.
This can be solved, but requires either significant architecture change to provide more direct feedback, or profiling of every spool.
Much the same is true for printing temperatures, which is why those numbers are always "guidelines" rather than "rules"