Palette 3 (Pro} experience

I love my Gridbot. It’s working great with the Mutant tool switcher and I currently have three different hotends on rotation (e3d v6, e3d volcano, e3d revo micro).

As far as I know, @stewart has successfully integrated a Palette 3 (Pro?).
I wanted to hear a little more about it. How smoothly does it work? Is it a pretty painless solution for multi color? What about multi material? I know Kiri:moto has support for it, but it’s there anything I would have to drastically change in my workflow or thinking process when switching to a Palette?
I don’t have any immediate plans to get a Palette, but it’s pretty intriguing and I might put it on my wishlist.

1 Like

When you setup Palette mode for a device in KM, extruders map to color/drives in the Palette. So using the Palette is just like using any multi-color / multi-extruder printer that requires purge blocks on material changes.

Calibration was difficult and tedious. Bowden printers are much harder to calibrate than direct drive. The ping system is also harder to keep in range with the added variability of a bowden drive. And that means larger purge areas (wasted filament) are needed.

I find the Palette approach to multi-filament highly wasteful and unreliable. It’s incredibly easy to blow an entire print even when you do everything right. If even a single splice out of hundreds or thousands fails (not entirely uncommon), days or work and kilograms of filament are gone. The same is true if filament breaks anywhere in the path, on any internal retraction, or if a fragment jams the splice drive (which seems to happen a lot).

I think it would be better with printers that have off-bed purge areas. This would mean less waste.

The most compelling case I have for it is “infinite spool” that can auto-splice a bunch of rolls together for big prints without any pause/resume. This doesn’t require purge blocks and greatly minimizes failure modes.

1 Like

Thanks for this summary. That was very useful. I thought it sounded too good to be true. I don’t really have too much use for the spooling capability, so I think it’s probably not worth the investment for me. For now…

Same problems with DIY MMU2S clone by kakou_fr - Thingiverse?
Or another approach: Tool Switching - Multi Extrusion | Hackaday.io

Or do you have any other suggestions @stewart ?
Edit: for an automated system that is.