Notifications
Clear all

Randomised Colour Changes  

  RSS
Paul Arden
(@paul-arden)
Eminent Member
Randomised Colour Changes

I am wondering if the MM Upgrade could be used to generate relatively rapid and/or random colour changes. This would specifically not use a purge tower so the changes would be blended (intentionally).

While there are colour changing filaments on the market (e.g., UniCoFil Rainbow PLA), they tend to change colours very slowly and on most size prints colour changes will not manifest within a single layer, usually you just get a gradient in the layers.

Since no one seems to be making filaments with a more rapid change I was wondering if the MM could possibly be used to achieve this (obviously limited to 4 colours). The desired effect would essentially be random rather than specific.

Since the MM Slic3r is based around assigning extruders to STL components I don't see a way to do this but I think it would be a cool feature. I imagine I could build a gcode processor to achieve it but it would be a nice built in feature.

I asked an almost identical question on the Mosaic Manufacturing forums about the Palette+ https://mosaicmanufacturing.zendesk.com/hc/en-us/community/posts/115008419728-Random-Colour-Changes-and-Frequency and they actually have an option for that and for gradients.

Posted : 10/09/2017 3:03 am
AJS
 AJS
(@ajs)
Noble Member
Re: Randomised Colour Changes

This is a very interesting idea. It might be worth posting as a feature request on GITHUB (?)

Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage or loss. If you solve your problem, please post the solution…

Posted : 10/09/2017 3:08 am
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: Randomised Colour Changes

This is something which could look good when printing in vase mode.

Unfortunately, any infill could cause more sudden and not particularly good colour changes.

It would be fairly easy to test by pasting a little extra G-Code into the "Change Tool" section. The only thing I would worry about is whether you would get a "clean" unload.

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Posted : 10/09/2017 10:14 am
Paul Arden
(@paul-arden)
Eminent Member
Topic starter answered:
Re: Randomised Colour Changes

Since I haven't got my MM upgrade together yet I have just been looking at some of the gcode output from the MM Slic3r to see if I could work out how I might be able to do this. It looks like some pretty complex things are happening on tool change, for example this sequence of retractions:


G1 E-15.0000 F5000
G1 E-50.0000 F5400
G1 E-15.0000 F3000
G1 E-12.0000 F2000

Differing amounts at different rates then this sequence of extrude and retract after setting the extruder temperature.


G1 X181.250 Y149.350 F4000
G1 X239.250 E2.0706
G1 Y148.750 E0.0109 F4600
G1 X180.750 E2.1353
G1 X181.250 Y148.150 E0.0109 F5200
G1 X238.750 E2.2518
G1 X239.750 Y147.750 F7200
G1 X180.250 E3.0000 F1600
G1 X239.750 E-5.0000
G1 X180.250 E5.0000 F2000
G1 X239.750 E-5.0000
G1 X180.250 E5.0000 F2400
G1 X239.750 E-5.0000
G1 X180.250 E5.0000
G1 X239.750 E-3.0000

For some reason there are M907's just before the sequence starts as well. I assume the T0, T1, T2, T3 lines are what is actually doing the change, I can't really investigate too deeply since I don't see the MM version of the firmware released yet (I could be wrong there). That is just the unload, there is a load and wipe sequence that are equally complex.

The above smells like something that involved a huge amount of trial and error to get right (judging by the posts leading up to release I am guessing so). It makes it pretty opaque to tinker with unfortunately. I guess I could try the unload and load sequences and leave out the wipe.

Interestingly if you turn off the wipe tower you just seem to get the T0, T1, T2 and T3 lines, a retraction afterwards and then off to the races. Will have to continue looking to see what might be possible. Can't really actually try anything until I get my upgrade running though.

Posted : 11/09/2017 1:13 pm
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: Randomised Colour Changes

There is a sequence of events happening on unload/load as follows:

1. Move to tower and print some lines to pressurise (and cool) the nozzle - this makes for a less stringy unload
2. Unload - this is 4 retracts, using different speeds to get past the various "obstacles" and to form the tip for the next load
3. Cooling movements - within the cooling tubes to prevent the filament tip from sticking. The final cooling move should be positive
4. Load the new filament, again at various speeds
5. Purge lines - print these to clean out the old filament
6. Return to part and continue.

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Posted : 11/09/2017 1:29 pm
Paul Arden
(@paul-arden)
Eminent Member
Topic starter answered:
Re: Randomised Colour Changes

Thanks for the clarification on those steps, the gcode makes more sense now. I wonder if omitting steps 1, 5 and 6 would allow the filament to relatively cleanly switch. Well, by cleanly I don't mean a sharp colour change (since a gradient is fine for my use case) but rather that it functionally works. Definitely have to get my kit together and try some things.

Posted : 12/09/2017 8:20 am
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: Randomised Colour Changes


I wonder if omitting steps 1, 5 and 6 would allow the filament to relatively cleanly switch.

No you cannot omit step 1; it is very important in unloading the filament cleanly and without strings. You need to get cooler filament into the nozzle as quickly as possible.

Prometheus system does much the same; it moves to an unused area of the bed, quickly extrudes (I think) 9mm of filament and unloads.

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Posted : 12/09/2017 9:32 am
Paul Arden
(@paul-arden)
Eminent Member
Topic starter answered:
Re: Randomised Colour Changes

Finally got the MMU together and have been printing various things. So much easier to understand what is happening when you can actually watch it run. I don't see an obvious way to achieve what I was trying to do without upsetting the very delicate process being done to ensure everything says clear and unjammed. Will keep thinking about it but it would still be a great feature to see officially implemented I think.

Posted : 11/10/2017 10:37 am
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: Randomised Colour Changes

You will still require a tower of sorts to prepare for the unload, but if you make the tower very small you should then be able to purge on the part itself.

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Posted : 11/10/2017 12:33 pm
Share: