Palette 2 / Canvas Hub Announcement -- Duet compatible?



  • Mosaic made some significant announcements for their multi-material / multi-color printing system.

    https://www.mosaicmfg.com/products/palette-2

    Palette 2 - looks like significant improvement in functionality and price.
    Canvas Hub - controller box for printer and Palette 2

    Wondering if anyone has any thoughts and how well it would work with a Duet-based printer. Mosaic said if it has a USB port and can take streaming G-Code, it should work. Canvas Hub helps keep the Palette and printer in sync.



  • I asked Mosaic, they said it's tested and works with the Duet.



  • Jonny & Adalyn

    Hey Joshua, we haven't gotten a chance to test this yet, but because people have expressed interest in having Palette 2 or CANVAS Hub work with Duet3D we will be considering developing a plugin for it. May not be the near future, but we're definitely paying attention to the fact that others have brought it up as well.

    from chat with Mosaic



  • A Palette has always been part of my plan for my CoreXY.


  • administrators

    It will work with the Duet over USB from what i can tell. Network integration should be possible using the canvas hub but that is still USB to the printer from a RaspI zero board + their custom electronics (from what I can see from Joel's video)



  • @t3p3tony Actually, it wouldn't surprise me if that other board is just the WiFi. He said if you have an Octoprint you don't need the hub. They designed it to integrate with Octoprint. So even though the Duet already has wireless and much of the Octoprint functionality, at least in the short term you probably need either the hub or an Octoprint Pi for this to work over network. One way or another you need a USB connection from Palette2 or Pi. I'm not really a fan of depending on USB for expensive multi-day print jobs. I fully understand why they went this way. For the Pro I believe they will eventually need the ability to have all communication over wired ethernet, or something equally robust.



  • I was looking into this, as well. From what I can tell, it basically would work as long as you use the Duet as a "dumb" printer controller (no networking, etc). You basically give up what makes the Duet a Duet (the web UI, the touchscreen control, etc).

    I was interested in it, but unless they add native support for RRF, I don't think its on my future purchase list. (I don't have any interest in going back to Pi-based/Octoprint control, nor do I want to have wasted a couple hundred dollars worth of stuff going with the Duet).

    Crossing my fingers, given that the Duet is so common now on high-end printers, that they eventually support it natively.



  • @dotorg said in Palette 2 / Canvas Hub Announcement -- Duet compatible?:

    I was looking into this, as well. From what I can tell, it basically would work as long as you use the Duet as a "dumb" printer controller (no networking, etc). You basically give up what makes the Duet a Duet (the web UI, the touchscreen control, etc).

    I was interested in it, but unless they add native support for RRF, I don't think its on my future purchase list. (I don't have any interest in going back to Pi-based/Octoprint control, nor do I want to have wasted a couple hundred dollars worth of stuff going with the Duet).

    Yeah, that would be a sad day. And given the way they seem to architecting Palette/Canvas I'm not super optimistic about native support. Their plan is to print from the web based Canvas interface. And to monetize Canvas. It's a good business plan, but it irritates the shit out of Makers and hobbyists. They are focused on end to end turn-key solutions. At least the kid who is CEO seems to be. And of course it's all closed source.

    I think our only hope is if they offer an API for the hardware that the open source community can run with. So you can either buy the box only, or the bundle. So you go the expensive turn-key road, or the tinkerer route.



  • Maybe the mosaic setup might be better for a duet 3. Which I think I heard is supposed to have a USB host?
    What about the prusa multi material mk2? I think it communicates over spi or can bus and is half the price?



  • Just FYI... I signed up for the public beta of the Canvas cloud based slicer and was pleasently surprised. For basic printing, it's very usable and it's actually giving me better prints with difficult material than S3D. The only thing I had to do specifically for the Duet was set the fan on and off commands correct. Other than that, the gcode it generated worked perfectly with the Duet in native mode.

    Now that doesn't say anything about how easy or hard future integration might be but I was able to set up a simple script that watches my gcode directory for new files, unzips them and sends the gcode to the printer just like my S3D post processing script did.

    Oh, I'm not saying the slicer is perfect by any means. There a few bugs and it needs some additional advanced control but for being in public beta for 2 days, it's usable.



  • Well there is nothing stopping you printing via the Duet web interface when not using the the pallet 2.

    Unfortunately the duet interface has been so buggy in the last couple of releases (connection drop outs, won't upload over Wifi when larger files already on duets SD card, stops counting the layer number if the layer number is displayed at all, fails pretty badly at estimating times, drops out continually when monitoring a print). That is all before you account for the fact you can't access the duet externally natively , there is no image capture to make doing timelapse easily, most of the menu items are a pain to use even on the 7" screen due to little to no UX methodology being used in it's design. Don't get me wrong, there are some great features in the Duet web design, especially when it comes to updating the printers firmware and it is my go to when doing something simple but Octoprint just offers a better overall solution despite looking a little less polished on the surface.

    I can see using the Pallet 2 without the Canvus hub/Pi would be a pain and this disconnect between the the Pallet and Printer is what stopped me purchasing it in the first place, but I signed up immediately to a couple of the new Pallet 2 with the Hub.



  • @redemptioner said in Palette 2 / Canvas Hub Announcement -- Duet compatible?:

    Well there is nothing stopping you printing via the Duet web interface when not using the the pallet 2.

    This is precisely what I'm thinking - use the Pallet for multi-whatever jobs, but swap bowden tubes over to my original extruder for single filament jobs if needed/desired.

    My delta has an ethernet Duet, as I'm not a fan of wifi at all, or streaming via USB. However, the Octoprint worked well for me before I bought the Duet.

    Please correct me if I'm wrong, but:

    1. all Palette models, including the 2 & 2 Pro, can operate without a Hub/ Pi interface
    2. the Hub gives
      a) an Octoprint connection
      b) allows for better extrusion syncing
    3. If I already have a Pi, I can use it to get the identical connectivity/benefits that the Hub offers

    Are these correct?



  • @plasticmetal said in Palette 2 / Canvas Hub Announcement -- Duet compatible?:

    Please correct me if I'm wrong, but:

    1. all Palette models, including the 2 & 2 Pro, can operate without a Hub/ Pi interface
    2. the Hub gives
      a) an Octoprint connection
      b) allows for better extrusion syncing
    3. If I already have a Pi, I can use it to get the identical connectivity/benefits that the Hub offers

    Are these correct?

    2 and 3 are correct. Basically if you have a Pi you can use it. If you don't have one and not willing to buy one (out of whatever reason) you can use the Hub instead.
    I don't know enough about 1 to answer that.



  • @3dmntbighker said in Palette 2 / Canvas Hub Announcement -- Duet compatible?:

    ne way or another you need a USB connection from Palette2 or Pi. I'm not really a fan of depending on USB for expensive multi-day print jobs. I fully understand why they went this way. For the Pro I believe they will eventually need the ability to have all communication over wired ethernet, or something equally robust.

    I am also interested in the Palette 2, but I have to say... the current prints they have showcased; in particular the.... weird multicolored t-rex skull, have me somewhat confused; it all just seems very random. Maybe that was the idea? The active mechanism they have devised for mixing seems like it would work; but... now, I also base this back on the Video that was posted of the Palette2 in use. I was counting for the color to visibly fade between the next; instead it seems to be 'purged' from within somehow in the video as the colors shift immediately.

    I have to echo what @3dmntbighker is saying here... and maybe add a little bit. The most concerning thing I see here is the wifi integration; that is a weak point unless it is devised to work in a way in which it delivers the necessary data in a chunk and then utilizes it locally -- if this is for syncing or for any sort of gcode stream/buffer, I would wait for a version with an Ethernet or USB connection. As far as working with the Duet? As @T3P3Tony has indicated, the video from Joel should answer most all of your questions. A good measure of the chance of this becoming problematic, for me, would be this: is the rasberry pi a zero, or a zero w? If it's a zero, you're in the clearo.... if it's not, your gcode is.... okay I give up. Also, to echo Tony a bit

    There are already solutions out there that could have easily chained the palette together with any printer... especially with a Duet board supporting it. I don't know what to make of that part in particular... Based on my experience with the Palette+, it's a very unique system. I think it has a lot of potential; however, I'm not sure if the same kind of person that would build a printer with a Duet at its heart would find much use in this. Don't let me discourage you however! Color mixing/material mixing is what many of us are laser focused on right now, and I am of the opinion that any product and associated user experiences can help us get closer to that goal.

    One more note @Redemptioner -- I am not sure which Duet version you're utilizing, but, I think you will find that DWC is a lot more reliable and faster than OctoPrint running on a Raspberry Pi when plugged in via Ethernet on any recent board. I am a recent convert to the realm of Duet boards after having previously developed firmware that was for use solely through OctoPrint(That was the Delivery Method chosen for Klipper initially, and still remains the most common implementation.)... and while I will always use OctoPrint, a Duet 2 running DWC is an amazingly fast experience for me.

    In conclusion: I'm sure you can make it work, you shouldn't buy it anyway(but you really should), and I will be happy to gather any data you're willing to supply in terms of attempting to mix other filaments. This goes toward a larger effort that we are currently looking at: making a publicly accessible and curated filament database. No matter how we choose to go forward with color/material mixing, it's not an easy problem to solve and I believe as many companies attempt this, the better. Now, get ready to help us solve the real problem -- intuitive color slicing with same-layer color shifts!

    Thanks,

    David



  • @streamliner said in Palette 2 / Canvas Hub Announcement -- Duet compatible?:

    The active mechanism they have devised for mixing seems like it would work; but

    There's no active mixing, or any ability to mix at all with the Palette -- it just snips and fuses filament. Its completely binary, all of one of the four input filaments.

    You need a mixing hotend like a Diamond to get any blending at all.



  • @dotorg Interesting -- I simply watched the Palette 2 promotional video and it seemed to give me a different impression. I am actually curating the Crane Line of 3d Printers from M3D, including the Crane Quad; it utilizes our QuadFusion Mixing Head and we offer a variety of color patched CMYK(process color) filaments. We've been able to produce around 55,000 colors so far... this is driven by a Duet Maestro with a 2-stepper expansion. That said, I have a particular interest in mixing... but from what you've said, and from the prints I've seen... now it just seems like the Palette 2 is a more sleek version of the Prusa MMU.

    Cheers,

    David



  • @streamliner said in Palette 2 / Canvas Hub Announcement -- Duet compatible?:
    ...................... now it just seems like the Palette 2 is a more sleek version of the Prusa MMU.

    Indeed. That is exactly what it is.



  • @deckingman Yep! How are you doing Ian? We've been insanely busy preparing for the Crane launch... it has pulled me away from the Quad for a bit unfortunately, but, it will soon be back in focus. Things are going really well otherwise. We had a very, very interesting brainstorming session on active mixing last night... I will have to email the details to you. πŸ˜ƒ

    I have the community refocused a bit to pick up on the 3MF standard, as it seems AMF has splintered and come nearly non-existent. On the other hand, the 3MF standard has been picked up by some very large companies that have put some very big promises behind it. I have to say in this occasion, in order to move past STL we do need some type of consortium/entity to truly take this to the next level.

    Cheers

    David@M3D



  • I talked to Mosaic support and they confirmed it works with a Duet:

    β€œYes, it will work with a Duet, and you have the option of connecting Palette 2 to the printer via USB using the CANVAS Hub, or you can simplify load the files to your printer's SD card, similar to how it's done with a Palette+.”



  • @streamliner Yeah I'm fine. Working on some ideas of my own - nothing that would impinge on what you guys are trying to do with the quad. Don't forget, I'm happy to help if I can but unless you talk to me and /or share information, there's nothing I can do.



  • I've got a question or two comparing the Palette 2 with the Crane Quad head.

    Distinct switching from one extruder to another (color or material) is generally going to be the rule for me. Both systems have to be actively purged for this to happen. Is there a difference between the two in their abilities to do this?

    Filament waste due to purging when switching is unavoidable, unless a transition is allowable or desired. Easy software control of this purge is highly desirable; it would be great to use as infill, but some of my products will be thin-walled with minimal infill. I realize the slicer is key to making this happen, but broadly, is either system better than the other for minimizing the purge volume?

    I like the versatility of the E3D interchangeable tool head system, priced (for mechanism only, no hot-ends) between the Crane head and Palette 2. Clearly a benefit of it is no purge waste at all. None of these are currently shipping.

    I can buy an entire Crane printer, with Quad head, for less than the Palette 2 Pro, even though it's print area is fairly small. My thoughts would be to use it as is while figuring it out, while building/buying a larger printer, and then just move the Quad head later.

    Thoughts?



  • @plasticmetal You could also throw the Prometheus system into the mix.

    Purging can be done a number of ways. I use 3 different techniques with my Diamond mixing hot end depending on the model I'm printing. If the colour change occurs on layer change, then you can often get away with no purge simply but printing inner perimeters before outer ones. If it's a small model or the colour change happens within the same layer, then it is necessary to extrude some filament. Most systems print a sacrificial tower of some sort but this is often wasteful as the tower has to be printer on every layer change, even if there is no colour change. I prefer to move the head outside the bed and simply extrude into a "bucket" which is less wasteful (in terms of time as well as filament). details here https://somei3deas.wordpress.com/2017/12/06/5-colour-printing-with-a-purge-bucket/
    The final approach I use is to advance the tool change point with the gcode file so effectively the filament is changed before it is needed. The easiest way to explain that is to point you to this part of my blog https://somei3deas.wordpress.com/2017/01/06/multi-colour-printing-without-wipe-or-prime-towers/


Log in to reply