Thank you for clarifying. I had some of the naming/acronyms wrong.
If I had ACE/ACE 2, the other requirements would be:
Is it possible to get everything needed in #1 and are the required cables the factory ones that come with the ACE/ACE2, or something new/custom?
Thank you!
You can easily modify the ACE and ACE 2 cables to connect to the U1 USB port, no real need to produce/buy new cable. You can even buy a plug from Amazon to do it without any soldering.
EDIT:
Here is a video explaining how to modify the cable:
Hi, can you share some hints hoe to connect BTT ViVid? Thank you.
If you want to just connect them to the U1 for “slow swapping”, yes you can use a Micro fit-USB cable or one of those adapter.
Thank you for sharing the excellent video resource. It’s very comprehensive.
Will be a day 1 adopter, love this project.
He said his primary focus was speed, which directly correlates to retraction distance. Only having to retract a foot or so vs four or five (visually) makes a huge difference for a lot of swaps. Plus, having the bend closer to the unit means less friction/resistance while pushing the filament to that junction, and when the toolhead eventually ‘grabs’ it and starts pulling, less stretching too. It seems the only benefit of swapping closer to the sidecar units would be less tubing, which hopefully will be out of view anyway. All my opinion so not speaking for him, but this does seems to make sense.
I’ve been trying to follow what’s going on here with your project so I can understand what is your piece (i.e. the Sidecar) vs. what else is out there. Originally I thought yours was the entire system… the filament spool feeders (8 colors each), the ‘quick purge-while-print’ components to mount under the printheads (two-screws installation each) then any needed electronics to connect it all together. But then I got confused because I saw mentions of all this working with the ACE Pro units (of which I have two) and then I thought your project was what I had described but the filament feeders were interchangeable, meaning you could use yours, or swap in some of the third-party ones.
But then I read no, that’s really UMMU and the print-while-purge add-ons as separate things and your ‘Sidecar’ was the filament feeder piece. You just worked with other existing parts (again, like UMMU).
But reading more, it again sounded like that other stuff would work, but would be slow and yours however is much faster.
My head is spinning trying to keep track of what’s what—what’s Sidecar, what’s third-party that Sidecar is using, etc. Again, I’ve gone up and down this thread a few times and I’m still confused. Can you please elaborate?
there is one thing i want to know，can ams lite working on ummu？（not ams，it’s ams lite
for about），thank you
I am also confused as to what this will take to make the system complete.
The project was split after I got a ton of comments saying “but I already have a (insert MMU here) unit, I don’t want to buy yet another one(s)”, and “but it’s just a customized Box Turtle”, along with the fact that manufacturing/assembling/packaging the complete system/units is a giant hurdle to undertake and will take a lot of time and overhead to get going.
So to differentiate Sidecar/UMMU/Current open source options
The Sidecar is the complete system, which consists of:
Quickpurge hardware that installs in the U1
Main MMU unit (toolhead 1 MMU with touchscreen/processor)
All the cables/PTFE required for installation.
The UMMU consists of:
Quickpurge hardware that installs in the U1
Processor/touchscreen in a small enclosure
Cables required for installation
You supply whatever MMU units you want/have (ACE/ACE2/CFS/Box Turtle/Sidecar Aux unit/AMS (eventually))
The Quickpurge stuff is the focus point of both projects. It’s what allows the asynchronous swapping/purging (swapping/purging parked toolheads) which is the big thing that differentiates Sidecar/UMMU from the currently available open source projects, as they use the “normal” pause-to-swap method, like any other single extruder MMU type machine (AMS etc) so you get more colors, but the U1 just sits there for 60-90 seconds for every swap, eating into the time savings the U1 offers (and again, was the whole focus of this project).
All of the pre-made MMUs like ACE/ACE2/CFS/ViVid etc feed slower than the Sidecar/Box turtle, simply because you can’t control them as much and are limited (for now) to their firmware limitations, and some, like the CFS and AMS, have a built in hub, that makes the physical travel distance farther as well. The closest thing to the Sidecar in the videos that will be available first and soon, would be the UMMU with regular Box Turtle MMU units you can buy/print/build yourself. The actual Sidecar units in the video will still be produced, just will take longer.
I’ve been following this thread since I first learned of it and it’s an amazing accessory/must-have for the U1.
I’m going to have to hold out for the turnkey version, “just plug it in and it works.” I have a feeling, with all the Snapmaker U1 printers in the wild, that you may be swamped with deposits/orders, etc.
Can we make a deposit to get on the early distribution list?
Ok, I think I got it. Tell me if I have this right.
Sidecar MMU (let’s call this SMMU) → Your custom first-party four-filament MMU solution with a built-in touchscreen+processor plus separate quick purge hardware that gets installed inside the U1. This setup controls a single U1 tool head and gives you all the features you’ve described, but for that tool head only.
Auxiliary Sidecar MMU (let’s call this AMMU) → Your same custom first-party four-filament MMU but without the touchscreen+procesor (instead with a small number display to identify the unit), but still with the quick-purge hardware. This is specifically an add-on for the SMMU for controlling tool heads 2-4 (one needed for each). This doesn’t need the screen or processor because the main SMMU handles all of them together (i.e. it can manage 1-4 units)
‘BYO 3rd-Party MMU’ (i.e. what you’re calling UMMU) → The touchscreen+processor from your SMMU but in a stand-alone box, plus the quick purge hardware. This is meant to be used with your own third-party MMU (like Anycubic ACE Pro, etc.) instead of your custom one. Not as feature-rich (i.e. the aforementioned limited speeds, etc.) but cost-effective as you’re repurposing your existing MMUs.
I’m also guessing all of this is interchangeable, meaning theoretically I could use a UMMU for an ACE Pro, then add one of your auxiliary MMUs without the screen (i.e. an AMMU) to combine them with the UMMU’s touchscreen+processor being the coordinator for them all.
And what about multiple third-party MMUs? Each of course needs the quick-purge hardware, but if the architecture is the same as your first-party solution, I’m guessing units 2-4 don’t need the touchscreen+processor either, but I’m not sure how they’d all connect then. (Basically I’m wondering if there’s an ‘AUMMU’ if you will.)
And regardless of whichever solution above is chosen, what about the filament hub itself (i.e. the four-in-one-out) thing? Do we need to BYO those for units like the ACE, or are they covered by your work as well?
And is there a color limit per toolhead? Like could I theoretically hook up both of my ACE Pros into their 8-in-1-out hub, connect them together then expose them as a single 8-color MMU feeding a single tool head (which is essentially what they are now in the Anycubic Kobra S1 as they all feed a single tool head there.) If so, that means this could technically be a 32-color system with eight per toolhead!!
Sorry for the long post, but just trying to make sure I have this right, as well as getting those questions answered.
Of course the biggest question… what’s available for purchase right now? Or if it’s all open-source, where can we get the files/information about what hardware we need to buy (i.e. touchscreen, MCU, hardware, etc.) And can we contribute? (I work for a very large ‘fruit’ company by day and love getting my hands dirty in the code for things like this, helping wherever I can.)
Just sharing an idea for the naming of your product that may help clarify things much better than what I attempted above.
You could simplify this further by making only one MMU (the auxiliary one), then making them all connect to the stand-alone controller which can be physically attached to the front of one of the MMUs (matching your current design) or if you prefer, you can place the controller in a more convenient location, like down by the printer itself while putting the MMUs up on shelves or under your desk/out of the way.
It also has the added benefit of keeping the story simpler when describing packages. e.g.
Anyway, just sharing some thoughts. Really excited about what you’re doing here. Can’t wait to ‘support the cause’!
If you do this mod (which seems trivial), does Snapmaker’s OrcaSlicer just ‘pick it up’ or do you need to do anything else to control it?
You need to modify the slicer to be able to optimize the color switches like Bambu H2D does. You want most of the color switches to happen on the separate toolheads and only switch filament in the MMU as little as possible.
I know you’re excited but I think you are essentially jumping the gun. As per the YouTube video I linked to above, the final direction of travel is yet to be finalised and no doubt involves discussion between various parties with commercial interest. IMO we need to be patient and see what happens; for my part, the knowledge that an AMS option will be available at some point is sufficient.
Producing the entire ecosystem from scratch for retail, with all the validation that entails, would be an enormous task requiring a development and production team. Far less demanding, more efficient, quicker, and probably cheaper retail price, to produce the controller only and rely on leveraging existing off-the-shelf AMS/MMU/… units, possibly manufactured under licence. I am sure this is the sort of thing being discussed right now.
Being able to stack more than 4x4-reel units has been mentioned. As for suggesting nomenclature… leave that to Eric!
Understood. So the project is now split into two parts: one is the sidecar solution, which offers the best performance, and the other is Plan B, supporting other third-party multi-color systems. Personally, I suggest ignoring other ideas for now and focusing on perfecting and launching the sidecar first. Only afterward should we consider supporting other versions, since maintaining a single system is already no small task. As a Bambu Lab user, I can use the AMS, so I’m not in a hurry for third-party solutions. However, U1 users are eager to have a multi-color system… It’s hard to believe that the U1 developer didn’t include a multi-color system from the start of the project.
I think having 4 heads for 4 filaments was seen as an alternative to 1 head with 1 filament path + a 4-way AMS, while simultaneously reducing waste, time, and being able to handle multi-material without compromise. That was Snapmaker’s focus for the Kickstarter campaign and development, and I don’t think they even considered extending that with AMS, nor realised it would be wanted by the community.
