The CAN tutorial I put together should still work fine. The only thing that has changed is that the CanBoot Bootloader is now Katapult. However, all of the legacy commands still work as of a few weeks ago. (I used it when setting up CAN on another printer). CAN in general has quite a few pieces. Theres the method of setting up the network so U2C or CanBridge, the configuring the actual network, flashing the bootloader, and then flashing the firmware. Also make sure you have the 120ohm resistor in place or your toolboard will not be detected.
Thanks man. Sorry if it felt like I was calling out your video. It turns out I was wrong, it wasn’t that the guide was out of date… something was wrong with my equipment. I think it was the BTT U2C, because even an MKs board didn’t work till I swapped to another MKS U2C! I’ll pin your comment
The question that the video title posits: "Why Don't More People Build This Open Source AMS/MMU?" Is answered in the first 2 minutes with the quote "I really don't know how much longer it's going to take to get this thing running perfectly..." That sentiment isn't just for TH-cam content creators. It also applies to most people into 3D printing. We just want something that works. As evidenced by the landslide popularity of the Bambu Lab printers. I want to 3D print to support my **other** hobbies. I don't want to take on another hobby of endlessly building and tuning and tweaking a 3D printer. I just want the thing to work.
I started 3D printing in 2013 and back then it was ALL fiddling. Because it was so new I didnt mind. Now I do way too many other things to want to screw around with it as much as I used to. Its why I have Bambu Labs X1 Carbons. Just turn it on, load filament and print. 45 minutes spent leveling the bed for the first layer to still screw up is not my idea of a good time anymore.
I also think the name makes it harder to find, carrot feeder? Really? If it was called something like Open Material Switcher or something to that effect more people would find it. You have to remember most people are searching for specific terms, not using word of mouth.
I think you hit the nail on head there. I have a complete V1.1 kit with printed parts and it hasn't left the boxes. Shortly after getting the V1.1, the V2 came out and then I got a P1S. It's a workhouse and my 3D printing needs are more as a tool rather than a hobby now.
yep same here, I started with an ender 3 back in like 2017. It was great to learn on and figure out the ins and outs of 3d printing, for the past couple of years it's just been sitting gathering dust because I got tired of tinkering with it and just bought me an x1c. I spent more time tinkering with the damn thing than I did actually printing parts for my actual hobbies.
We choose to build an ERCF in this decade and do the other things, not because they are easy, but because they are hard; because that goal will serve to organize and measure the best of our energies and skills, because that challenge is one that we are willing to accept, one we are unwilling to postpone, and one we intend to win, and the others, too.
the encoder is to detect jams, the switches only detect there is filament but not if its moving the right direction etc, the encoder allows monitoring that, so if it isnt moving but all other switches are actuated then it knows there is a jam
You asked why anyone would want to watch a live stream after the fact. They are perfect to play in the background while doing the same project. It let's you know what is coming up next and you can pause/rewind/fast forward as necessary.
I love that you even covered the Blobifier! It's just a small addition to the system but seeing it appear in other people's videos makes me really proud! Thank you!
Use your RPI to communicate with the ERB over USB, significantly easier. As for setting up the firmware, I recommend Happy Hare. Happy Hare is so easy to make everything work and their documentation is pretty good, cut down the set up time so much and gives plenty of options for material swapping procedures. Happy Hare uses retract before cut, which I am sure Bambu is going to try and patent so other manufacturers don't use it.
The thing with tip tuning is, that it is not just about the tip. Bambu printers are known for being VERY wasteful when swapping, because they have to push out the filament piece, that gets left behind after the cut. If you can cleanly extract and reinsert your filament, you are saving material and time on every single swap you do.
I think the complexity sort of answers the question of why no one's heard of the ERCF. I don't think I've ever seen one up and running without the owner being extremely nervous about how long it would stay that way. The rabbits are too enraged, methinks.
@@gatling216 its not that complicated though. But you kinda need a filament-cutter like on bambulab... once I installed one I have had lots of succesful prints on my ercf, and the ones that failed wasnt because of the ercf 🙂
O dont know how few people have heardbabout this but the fact that people think bambu invented core xy makes me think that most people dont know much about project in 3d printing
@@joeltorres3202 Most people don't. 3D printing has always been a niche hobby. It's easy to forget that, from the outside looking in, it might as well be black magic. The average person looking into printing wants an appliance they can plug in and will work, just like a microwave or a toaster. The time, effort, and skill it takes to put together something like a Voron or an ERCF is literally incomprehensible if you're not deep in the hobby, and most people have no desire to get that deep.
Dude, thanks for the video! I was there for a chunk of the live streams, you've put a ton into this and I know it's taken away a lot of your time from other videos. I bought one of these the day of your first live stream and I really appreciate all that you put into this. Good luck with the remaining work!
Not owning a Voron or an ERCF, you questioned the need for two runout sensors so close to eachother: I'm fairly sure it's because of the retraction. If you *only* have the one runout sensor at the multiple different units, when you're retracting to swap filaments you'd end up pulling up too far and not be able to use that one slot until manually reloading it. Having a secondary sensor in the actual selector allows you to figure out that "Hey, I'm now *out* of the selector, and the second sensor's not tripped yet. Good time to switch." The other side works the other way, it tells you you're running out at a point where you can still retract to swap and not get stuck.
Pro Tip: you can actually change the Bambu gcode for the AMS/AMS lite to also retract the filament before cutting to reduce waste! I even went a step further with my A1 mini with AMS and also changed the purge speeds slightly to make the color changes faster, still haven’t tested it tough but the filament retraction has been tested and there’s a makerworld post that has the costum gcode for it. You should give it a try
It's been around for most of this year unofficially and officially supported in May 2024 in the slicer. There was a test that got it down to 70% reduction or 30% purge multiplier.
@nunovicente34 sorry, actually meant to reply to the video, not your comment. I am currently in the process of this mod and wondered about their thoughts on it.
Subscribed. I started getting into printer building and modifying after breaking my back over the last decade of building cars. I've got tinkering experience and a basic understanding of circuits as I was a master of automotive electrical systems. I started with an Ender 3 in 2022 and recently moved up to a Bambu P1S to print high-quality parts for the mods with the intent of using my Ender 3, which is no longer Creality except the frame and the bed, which just got the BTT SKR 3 EZ so I can run more steppers. I recently converted an old FlashForge Adventurer 3 to an open source setup running a BTT E3 Mini v3 and Klipper. I'm almost done with that build and looking forward to my first print on it. Next, I think I'm going to take on a Voron build or modify an old Ender 5 Plus for multi material printing.
Yeah. It’s a lot. I wish the software had a very clear manual like the hardware does. The wiki is a lot of info but I don’t think it’s comprehensive or linear yet.
I've got absolutely nothing about the ERCF, it seems like an awesome mmu, but now that the voron v2 has a really decent toolchanger mod which has the incredible benefits of NOT NEEDING to purge NEARLY as much (in some cases not at all if you tune your setup really well) which SIGNIFICANTLY reduces waste, as well as tool changes are WAY faster than a standard mmu (because you only have to retract a tiny ammount (a mm or two) just as you would when parking the toolhead after printing so multimaterial or multicolor parts print way faster.
The filament movement tracking is used by some printers to monitor for jams or other extrusion issues. If a filament should be moving at X speed, and its moving to fast it may be due to printing spaghetti on a failed print, while moving too slow likely means a jam.
I recently ordered a Voron 2.4 kit to upgrade from my large bedslinger running Marlin. I've been curious about the ERCF as an upgrade later down the road (using funds from selling my current printer after the Voron is complete). Thanks for the detailed explanation of the ERCF.
This is an ambitious project I have no business doing but want to now that I've just gotten Klipper running on my mostly OEM old Ender 3 Pro with the original 8 bit board lmao
If you have an Octopus, you do not need the U2C. When configuring the firmware for the octopus, you can choose to use the octopus as the canbridge. SInce canbus is a network, all canbus devices can be connected to the same single canbus network as nodes.
Awesome video! I've seen stuff about and looked into the ERCF before and I was always curious how it actually worked. Your explanations in this video were perfect! Thanks a ton! Looking forward to seeing it when you get it all working!
I get why some want open or full control over the system but when comparing a multi hour/day/week endeavor vs plug in some cables and ptfe tubes and be up in running in 20 minutes. There is a market here. Why are there no plug and print solutions in the open source world?
It's frustrating for sure. I can appreciate the level of community input getting the open source instructions/parts out there but after seeing connect and go solutions it's world apart. You are right, making it easy is secondary to making it work. Then you have competing personalities, implementations, limited resources and printer hardware/software. Every time I see the ercf topic show up it reminds me of a Rube Goldberg machine of parts and contraptions that one errant filament line will make the whole thing come crashing down. But if an outfit could make a polished product that connects without fuss they would shake up the open source community.
The Open community thinks making money is anathema to maker ethos. That's why companies can swoop in and do the unthinkable like.....make a good product that meets consumer needs. Then the OS folks clutch their pearls that ANYONE DARE make a product for the masses who should be sacrificing raspi on a pentium processor, read miles of code and search outdated forums for bespoke solutions to a problem that was fixed 3 firmware revisions back but the documentation hasn't been updated 6 revisions ago. 🤣
@@wyattutz Hmm. I'll ignore the insult and dig into the heart of your statement. The Open source community had a several year head start and it is still a mess. They have seen a working design as of 2 years ago and still double down on the mess they had years prior. You use the term idiot proof or idiot friendly. I say "if it works, regardless of the audience.....it still works" A mostly closed source company had 2 working models with the AMS and AMS lite and let's be frank it is a *gold standard* of working within minutes of setup. The ams copies from creality, anycubic aim for plug and play, not build and pray. Name a better system that can be up and running with quality within 20 min. What system do you consider the "GOLD STANDARD" I am actually curious for your answer. All of the OS mmu types require a build, tune and a hours upon hours of crap to get working reliably. You should know off the shelf doesn't mean best for the job. Of the thousands of engineers, software devs, talented hobbyists NO ONE saw a need and could provide a commercial/retail solution? This is the 2nd(3rd?) revision of the ercf...and how is that working out? "you can already get better reliability, performance etc if you are capable or sufficiently persistent." You reaffirmed my original argument. There was money to be made here, you are telling me all of the teams, OS companies couldn't pull together to be first to market with an "idiot proof" product? The guy above was right. the Open source sphere want to make things needlessly complicated.
The filament switch sensors at the erct are just for detection if filament is in the buffer. As for the encoder, you can go several ways. The stock voron build has a sensor at the toolhead and the encoder, which mean redundancy as with a encoder you dont need any sort of toolhead sensor as the encoder detects when the filament bottoms out at the extruder gears. You can also forgo the encoder and only use switches like the annex tradrack or prusa mmu, the software behind it is really configurable. Some people also use multiple sensors at the toolhead with the encoder, one sensor above and another one below the extruder gears. In my two builds i always skipped the toolhead sensors, above the gears is in my opinion not repeatable enough depending on bowden tube length and diameter. The sensor below gears is for me too big of a compromise for tpu as i generally print 85a and softer. Also neither of my toolheads have enough space for one. I also built a ercf 1.1, ercf 2, and public beta tradrack. A modded ercf v1.1 is basically a v2, so if anyone got a v1.1 with springy, binky and sturdy bunny mods, there is no point in upgrading. The tradrack is a vastly different in construction, significantly simpler and cheaper to build with only one downside, you have less grip at the mmu feeder since you use a single gear instead of two. The build is also significantly quicker, it took me 3 to 4 hours, with mods such as encoder and filament brakes. The other advantage is that it can feed quicker since it only has a 3 to 1 gearing instead of 4 to 1. Last but not least its much easier to self source in my opinion, no super long d cut shaft, only one set of bmg gears, no closed loop belt
i like the state of development this 'home brew' ams enough to forgive the 2 tips. when i 1st saw the bambu ams i wondered how long before the maker community out did it, a common theme for the maker community in my experience, and to my immense surprise and delight it was apparently done far earlier than i thought. i look forward to seeing this in action
Far sooner but never a home run in terms of implementation. The AMS while simple looking is a engineering marvel compared to what was available before 2022. MMU, Palette ercf etc. With so many moving parts standardization should be a thing. Cutter, tip forming, buffer, and literally down the rabbit hole(pun intended) we go.
Thank you for covering this, I've looked into making one awhile ago but never found a vid i enjoyed with some one showing there journey building it. Great Vid
Amazing video! I have almost everything I need to add an ERCF, my hotend (XOL w/ ebb36) even has some filament cutters designed for it. Looking forward to you showing me how to troubleshoot it too!
I'm looking forward the seeing your ERCF in action. My Voron 2.4 is pending some planned upgrades one of which is a canbus tool head, which will also be going with tap from standard omron sensor and relocating the end stops; x on toolhead and y at the rear right. Then I can look to get my Siboor ERCF v2 connected connected to the same canbus.
I strongly advise you to install the second filament sensor after the Gear extruder, this way Happy Hare knows if the filament has been loaded correctly and will try again if there is an error. I'm a contributor of the team (mcu localisation, and some minors ideas)
I had my eyes on it for a few month now. It's an awesome video, will totally build one, but i will also take my time, the complexity and cost tell me this isn't a quick weed end project.
There's a key difference between this and the AMS - the AMS uses a dedicated feeder motor for each channel. IMHO this would be a better approach as it results in a much less complex motion system and gives you more control over buffering and spool tension
I believe the one who burnt the ships was Hernán Cortés when arrived to America. Vasco de Gama tried to be the first to go all the way around the earth, that would be a bit hard with no ships :)
Since I got my Sovol SV06 earlier in the year I have wondered if it would be possible to upgrade it to be able to print multiple colours. It's nice to know that it is in the realm of possibility. Maybe one day.
Great video! I was considering the enraged rabbit when the new MMU came out. Non printing related note: if you wanted to try tracking points for the labels you added but in a new way using After Effects, it’s a more simple process than you’d think! Obviously this assumes you’re using adobe which hopefully you aren’t, but as someone who has only a basic understanding of motion graphics etc, tracking points for labels/tags comes out amazing with very limited effort. If you have the patience to build vorons and this project, you’re more than capable of figuring out the pick whip and crazy way layers work in AE, or whatever 3rd party option there is in Davinci or Final Cut. Your production quality clearly doesn’t need any help, but maybe this will be a new, fun little skill for your workflow!
Hi! Nice video! A thought and friendly suggestion: I really think nixing the innuendo entirely would greatly improve your videos. Said as someone with a brain always in the gutter, this isn't coming from a puritan or prude here, it just made me second guess subscribing to your channel. I think you're doing really good work here, that just seems like an easy tweak that would make your content a lot more approachable (and non- offputting) to a wider variety of people.
As a bambu x1c owner with ams, coming from a diy core xy, mainly for ams, should i ever want to build or buy a new printer it would be an IDEX or anysort of tool changer. Got ams mainly for multi material prints (support + part) is about best i can tolerate, and appreciate. However even when switching between two materials, and support interface layer only supports of different material, depending on prints it can add hours of extra time. Hence why an idex type would be more suitable even for this 2 material use case. One thing i haven’t seen yet on vorons or others is the lidar sensor from bambu - that thing is a godsend - chuck in and brand of filament and off you go with zero manual tunning needed.
The Trad Rack by Annex team is a better option for a MMU in my opinion as i have experience building both, the trad rack is just more simpler to assemble,supports more filaments than ercf and also has fewer parts, and has higher repeatability
Was planning to build one but now I'm much more interested in multi head solution that allows me to use multiple materials on single print. ERCF doesn't solve a major problem - slowness of filament switching.
50-50 on this. My ERCF can swap a filament in 30 seconds + purging time (another 15 seconds or so). A toolchanger still needs to move the gantry to the parked tool location, heat up the nozzle, dock, undock and move back down again. So I'm not sure how much faster it will be. However filament waste is another matter... This is I think the biggest benefit; however the drawback is much more complex calibration and alignment of the tools to each other.
@@igiannakas Prusa XL needs ~5s for tool change (from interrupting print, moving gantry, putting old tool, picking up new tool, and going back to the print). 1000 changes with 45s change is over 12.5h for tool changes only. On XL it is 1.5h only. 9x faster. Even assuming it would do twice slower changes (10s) it would still complete all changes in 3h only. It's a major deal for prints with multiple color changes.
@@arekx exactly , that is why when doing a print with multiple color changes a Prusa XL with the toolheads will be *way* faster than a Bambulab with AMS , plus you don't have waste at all with the XL , so the multiple toolheads will save up quite alot of time
I could never get the "tip tuning" thing to happen to make a phaetus taichi work properly, that would have been awesome, but I guess for now I just do manual colour changes when they're needed.
Looks interesting though I don't need or want a big thing like that which takes 12 spools. If there was a smaller version or if you could configure this one to only use 2,3 or even 4 spools I would be interested.
subbed to make sure I don't miss that 7-tool sovol build. Hope you can get different nozzle sizes slicing together, which is still a shortcoming of the Prusa XL. Different layer heights seems even more challenging, but if both of those capabilities get working, you can do detailed outer layers and chonky infill. Stronger, more beautiful parts in multimaterial, multicolor, and minimum print time... the future is toolchanging
i almost got a voron 2.4 with an ERCF last year but ultimately decided against it and to get a 2 head prusa XL instead as multi-head will outclass multi-color/material any day and the cost of a 2 or even 5 head XL is similar in cost to a voron 2.4 with an ercf. i would still love to get one but voron and the ERCF is a project that you need to build and tweak yourself while prusa machines are a ready to go product that you can still build but require almost no tweaking.
Can't watch at work, but just wanted to point out that the ERCF isn't the only open source MMU. It was based on the SMuFF which is still actively developed, and may be a better option for some people.
At least for the Sovol SV08, I'm more interested in the mods coming out for changing the entire tool head for faster filament change times and no filament waste or dealing with the poop. I almost never print in color but I'd like to have two tool heads so I can print the bottom of a part in PETG-CF, a 4 mm interface layer that interlocks the PETG-CF with TPU, and the top of the part printed in TPU.
Interesting build and congrats for showing this to us. BUT !! :)) regarding the design. I really can't understand why not just "prepare" the filament before printing so when you print all the right colours are already on a continuous filament so there's no need to retract and cut the tip .....is automated filament welding such a difficult thing?? I only print in white but I imagine on a print farm it would be a powerhouse if you use multi colour print......or is multi colour print just a gimmick? Sorry for the long subject, this video seemed to go into the nitty gritty and the DIY atitudine seemed perfect. Thanks.
hmm this one has a big potential...if all the stars align. It could reduce filament purging if you would add a filament cutter. If you know you need 10g of filament than you would measure that amount of filament CUT (it not retract it) and than add the next color. This would mean that the amount of filament that needs to be purged from the previous color is much smaller and determined because you cut it not retract the filament and not knowing how much is there left in the head. It would also reduce the time to retract each spool.
if you swap out the octopus with the latest version of the manta m8p (make sure it's v2), you can get rid of the U2C board altogether and go directly from the manta m8p to the ebb36 (via can bus still)... and, from personal experience in my zero-g mercury one build, it's far more reliable.
I had some problems setting up can bus aswell, but the Manta M8P+EBB CANbus Setup Tutorial from Stacking Layers helped me quite a bit. Its fairly up to date and covers most of the setup (except for the u2c) but maybe it helps you aswell.
I think the biggest problems with this are twofold, one is self assembly, the others a bit more complicated. Self assembly is a huge turn off for a lot of people. Even with help, if you screw up and it partly works, it's still on you to fix it. A prebuilt product on the other hand generally comes with support and is ultimately less work on your end. That means ALOT to the average person. The second part is the real usefulness of multi-material printing. If your prints have been designed and optimized to work in a specific material then using a different material isn't what you really want to mess with. It's another hassle and something you have to start with, if you aren't using MM to print now, then going forward you likely wont. For instance I print alot of toys. I can design for TPU, PLA, ABS, or PETG, but generally speaking I don't NEED to print in more than one of those. So these are ultimately pointless for my design process. So why waste time and money building this?
I've been wanting to add an MMU to my printers and have been looking at the 3D Chameleon but just haven't been completely set on it, I think the ERCF is more what I am looking for and will be my winter project. Having modded my Ender 3 Pro and 3 Max to use a Sherpa Mini DD with a Revo CR with the Hero Me 7, adding in a KlackEnder Klicky Probe, Z Belts (BeltDrivenEnder3), with both running Klipper this hopefully won't give me too much trouble. For those wondering why I would mod my Ender 3s like that instead of buying/building a Voron, well I've had the Enders for years now, they print pretty decently, it's been cheaper to mod them than buy a Voron Kit and well I love to tinker much to my wife's dismay sometimes 😁
I found some video somewhere on the history of multicolor printing. and one of the first approaches had a system that mixed filament. So almost like a ink printer. You would have CMYK and black and white. rolls. and then basically it would be like mixing on the fly. The filament was melted together and extruded and then re-fed into the print head (if I recall correct( . Its very hard to control that sort of mixing. But I still think that approach should be re-checked into. maybe also, consider that filament might not be the best option to do something like that. What about printing with pellets? it would be much easier to control the mixing this way. maybe. idk. Its cool to see people working on these projects.
Good job on the video, and I hope you can get those communication issues sorted out. Is there a smaller version of this, like maybe something in the range of 3-5 different filaments? I usually only have 2-4 spools of filament at any one time, and controlling moisture here in the Orlando area is practically a nightmare.
Tip tuning was hell. Never really got mine dialed in. Was starting to get into the new things (i have a prototype Binky) when life for in the way. My V1.1 is still sat on my desk waiting for me to work out what i need to upgrade it. Tempted to just get a kit.
Where did you get your apron's LED strip? The one named "DIGI_LED_12 V1.0" ? It seems sturdier than the PCB LED I used (and soldered a million times 😅)
SOOO COOOL I want one :D BUTTTTTT is there any tutorials to add the cutter to a elegoo neptune 3 printer assembly I'm new to the hoby but ive always wanted to tinker so this is right up my ally :) wanted to ask aswell the back bit you didnt specify you called it a buffer so it would need to run from my drybox through the buffer to the selector ? sorry tons of questions just passionate
😂 glad the hackerman bit made it in. And yeah that multimeter stuff was a drag, but the video came out great at least! Can’t wait to see the rest of the project come together
Lane based systems are the way. This method of changing colors needs everything to go exactly right every time. And 3d printers are bad at always doing things exactly right... Exxcellent overview of this project though, good work!
I'm very curious if this will work with Bambu Lab or another non-Voron printer (Elegoo, or Creality?). Voron price and work involved to put together is a bit much for me.
I have an IDEX, and do multi material prints on occasion, and would LOVE an MMU for one of the two heads, but realize nobody will ever build one to just work with any ol' slicer.
Material switching is convinient, I guess, but I really never got the whole multi color thing. Can´t you just paint the prints and way more accurate, too?
My canbus hang up was not knowing i had to reflash klipper onto the printers main mcu with canbus turned on. It should defiantly mention this in the instructions.
While this is definitely cool and can definitely give some more options and choices over an ams. This id still definitely not an enclosed system like the AMS is which limits what you can use it with compared to the AMS
Excellent video. This is an awesome project, and I'm REALLY looking forward to the 7 head tool changer. I had not heard of this before, and I'm glad you made a video of this. Thanks!
The CAN tutorial I put together should still work fine. The only thing that has changed is that the CanBoot Bootloader is now Katapult. However, all of the legacy commands still work as of a few weeks ago. (I used it when setting up CAN on another printer). CAN in general has quite a few pieces. Theres the method of setting up the network so U2C or CanBridge, the configuring the actual network, flashing the bootloader, and then flashing the firmware. Also make sure you have the 120ohm resistor in place or your toolboard will not be detected.
Thanks man. Sorry if it felt like I was calling out your video. It turns out I was wrong, it wasn’t that the guide was out of date… something was wrong with my equipment. I think it was the BTT U2C, because even an MKs board didn’t work till I swapped to another MKS U2C!
I’ll pin your comment
@@thenextlayer No worries. Glad you got it figured out. Just wanted to clarify 👍.
That little resistor jumper has got everyone I reckon. Why doesn’t it just come populated?
@@TheButchersbLockyou can daisy chain nodes on a CAN network so you only want it at the end and beginning of the network.
Aaawhhh huu huh... He said Rods😂
The question that the video title posits: "Why Don't More People Build This Open Source AMS/MMU?"
Is answered in the first 2 minutes with the quote "I really don't know how much longer it's going to take to get this thing running perfectly..."
That sentiment isn't just for TH-cam content creators. It also applies to most people into 3D printing. We just want something that works.
As evidenced by the landslide popularity of the Bambu Lab printers.
I want to 3D print to support my **other** hobbies. I don't want to take on another hobby of endlessly building and tuning and tweaking a 3D printer. I just want the thing to work.
I think a lot of these people have no idea how easy they make this stuff look.
16:00 -- "Unless you're an engineer, or a machining hobbyist..."
I started 3D printing in 2013 and back then it was ALL fiddling. Because it was so new I didnt mind. Now I do way too many other things to want to screw around with it as much as I used to. Its why I have Bambu Labs X1 Carbons. Just turn it on, load filament and print. 45 minutes spent leveling the bed for the first layer to still screw up is not my idea of a good time anymore.
I also think the name makes it harder to find, carrot feeder? Really? If it was called something like Open Material Switcher or something to that effect more people would find it. You have to remember most people are searching for specific terms, not using word of mouth.
I think you hit the nail on head there. I have a complete V1.1 kit with printed parts and it hasn't left the boxes. Shortly after getting the V1.1, the V2 came out and then I got a P1S. It's a workhouse and my 3D printing needs are more as a tool rather than a hobby now.
yep same here, I started with an ender 3 back in like 2017. It was great to learn on and figure out the ins and outs of 3d printing, for the past couple of years it's just been sitting gathering dust because I got tired of tinkering with it and just bought me an x1c. I spent more time tinkering with the damn thing than I did actually printing parts for my actual hobbies.
We choose to build an ERCF in this decade and do the other things, not because they are easy, but because they are hard; because that goal will serve to organize and measure the best of our energies and skills, because that challenge is one that we are willing to accept, one we are unwilling to postpone, and one we intend to win, and the others, too.
😂🤣
🌛
what he said
@@gorgonbert Second!
Cringe.
the encoder is to detect jams, the switches only detect there is filament but not if its moving the right direction etc, the encoder allows monitoring that, so if it isnt moving but all other switches are actuated then it knows there is a jam
Thanks!!!
You asked why anyone would want to watch a live stream after the fact. They are perfect to play in the background while doing the same project. It let's you know what is coming up next and you can pause/rewind/fast forward as necessary.
I love that you even covered the Blobifier! It's just a small addition to the system but seeing it appear in other people's videos makes me really proud! Thank you!
I can't imagine how hard it is to build AND film all this. Thank you for doing it.
Use your RPI to communicate with the ERB over USB, significantly easier. As for setting up the firmware, I recommend Happy Hare. Happy Hare is so easy to make everything work and their documentation is pretty good, cut down the set up time so much and gives plenty of options for material swapping procedures. Happy Hare uses retract before cut, which I am sure Bambu is going to try and patent so other manufacturers don't use it.
The thing with tip tuning is, that it is not just about the tip. Bambu printers are known for being VERY wasteful when swapping, because they have to push out the filament piece, that gets left behind after the cut. If you can cleanly extract and reinsert your filament, you are saving material and time on every single swap you do.
I think the complexity sort of answers the question of why no one's heard of the ERCF. I don't think I've ever seen one up and running without the owner being extremely nervous about how long it would stay that way. The rabbits are too enraged, methinks.
@@gatling216 its not that complicated though. But you kinda need a filament-cutter like on bambulab... once I installed one I have had lots of succesful prints on my ercf, and the ones that failed wasnt because of the ercf 🙂
O dont know how few people have heardbabout this but the fact that people think bambu invented core xy makes me think that most people dont know much about project in 3d printing
@@joeltorres3202 Most people don't. 3D printing has always been a niche hobby. It's easy to forget that, from the outside looking in, it might as well be black magic. The average person looking into printing wants an appliance they can plug in and will work, just like a microwave or a toaster. The time, effort, and skill it takes to put together something like a Voron or an ERCF is literally incomprehensible if you're not deep in the hobby, and most people have no desire to get that deep.
I don't have time for a project this complex.
Dude, thanks for the video! I was there for a chunk of the live streams, you've put a ton into this and I know it's taken away a lot of your time from other videos. I bought one of these the day of your first live stream and I really appreciate all that you put into this. Good luck with the remaining work!
Glad you enjoyed it!
Oh really looking forward to the sv08 tool changer series - I'm patiently waiting for my SV08 to arrive.
I didn't know this existed! I was taking a shot in the dark that there might be a DIY ams build and voila! Thank you. Will be reviewing your streams.
For CAN setup, refer to "Esoterical CAN BUS Guide"; is up to date and really comprehensive.
Not owning a Voron or an ERCF, you questioned the need for two runout sensors so close to eachother:
I'm fairly sure it's because of the retraction. If you *only* have the one runout sensor at the multiple different units, when you're retracting to swap filaments you'd end up pulling up too far and not be able to use that one slot until manually reloading it.
Having a secondary sensor in the actual selector allows you to figure out that "Hey, I'm now *out* of the selector, and the second sensor's not tripped yet. Good time to switch."
The other side works the other way, it tells you you're running out at a point where you can still retract to swap and not get stuck.
Pro Tip: you can actually change the Bambu gcode for the AMS/AMS lite to also retract the filament before cutting to reduce waste! I even went a step further with my A1 mini with AMS and also changed the purge speeds slightly to make the color changes faster, still haven’t tested it tough but the filament retraction has been tested and there’s a makerworld post that has the costum gcode for it. You should give it a try
It's been around for most of this year unofficially and officially supported in May 2024 in the slicer.
There was a test that got it down to 70% reduction or 30% purge multiplier.
@@hawtdayum for now it’s only been implemented for the X1 printers, not P1 or A1
Have you looked into the Python AMS mod?
@@Belenosdj I know about it but haven’t looked into it why? I have the A1 mini so not useful for me
@nunovicente34 sorry, actually meant to reply to the video, not your comment.
I am currently in the process of this mod and wondered about their thoughts on it.
Subscribed. I started getting into printer building and modifying after breaking my back over the last decade of building cars. I've got tinkering experience and a basic understanding of circuits as I was a master of automotive electrical systems. I started with an Ender 3 in 2022 and recently moved up to a Bambu P1S to print high-quality parts for the mods with the intent of using my Ender 3, which is no longer Creality except the frame and the bed, which just got the BTT SKR 3 EZ so I can run more steppers. I recently converted an old FlashForge Adventurer 3 to an open source setup running a BTT E3 Mini v3 and Klipper. I'm almost done with that build and looking forward to my first print on it.
Next, I think I'm going to take on a Voron build or modify an old Ender 5 Plus for multi material printing.
At the price of like 200 USD, tbh the elbow grease requied is definitely worth it. Thanks for introducing this to me.
Enjoy... but I STILL don't even have mine working :)
Filametrix designer here. Thank you for that cool video ✌🏻
Building one too and have been for the last month, glad I am not the only one having trouble getting it to work lol
Yeah. It’s a lot. I wish the software had a very clear manual like the hardware does. The wiki is a lot of info but I don’t think it’s comprehensive or linear yet.
@@thenextlayer well good to know, just not going to get it then. Hope the creality one works with any klipper printer
I've got absolutely nothing about the ERCF, it seems like an awesome mmu, but now that the voron v2 has a really decent toolchanger mod which has the incredible benefits of NOT NEEDING to purge NEARLY as much (in some cases not at all if you tune your setup really well) which SIGNIFICANTLY reduces waste, as well as tool changes are WAY faster than a standard mmu (because you only have to retract a tiny ammount (a mm or two) just as you would when parking the toolhead after printing so multimaterial or multicolor parts print way faster.
I didn't even know you can build a 3D Printer from scratch. It sounds fun!
The filament movement tracking is used by some printers to monitor for jams or other extrusion issues. If a filament should be moving at X speed, and its moving to fast it may be due to printing spaghetti on a failed print, while moving too slow likely means a jam.
Can’t wait to see the 12 color prints.
I would also recommend building a Trad Rack MMU to make the full comparison more comprehensive!
If I ever recover from this build
This project seems incredible
maybe you should make a video covering this and maybe try building it and rating how hard it is 1/10?
So I'm convinced to just do multi color on my Bambu
I'm genuinely debating taking on this project... this seems fun
I recently ordered a Voron 2.4 kit to upgrade from my large bedslinger running Marlin. I've been curious about the ERCF as an upgrade later down the road (using funds from selling my current printer after the Voron is complete). Thanks for the detailed explanation of the ERCF.
This is an ambitious project I have no business doing but want to now that I've just gotten Klipper running on my mostly OEM old Ender 3 Pro with the original 8 bit board lmao
If you have an Octopus, you do not need the U2C. When configuring the firmware for the octopus, you can choose to use the octopus as the canbridge.
SInce canbus is a network, all canbus devices can be connected to the same single canbus network as nodes.
finally a video about ercf its like ams but on steroids. I love the idea and im suprised youre so late on this.
I built the other one but held off because v2 came out
Awesome video! I've seen stuff about and looked into the ERCF before and I was always curious how it actually worked. Your explanations in this video were perfect! Thanks a ton! Looking forward to seeing it when you get it all working!
Awesome, thank you!
This is amazing. Im sure all the kinks will get worked out with all the smart people working on it.
I get why some want open or full control over the system but when comparing a multi hour/day/week endeavor vs plug in some cables and ptfe tubes and be up in running in 20 minutes.
There is a market here. Why are there no plug and print solutions in the open source world?
I know, right?
You would be surprised by how many people within the open source sphere are against making things easy to use.
It's frustrating for sure. I can appreciate the level of community input getting the open source instructions/parts out there but after seeing connect and go solutions it's world apart. You are right, making it easy is secondary to making it work. Then you have competing personalities, implementations, limited resources and printer hardware/software.
Every time I see the ercf topic show up it reminds me of a Rube Goldberg machine of parts and contraptions that one errant filament line will make the whole thing come crashing down.
But if an outfit could make a polished product that connects without fuss they would shake up the open source community.
The Open community thinks making money is anathema to maker ethos.
That's why companies can swoop in and do the unthinkable like.....make a good product that meets consumer needs.
Then the OS folks clutch their pearls that ANYONE DARE make a product for the masses who should be sacrificing raspi on a pentium processor, read miles of code and search outdated forums for bespoke solutions to a problem that was fixed 3 firmware revisions back but the documentation hasn't been updated 6 revisions ago. 🤣
@@wyattutz Hmm. I'll ignore the insult and dig into the heart of your statement.
The Open source community had a several year head start and it is still a mess. They have seen a working design as of 2 years ago and still double down on the mess they had years prior.
You use the term idiot proof or idiot friendly. I say "if it works, regardless of the audience.....it still works"
A mostly closed source company had 2 working models with the AMS and AMS lite and let's be frank it is a *gold standard* of working within minutes of setup. The ams copies from creality, anycubic aim for plug and play, not build and pray. Name a better system that can be up and running with quality within 20 min.
What system do you consider the "GOLD STANDARD"
I am actually curious for your answer.
All of the OS mmu types require a build, tune and a hours upon hours of crap to get working reliably. You should know off the shelf doesn't mean best for the job. Of the thousands of engineers, software devs, talented hobbyists NO ONE saw a need and could provide a commercial/retail solution?
This is the 2nd(3rd?) revision of the ercf...and how is that working out?
"you can already get better reliability, performance etc if you are capable or sufficiently persistent."
You reaffirmed my original argument. There was money to be made here, you are telling me all of the teams, OS companies couldn't pull together to be first to market with an "idiot proof" product? The guy above was right. the Open source sphere want to make things needlessly complicated.
The filament switch sensors at the erct are just for detection if filament is in the buffer.
As for the encoder, you can go several ways. The stock voron build has a sensor at the toolhead and the encoder, which mean redundancy as with a encoder you dont need any sort of toolhead sensor as the encoder detects when the filament bottoms out at the extruder gears. You can also forgo the encoder and only use switches like the annex tradrack or prusa mmu, the software behind it is really configurable. Some people also use multiple sensors at the toolhead with the encoder, one sensor above and another one below the extruder gears. In my two builds i always skipped the toolhead sensors, above the gears is in my opinion not repeatable enough depending on bowden tube length and diameter. The sensor below gears is for me too big of a compromise for tpu as i generally print 85a and softer. Also neither of my toolheads have enough space for one.
I also built a ercf 1.1, ercf 2, and public beta tradrack. A modded ercf v1.1 is basically a v2, so if anyone got a v1.1 with springy, binky and sturdy bunny mods, there is no point in upgrading. The tradrack is a vastly different in construction, significantly simpler and cheaper to build with only one downside, you have less grip at the mmu feeder since you use a single gear instead of two. The build is also significantly quicker, it took me 3 to 4 hours, with mods such as encoder and filament brakes. The other advantage is that it can feed quicker since it only has a 3 to 1 gearing instead of 4 to 1. Last but not least its much easier to self source in my opinion, no super long d cut shaft, only one set of bmg gears, no closed loop belt
i like the state of development this 'home brew' ams enough to forgive the 2 tips. when i 1st saw the bambu ams i wondered how long before the maker community out did it, a common theme for the maker community in my experience, and to my immense surprise and delight it was apparently done far earlier than i thought. i look forward to seeing this in action
Far sooner but never a home run in terms of implementation. The AMS while simple looking is a engineering marvel compared to what was available before 2022. MMU, Palette ercf etc.
With so many moving parts standardization should be a thing.
Cutter, tip forming, buffer, and literally down the rabbit hole(pun intended) we go.
Thank you for covering this, I've looked into making one awhile ago but never found a vid i enjoyed with some one showing there journey building it. Great Vid
Amazing video! I have almost everything I need to add an ERCF, my hotend (XOL w/ ebb36) even has some filament cutters designed for it. Looking forward to you showing me how to troubleshoot it too!
Je ne dirais qu une chose ...super travail👍et j ai bien envie de regarder la video de 16h...
Yep, love the ERCF v2 (much better than v1) however I am very excited for "Armoured Turtle modular AMS" looks even more slick than Bambu AMS
I'm looking forward the seeing your ERCF in action. My Voron 2.4 is pending some planned upgrades one of which is a canbus tool head, which will also be going with tap from standard omron sensor and relocating the end stops; x on toolhead and y at the rear right. Then I can look to get my Siboor ERCF v2 connected connected to the same canbus.
You can do sensorless homing on the 2.4. That's what I did :)
I'd actually like to see the in-action comparison of all the multi-material machines, but with the Prusa MMU2 and MMU3 included!
I strongly advise you to install the second filament sensor after the Gear extruder, this way Happy Hare knows if the filament has been loaded correctly and will try again if there is an error.
I'm a contributor of the team (mcu localisation, and some minors ideas)
I had my eyes on it for a few month now.
It's an awesome video, will totally build one, but i will also take my time, the complexity and cost tell me this isn't a quick weed end project.
I have the ERCF 1.0 and it worked great. Only issue is the missing filament box with rewinder.
There's a key difference between this and the AMS - the AMS uses a dedicated feeder motor for each channel. IMHO this would be a better approach as it results in a much less complex motion system and gives you more control over buffering and spool tension
I believe the one who burnt the ships was Hernán Cortés when arrived to America.
Vasco de Gama tried to be the first to go all the way around the earth, that would be a bit hard with no ships :)
True. De Gama burned villages… mixed them up. He was a real psychopath.
@@thenextlayer well.. every civilisation has it lights and shadows …
Since I got my Sovol SV06 earlier in the year I have wondered if it would be possible to upgrade it to be able to print multiple colours. It's nice to know that it is in the realm of possibility. Maybe one day.
Great video! I was considering the enraged rabbit when the new MMU came out.
Non printing related note: if you wanted to try tracking points for the labels you added but in a new way using After Effects, it’s a more simple process than you’d think! Obviously this assumes you’re using adobe which hopefully you aren’t, but as someone who has only a basic understanding of motion graphics etc, tracking points for labels/tags comes out amazing with very limited effort.
If you have the patience to build vorons and this project, you’re more than capable of figuring out the pick whip and crazy way layers work in AE, or whatever 3rd party option there is in Davinci or Final Cut.
Your production quality clearly doesn’t need any help, but maybe this will be a new, fun little skill for your workflow!
This man is doing the lords work!
Hi! Nice video! A thought and friendly suggestion: I really think nixing the innuendo entirely would greatly improve your videos. Said as someone with a brain always in the gutter, this isn't coming from a puritan or prude here, it just made me second guess subscribing to your channel. I think you're doing really good work here, that just seems like an easy tweak that would make your content a lot more approachable (and non- offputting) to a wider variety of people.
As a bambu x1c owner with ams, coming from a diy core xy, mainly for ams, should i ever want to build or buy a new printer it would be an IDEX or anysort of tool changer. Got ams mainly for multi material prints (support + part) is about best i can tolerate, and appreciate. However even when switching between two materials, and support interface layer only supports of different material, depending on prints it can add hours of extra time. Hence why an idex type would be more suitable even for this 2 material use case. One thing i haven’t seen yet on vorons or others is the lidar sensor from bambu - that thing is a godsend - chuck in and brand of filament and off you go with zero manual tunning needed.
Hi, could I use this with my Bambu P1S and Bambu Slicer?
i ran my prusa mk3s+ with a MMU2s for 1 year stright before i took it off to just have some good reliability.
The Trad Rack by Annex team is a better option for a MMU in my opinion as i have experience building both, the trad rack is just more simpler to assemble,supports more filaments than ercf and also has fewer parts, and has higher repeatability
Was planning to build one but now I'm much more interested in multi head solution that allows me to use multiple materials on single print. ERCF doesn't solve a major problem - slowness of filament switching.
Coming soon. So like, judging by how long it takes me to finish projects… 2026?
50-50 on this. My ERCF can swap a filament in 30 seconds + purging time (another 15 seconds or so). A toolchanger still needs to move the gantry to the parked tool location, heat up the nozzle, dock, undock and move back down again. So I'm not sure how much faster it will be. However filament waste is another matter... This is I think the biggest benefit; however the drawback is much more complex calibration and alignment of the tools to each other.
@@igiannakas Prusa XL needs ~5s for tool change (from interrupting print, moving gantry, putting old tool, picking up new tool, and going back to the print). 1000 changes with 45s change is over 12.5h for tool changes only. On XL it is 1.5h only. 9x faster. Even assuming it would do twice slower changes (10s) it would still complete all changes in 3h only. It's a major deal for prints with multiple color changes.
@@arekx exactly , that is why when doing a print with multiple color changes a Prusa XL with the toolheads will be *way* faster than a Bambulab with AMS , plus you don't have waste at all with the XL , so the multiple toolheads will save up quite alot of time
I could never get the "tip tuning" thing to happen to make a phaetus taichi work properly, that would have been awesome, but I guess for now I just do manual colour changes when they're needed.
Looks interesting though I don't need or want a big thing like that which takes 12 spools. If there was a smaller version or if you could configure this one to only use 2,3 or even 4 spools I would be interested.
subbed to make sure I don't miss that 7-tool sovol build. Hope you can get different nozzle sizes slicing together, which is still a shortcoming of the Prusa XL.
Different layer heights seems even more challenging, but if both of those capabilities get working, you can do detailed outer layers and chonky infill. Stronger, more beautiful parts in multimaterial, multicolor, and minimum print time... the future is toolchanging
i almost got a voron 2.4 with an ERCF last year but ultimately decided against it and to get a 2 head prusa XL instead as multi-head will outclass multi-color/material any day and the cost of a 2 or even 5 head XL is similar in cost to a voron 2.4 with an ercf. i would still love to get one but voron and the ERCF is a project that you need to build and tweak yourself while prusa machines are a ready to go product that you can still build but require almost no tweaking.
Be nice to have options for how colors I'd like a small 3 color system
Can't watch at work, but just wanted to point out that the ERCF isn't the only open source MMU. It was based on the SMuFF which is still actively developed, and may be a better option for some people.
Consider looking at the tradrack, I've found it works much better than the original ERCF did
Tempting
Nice video... Quick question whats the Orange accent color filament on that Voron and ERCF? look really nice
Https://jle.vi/polymaker orange ASA
11:00 retraction tuning is also possible with AMS. Just FYI
Can you make a video installing an MMU on the SV08?
Great video and thank you for documenting your experience!
Glad you enjoyed it!
At least for the Sovol SV08, I'm more interested in the mods coming out for changing the entire tool head for faster filament change times and no filament waste or dealing with the poop. I almost never print in color but I'd like to have two tool heads so I can print the bottom of a part in PETG-CF, a 4 mm interface layer that interlocks the PETG-CF with TPU, and the top of the part printed in TPU.
5:20 i would think the sensor is for knowing how much fillament you use, no? Or maybe it's to check for jamming?
@@thechuckinator both
Interesting build and congrats for showing this to us.
BUT !! :)) regarding the design.
I really can't understand why not just "prepare" the filament before printing so when you print all the right colours are already on a continuous filament so there's no need to retract and cut the tip .....is automated filament welding such a difficult thing??
I only print in white but I imagine on a print farm it would be a powerhouse if you use multi colour print......or is multi colour print just a gimmick?
Sorry for the long subject, this video seemed to go into the nitty gritty and the DIY atitudine seemed perfect.
Thanks.
hmm this one has a big potential...if all the stars align.
It could reduce filament purging if you would add a filament cutter.
If you know you need 10g of filament than you would measure that amount of filament CUT (it not retract it) and than add the next color. This would mean that the amount of filament that needs to be purged from the previous color is much smaller and determined because you cut it not retract the filament and not knowing how much is there left in the head.
It would also reduce the time to retract each spool.
if you swap out the octopus with the latest version of the manta m8p (make sure it's v2), you can get rid of the U2C board altogether and go directly from the manta m8p to the ebb36 (via can bus still)... and, from personal experience in my zero-g mercury one build, it's far more reliable.
Excited for this project may do one for myself
I had some problems setting up can bus aswell, but the Manta M8P+EBB CANbus Setup Tutorial from Stacking Layers helped me quite a bit.
Its fairly up to date and covers most of the setup (except for the u2c) but maybe it helps you aswell.
I think the biggest problems with this are twofold, one is self assembly, the others a bit more complicated.
Self assembly is a huge turn off for a lot of people. Even with help, if you screw up and it partly works, it's still on you to fix it. A prebuilt product on the other hand generally comes with support and is ultimately less work on your end. That means ALOT to the average person.
The second part is the real usefulness of multi-material printing. If your prints have been designed and optimized to work in a specific material then using a different material isn't what you really want to mess with. It's another hassle and something you have to start with, if you aren't using MM to print now, then going forward you likely wont.
For instance I print alot of toys. I can design for TPU, PLA, ABS, or PETG, but generally speaking I don't NEED to print in more than one of those. So these are ultimately pointless for my design process. So why waste time and money building this?
I've been wanting to add an MMU to my printers and have been looking at the 3D Chameleon but just haven't been completely set on it, I think the ERCF is more what I am looking for and will be my winter project. Having modded my Ender 3 Pro and 3 Max to use a Sherpa Mini DD with a Revo CR with the Hero Me 7, adding in a KlackEnder Klicky Probe, Z Belts (BeltDrivenEnder3), with both running Klipper this hopefully won't give me too much trouble. For those wondering why I would mod my Ender 3s like that instead of buying/building a Voron, well I've had the Enders for years now, they print pretty decently, it's been cheaper to mod them than buy a Voron Kit and well I love to tinker much to my wife's dismay sometimes 😁
I found some video somewhere on the history of multicolor printing. and one of the first approaches had a system that mixed filament. So almost like a ink printer. You would have CMYK and black and white. rolls. and then basically it would be like mixing on the fly. The filament was melted together and extruded and then re-fed into the print head (if I recall correct( . Its very hard to control that sort of mixing. But I still think that approach should be re-checked into. maybe also, consider that filament might not be the best option to do something like that. What about printing with pellets? it would be much easier to control the mixing this way. maybe. idk. Its cool to see people working on these projects.
Good job on the video, and I hope you can get those communication issues sorted out.
Is there a smaller version of this, like maybe something in the range of 3-5 different filaments? I usually only have 2-4 spools of filament at any one time, and controlling moisture here in the Orlando area is practically a nightmare.
they make a 4 color version I believe
Could you print in any material if you were to print them in a vacuum chamber?
great introduction ^^ I was wondering if it would be possible to do a cold pull for automated changes
Tip tuning was hell. Never really got mine dialed in. Was starting to get into the new things (i have a prototype Binky) when life for in the way.
My V1.1 is still sat on my desk waiting for me to work out what i need to upgrade it. Tempted to just get a kit.
I purchased the ERCF V2 kit on West3D. haven’t started building it yet, but I hear it’s a fairly difficult build.
Where did you get your apron's LED strip? The one named "DIGI_LED_12 V1.0" ? It seems sturdier than the PCB LED I used (and soldered a million times 😅)
Came with the FYSETC kit
SOOO COOOL I want one :D BUTTTTTT is there any tutorials to add the cutter to a elegoo neptune 3 printer assembly
I'm new to the hoby but ive always wanted to tinker so this is right up my ally :)
wanted to ask aswell the back bit you didnt specify you called it a buffer so it would need to run from my drybox through the buffer to the selector ?
sorry tons of questions just passionate
You probably want to do the encoder cutter otherwise you need to modify or replace the tool head altogether
more interested in the Voron tool changers so that multicolor doesn't require purging. They already work, only hold up is slicer support
outfox zero is a design on thingiverse that first started the filament cutting for multi-color
😂 glad the hackerman bit made it in. And yeah that multimeter stuff was a drag, but the video came out great at least! Can’t wait to see the rest of the project come together
Thanks again!
Lane based systems are the way. This method of changing colors needs everything to go exactly right every time. And 3d printers are bad at always doing things exactly right... Exxcellent overview of this project though, good work!
I feel that.
I'm very curious if this will work with Bambu Lab or another non-Voron printer (Elegoo, or Creality?). Voron price and work involved to put together is a bit much for me.
Only Klipper machines, no go for Bambu
I have an IDEX, and do multi material prints on occasion, and would LOVE an MMU for one of the two heads, but realize nobody will ever build one to just work with any ol' slicer.
I wonder if this would work with the Elegoo Neptune 4 Max?? It runs Klipper
Material switching is convinient, I guess, but I really never got the whole multi color thing. Can´t you just paint the prints and way more accurate, too?
@@KiffgrasConnaisseur Yes, but… thin paints and inks wick along layer lines, and that’s assuming you got the painting gene, which sadly I did not.
I haven't seen anybody build the annex engineering tradrack though. It's another open source AMS.
I hear it’s simpler too.
And what about non klipper machines?
How about we add usability for oem brands, and non klipper printers
My canbus hang up was not knowing i had to reflash klipper onto the printers main mcu with canbus turned on. It should defiantly mention this in the instructions.
I hated the BIGTREETECH EBB36 CAN, it was unstable when it got too hot.
Eventually I switched back to the hub board.
While this is definitely cool and can definitely give some more options and choices over an ams.
This id still definitely not an enclosed system like the AMS is which limits what you can use it with compared to the AMS
Excellent video. This is an awesome project, and I'm REALLY looking forward to the 7 head tool changer. I had not heard of this before, and I'm glad you made a video of this. Thanks!
Chuck the ercf in the bin and do the tool changer, it makes it redundant.
Thanks for putting this out. It's a really impressive project!