I've been using creality printers for a while now, and I have to say their advancements in features and ease of use are truly impressive. the k2's precision and its integration of technologies like RFID filament detection really streamline the printing process. for someone like me who enjoys experimenting with multi-material projects, this level of innovation feels like a game-changer. that said, I've also had the chance to try delta-style printers like the flsun, and their speed for taller prints is genuinely appreciated. while creality remains my go-to for reliability and everyday projects, delta machines like the flsun bring unique strengths to the table. together, they make handling diverse 3D printing projects so much easier
This seems like a lot more refined design. Not sure what the vision for the future is, for the industry, but Creality is upping its game. That's a good thing.
I would like to know a lot more on how this system works, because it seems like a very risky system that if not implemented 100% properly could cause lots of issues
@@53Aries all you need is a motor to turn a screw in and out that is connected to the pulley thus applying or loosening tension simply by a motor turning a screw. really not complex at all. also it would not cause any extra problems.... if the motor doesnt turn..... you just go back to hand tightening... even worse cause scenario it wouldnt affect the printers ability to print at all
@Mr.Thermistor7228 The important part is how they are determining tension and controlling the motor's. That's what I want to know. If implemented poorly belt tension could be easily messed up. Are they relying on the stall torque of the motor at a given voltage, or what.
@@53Aries, they are probably reading the amperage draw of the motors and stopping at some set point that gives them the right tension. The chip to detect amperage draw and voltage is dirt cheap so it's a very feasible solution.
It looks like it has belt tensioning motors, and a screw to lock in the adjustment. There is a black screw accessible on the backside of the frame on both sides in line with the pulley system. It's interesting for sure, but would require more disassembly to get a good look at.
@NathanBuildsRobots yeah you showed it, i have to say that part made me really curious and its good for the vid that you went on and not startin to hyper focus on that and how it works.. wich i would do and love 🤣
@@DingleBerryschnapps I've built several printers, and I have to agree :p I'm excited about auto tensioning, because I have always been confused by the plethora of different info regarding proper tensioning. I have always done it by feel, and it seems to work out well. I'm just so used to working on German cars that have manuals with torque specs and tensioning guides!
Not sure if the questions were all real but here is what I think are the answers. 17:06 yes that one is a RPi equivalent board (better called "SBC"). 17:10 that looks like an expansion board, it has the connectors that the SBC don't have to connect to the other components, the external usb and also a power supply for the SBC. 18:19 I'm not totally sure about that, judging by the connectors, it looks like a module for the powering and driving of the RFID and the connector with 2 wires sure goes to the antena. 19:12 it doesn't have drivers cause many servos uses their own internal drivers, look like they made a universal controller board, maybe for a lower spec printer? 20:59 that indeed is a motor, a two phase stepper. 23:27 those are driver boards for the steppers. 28:24 using that zoom, those are stepper drivers, pretty similar to TMC drivers but a chinese variant, wires suggest a step/dir interfase and each one has it's own power supply. I like how modular this printer is and if Creality does it pretty well on the software side, this can be huge.
This looks like they just let an engineer go to town on the design and just said, "we need to sell it for under $1500 and make a profit. We want something to turn around our image with."
@@NathanBuildsRobots It seems like so many things a bean counter would cross off the list for a nearly zero-dollar solution: Motorized belt tensioning instead of a spring/screw; dual z motors instead of a belt; dual LED lighting, dual exhaust fans. I think you definitely nailed it - the complexity worries me if it stops "just working", but with the size and quality of this, the SV-08 may have been bumped off my next printer list.
@@See-essEll - I was planning on the SV08 being my next 3D printer and building an enclosure for it, but pending more long term reviews, I may be purchasing a K2 Plus, but not one of the earliest units. I'll wait for the inevitable rolling improvements that such a complex product will have, but it does look like Creality did their homework on this one. Bambu Lab just announced a TPU filament that'll work with their AMS so hopefully we can get a higher durometer TPU from a filament manufacturer that will print better/faster that will feed in the CFS, so I can manufacture products printed in place with ABS and TPU.
Nice to see Creality doing the chamber heater right with an insulated heater instead of the cheaper live versions and what looks like some overheat / thermal runaway protection. The overall construction and design looks very good. I still have grave concerns about the serviceability and repairablity of these though. Belts and bearings are consumables. If the open source community can make them serviceable then these companies should be able to as well.
Nice that creality got their ducks in a row. K1 seemed like it was hurried to the market leaving a lot to be desired in terms of quality and reliability. Competition is always a good thing.
21:59 Those are EPS connectors, available in 4 and 6 pins. They’re used not only in PC power supplies and motherboards but also in many industrial applications, so it’s unlikely they copied anyone. Maybe inspired
Sure, but they could have used any connector. So, I think he is pointing out that it seems likely they used a compatible connector. Perhaps just coincidence, perhaps copied because it works, or perhaps they wanted to follow it as if it were a standard (assuming pinout is also copied).
@@Jeditiger05 I think he was specifically talking about the AMS and CFS connectors. Creality has been using those connectors internally for ages, just like any other printer manufacturer. Saying they copied Bambu might be a bit of a stretch-it could just as easily be inspiration or maybe even some kind of standardization.
As someone who works on large CNCs and Industrial FDM printers, I always forget that people think 110vac is high voltage. I deal with 480vac when installing machines.
Well, it is the context of 3D printers. Just like 600mm/s is considered high speed in 3D printing, but a car with a max speed of 600mm/s would be laughable
But I’ve also heard some crazy stories about the real high voltage stuff. DON’T UNPLUG THE MACHINE WHILE IT’S RUNNING! It’s not for the machines protection, it’s for yours
@NathanBuildsRobots Right, I just forget from time to time because I am around these bigger units when I get home and play with mine. I just kind of have that muscle memory on not doing things to get shocked while live, lol.
@NathanBuildsRobots by the way, I do like the video and definitely agree with the way you stress the "high voltage" dangerous for people that don't know better.
This looks like more than a clone. It's a solid improvement and open alternative to the closed bambu ecosystem. I'm very excited for the future of these machines.
Best engineered Creality printer thus far... Only real flaw i see is the z axis bottom homing tabs/slots. If some filament falls in there it could cause issues. Hey Nathan, can you do me a favor and take a look at the heat bed and tell me if the heating element covers the entire bed or if it is smaller than the build plate? I noticed there is a smaller cutout box under the bed, so not sure if that is the size of the heating element or what.
Yeah that's a weird choice (The Z axis sensor) because they could have used the advanced motors itself for homing purposes. Still, this is such a massive leap in engineering and throughful design. The belt tensioners wasn't expected, but very much welcomed.
There is a big empty cavity under the holes, and the tabs that trigger the sensors are very stiff, solid aluminum or steel, so I doubt there would be any problems unless you had something big stored under the bed, which you're not supposed to do!
@@RetroDaddyPH using sensor less homing for z axis is a terrible idea. Even on the x and y axis sensor less homing causes excessive belt wear if the tripping force is too high and is often the root cause for large prints with a failed layer from the axis not being properly registered as zero during the initial homing but doesn't manifest until much later when the tool head has to move to the opposite extreme.
@@NathanBuildsRobots I would assume the nozzle sensor is used as actual homing sensor. Since the bed drops on power out they can't know for certain where the bed is at when powered on. By first "homing" to the bottom (which is the most likely position) they can then move the bed very speedy to the top as there is a known distance between bottom and nozzle. (edit: since they have 2 sensors at the bottom, they could actually use it for z-tilt, but in that case they are probably also used for homing/z-offset)
I got a K1 Max and now I'm ITCHING to get a K2. The max is modded to the wazoo with linear rails on X/Y, watercooled heatbreak and other goodies. It's a dream of a machine now, but the K2 with those servos and the CFS is just so juicy
You listed off the bells and whistles of this thing but is any of that actually needed? I have a K1 Max too and it prints fine. I mean, there's nothing on this machine that interests me except the bigger build volume. I don't even care about the multicolored printing. I don't even use the AMS for that on my x1c. I find it to be a waste of material and time.
How well does the watercooled heat break work on your K1 max? I've been having some heat creep issues on mine (stock heat break, Microswiss Flowtech hotend), causing jams.
Regarding the electronics.. card at the front is the CPU, a 1.2GHz 64 bit dual core ARM that can run in 85 degrees celsius without heat sink (basically a mobile phone CPU) . It has plenty of power and can run the printer with its pinky finger probably only manages the screen, IO, wifi and such... . The board on the backside controls the motors, however it seems like the actual driver is on the small breakout boards next to the motors. The small pcb at the back in top is probably the LED light driver. Feels like this printer is a very modular build and changes are to be expected in upoming versions to reduce costs.
That bottom bay may not get used very much by the average person but I'm sure the easiest use for that extra space could be desiccant storage. A more advance use could also run air across the desiccant and route the air all the way up to the AMS potentially preheating the filament to get 5% extra max flow rates while actively dehydrating the filament. This printer is looking like it might be the new standard, I just might have to get my hands on it and see how easy it is to do a 60v stepper conversion with these steppers.
I love these hardware tear down videos. Nobody else does this. Thanks! Nathan Builds Robots is the definitive source for 3D printer hardware hacking information. 19:00 - Maybe the stepper motor drivers are on the back side of the PCB and what looks like unpopulated surface mount IC pads are test probe pads? 28:25 - That's the X axis microstepping stepper motor driver daughter board, which might explain why the stepper motor drivers seem to be missing from the larger motion control board at 19:00. Bambu Lab is apparently making a TPU filament that feeds in their AMS, so maybe filament manufacturers will finally give TPU some love and my years of begging for a higher durometer TPU that prints better, easier and more consistently between manufacturers will finally be answered and I can use the CFS on a K2 Plus to print parts that combine ABS and TPU.
I'm excited about the bigger build volume. I wish Bambu would do this. But hopefully this will be available before they do and then I'll buy one. One of my large printers now is a heavily modified Ender 3 that's idex and 400x400x500. This could replace that nicely.
I also thought the closed loop motors had their own integral motor drivers, but the small board that Nathan showed at 28:25 is apparently the X axis microstepping stepper motor driver board.
They look like closed loop indeed. Look at the servo42d, very similar. Those also don't need separate drivers on the MB. I've used both the normal driver on my Arduino CNC shield and the servo42d on the back of the motors.
@@Liberty4Ever That is for the Belt Tensioning motor, not X axis motor (remember Core-XY combines both motors for X and Y movements) The main motors have 9 + X wires to each board at the end of the stepper motor. I think there might be a strain gauge and driver for the small silver motors.
@@alexa5231 - I assumed any belt tensioning motors like the small silver motor shown in the video are DC gear motors, and the calibration might not need a sensor such as a strain gauge because it may be possible to use the input shaper accelerometer to detect a loose belt. As for an X stepper motor on a core XY system, the main motion control board has unpopulated IC locations labeled X, Y, Z1 and Z2, so I was trying to use the Cartesian coordinate nomenclature that seems to be an inaccurate holdover from previous printers.
i really hope the remaining filament is measured using the weight of the spool and not writing a value to the rfid tag on the spool. that would allow the rfid tags to be placed on third party spools.
a very complex design for a printer expecting the users to operate and maintain themselves. it looks like a solid and on paper everything we wanted from a next gen printer, but yet after seeing this design, and from my experience with creality, i will rather wait to see what issues this printer might have and how hard will it be to solve them. also cant wait to see the process of fixing a clogg in this printer and how to switch/disassemble the hotend and print head
Hotend swaps should be relatively easy. Cut filament, unscrew nozzle, done. The complexity and number of PCBs does make me a little worried. Having worked in a factory manufacturing robots, one of the big challenges is diagnosing and figuring out which board needs to be replaced when there is an electronics failure. If it's designed and built properly, it should rarely be an issue. Will have to see how reliable they end up being long term.
Designing a closed heated chamber, sort of hinders serviceability. You need to work really hard as a designer, to provide heated chamber and ease of service. Since the market is stupid, and demands stuff to "feel" solid on top of that, and look "great", stuff like feel and form, is crashing at the design stage. You will also find, that the people reviewing this thing, tend to reflect on understanding of that aspect of designing, and thus never call out the insane priorities products tend to have. The security issues of printing with filaments that produce toxic gasses, and the usage of of like 2000W of power, sort of require a different mindset if you want to produce secure products for the public. Not to mention, the insane vibrations introduced by these machines. This machine shakes like a horny vibrator doing an overdose of Viagra, with a heater cooled with a el-cheapo fan. What could possibly go wrong?
@Zamugustar Yeah, but bambulab have proven themselves as manufacturing reliable machines, creality hasn't successfully done it yet. also, look at the number of bambulab models. it's exactly 4. Now look at creality. I seriously stopped counting. this is to show you where the focus of the company lays. It's not about being a fan of bambulab. it's about getting a refined product that has been tested thoroughly that will not go deprecated after a few months cause a newer and better model is out
as a k1 max owner that has had great success with mine this looks like an upgrade in every single department...I'll still wait for a few revisions like I did with my k1
@@NathanBuildsRobots The things I read it were probally cancelled preorders because there were like 30 in stock Also with a certain Halloween coupon you could get 20% off.
I'm confused about the chamber heater. Where it is pulling air from? Inside the chamber or is it constantly pulling fresh air from outside the chamber?
@@NathanBuildsRobots Seems like a big waste of energy to be constantly heating new air rather than just keeping it in the chamber. It also means more expelling of printing fumes out into the room. If you covered the back air intake/exhaust holes, do you think the chamber heater would effectively recirculate air by pulling it in from the fans to the left and right?
IMO of printing for 10+ years now... The heater and filters should never be exposed to the outside. That would make heating the chamber inefficient and most filters dont catch everything in one go, but if you cycle the air it will catch most AND more importantly wont dump toxic fumes outside the printer Other side note, if that's a load cell on the extruder then it would be able to also compensate in real time for any variance in melt speed. No more max flow tests either and perfectly melted melty melts. I would LOVE to see the g-code for it. And make different profiles for different applications. Like if I want to ensure mold like strength I would be able to set back pressure to a small value and draft to a larger value, etc. I know Bambu Labs has it on some of their printers but it's closed source. Otherwise, this is seriously convincing me to drop $1,500 at my first new printer in years. Maybe it'll be the first printer I've ever had that I haven't had to mod 🥲 I've never paid more than $300 for any of my printers, I've always gotten them open box at Micro Center or on liquidation 😅 and the belts actually look straight for the first time on a creality printer 🎉
I don't know if I'm right but i think the chamber heater should take air from the inside of the chamber (recirculate and warm it up) not push air from outside, for me it makes more sense...
You do lose some control then. If it is too hot, where is the cold air going to come from. Recirculating is good for heating, but not for controlling the temperature.
@@alexa5231 Maybe use the exhaust fans to cool the chamber, only if needed. By doing that you save some energy. Like a PID control with two outputs, if temperature is less than the set point the heater is on and chamber exhaust fans are off, if the temperature is higher than the set heater is off and chamber fans are on... Split range.
My first mod would be a cover over all three fans causing them to recirculate rather than intake/exhaust. I would prefer to have both fumes and heated air stay in the chamber.
Would you consider doing a ringing/VFA comparison between the K1, K2, and X1C? I know ringing has been an inherent issue with previous generations of Core XY printers, so it’d be great to see how the new K2 with the linear rails handle it!
@@madmatracYes, the motor may be the source of excitation but what is self-resonating ? I believe it's the mass of the printhead resonating on mass on the compliance of the supporting rails. I saw a RUclips recently of what I believe was the input shaping and while it was doing that it was moving to various spots over the print bed. That's because the resident frequency would depend greatly on how far the print head is from the support of the rail it's on, with the lowest frequency being when it's in the center and higher frequency when it's closer to the supports. I'm a mechanical engineer but I'm new to 3D printing so I could be wrong. You could verify by comparing VFA's in something printed in the center of the bed with something printed in the corner.
@nlkatz self resonant frequencies is that motor has on its own. U can just hang it floating in air without any load and it still would have rotation speeds when it make enormous noise.
I love how Nathan gets a brand new printer and rips off all the covering to expose the definitive source of under the hood Printer news. @18:34 Love it!
Nathan, maybe I missed it. You aluded to buying this printer, but you haven't said much about how you obtained it so early. No one else seems to have one. Did you order it normally? Did they send it to you for evaluation? Are you confident this is a typical unit, and not something special for a RUclipsr? It would be informative to know more about its journey to you, the unboxing, etc.
Nathan received his unit at the same time other RUclipsrs received theirs. He published his review first because he didn’t respect the embargo. You will see more reviews at November 11th
You were right to assume that the right board is the RPi clone, unless I am mistaken that board has an Allwinner T113i, which is a 1.2GHz dual core ARM chip. It is a bit low spec but will do the job.
In the past was used to seeing everyone copy Creality's printers, now everyone is copying Bambu's. Lets wait and see what Bambu puts out after the first of the year, we will have a hole new standard for printers I'm betting.
I’m very concerned about our the gantry serviceability as those panels not being easily removable is definitely concerning and is very similar to what Bambu did on their machines.
2 things .. 1: that empty bottom chamber is begging to be used. Either with a slide out drawer for storage, or an extra place to put fans for cooling for PLA purposes. 2: any idea on the wattage this thing pulls when it's at full use with the heated chamber and higher temps? Have a feeling it's going to be substantially more than lower tier printers.
One thing that I suspect the bambu has over this printer is the top layer quality. On the bambu it looks ironed. The reason for this is the x axis torsional stiffness. Bambu has two rods which makes it really torsionally stiff which results in very nice top layers.
That’s an interesting theory, I noticed that the top layers looked a little bit dragged, like they were painted on. However it’s only on small top features on the little figurines. Large flat top features were very clean. I think it has more to do with the larger meltzone. The longer the meltzone, the less control you have over fine extrusions and retractions. It’s a trade-off
@@NathanBuildsRobots It's not a theory, it's well known in the voron discord channels at this point. You can see this most easily in the top layers. There are several things, first the toolhead needs to be balanced with it's center of mass in the spot where the rail attaches to the toolhead. Second is you want the X beam to be as rigid as possible when it comes to torsion. Three things are important here. The XY joints, the x beam it self and the attachment of the x rail to the x beam (rail can rotate on the beam). Those all need to be very stiff and resist rotational forces (torsion). If you have an x axis like the bambu, there are two rods in stead of just a single rail. The double rods resist the toolhead from rotating along the x axis which is what we want. Basically you want the toolhead to resist "nodding" on quick accels in the y-direction. This is most noticeable on top layers.
It's not about the ambient temperature of the PCB. Every chip has a junction temperature. This temperature might not be exceeded! So if T(j) is 85°C, your chip produces 30° heat by itself and the ambient temperature is 60°C, your core temperature would be at least 60°C + 30°C = 90°C in the best case. But due to packagesizes and heat transfer materials it might easily be hotter than this...
Bambu suggest they are working on a new 3D printer... Look how the changed the whole 3D printing world so it will be of interest, I can't think of many improvements, size maybe
The board inside the front chamber is called "host" board, the small one in the back is called RFID sensor board and the one in the middle I don't know. Creality After Sales RUclips channel has repair/replacement tutorials for each component of the K2. The vids dealing with the main board and other electronics they show motherboard labeled map.
Your video plus theirs is really awesome stuff. I just wonder what your thoughts are about customizing the machine. I want to get the chamber/bed temps higher in an attempt to print PPS-CF and maybe even PEKK. Creality has stated that it will be open sourced and able to be rooted and Klipper run on it
@@RaRa-xg7le - I prefer plain Jane Klipper. I'd also prefer that 3D printer manufacturers publish profiles for their printers for the common slicers instead of making a custom version of Cura from two years ago.
@@Liberty4Ever Creality Print 5.1 is currently Orca based, and a first run profiles for Orca has already been released, but needs tuning, based on comments I have seen.
Curious how easy / complicated it is to get to the Extruder gears in case of TPU-salad and similar problems. Reason I didn‘t buy the Qidi, because they don‘t get it to build a reliable Extruder, and it’s a hassle to fix problems. Or, hopefully, the Extruder is just as reliable as on my four A1 minis, where I never had to open the extruder.
Is the spool drive mechanism just like Bambu AMS? That is a definite weak point of the BBL AMS with cardboard spools or when spools get light closer to empty.
Honestly there have been so many models that I have started to colour in preparation for getting the K2 that would require five colours so I was trying to figure out a good way to get it so that I can manually load in a fifth as needed. Having a motorized spool holder that works as just a fifth colour would be amazing to have instead of having to infest in a second CFS.
You can't manually load the side mounted holder and CFS together, but you can load up to 4 CFS units together on the K2 Plus, once they are released, for up to 16 filaments.
Thanks for the video. Frame on K2 is realy should become industry standard. It looks overengineered, but it's perfectly fine for 1.5 grand. Lack of dryer in the Creality AMS implementation is my biggest complaint for now. Anycubic ACE Pro has that. Last hanging questiosn are: - Is extruder motor closed loop? (has encoder on the motor) - Does hotend uses thermal grease? (did creality put additional thermal grease package in the box
The Mainboard doesn't need stepper drivers as the drivers are integrated into the closed loop control system of the motors, so the wires you see are most likely power, step, direction, enable and uart for sensorless homing. That's probably the same reason why they are outside the chamber, the driver would need active cooling. I find it very odd for the z axis to go down on its own, i have only ever seen that on belted z since there is less and different friction involved. I couldn't really tell, but were the leadscrews 4 or 8mm pitch? 4mm would be self locking if im correct, so impossible to backdrive. 8mm could be back driven, its not that hard with POM leadscrew nuts, not sure about brass ones, haven't had one of those in three years. Certainly interesting design with all those servo motors, although i still dont get it why manufacturers use smooth idlers on the teethed side of the belt. I would understand bearing stacks to a certain degree since they are sturdier, but not smooth idlers. The belts are 9mm, right? I wasn't sure as they look a bit thin compared to the ones i have on my 350 trident, but 6mm would look even thinner, ao i guess that answers my question already.
Are there any marks on the back plastic part of the printer to indicate what type of plastic it is? Are the electronics screwed into the plastic? Do you think making a back plate made of pet-cf would be possible? How about covers for electronics inside the printer made of pet-cf? I'm thinking of high temp plastics in place of the stock plastic for esthetic purposes as well as making it able to handle higher chamber temps with additional chamber heater
Well, I'm starting to be a Creality fanboy again; it's a big step up from the Ender 3 Pro I've had for a long time. Sadly this wasn't available when I bought A1
I think they could have consolidated it a little more. Having separate modules is a common way to break systems into easier to design pieces, and has some benefits in servicing and repair. Like if the main computer goes out you only have to replace one medium cost board instead a giant monolithic board. It is especially relevant if they want to use the same boards across multiple products, and lets say they make a smaller K2 that doesn't need to have all the same parts, but the main computer is the same.
@@NathanBuildsRobots - I think the unpopulated stepper motor drivers at 19:00 and the external microstepping stepper motor driver board at 28:25 is an example of this. Maybe the bigger motion control board gets smaller stepper drivers for a K2 with smaller motors but larger stepper drivers are used on daughter boards for the larger motors on a K2 Plus? This would allow more part commonality and shared design across a product line.
spray some WD40 on paper towel it will take the glue residue lots easier then IPA, looks like the load cell working like the pursa printers? I do hope they will come out with longer cables to put the CFS on the side maybe if Bambu has same cables? wonder if that would work putting the CFS on the side...
What has been said is that the CFS cables in the K2 Plus Combo basically force you to top mount it, but a retail CFS will have the longer cables, and they will also be available in the store soon-ish as well. Remember, you will be able to chain up to 4 CFS units together.
I don't have the printer yet but looking at the pictures I think you can put the CFS on the side if you mount the shock module on the CFS itself. The only thing that would be missing is one of the Bowden tubes which would need to be longer, but that is easy to acquire.
God dang creality came in strong whit this one holy moly a bigger machine, RFID for the ams and outside the printer whit filament quantity display, filament dryer, easy parts replacement on the hot end,heated chambre and belt tension motors WOW this might actually dethrone the bambu.
I've been trying to find out what the new clippy connectors are on the boards. I yanked them off my V3 XZ when I needed to replace a globbed hotend before replacements were available. Now I've lost them and I don't know what they're called. The K2 Plus is looking very cool, but it is way too big for my purposes. If they backport the technology to a K1 Max or K1/C size I'd snag one instantly.
Motors for belt tensioning is unnecessarily complex. A properly designed spring made from quality material will be much cheaper and more reliable, and give a very consistent preload force which is esily adjustable with a simple screw and nut.
A spring will change tension based on hookes law. A motor has the potential to be more accurate, but for the most part a spring is sufficient and much cheaper.
@@NathanBuildsRobots Right; by "properly designed" I meant a spring with a low rate that will have a small change in tension with the amount of deflection we're talking about.
@@motordude67 But with wear and tear, and the sudden changes in accelerations of the belt, as well as belt stretch/component wearing over time, is it the same consistency as belts seen in cars? I imagine in the short run, springs are fine, but machines with heavy usage/less maintenance, an active system compensates better in the long run?
Hey @NathanBuildsRobots, I'd be interested to know the exact limits of the print size. Is it 350x350 exactly or can it be pushed further a mm or two? Sometimes these companies advertise the size of the bed but the print size is actually smaller. Important for me as I have a business that mass prints a specific part which is 350x350mm circular. Also I'd like to know the exact chamber temp it can get up to and if the temp is even inside the whole chamber.
In my experience with the k1 max and printing almost exclusively in ASA, the electronics can not handle the heat. Once the chamber is 45c on my k1 max, things start failing. Camera usually turns off. Im currently adding a fan to the heatsinks i already added...
In theory, an IDEX version of the K2 Plus would rock - much faster filament changes and no poop. IDEX is good for doubling print speed by printing two parts at the same time but that'd play hell with input shaping calibrations needed to print at high speeds. It could be possible to print mirror image parts and have the motions cancel each other and print better/faster, but that's a special case. IDEX for faster filament switching with no poop would be worth it for some people. A mirror print mode with its own input shaping calibration could be very useful as well, but the complexity of the K2 Plus is already concerning and IDEX may be more complex/expensive than it's worth.
This looks like a solid printer. Seems Creality wanted to prove they can hang with the best when it came to this printer. As far as connectors and the like... Its all industry standard parts, so companies are bound to end up using the same ones in many cases. As long as the pinout is the same as the Bambu, this allows flexibility on the part of the user, should they need to aquire a new cable. They can run to their local MicroCenter or wherever and pick up whichever brand is in stock. Did they look at Bambu and apply what they saw, yeah definitely, but it also seems they made real effort to improve on the design as well, rather than a simple copy. Having the leveling load cell on the toolhead is a great change. I notice that when my K1 Max is doing a bed mesh, that there is more flex of the rods and gantry when it is probing the middle of the bed, that when it is probing the corners and edges. I believe that is because the load cells are sharing the pressure more evenly, so the signal is smaller. When probing the edges, only one or two cells take the majority of the force, and would make a stronger signal. The single load cell in the toolhead means the activation force will be even across the bed. It could also allow for better flow compensation, as it would allow the toolhead to sense how much force is on the nozzle while printing.
That's a good point about bed leveling. This should be able to monitor extruder back pressure as well, since the extruder pushes on the hotend, and they are both on separate sides of the loadcell
@NathanBuildsRobots Yup, looks like a ton of capability in this printer. Now Creality needs to nail the software side of things. Well tested and tuned features. I felt the K1 series had issues with some of the extra features like the "LIDAR AI" pressure advance function on the Max. I never used it, and even removed the sensor, as it seemed to not really do much. That and it doesn't work in Orca Slicer... Unless that has changed. There have been firmware updates since I stopped using it, so maybe there is improvement, but i am not switching to the Creality slicer to test it. Even though they use Orca (or Prusia, whichever) as the base now.
Any ideas how hard it would be to make the vent fans actually vent outside? With my K1 Max someone designed a clip on adapter that allowed you to attach a 4" duct and fan to vent the printer outside... I print a lot of ABS and ASA. I figured someone might be able to make something magnetic perhaps? I thought I saw a magnetic poop catcher. I invision it being like a Y shape... connect over each exhaust fan and then at the other end of the Y they converge to a 4" duct to connect for propper venting. I print in an area I occupy and the ABS fumes are not something I want in the air.
Adding motors just to tension the belts sounds a bit overengineered. By the way, how does it measure belt tension? Is there a load cell somewhere under the tension motor?
And they still might be right, this printer has a lot to prove before you can call it a success. (The print quality from other sources has been quite bad)
I mean, Creality is still holding onto 47% of the total entry-level 3d printer market, compared to Bambu's 26%- as of Q2 2024. People continue to buy Creality, despite the "Tinkering required" reputation. Kinda surprising really. Elegoo: ~10.5% Anycubic: ~10.5% Everyone else (Including Prusa): ~6%
I've been using creality printers for a while now, and I have to say their advancements in features and ease of use are truly impressive. the k2's precision and its integration of technologies like RFID filament detection really streamline the printing process. for someone like me who enjoys experimenting with multi-material projects, this level of innovation feels like a game-changer. that said, I've also had the chance to try delta-style printers like the flsun, and their speed for taller prints is genuinely appreciated. while creality remains my go-to for reliability and everyday projects, delta machines like the flsun bring unique strengths to the table. together, they make handling diverse 3D printing projects so much easier
Yes, FLSUN good. Very true, definitely a very human thought I often have
You know Nathan really likes a product when he doesn’t tear it apart or mod it within 24 hours of receiving it.
This seems like a lot more refined design. Not sure what the vision for the future is, for the industry, but Creality is upping its game. That's a good thing.
The motors for belt tension is the most interesting thing I've seen in awhile.
I would like to know a lot more on how this system works, because it seems like a very risky system that if not implemented 100% properly could cause lots of issues
@@53Aries all you need is a motor to turn a screw in and out that is connected to the pulley thus applying or loosening tension simply by a motor turning a screw. really not complex at all. also it would not cause any extra problems.... if the motor doesnt turn..... you just go back to hand tightening... even worse cause scenario it wouldnt affect the printers ability to print at all
@Mr.Thermistor7228 The important part is how they are determining tension and controlling the motor's. That's what I want to know. If implemented poorly belt tension could be easily messed up. Are they relying on the stall torque of the motor at a given voltage, or what.
@@53Aries, they are probably reading the amperage draw of the motors and stopping at some set point that gives them the right tension. The chip to detect amperage draw and voltage is dirt cheap so it's a very feasible solution.
Lets wait few months it will not take long to hear real stories how it performs. If it will be reliable 350x3 is very attractive.
350^3 ;-)
The printer Stratasys dreamed of. Those motors do the belt tension, no more manual belt tensioning ❤
It looks like it has belt tensioning motors, and a screw to lock in the adjustment. There is a black screw accessible on the backside of the frame on both sides in line with the pulley system. It's interesting for sure, but would require more disassembly to get a good look at.
@NathanBuildsRobots yeah you showed it, i have to say that part made me really curious and its good for the vid that you went on and not startin to hyper focus on that and how it works.. wich i would do and love 🤣
Even my Bambu A1 Mini can test the belts and tell me I need to adjust them. Being able to do it automatically is a nice new feature to see!
Yeah, I'm thinking back about all the hours and days I spent tensioning my belts lmao.
People are so lazy now.
@@DingleBerryschnapps I've built several printers, and I have to agree :p
I'm excited about auto tensioning, because I have always been confused by the plethora of different info regarding proper tensioning. I have always done it by feel, and it seems to work out well. I'm just so used to working on German cars that have manuals with torque specs and tensioning guides!
Not sure if the questions were all real but here is what I think are the answers.
17:06 yes that one is a RPi equivalent board (better called "SBC").
17:10 that looks like an expansion board, it has the connectors that the SBC don't have to connect to the other components, the external usb and also a power supply for the SBC.
18:19 I'm not totally sure about that, judging by the connectors, it looks like a module for the powering and driving of the RFID and the connector with 2 wires sure goes to the antena.
19:12 it doesn't have drivers cause many servos uses their own internal drivers, look like they made a universal controller board, maybe for a lower spec printer?
20:59 that indeed is a motor, a two phase stepper.
23:27 those are driver boards for the steppers.
28:24 using that zoom, those are stepper drivers, pretty similar to TMC drivers but a chinese variant, wires suggest a step/dir interfase and each one has it's own power supply.
I like how modular this printer is and if Creality does it pretty well on the software side, this can be huge.
This looks like they just let an engineer go to town on the design and just said, "we need to sell it for under $1500 and make a profit. We want something to turn around our image with."
There is much better attention to detail on this machine compared to their previous designs, I can say that much.
@@NathanBuildsRobots It seems like so many things a bean counter would cross off the list for a nearly zero-dollar solution: Motorized belt tensioning instead of a spring/screw; dual z motors instead of a belt; dual LED lighting, dual exhaust fans.
I think you definitely nailed it - the complexity worries me if it stops "just working", but with the size and quality of this, the SV-08 may have been bumped off my next printer list.
I'm pretty happy with the K1C. But I am definitely K2 Plus curious!
@@See-essEll - I was planning on the SV08 being my next 3D printer and building an enclosure for it, but pending more long term reviews, I may be purchasing a K2 Plus, but not one of the earliest units. I'll wait for the inevitable rolling improvements that such a complex product will have, but it does look like Creality did their homework on this one.
Bambu Lab just announced a TPU filament that'll work with their AMS so hopefully we can get a higher durometer TPU from a filament manufacturer that will print better/faster that will feed in the CFS, so I can manufacture products printed in place with ABS and TPU.
A definite buy for me but maybe 6 months or so down the line.
Nice to see Creality doing the chamber heater right with an insulated heater instead of the cheaper live versions and what looks like some overheat / thermal runaway protection. The overall construction and design looks very good.
I still have grave concerns about the serviceability and repairablity of these though. Belts and bearings are consumables. If the open source community can make them serviceable then these companies should be able to as well.
chamber heater should be close to the bottom though. hot air rises....
Nice that creality got their ducks in a row. K1 seemed like it was hurried to the market leaving a lot to be desired in terms of quality and reliability.
Competition is always a good thing.
21:59 Those are EPS connectors, available in 4 and 6 pins. They’re used not only in PC power supplies and motherboards but also in many industrial applications, so it’s unlikely they copied anyone. Maybe inspired
Sure, but they could have used any connector. So, I think he is pointing out that it seems likely they used a compatible connector. Perhaps just coincidence, perhaps copied because it works, or perhaps they wanted to follow it as if it were a standard (assuming pinout is also copied).
@CL-gq3no Creality have been using these connectors since the CR10 so if anything Bambu copied them!
@@Jeditiger05 I think he was specifically talking about the AMS and CFS connectors. Creality has been using those connectors internally for ages, just like any other printer manufacturer. Saying they copied Bambu might be a bit of a stretch-it could just as easily be inspiration or maybe even some kind of standardization.
As someone who works on large CNCs and Industrial FDM printers, I always forget that people think 110vac is high voltage. I deal with 480vac when installing machines.
Well, it is the context of 3D printers. Just like 600mm/s is considered high speed in 3D printing, but a car with a max speed of 600mm/s would be laughable
But I’ve also heard some crazy stories about the real high voltage stuff. DON’T UNPLUG THE MACHINE WHILE IT’S RUNNING! It’s not for the machines protection, it’s for yours
@NathanBuildsRobots Right, I just forget from time to time because I am around these bigger units when I get home and play with mine. I just kind of have that muscle memory on not doing things to get shocked while live, lol.
@NathanBuildsRobots by the way, I do like the video and definitely agree with the way you stress the "high voltage" dangerous for people that don't know better.
Well, you simply should not work with industrial machines, if you do not classify 110V as heigh voltage. Because it is.
This looks like more than a clone. It's a solid improvement and open alternative to the closed bambu ecosystem.
I'm very excited for the future of these machines.
As far as Clones go, perhaps this is a Clone Commando? Like the ones from Republic Commando...
good thing they brought back the leveling knobs. That was the first thing I had to add to my K1 Max
I’m glad you liked that little detail, almost forgot to mention it!
Best engineered Creality printer thus far... Only real flaw i see is the z axis bottom homing tabs/slots. If some filament falls in there it could cause issues. Hey Nathan, can you do me a favor and take a look at the heat bed and tell me if the heating element covers the entire bed or if it is smaller than the build plate? I noticed there is a smaller cutout box under the bed, so not sure if that is the size of the heating element or what.
Yeah that's a weird choice (The Z axis sensor) because they could have used the advanced motors itself for homing purposes. Still, this is such a massive leap in engineering and throughful design. The belt tensioners wasn't expected, but very much welcomed.
There is a big empty cavity under the holes, and the tabs that trigger the sensors are very stiff, solid aluminum or steel, so I doubt there would be any problems unless you had something big stored under the bed, which you're not supposed to do!
Motor homing won't be as accurate or repeatable as the optical sensors they are using. I think they made the right choice
@@RetroDaddyPH using sensor less homing for z axis is a terrible idea. Even on the x and y axis sensor less homing causes excessive belt wear if the tripping force is too high and is often the root cause for large prints with a failed layer from the axis not being properly registered as zero during the initial homing but doesn't manifest until much later when the tool head has to move to the opposite extreme.
@@NathanBuildsRobots I would assume the nozzle sensor is used as actual homing sensor. Since the bed drops on power out they can't know for certain where the bed is at when powered on. By first "homing" to the bottom (which is the most likely position) they can then move the bed very speedy to the top as there is a known distance between bottom and nozzle. (edit: since they have 2 sensors at the bottom, they could actually use it for z-tilt, but in that case they are probably also used for homing/z-offset)
I got a K1 Max and now I'm ITCHING to get a K2. The max is modded to the wazoo with linear rails on X/Y, watercooled heatbreak and other goodies. It's a dream of a machine now, but the K2 with those servos and the CFS is just so juicy
You listed off the bells and whistles of this thing but is any of that actually needed? I have a K1 Max too and it prints fine. I mean, there's nothing on this machine that interests me except the bigger build volume.
I don't even care about the multicolored printing. I don't even use the AMS for that on my x1c. I find it to be a waste of material and time.
How well does the watercooled heat break work on your K1 max? I've been having some heat creep issues on mine (stock heat break, Microswiss Flowtech hotend), causing jams.
Thank goodness they actually redesigned the extruder.
"Can I copy ur homework?"
"Sure, but don't make it obvious"
I hope they open source the filament monitoring rfid system because it would be awesome to have custom rfid tags per custom non Creality filament
Thank you to show us.
That space in the bottom, one good mod, i think. Maybe design a tray that fit in the bottom for storage tools and other things.
I’ve already started storing things down there. Good idea to make it a proper tool box!
You could store an A1 mini down there.
🤣
Ideas...
1) Tray for a stack of various build plates.
2) HEPA/carbon air filter similar to bento box or nevermore.
3) Hot tub.
This looks good. Hopefully there will be no post-release surprises. I like the direction Creality is going, I hope they keep this trend.
Regarding the electronics.. card at the front is the CPU, a 1.2GHz 64 bit dual core ARM that can run in 85 degrees celsius without heat sink (basically a mobile phone CPU) . It has plenty of power and can run the printer with its pinky finger probably only manages the screen, IO, wifi and such... .
The board on the backside controls the motors, however it seems like the actual driver is on the small breakout boards next to the motors.
The small pcb at the back in top is probably the LED light driver.
Feels like this printer is a very modular build and changes are to be expected in upoming versions to reduce costs.
Thanks for this teardown. Very cool to see the amount of changes on this vs the K1 series.
Looks like a massive leap in how it works.
That bottom bay may not get used very much by the average person but I'm sure the easiest use for that extra space could be desiccant storage. A more advance use could also run air across the desiccant and route the air all the way up to the AMS potentially preheating the filament to get 5% extra max flow rates while actively dehydrating the filament.
This printer is looking like it might be the new standard, I just might have to get my hands on it and see how easy it is to do a 60v stepper conversion with these steppers.
I love these hardware tear down videos. Nobody else does this. Thanks! Nathan Builds Robots is the definitive source for 3D printer hardware hacking information.
19:00 - Maybe the stepper motor drivers are on the back side of the PCB and what looks like unpopulated surface mount IC pads are test probe pads?
28:25 - That's the X axis microstepping stepper motor driver daughter board, which might explain why the stepper motor drivers seem to be missing from the larger motion control board at 19:00.
Bambu Lab is apparently making a TPU filament that feeds in their AMS, so maybe filament manufacturers will finally give TPU some love and my years of begging for a higher durometer TPU that prints better, easier and more consistently between manufacturers will finally be answered and I can use the CFS on a K2 Plus to print parts that combine ABS and TPU.
Damnit Creality. I swore never again with you but now Im tempted by this thing
I'm excited about the bigger build volume. I wish Bambu would do this. But hopefully this will be available before they do and then I'll buy one. One of my large printers now is a heavily modified Ender 3 that's idex and 400x400x500. This could replace that nicely.
lol how can you even call that an Ender 3 at that point 🤣
@NathanBuildsRobots well the build plate under the 400x400 plate is still the Ender one. Lol
the motors are closed loop afaik, so they don't need drivers on MB
I also thought the closed loop motors had their own integral motor drivers, but the small board that Nathan showed at 28:25 is apparently the X axis microstepping stepper motor driver board.
They look like closed loop indeed. Look at the servo42d, very similar. Those also don't need separate drivers on the MB. I've used both the normal driver on my Arduino CNC shield and the servo42d on the back of the motors.
@@Liberty4Ever That is for the Belt Tensioning motor, not X axis motor (remember Core-XY combines both motors for X and Y movements) The main motors have 9 + X wires to each board at the end of the stepper motor. I think there might be a strain gauge and driver for the small silver motors.
@@alexa5231 - I assumed any belt tensioning motors like the small silver motor shown in the video are DC gear motors, and the calibration might not need a sensor such as a strain gauge because it may be possible to use the input shaper accelerometer to detect a loose belt.
As for an X stepper motor on a core XY system, the main motion control board has unpopulated IC locations labeled X, Y, Z1 and Z2, so I was trying to use the Cartesian coordinate nomenclature that seems to be an inaccurate holdover from previous printers.
I would like to see the auto level process, time to heat the bed to 110, time to heat the nozzle to 300c
i really hope the remaining filament is measured using the weight of the spool and not writing a value to the rfid tag on the spool. that would allow the rfid tags to be placed on third party spools.
I would almost guarantee it's writing to the RFID.
@KJMcLaws probably, yeah
Yeah that is the most likely scenario. Adding 4 additional weight sensors would be expensive
a very complex design for a printer expecting the users to operate and maintain themselves.
it looks like a solid and on paper everything we wanted from a next gen printer, but yet after seeing this design, and from my experience with creality, i will rather wait to see what issues this printer might have and how hard will it be to solve them.
also cant wait to see the process of fixing a clogg in this printer and how to switch/disassemble the hotend and print head
Hotend swaps should be relatively easy. Cut filament, unscrew nozzle, done.
The complexity and number of PCBs does make me a little worried. Having worked in a factory manufacturing robots, one of the big challenges is diagnosing and figuring out which board needs to be replaced when there is an electronics failure. If it's designed and built properly, it should rarely be an issue. Will have to see how reliable they end up being long term.
Yes but this is the same thing Bambu has done and their fans worship them for it. The days of easy to work on printers are numbered.
Designing a closed heated chamber, sort of hinders serviceability. You need to work really hard as a designer, to provide heated chamber and ease of service. Since the market is stupid, and demands stuff to "feel" solid on top of that, and look "great", stuff like feel and form, is crashing at the design stage. You will also find, that the people reviewing this thing, tend to reflect on understanding of that aspect of designing, and thus never call out the insane priorities products tend to have.
The security issues of printing with filaments that produce toxic gasses, and the usage of of like 2000W of power, sort of require a different mindset if you want to produce secure products for the public. Not to mention, the insane vibrations introduced by these machines. This machine shakes like a horny vibrator doing an overdose of Viagra, with a heater cooled with a el-cheapo fan. What could possibly go wrong?
@Zamugustar Yeah, but bambulab have proven themselves as manufacturing reliable machines, creality hasn't successfully done it yet.
also, look at the number of bambulab models. it's exactly 4. Now look at creality. I seriously stopped counting.
this is to show you where the focus of the company lays.
It's not about being a fan of bambulab. it's about getting a refined product that has been tested thoroughly that will not go deprecated after a few months cause a newer and better model is out
as a k1 max owner that has had great success with mine this looks like an upgrade in every single department...I'll still wait for a few revisions like I did with my k1
Thanks, appreciate the honest and educational reviews. Ordered mine a few days ago after seeing your unboxing video.
I didn't realize they are taking new preorders yet?
@@NathanBuildsRobots The things I read it were probally cancelled preorders because there were like 30 in stock
Also with a certain Halloween coupon you could get 20% off.
Love the possibilities here! Another level up that is a result of free market competition!
this machine is SOLID, hype for it's trickle down effect on other manufacturers
I'm confused about the chamber heater. Where it is pulling air from? Inside the chamber or is it constantly pulling fresh air from outside the chamber?
heater pulls in air from outside, it has a 60C limit.
If they needed it to get hotter or run more efficiently they could recirculate air
@@NathanBuildsRobots Seems like a big waste of energy to be constantly heating new air rather than just keeping it in the chamber. It also means more expelling of printing fumes out into the room. If you covered the back air intake/exhaust holes, do you think the chamber heater would effectively recirculate air by pulling it in from the fans to the left and right?
The 100% trusted reviewer.
IMO of printing for 10+ years now... The heater and filters should never be exposed to the outside. That would make heating the chamber inefficient and most filters dont catch everything in one go, but if you cycle the air it will catch most AND more importantly wont dump toxic fumes outside the printer
Other side note, if that's a load cell on the extruder then it would be able to also compensate in real time for any variance in melt speed. No more max flow tests either and perfectly melted melty melts. I would LOVE to see the g-code for it. And make different profiles for different applications. Like if I want to ensure mold like strength I would be able to set back pressure to a small value and draft to a larger value, etc. I know Bambu Labs has it on some of their printers but it's closed source.
Otherwise, this is seriously convincing me to drop $1,500 at my first new printer in years. Maybe it'll be the first printer I've ever had that I haven't had to mod 🥲 I've never paid more than $300 for any of my printers, I've always gotten them open box at Micro Center or on liquidation 😅 and the belts actually look straight for the first time on a creality printer 🎉
Nice! Looks a lot more refined thatn the Max. Good Job!
I don't know if I'm right but i think the chamber heater should take air from the inside of the chamber (recirculate and warm it up) not push air from outside, for me it makes more sense...
You do lose some control then. If it is too hot, where is the cold air going to come from. Recirculating is good for heating, but not for controlling the temperature.
@@alexa5231 Maybe use the exhaust fans to cool the chamber, only if needed. By doing that you save some energy. Like a PID control with two outputs, if temperature is less than the set point the heater is on and chamber exhaust fans are off, if the temperature is higher than the set heater is off and chamber fans are on... Split range.
My first mod would be a cover over all three fans causing them to recirculate rather than intake/exhaust. I would prefer to have both fumes and heated air stay in the chamber.
@@CL-gq3no Yeah, there are fumes also, you are right, I forgot about them.
Would you consider doing a ringing/VFA comparison between the K1, K2, and X1C? I know ringing has been an inherent issue with previous generations of Core XY printers, so it’d be great to see how the new K2 with the linear rails handle it!
u have no vfa on servomotors
@@madmatrac I thought one of the causes of VFA is vibration of the X and Y rails, and that that is what is being characterized by the input shaping.
@@nlkatz the one and only vfa reason is stepper motor stepping over self resonant frequencies.
@@madmatracYes, the motor may be the source of excitation but what is self-resonating ? I believe it's the mass of the printhead resonating on mass on the compliance of the supporting rails.
I saw a RUclips recently of what I believe was the input shaping and while it was doing that it was moving to various spots over the print bed. That's because the resident frequency would depend greatly on how far the print head is from the support of the rail it's on, with the lowest frequency being when it's in the center and higher frequency when it's closer to the supports.
I'm a mechanical engineer but I'm new to 3D printing so I could be wrong.
You could verify by comparing VFA's in something printed in the center of the bed with something printed in the corner.
@nlkatz self resonant frequencies is that motor has on its own. U can just hang it floating in air without any load and it still would have rotation speeds when it make enormous noise.
I love how Nathan gets a brand new printer and rips off all the covering to expose the definitive source of under the hood Printer news. @18:34 Love it!
Nathan, don't yank on it, swing on it with both feet off the ground. It is your signature move when testing printers.
Looks like a beast.
I want one ... maybe two :)
So I have an X1C and I have to say I want one of these. This seems to have taken everything a step further.
Nathan, maybe I missed it. You aluded to buying this printer, but you haven't said much about how you obtained it so early. No one else seems to have one. Did you order it normally? Did they send it to you for evaluation? Are you confident this is a typical unit, and not something special for a RUclipsr? It would be informative to know more about its journey to you, the unboxing, etc.
Nathan received his unit at the same time other RUclipsrs received theirs. He published his review first because he didn’t respect the embargo. You will see more reviews at November 11th
@@GeekDetour Ah. That would explain it.
@GeekDetour I’m not disrespecting anything
@@NathanBuildsRobots But all I was wondering was, did you order it normally? Or did they send it for evaluation (with explicit or implied conditions)?
Really hope they make a cr10max or 500mm version of this printer.
You were right to assume that the right board is the RPi clone, unless I am mistaken that board has an Allwinner T113i, which is a 1.2GHz dual core ARM chip. It is a bit low spec but will do the job.
I figured it was a little lower in spec. The touchscreen isn't the most responsive thing, but it gets the job done.
any lowspec arm chip would be planet ahead of k1 chip of wifi router
In the past was used to seeing everyone copy Creality's printers, now everyone is copying Bambu's. Lets wait and see what Bambu puts out after the first of the year, we will have a hole new standard for printers I'm betting.
There 1500$printer doesnt come near this one tho. Spec wise.
Bambulab wasted time .....and the others moved on....
K2 plus much more for the same price as the X1C (without discount).
I’m very concerned about our the gantry serviceability as those panels not being easily removable is definitely concerning and is very similar to what Bambu did on their machines.
Aren't Stratasys suing Bambu over the RFID spools?
Yes, lawsuits are expensive, but if they view Creality as a threat they'll probably sue them too.
2 things ..
1: that empty bottom chamber is begging to be used. Either with a slide out drawer for storage, or an extra place to put fans for cooling for PLA purposes.
2: any idea on the wattage this thing pulls when it's at full use with the heated chamber and higher temps? Have a feeling it's going to be substantially more than lower tier printers.
The stated specs is 1200W, no idea of actual power draw yet.
I'm guessing it draws ~500W max on 120V, and 1200W max on 240v.
One thing that I suspect the bambu has over this printer is the top layer quality. On the bambu it looks ironed. The reason for this is the x axis torsional stiffness. Bambu has two rods which makes it really torsionally stiff which results in very nice top layers.
That’s an interesting theory, I noticed that the top layers looked a little bit dragged, like they were painted on.
However it’s only on small top features on the little figurines. Large flat top features were very clean. I think it has more to do with the larger meltzone. The longer the meltzone, the less control you have over fine extrusions and retractions. It’s a trade-off
@@NathanBuildsRobots It's not a theory, it's well known in the voron discord channels at this point. You can see this most easily in the top layers. There are several things, first the toolhead needs to be balanced with it's center of mass in the spot where the rail attaches to the toolhead. Second is you want the X beam to be as rigid as possible when it comes to torsion. Three things are important here. The XY joints, the x beam it self and the attachment of the x rail to the x beam (rail can rotate on the beam). Those all need to be very stiff and resist rotational forces (torsion). If you have an x axis like the bambu, there are two rods in stead of just a single rail. The double rods resist the toolhead from rotating along the x axis which is what we want. Basically you want the toolhead to resist "nodding" on quick accels in the y-direction. This is most noticeable on top layers.
It's not about the ambient temperature of the PCB. Every chip has a junction temperature. This temperature might not be exceeded! So if T(j) is 85°C, your chip produces 30° heat by itself and the ambient temperature is 60°C, your core temperature would be at least 60°C + 30°C = 90°C in the best case. But due to packagesizes and heat transfer materials it might easily be hotter than this...
Bambu suggest they are working on a new 3D printer... Look how the changed the whole 3D printing world so it will be of interest, I can't think of many improvements, size maybe
Size and tool changer probably, imagine a 4 or 5 head system like the Prusa XL with up tom 4 AMS units per head...
The board inside the front chamber is called "host" board, the small one in the back is called RFID sensor board and the one in the middle I don't know. Creality After Sales RUclips channel has repair/replacement tutorials for each component of the K2. The vids dealing with the main board and other electronics they show motherboard labeled map.
Your video plus theirs is really awesome stuff. I just wonder what your thoughts are about customizing the machine. I want to get the chamber/bed temps higher in an attempt to print PPS-CF and maybe even PEKK. Creality has stated that it will be open sourced and able to be rooted and Klipper run on it
Ahh that's awesome. They really need to step up on the support documents to match the complexity of this machine.
@@RaRa-xg7le - I prefer plain Jane Klipper. I'd also prefer that 3D printer manufacturers publish profiles for their printers for the common slicers instead of making a custom version of Cura from two years ago.
@@Liberty4Ever Creality Print 5.1 is currently Orca based, and a first run profiles for Orca has already been released, but needs tuning, based on comments I have seen.
Curious how easy / complicated it is to get to the Extruder gears in case of TPU-salad and similar problems. Reason I didn‘t buy the Qidi, because they don‘t get it to build a reliable Extruder, and it’s a hassle to fix problems.
Or, hopefully, the Extruder is just as reliable as on my four A1 minis, where I never had to open the extruder.
Is the spool drive mechanism just like Bambu AMS? That is a definite weak point of the BBL AMS with cardboard spools or when spools get light closer to empty.
Honestly there have been so many models that I have started to colour in preparation for getting the K2 that would require five colours so I was trying to figure out a good way to get it so that I can manually load in a fifth as needed. Having a motorized spool holder that works as just a fifth colour would be amazing to have instead of having to infest in a second CFS.
You can't manually load the side mounted holder and CFS together, but you can load up to 4 CFS units together on the K2 Plus, once they are released, for up to 16 filaments.
With the CMS on top, does that mean they've fixed the issue with it getting too hot inside to print PLA and the extruder clogging?
Thanks for the video.
Frame on K2 is realy should become industry standard. It looks overengineered, but it's perfectly fine for 1.5 grand.
Lack of dryer in the Creality AMS implementation is my biggest complaint for now. Anycubic ACE Pro has that.
Last hanging questiosn are:
- Is extruder motor closed loop? (has encoder on the motor)
- Does hotend uses thermal grease? (did creality put additional thermal grease package in the box
I touched grass right before starting the video, but it’s still for me!
but did you smoke it?
@@jud-asinsmith-stansell2022 the step most people forget!
The Mainboard doesn't need stepper drivers as the drivers are integrated into the closed loop control system of the motors, so the wires you see are most likely power, step, direction, enable and uart for sensorless homing. That's probably the same reason why they are outside the chamber, the driver would need active cooling. I find it very odd for the z axis to go down on its own, i have only ever seen that on belted z since there is less and different friction involved. I couldn't really tell, but were the leadscrews 4 or 8mm pitch? 4mm would be self locking if im correct, so impossible to backdrive. 8mm could be back driven, its not that hard with POM leadscrew nuts, not sure about brass ones, haven't had one of those in three years. Certainly interesting design with all those servo motors, although i still dont get it why manufacturers use smooth idlers on the teethed side of the belt. I would understand bearing stacks to a certain degree since they are sturdier, but not smooth idlers. The belts are 9mm, right? I wasn't sure as they look a bit thin compared to the ones i have on my 350 trident, but 6mm would look even thinner, ao i guess that answers my question already.
Are there any marks on the back plastic part of the printer to indicate what type of plastic it is? Are the electronics screwed into the plastic? Do you think making a back plate made of pet-cf would be possible? How about covers for electronics inside the printer made of pet-cf? I'm thinking of high temp plastics in place of the stock plastic for esthetic purposes as well as making it able to handle higher chamber temps with additional chamber heater
Also make sure that the UI on the touch screen is green.
Good to see creality getting out of the one trick bed slingers and doing well.
Don't touch high voltage, touch grass instead, but only if it's less than 24 hours before the next NBR video drop.
Well, I'm starting to be a Creality fanboy again; it's a big step up from the Ender 3 Pro I've had for a long time. Sadly this wasn't available when I bought A1
Thanks for this deep dive into the internals. Did creality make this machine more complicated then it needed to be?
I think they could have consolidated it a little more. Having separate modules is a common way to break systems into easier to design pieces, and has some benefits in servicing and repair. Like if the main computer goes out you only have to replace one medium cost board instead a giant monolithic board. It is especially relevant if they want to use the same boards across multiple products, and lets say they make a smaller K2 that doesn't need to have all the same parts, but the main computer is the same.
@@NathanBuildsRobots those are some good points! I would be surprised if they dont release a normal K2 before second half of 2025.
@@NathanBuildsRobots - I think the unpopulated stepper motor drivers at 19:00 and the external microstepping stepper motor driver board at 28:25 is an example of this. Maybe the bigger motion control board gets smaller stepper drivers for a K2 with smaller motors but larger stepper drivers are used on daughter boards for the larger motors on a K2 Plus? This would allow more part commonality and shared design across a product line.
spray some WD40 on paper towel it will take the glue residue lots easier then IPA, looks like the load cell working like the pursa printers? I do hope they will come out with longer cables to put the CFS on the side maybe if Bambu has same cables? wonder if that would work putting the CFS on the side...
What has been said is that the CFS cables in the K2 Plus Combo basically force you to top mount it, but a retail CFS will have the longer cables, and they will also be available in the store soon-ish as well.
Remember, you will be able to chain up to 4 CFS units together.
I don't have the printer yet but looking at the pictures I think you can put the CFS on the side if you mount the shock module on the CFS itself. The only thing that would be missing is one of the Bowden tubes which would need to be longer, but that is easy to acquire.
Regarding the motor controller, I would assume it has something to do with the fact that they are FOC motors, not stepper motors
I also find that a doubly supported shaft is best.
God dang creality came in strong whit this one holy moly a bigger machine, RFID for the ams and outside the printer whit filament quantity display, filament dryer, easy parts replacement on the hot end,heated chambre and belt tension motors WOW this might actually dethrone the bambu.
I've been trying to find out what the new clippy connectors are on the boards. I yanked them off my V3 XZ when I needed to replace a globbed hotend before replacements were available. Now I've lost them and I don't know what they're called.
The K2 Plus is looking very cool, but it is way too big for my purposes. If they backport the technology to a K1 Max or K1/C size I'd snag one instantly.
Great video 👍🏻 I can’t see any microswitches in the gantry, isn’t it run in sensorless homing approach? Klipper firmware?
Yes
It looks much higher quality than the X1C. I hope they are able to lower the price a bit maybe to $1199 or so.
Nathan, could you compare the poops between the Creality and Bambu?
Great info
How long do you think it will take for filament bits to fall into the homing holes?
The holes are really small, and even if they did fall in there is a massive cavity in the base to absorb the strays.
Is on the right board really a Micro-USB Connector.
Why the board for Motors need Stepper-Drivers, because these are Servos
I have a ender5 s1 and the k2 will definitly be my upgrade.
Motors for belt tensioning is unnecessarily complex.
A properly designed spring made from quality material will be much cheaper and more reliable, and give a very consistent preload force which is esily adjustable with a simple screw and nut.
A spring will change tension based on hookes law. A motor has the potential to be more accurate, but for the most part a spring is sufficient and much cheaper.
@@NathanBuildsRobots Right; by "properly designed" I meant a spring with a low rate that will have a small change in tension with the amount of deflection we're talking about.
@@NathanBuildsRobots, belt tensioning by springs is the norm in millions of car engines. This is an unnecessary complication of a simple problem.
@@motordude67 But with wear and tear, and the sudden changes in accelerations of the belt, as well as belt stretch/component wearing over time, is it the same consistency as belts seen in cars? I imagine in the short run, springs are fine, but machines with heavy usage/less maintenance, an active system compensates better in the long run?
After being burnt by the K1, IDK if I can go back to it.
Hey @NathanBuildsRobots, I'd be interested to know the exact limits of the print size. Is it 350x350 exactly or can it be pushed further a mm or two? Sometimes these companies advertise the size of the bed but the print size is actually smaller. Important for me as I have a business that mass prints a specific part which is 350x350mm circular. Also I'd like to know the exact chamber temp it can get up to and if the temp is even inside the whole chamber.
Nice! I am definitely needed of a printer upgrade and this might be it
In my experience with the k1 max and printing almost exclusively in ASA, the electronics can not handle the heat.
Once the chamber is 45c on my k1 max, things start failing. Camera usually turns off. Im currently adding a fan to the heatsinks i already added...
NGL im pretty impressed
Make that IDEX and I'd be interested. Still a very promising start, good to see some movement in the market.
Unfortunately I don’t think that would fit into this design, unless they moved the top electronics bay into the rear. Maybe in a K2 Pro model?
In theory, an IDEX version of the K2 Plus would rock - much faster filament changes and no poop. IDEX is good for doubling print speed by printing two parts at the same time but that'd play hell with input shaping calibrations needed to print at high speeds. It could be possible to print mirror image parts and have the motions cancel each other and print better/faster, but that's a special case. IDEX for faster filament switching with no poop would be worth it for some people. A mirror print mode with its own input shaping calibration could be very useful as well, but the complexity of the K2 Plus is already concerning and IDEX may be more complex/expensive than it's worth.
How many companies will stratysis have to sue before they go bankrupt to legal fees and a lack of sales?
It’s NOT what a printer can do, it’s what it DOESN’T do that counts! I will go with the printer that messes up the least.
This looks like a solid printer. Seems Creality wanted to prove they can hang with the best when it came to this printer.
As far as connectors and the like... Its all industry standard parts, so companies are bound to end up using the same ones in many cases. As long as the pinout is the same as the Bambu, this allows flexibility on the part of the user, should they need to aquire a new cable. They can run to their local MicroCenter or wherever and pick up whichever brand is in stock.
Did they look at Bambu and apply what they saw, yeah definitely, but it also seems they made real effort to improve on the design as well, rather than a simple copy.
Having the leveling load cell on the toolhead is a great change. I notice that when my K1 Max is doing a bed mesh, that there is more flex of the rods and gantry when it is probing the middle of the bed, that when it is probing the corners and edges. I believe that is because the load cells are sharing the pressure more evenly, so the signal is smaller. When probing the edges, only one or two cells take the majority of the force, and would make a stronger signal. The single load cell in the toolhead means the activation force will be even across the bed.
It could also allow for better flow compensation, as it would allow the toolhead to sense how much force is on the nozzle while printing.
That's a good point about bed leveling. This should be able to monitor extruder back pressure as well, since the extruder pushes on the hotend, and they are both on separate sides of the loadcell
@NathanBuildsRobots Yup, looks like a ton of capability in this printer.
Now Creality needs to nail the software side of things. Well tested and tuned features. I felt the K1 series had issues with some of the extra features like the "LIDAR AI" pressure advance function on the Max. I never used it, and even removed the sensor, as it seemed to not really do much. That and it doesn't work in Orca Slicer... Unless that has changed. There have been firmware updates since I stopped using it, so maybe there is improvement, but i am not switching to the Creality slicer to test it. Even though they use Orca (or Prusia, whichever) as the base now.
22:21 lol, that made me laugh, copy and paste.
Can't wait for mine to arrive 😅
Any ideas how hard it would be to make the vent fans actually vent outside? With my K1 Max someone designed a clip on adapter that allowed you to attach a 4" duct and fan to vent the printer outside... I print a lot of ABS and ASA. I figured someone might be able to make something magnetic perhaps? I thought I saw a magnetic poop catcher. I invision it being like a Y shape... connect over each exhaust fan and then at the other end of the Y they converge to a 4" duct to connect for propper venting.
I print in an area I occupy and the ABS fumes are not something I want in the air.
Thank you
I walked on some gravel but avoided the grass. Does that count?
Unsure, in cases like this it’s best to try a coin flip
Do you think Bambu labs new printer is going to be better? And the same size?
Adding motors just to tension the belts sounds a bit overengineered.
By the way, how does it measure belt tension? Is there a load cell somewhere under the tension motor?
It actually looks like an appliance.
Really hope we don't start hearing about issues in the next months.
That heatsink looks like a step back
I bet people will still buy a Bambu Lab X1 or P1 over this because of Creality's reputation
And they still might be right, this printer has a lot to prove before you can call it a success. (The print quality from other sources has been quite bad)
Well yea. I want a printer that works continuously with zero issues, not one that'll break a couple months after I get it.
I mean, Creality is still holding onto 47% of the total entry-level 3d printer market, compared to Bambu's 26%- as of Q2 2024. People continue to buy Creality, despite the "Tinkering required" reputation. Kinda surprising really.
Elegoo: ~10.5%
Anycubic: ~10.5%
Everyone else (Including Prusa): ~6%
@@nickdarrow4873 bambu doesn't really have ender 3 priced printers (at least without the current massive sale lol)
Where are you getting those stats? Could be interesting to browse through them