Are Carbon Fiber 3D Printer Beds the Future?
ฝัง
- เผยแพร่เมื่อ 3 มิ.ย. 2024
- They're certainly an interesting concept. But how promising are they really?
Check out PCBWay! I'm extremely pleased with their SLM services.
pcbway.com/g/2EZ6k2
Music ►► elements.envato.com/
Business Contact ►► tommy@tommyhoughton.com - วิทยาศาสตร์และเทคโนโลยี
Its a combination of rigidity and weight. I rebuild a Makergear m2 -- which already has a very rigid steel frame, but lacks the Y axis rigidity. I replaced the steel bed carriage with a 6mm aluminum carriage -- i made it bigger for the same weight, and add 2 linear rails instead of one and also added a 2nd Z axis lead screw attachment to the bed stage. I also replaced the steel bed with a 24v heater with a mic6 aluminum bed with a mains powered heater. End result is net zero weight change, but significant increase in rigidity. Accelerations on Y went from 1500 to 8000 and max speeds are whatever you want them to be -- factor of motors a voltage, I can hit 500-600 mm/sec with no issues. Light weight beds which you fling around can lead to added micro vibrations. I found out the hard way -- I tried a skeletonized X gantry on my voron, and input shaper was really hard to tune, I went to a unibody light weight beam, which is heavier than the skeletonized X gantry by a lot, but is also WAY more rigid, and my accelerations and speed doubled -- with very clean input shaper curves. I'm going to be rebuilding my v2.4 to a doomed cube trident to improve rigidity -- shaving 200-300g is not as important as you think, make the frame more rigid, replace the v-wheels on Y with 2 linear rails, and note the difference.
Thanks for sharing! That's quite interesting about the micro-vibrations, I've been working on some other lightweight parts, so that could be an important variable when testing. I'll also do research on improving the rigidity for future projects. Thanks!
Came for the carbon fiber, stayed for the ender of theseus debate.
I look forward to the next stage of it! Cheers
Good video! Glad to see someone taking the adventurous path rather than just slapping more motors on the printer (although I'd still like to see an AWD Y axis video :D). Looking at your graphs at the end, I think you still have a lot of room for improvement in that regard. It looks like you have three distinct peaks, so other things on your machine are rattling and throwing off your results.
I'd absolutely love to do AWD on the Y axis, I've been a bit intimidated to do that though. It could make an excellent video, I'll try and reach out and see if LDO would be keen to help! There's definitely room for improvement, the machine was wobbling quite a bit, and I'm pretty sure a few screws are loose somewhere. Once I get the sheets machined I'll refine everything. Thanks for your support as always!
I suspect a CF bed will have some temperature limitations depending on the resin used to make it.
I gutted an old Ender 3 Pro and redesigned it for use as an enclosed DD IDEX for certain projects where I don't care about speed. The thing I found out in that process was how bad those stamped steel mounts and pom wheels are with resonance. It ended up with full linear rails, went with dual side mounted Z motors and rails. The real game changer was I machined a pair or HD 1/8" (~3mm ) thick steel Z to X gantry mounts, the rigitity of those made a huge difference on X axis resonance. The Y axis will always be a limiting factor on any bed slinger. I just try to avoid as much Y motion as possible when laying out parts to print in slicer.
Honestly CoreXY machines are getting cheaper and chasing speed on a bed slinger can end up costing you more. We're reaching a point where some entry level CoreXY are hitting the $500 price point (IE: Sovol SV09). It's always fun to tinker though, that's how you learn. Cool video !
That’s true. I’m unable to tell what resin was used, but I’m sure it’s more than suitable for PLA. I will do testing in a follow up video once I have the parts machined.
You’re machine sounds awesome, I’ve heard several comments that increasing the rigidity is highly important, I’ll definitely look into something similar.
While bed slingers are fun to push, you’re right about the costs. I’d love to get Sovol’s new machine, but after shipping costs/or local markup it wouldn’t be feasible for me just yet. I’ll be converting my Ender 5 to CoreXY in July, so I’m hoping it holds up to my speed printing ambitions. I really appreciate your comment! Cheers
Я переделывал свой Эндер 3 Про, уже несколько раз.Последнее изменение было связано с изменением привода ремня к кровати, и я сразу получил более стабильные результаты. Далее я стал думать как уменьшить вес кровати и также пришел к выводу, что неплохо использовать карбон. Стал искать кто такое делал и нашел ваше видео. У меня будет немного другая схема, но принцип такой же. Я не то что бы гонюсь за большими скоростями, скорее я хочу поднять качество печати, а тяжелая нагревательная кровать одна из причин появления дефектов. Кстати я тоже заказал силиконовую грелку, но только потому что из продажи практически исчезли грелки на основе каптоновой пленки (я не нашел подходящую по параметрам). Нагревательный элемент на основе каптоновой пленки намного легче силикона. Удачи в вашем проекте, я тоже попробую завершить свой, сейчас жду когда мне вырежут нужные детали.
I've heard good things about that mod, I must try it myself. As nice as high print speeds are, higher-quality parts are always the goal. I'll keep a look out for these Kapton heating pads, they seem very promising. I've noticed the silicone heating pad is much heavier than initially expected. I wish you the best with your project! I will try this again later on in the year but with some better changes. Cheers
A couple of comments. A silicone heater also adds quite some weight - so I wouldn't be surprised if the weight of the CF plate together with a silicone heater is more or less the same as the aluminium PCB heater. The thought you had, replacing the steel bed frame with a CF frame really has merit as there should be a substantial amount of weight saved.
Good point, I didn't weigh the silicone heater, I just assumed it would be much lighter with the CF plate compared to stock. Once I get the bed frame and the bed properly machined, I'll do this project again with the feedback I've received. Thanks for your comment! Cheers
I've been waiting 5 years for a video like this haha, carbon fiber beds are a game changer with nylon
I’ll try do some experiments with it in the near future, I’ve never worked with nylon but I hear it’ll be a challenge. Thanks!
@TommyHoughton it's very challenging, it hates moving air. Liquid cooled extruder and heat lamps for temp stability are your friends. But the carbon fiber beds act like garolite beds on crack, plus they are very flat and smooth. I print with a .25" thick carbon fiber bed.
@@elvo_racing Thanks for the info!
Well explained man. Now I want a completey unreasonable carbon fibre bed too 😂
Thanks Scott!
Raaaaa pagani mentioned raaaaa
I really like the animations/editing style little things you used in this video very nice
also bed move fast, really fast
Ok so I've got 2 questions. Good video BTW. Firstly, what thickness is the carbon fibre sheet that you are using. And secondly, what command do I use in klipper to test acceleration. I know how to do input shaping, but I didn't know there was a command to test until failure
Thanks for watching! The sheet is 2.5mm. I used this macro to test the max speed, however, I got the process a little confused with a video with MirageC, where he made a Gcode that increases the acells over time. I'll share that video as well as it was very helpful. ellis3dp.com/Print-Tuning-Guide/articles/determining_max_speeds_accels.html
th-cam.com/video/AqlP5ar3LkE/w-d-xo.html
All the best! Cheers
What kind of drivers are you running on this? Also, are they running in stealthchop or spreadcycle? TMC drivers skip steps a lot easier in stealthchop at higher speeds.
It's a stock Creality 4.2.2 with the A4988 drivers. Unfortunately I can't control the current (to my knowledge.) I wish I never fried my SKR Mini E3 V3, but some things can't be changed. Would definitely love to put some HV drivers on this machine, perhaps BigTreeTech could help with that. An AWD Y axis upgrade was requested, combining that with some TMC 5160's would be insane. Thanks for the info too! It may prove useful in the future, I've never really looked into the details of controlling TMC drivers. Cheers
@@TommyHoughton yeah you cant do much with that board, but you can in fact increase the VREF on each driver. Maybe get a volt meter and double check what they are set to so you can maybe get more out of them. Other than that, you should get a new board so you can use some TMC drivers and configure them in spreadcycle.
That’s a good idea. I have no immediate plans for a new mainboard, but I’ll put in on the list for future upgrades since I definitely want to play around with different drivers. If I can plan a good enough video, I’ll try to fit in as much as possible; Perhaps “Building a 48V AWD Ender 3” has potential, I’ll look into it. Thanks for your suggestions!
@@TommyHoughton I appreciate your positive attitude and how you listen to suggestions. You have a bright future on YT, bud. 🙂
That means a ton to me! The suggestions from everyone (you and a couple of others especially) these last couple of years have been invaluable, I'll try my best to read and respond to all of them for as long as I can. I look forward to hearing from you again in the future!
can you do a hardware klipper setup overview, I also have a pi zero 2 w and want to add a touchscreen. Nice setup btw!
Thanks for watching! I could certainly do a deeper dive in the future, but it isn’t something I’m able to focus on currently. I’m more than happy to try and help you if I can; I used KIAUH to install Klipper screen, it didn’t work straight away but a reinstall made it work properly. I’m using a 5” screen from BTT using an hdmi to mini hdmi adapter (or micro not sure which one) everything is connected using the required adapters (usb C to A to Micro using a splitter for more ports.) Teaching tech did a video on a cheap esp screen I think it was, it’s a better value option but I haven’t delved into it just yet or checked compatibility but should all be fine. Cheers
@@TommyHoughton Thanks! I have changed the Y-axis carriage on my setup with a smaller cheap aluminum one from AliExpress and saved +- 100 gr, but I've seen they also sell cf versions for around 40 euro. Keep up the good work!
great video! thanks for always putting out good content.
I appreciate your support! I'll keep working on improving the little mistakes here and there... Thanks!
Ditch levelling wheels and put plain nylon locking nut, ive got cheap 1/4 socket set for it from lidl... even handier now
Good idea! I might even have some in storage. Thanks!
@@TommyHoughton Yeah i removed levelling springs as well. I use a bare nut just below the bed just keeping that screw in, then two nuts in really flat lightweight 3D printed wheels one above the bed carrier and one below. They jam against the carrier, no nylok nut needed. But my printer is not an Ender3 but a DIY i3-variant.
The printer like that is susceptible to accidents though, like that one time when my DIY probe failed to deploy... the nozzle can hit pretty hard!
I replaced the Y carriage on my Ender 3v2 Neo with a carbon fiber one, it only weighs 95g
That’s awesome! How’s it working for you?
Is it possible to heat the bed after installing it?
Yep, you can still heat a CF bed
It sure is! I purchased a nice silicone heater to do that, but unfortunately the cf sheet was too small. When I get the sheet machined to the right size, I’ll install the heater and test the performance. Cheers
No it has to be a cf made with high temp epoxy and even then at ABS bed temps would still most likely make it warp. Do some research.
@@Tinkerer5888 Good point, I’ll look into getting a better sheet. I was assuming they were all relatively similar. Thanks for mentioning it.
I loved your video
It would awsome if you could replace the y-axis gantry with carbon fiber tooo.
Thanks, I’m glad you enjoyed it! I’ll definitely be replacing the gantry in the near future. Cheers
whats the name of this hotend?
nova ?
It’s the Biqu H2 V2S Lite. I believe the nova is higher end. Cheers
In this day and age - Why still bother chasing Speed with a Bedslinger when going CoreXY is the more sensible choice? 🤨
Besides taking up way more space due to the moving Bed the most obvious problem with a Bedslinger is that you _really_ cannot calibrate the Y-Axis unless everything you print remains in the realm of tiny as once you've reached a certain size / mass on your build plate all that calibration you did using an empty bed basically flies out the window - Unlike on a CoreXY where all the calibration is being done on Axes whose mass never changes as a print progresses 😏
I've been meaning to purchase one, but local imports have a decent markup and shipping costs from overseas are too high to justify currently. That's a good point with the calibration. I really can't wait to build/order a corexy. I'll be doing a CoreXY conversion soon on my Ender 5, so I'm hoping it's as good as expected. Thanks for your comment! Cheers
I might be silly and it wouldn't work, but how about using magnesium instead ? It has a lower density than carbon fiber
It's definitely possible, but there's too many variables. I'm not sure about the safety, useability and other things. I'll look into it in the future. Thanks for mentioning it!
Is it just me or did you read your input shred and graphs wrong? Your first recommended excel was 3800. Your second recommended input shaper was 1900. I can almost guarantee the print results will look horrible if you go past 4000 accel with the cf
I likely did. The recommended shapers and values are noticeably different, and I'm not sure why that is. I held off on printing as I couldn't install the heater and the sheet wasn't up to spec, I plan on redoing this video but better with properly machined parts, hopefully in a couple of months. I'll also be using the feedback I've received, as It's quite a new area for me. I appreciate your input! Cheers
you should make an ender 3 ng
I’ll hopefully be doing a similar conversion later this year!
Yeah I kinda want to do it but I need to make a H2 v2s lite adapter for it to work. But keep up the work