ฝัง
- เผยแพร่เมื่อ 7 พ.ค. 2021
- Exhibit entry for the #speedboatrace challenge ► • #SpeedBoatRace Challen...
🍺 SUPPORT ME
If you like my work and would like to support my efforts in making more open source 3D printing solutions, please consider a donation or becoming a Patron.
Patreon ► / miragec
PayPal ► www.paypal.com/donate?hosted_...
Equipment:
#HextrudORT ► page miragec79.github.io/HextrudORT/
#Klipper Firmware ► www.klipper3d.org/Overview.html
#NovaHotEnd ► 3dpassion.com/nova
#ClearPath Servo Motors ► www.teknic.com/model-info/CPM...
Material:
Magigoo bed adhesive ►amzn.to/3sUcGLa
You want to know more about the HevORT advanced DIY 3D printer?
WEBSITE ► www.hevort.com
You want to learn and contribute to the HevORT project?
FORUMS ► forums.hevort.com/index.php
FACEBOOK ► / hevort
DISCORD ► / discord
🍺 SUPPORT ME ► www.paypal.com/donate?hosted_...
If you like my work and would like to support my efforts in making more open source 3D printing solutions, please consider a small donation.
Thank you and thanks for watching,
MirageC - วิทยาศาสตร์และเทคโนโลยี
In a reprap sprit, I feel like competitions like this would do well to be separated into "weight classes" based on cost of off-the-shelf components.
Agree.
i would not see this a normal competition where somebody wins.
All this superspeed benchy look terrible - but they show what the limits are where you can improve.
Everybody is a winner here.
@@JoergGebhard That's obviously true, but the idea here is gamifying it a bit, making it a kind of "speedrunning category". Much like race car tech trickles down, this stuff will.
Quality classes would be another very reasonable category division.
@@JoergGebhard Exactly!
After all that's the spirit of the #speedboatrace - not to claim 'Oh I've got the fastest printer and it's the best!' - it's to show everyone what is possible and ignite the spark in us to improve our printers even further! :)
I feel like the ender 3's have their own league lol
Buddy!!! This thing rocks!!! Its always fun to see your stuff and I love how we are pushing each other to go further :) i have found a very good friend that shares more than just 3d stuff with me. Keep up the good work and cant wait to go for a cold beer soon
Brother from another mother! Looking forward to that one or twenty beers!! :)
I have a pair of those motors sitting aside waiting for something like this. Good stuff! It's not cheating, it's the future of 3D printing.
Now companies just have to come up with a new high speed printing filament to keep up with hevort printers :)
yes! that is definately one roadblock that we are starting to feel being very close, if not already present.
@@MirageC Yeah, serious research in material science has to be done.
I thought injection molding could do hundreds of parts an hour with the same material. Perhaps it is just the delivery method.
@@TheboysEpicshit the process and cost associated in the creation of molds for plastic injection is the draw back. You need very high production levels to justify that process. FDM is more appropriate for prototyping and small volumes.
You would almost certainly need a thinner diameter filament for thermal conductivity to not be the limiter. Hard to know whether that approach or a longer melt zone, like in a volcano hotend, is superior when factoring in issues from filament that becomes too flexible. Whether a larger heated area or thinner filament causes more issues there.
Here's a video that i watch at 1x speed rather than 2x and it STILL looks like it's sped up by several times :D
so try at 0.25 X speed ..
You will see it's just crazy fast !!
So cool and inspiring to see you pushing the limits of desktop FDM printing!
At first when I heard 1500 mm a second I thought "that's not fast!" Then I realized, that's 90,000 mm / minute 😮🚀🚀
It never reaches that speed tho
@@alejandroperez5368 Yes it does. Maybe not on a benchy because of how small it is, but it would on larger things. At 10k acceleration it would reach that speed after 112 mm.
@@Chris-hn4lp no, the max speed is 300mm/s. The limiting factor is the material melting rate, not the stepper motors.
@@alejandroperez5368 in this case this is a custom 3d printer. It’s got everything upgraded even the speed at which the filament melts. While he may not have that at the maximum yet the speeds this commenter said are possible with this printer. Your printer may have a max speed of 300mm/s but this one does not.
@@xiar5546 he even talks about it in the video? 2:53, 4:25
He didn't mention max speeds, but he said he printed the benchie with 300mm/s print speed and 1000mm/s travel speed
im glad to finally find someone to have put these crazy motors on a 3d printer
incredible! so satisfying to watch this thing move, and just amazing what these actually really tiny servo motors can do. absolutely beautiful. i have 700W servos sitting here for a cnc i'm planning, and i'm just realising how scared i should be.
must be a big CNC to warrant 700W servos (guessing its 750w). Make sure to limit the torque as otherwise those servos could bend/dent it if you crash into the vise or workpiece. My BF20 conversion is running 180W servos and they got plenty of torque for that tiny toy.
@@exol511 you're right, they are indeed 750W. thanks for the advice. they're probably overkill and i'll treat them with the utmost respect. i'm aspiring towards a granite cnc similar to Piotr fox wysocki (youtube), but may save the 750W servos for a cycloidal drive B and C axis arranged like a mini hermele.
or use one as a low rpm spindle initially, since a good ATC spindle starts at 3-5x what i've spent so far and it'll be a while before i can afford one.
i was a bit power greedy and they were very attractively priced.
if they're really too much i've also considered trying to make a large scara concrete printer where they might come in handy.
@@DMonZ1988 if you can get the faces for the linear components machined by someone I would just go UHPC with inserts like most manufacturers do these days.
To deal with excess vibrations as you get to higher and higher speeds, I would suggest:
- Setting the printer down on a rigid block
- Using vibration-dampening materials for all of your soft fittings.
- Any springs, if present, should be converted into J-Dampers (look them up, F1 used to use them).
- Use of input shaping via accelerometers would be helpful too.
Dang beat me to it with the mass dampers
Looking forward to the video about servo motors, I've thought about using servo motors on a printer before so I'm happy to see it works. Also using compressed air as cooling is absolutely ingenious!!
This is great work. I'm impressed by your continual improvement and excellent results. I'm going to follow you with the clear path servos once you drop your setup video. I use the larger version of those servos on my Tormach 1100MX and they are very high quality.
Dude, if pc overclockers can use liquid nitrogen to cool processors, then you can use compressed air and clearpath servo's.
They're really the future. Built-in path correction, overshoot correction, the works. And they have the power to use some pretty heavy setups, which adds to the rigidity / lower resonance of the whole system. They're worth their weight in gold. And the supplied software too! I love your setup so much. This is amazing!
Very nice!! Glad you finally found a setting that seems to work well! :D great job
I love the cooling, very creative and logical. Great job
Great to see you got the clearpath servos working Mirage!
Awesome printing! Looking forward to your vids on using them servos!
Hehe like the Project Farm music - the Benchy's awesome as well
Absolutely incredible! I didn't know such servos exist. I'd love to see a comparison between O-Drive brushless motors vs the industrial servos.
ODrive is good for the position-sensing/orientation-sensing. High precision and accuracy, basically.
Great to see this in action!! Agreed that overall speed isn't the clear answer, the insane acceleration is what will get us faster prints! I also don't understand why servos aren't "allowed" in this silly competition, it's a clear evolution of printers that are designed for speed, hence the race itself. All mechanical setups should be allowed.
This is outright amazing! I guess servos are the future at least for this type of 3D printing.
Nice work! very impressive!
Amazing work!
Congrats on getting it to work! Interested to see the follow-on video. Particularly how you powered them and what board you ended up using to get step/dir signals.
You are the best bro. Pioneer !!!!
Thanks for sharing
Indy car racing of 3d printers.... Thank you for your contribution and most importantly Thank you for sharing it for others to build on!
Finally someone used servo in a 3d printer, love it, they are so much overkill hahahah
Good point about safety. I'm so used to putting my hands on the machine while printing. I do it to save a few minutes, why restart a print if I can give a nudge here or there to save the print. I've gotten used to this because my motors would barely give my fingers a good squeeze before stalling. If I were to move to a much more powerful and faster machine there would be less incentive to try to "save" prints but by now my learned response is to touch the moving parts and that would be hard to unlearn. I think I would move to a fully enclosed machine and wire a door switch, I don't trust myself.
Hahaha, yeah, thats good thinking. Probably good idea.
servos just simply hit different.
looking forward to the servo followup
keep your fingers and have a great day!
This is plain crazy speed dam
Very impressive.
So I decide to build my own 3d Printer and started trolling TH-cam to see what others are doing. I find your channel. You have pretty much built what I had in mind and some. Here I was thinking I'd be the worlds first to use shop air lol. I was not thinking servo motors till now. I was thinking I'd CNC all my fixtures and custom machine the print head etc. It's the electronic considerations I'm still getting my head around. Ultimately I want to print PC reliably with an excellent appearance and finish and reasonably fast. I'm happy to sacrifice speed for finish.
Your videos are fantastic and saving me a massive amount of trial and error although I fully expect to make plenty of errors. I fully appreciate how much work you have put into this, I just built my CNC Mill. That was a very impressive Bency by the way especially as it was done 5 times faster than best time I currently have.
Now you are on a speedlevel of industrial millingmaschines. Awesome.
Gods work!
Looking forward to the next video. Im definitely considering these servos for our new print farm
On a print farm i am not sure this would be worth it. Make sure to check the price tag associates to them ;)
MirageC I spoke to their Techsupport a few months back and have been considering for some time. 1k/ printer for motion and control isn’t bad when you’re talking quality and precise movements at high speeds. We could double our output with this kind of motion!
@@MirageC did you ever get around to making the video you mentioned here about how you implemented these clearpaths? Im tackling my prototype finally in the new year! We have a VF-2 Haas VMC, if youd like any aluminum servo mounts, hit me up!
Amazing!
impressive !
I was about to add air tubes instead of fans great idea 💡 i was gonna use a small air pump still designing my hotend mount
New meaning to rapid prototyping!
this is so frikin cool
This is the future of manufacture using FDM printing, getting this neat of a print with those times is insane, it just needs a bit of tweaking, more cooling and a slight deacceleration to achieve beautiful prints in jaw dropping times, using this technique with hot technical materials like PEEK, could produce strong parts in mindblogging times.
i was planing on using a small pump to make a 2 nozzle cooler. hit it from either side slightly off axis, make it kind of spiral.
C'est grâce au téléphone sur la vidéo que je m'aperçois que ce génie est français! Tu pourrais m'être très utile! Très bonne vidéo!
I just subscribed and the first thing I see is that lmao
Nice work that’s FAST A F
I’m new in 3D printing bought a ender 3 last week
@MirageC great video! When can we expect the video on the ClearPath motors? This should be really interesting and I am really looking forward to seeing it once it comes out!
My god I love this printer so hard haha that is amazing. Coming from home made cnc mills I have always been surprised not more people were utilizing servos. But I guess now that all these corexy variations are becoming the norm and be slingers are gettings left behind, it will be more common.
This is insane
Holy... I have a small 3D printing business and this would really help even at 1/4 speed
Wow. U do great job :)
I'd love to see an updated cooling video like you did with the thermal camera showing the effectiveness of the new cooling setup. Are you also still using the side coolers?
Not on this run. Shop air was more than enough. :)
Better quality than my 3:30 hours benchy.
I have one of these clear path motors on my bench and they are pretty awesome. I only wish they made lower power motors that would be better suited to an application like this as all the additional driving hardware adds up in space/size
Got some on my lathe and I love them too. I asked if they were planning to make a Nema 17 format, but the answer that it is not cost efficient or easy to size down further the eectronics.
Salut Cousin, Magnifique !...
i know it is not sped up but it just looks like a timelapse... crazy what these motors can do
I can very much agree that those clearpath servos are realy strong. i have some ~400w nominal power nema34 clearpath servos on my cnc mill and it makes it shake a lot if im not carefull, which is kind of impressive as the thing weighs around 1000kg :o
Thats not cheating in any whay.
Great job, very nicely done 🤘😎👍
Ta vidéo est super!(COOL)
Awesome
Extrution head: i think tht is a bit too fast and im getting dizzy XD
Dang that's real time that thing is fast
Very interested in knowing how you got the motors to work. I have a box of ClearPath servo motors and was wondering about using them for one of my printers?
It's always exciting to see one of your videos in my notification list! I was wondering what kind of belts you use? I've seen some references somewhere to metal belts. Have you heard of that? Or am I just thinking of belts reinforced with stainless steel braid? You did a video recently on lead/ball screws. You noted how prohibitively expensive ballscrews are, but do you think you'll one day use them to match the performance that the rest of your printer is capable of, especially with these new servo motors? Keep up the great work.
I am using 12mm wide Gates 2GT belts (from china). They are using fiberglass inside as reinforcement. They are good, I find they will ring a lot less than my 6mm ones. But they are heavier and stiffer, something to consider when using stepper motors. As for ball screw, I am looking at the near future where I would try fast travel ball screws with those motors :) that will be fun!.
@@MirageC I can't wait to see ball screws with these servo motors! Soon you'll need a high-speed camera to trace your movements. Actually, now that I think about it, that would be a good way to measure the actual speed of the print head! Time to call the Slow Mo Guys. :))
I personally don't GAF about the contest, this is freaking amazing.... lol
Great work? How are positions determined? Optical? Thanks.
Well torque is not the only think to consider, inertia ratio is way more important for high dynamic systems, sometimes you end using 10 times more torque than required to keep the system stable
Awesome machine btw
oh man that's powerful! I've heard that odrive is also strong, but I'd love a nema17 servo package for any printer! I mean, I could be running like that at 80% and I'd be happy, just drop in
HevORT HD support Nema23 and Nema17. Just saying ... ;)
Amazed what a machine wish my ittke printer could do that
Awesome!
I wonder about trying to use linear rail steppers instead of belt drives. This is a whole other level. Another method is galvo servos that I have heard about but have no idea about their abilities.
Yes, a lot of work will be required to make these affordable and usable into what we do. But the idea of them sounds pretty cool.
That cooling tube system is awesome, do you plan on publishing info for it?
Always cool content with you,
The only reason we don't want servos in the SpeedBoatRace is the prohibitive cost, if you could do it cheap enough we would consider allowing it
Although this rule may sound weird to some it really forces me to work my stepper system to the max. And now joshmurrah is taking this even deeper. I would not be surprised his 48V may get close to servo times. So i see this rule similar as a hp limitation in F1. It forces to maximize every aspects.
Yeah, I like MirageC’s reasoning a bit better than the cost reason. Unless Hiwin, Moons and Slice engineering are outlawed from the race as well 😉 Quick math tells me crosses gantry with high cost parts (well engineered parts 😉) gets a bit $$$$
Are odrive bldc ok for the race? :p
@@MirageC Have you looked into using moteus or odrive instead? pretty experimental.. but cheap (er)
@@jotham123 Vez3D will do an update on ODrives shortly. On another hand, I am happily running my ClearPaths ;)
You should put a camera on the extruder assembly, I would love to see the footage
I love your videos, between you and zex yall inspired me to make a bigger, metter, faster printer than my prusa and to go for more.
what wattage heater cartridge is in that nova? Doing some research before planning out a new printer.
My Nova hotend has a 100W 24V cartridge.
This speed is beyond cool but at the end of the day what matters is quality of the prints. You need to tune your machine to fastest set up with fiber mixed filement to minimize layer lines and just start selling it, we are ready. 🤘🤣
Would the additional power of the servos be able to work with heavier gantry/gantry setups? If that is the case, that would mean more space for cooling and and stiffening without sacrificing speed and acceleration, right?
If your doing shop air for cooling, you should checkout vortex tubes to give even more cooling... 😎
Haven't you tried a dual 5015 blower fan duct for cooling like the hero me gen 3 ducts,i feel like running them at 100% power will provide adecuate cooling!
Go for a remote direct extruder. So you're able to loose weight and gain extrusion speed and torque.
Beside: really good video and really admirable machine and abilities! Thanks a lot for sharing!
How does a remote direct extruder work? Doesn't that make it more akin to a boden tube system?
@@theninjascientist689 you find examples when googling it. It has the advantages of both: Bowden tube and direct drive. And the weight is not even close to a direct drive motor.
What happened to the setup and config video for these, was looking forward to it!
How does printing at this speed affect layer adhesion? I'd like to see some stress tests.
racing 3d printer with a kompressor :D
holly hell, that's insane as always! Last video you said you were going with servos to push it even further, I was looking forward, and damn no disappointments at all.
Like, you're doing 1G acceleration on that benchy. The parts literally couldn't free fall faster. No wonder the THICCCC belts there! You still probably need quite some backlash compensation right? Even the pure movement probably twists the mounts some micrometers here and there enough to need it no matter how stiff (you said you tested 15000mm/s² right? 1.5G ain't no joke for the precision needed on a printer).
I ain't no polymer specialist engineer for sure, but my gut tells me that at the speeds you're extruding things don't really work the same anymore as in a normal print, like extrusion pressure and so on (actually, that much I'm 100% sure. Speed at which a deformation is applied to a polymer does change significantly it's mechanical properties). You're definitely threading into some very advanced polymer engineering here.
...meanwhile I'm trying to solve a resonance problem on my Y motor on the moving bed... No matter what current settings I use, specifically 3500 to 5500 mm/min. won't move no matter what, it can move faster tho. The motor locks up loaded or not, and it's Y only not X with same motor and driver. Running TMC2208s in all steppers, SpreadCycle on X and Y and NEMA 23s on those two (it refuses to even move those NEMA 23s in any current in StealthChop, only the dual Z NEMA 17s move on StealthChop).
Oh well after the last assignments are done, time to get the damn thing to work properly.
I was running 100 000 mm/s2 on that run. Control interface only displays 5 digits lol. I have pushed it to 150 000 already. Frame and mounts are not even sweating. All is safe and sound. :)
@@MirageC Damn, time to swap to m/s² then! Damn, 15G+ on that extruder and hotend... No matter how light, that's quite some force already for sure.
You should do a video running this in vase mode on a simple cylinder
I would like to suggest a vortex cooler for improved cooling rate.
This way you could drive higher delta t
how do you get the clearpath motors to work with klipper?
dude. holy shit.
well time to design a servo driven bowden setup
There are always people who are more elite. I mean, not for this guy. But for you and me.
That is crazy! Hahah, does the air PSI cause any deformation of the print? I wonder if some of your quality loss is due to that?
I have known, that you break the contest record ;-)
I am in the early stages of planning a large printer, and the X/Y is giving me lots of trouble. I can't keep the weight down, while keeping adequate damping on the print head with adequate rigidity.
I might have to revisit servos to solve the mass issue by brute force.
In some rare cases, nothing beats brute force. Also have you looked at Carbonfiber tubing?
Something else that might help is designing the print head to partially warp around the X axis, so the total center of mass of the head is as close as you can get it to the structural member
@@mr.thomas6128 The EVA 2 modular print head does exactly what you are talking about.
main.eva-3d.page/
It seems to be very well designed, and very easy to modify.
I am actually considering building the RatRig, it looks like it took rigidity past what the HeVort has, and has none of the overly complex design solutions that the Voron team have come up with to solve problems that should not be problems.
I should be able to extend the Y-axis to 1000mm. I think buying the kit is worth the cost vs my time designing it from scratch and machining all the parts for it. This video was actually pretty helpful seeing how well the 3030 extrusions performed.
Maybe the increased cooling for overhangs is something printer manufactures should look at. Fans that spin faster or even activated compressed air additional cooling. Could be something that is manually or automatically implemented in the slicer. Nobody wants a noisy printer but also nobody wants droopy overhangs ruining the print.
I could watch that thing go for an hour
I've only seen a few of your videos, very cool stuff, but I'm curious, how far can you bridge with that monster?
As far as any other printer if tuned properly. Same technologie is being used here.
When will you publish the video about the servo motors? I'm very curious!
I am working on it, but other life priorities are stepping in the way... Its coming though.
did you ever post the setup video? this is really awesome!!!
Not yet... Need to get back at it :)
What's the advantage of a servo motor?
Also, are you still using these same servo motors in your latest videos? And can you make a video of this benchy printing out at half the speed and acceleration? Btw, does HevORT support Klipper input shaping? I want to see the maximum quality achievable in the minimum amount of time with the latest and greatest setup you've got.
Sounds like you need larger diameter pulleys 😁. Maybe 24 tooth for crazy speed? Are you running klipper with input shaper?
Current pulleys will give me 2000mm/s at max motor RPM. Enough for now ;)
Have you considered linear servo motors?? Granted... price is... well... another level. Very nice job!!!! Keep it up!!
I am barely scratching the surface of what these clearpaths can do. So those Linear actuator will wait a bit ;)