ฝัง
- เผยแพร่เมื่อ 9 พ.ค. 2021
- Benchy Printed in 4:57 as per #SpeedBoatRace Challenge slicing specification.
ClearPath Servo Motors are being used making this and exhibit entry.
Printer Details:
#HextrudORT ► page miragec79.github.io/HextrudORT/
#NovaHotEnd ► 3dpassion.com/nova
#ClearPath Servo Motors► www.teknic.com/products/clear...
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
Gonna wait on a Summoning Salt video on "The history of 3DBenchy speedrunning"
"Years passed with no significant changes to the WR, until one day a TH-camr by the name of MirageC and his HevOrt printer utilizing Liquid Nitrogen cooling moved the goal to sub 4 Minutes!"
I want someone, who understands math, to figure out how many benchy's have been printed since it was introduced. And if they were stacked end to end, how many times around the world they would stretch. Same for the XYZ, and Chep Cubes. I bet millions have been printed...
@@juliankandlhofer7553 liquid cooling might be far more sensible than you might think. I am testing to print inside a liquid which is heated just below the glass transistion of the plastic resin. Instant hardening, why try to jet a gas onto the print with ever increasing pressure instead of just using a themally more suitable medium?
@@flipschwipp6572 Nifty! I expect the hard part is picking a liquid that doesn't mess up the layer bonding. care to share what you use?
@@Patrick_Cooper I haven't done the math but I think that maybe more benchys were printed in the previous 30 days, than all of history before that combined
posting pictures in a novice forum and asking "is my benchy looking good?" then posting the insane speed settings and video
I have done that =)
Wow!
I also love the fact that the three featured machines are all quite different!
Wow I never thought I would have enough patience to watch a whole benchy print. Sub 5min is insane!
Go buy a garbage printer. That will make you watch every single layer go down to make sure it doesnt fail.
Print in black for faster heat dissipation ;-)
Yes, new material inbound! ;)
😍😍😍seems that time lapses aren't needed anymore !
It's funny how a non time lapse print looks time lapsed due to the speed of it
This is GREEEEEAAATTT! Watched Josh, watched Vez, you BREAK 5min!! Excellent!!!! The four of you are my heros. Please keep going.
Amazing! My printer would shake itself into pieces if I tried this. Cheers 👍😎🇦🇺
I'm just utterly blown away by the fact that the ribs on those belts hold up to those stresses caused by such accelerations and speeds.
yeah but they are moving so little mass
12mm wide belts with light weigh gantry makes it possible. Vez is running 6mm belts on his VzBot.
Omg, this is insane💥
your Z-axis moves faster than my x and Y- axis.....
Absolute legend!
Will you have STL files (or step files) for the parts cooler? I'm interested in maybe hooking up an aquarium / aquaponics air pump (with like, 100GPH flow) to one those nozzles. Would be light weight tubing and should be great flow!
Benchy history!
i applaud you for using servos, it‘s the next logical step in fast printing, no sense in holding back. you need to try the MG94 ABS with it‘s high melt flow.
I was smiling before I even pressed play and by the end was laughing out loud 😍 How about leading the community back towards quality with speed. Also, despite the music, I think I could appreciate your machine is very quiet. Still you continue to put this within everyone’s reach. Love it!
Some Finnish music there, unless I'm totally inaccurate.
Honestly this speedboatrace is probably going to help discover best practices that improve speed without affecting quality, even if the leading edge produces garbage at insane print times
it's hard to measure quality...
@@JoneKone it’s Beethoven, so German
If a garbage quality print goes from 20 minutes to 5 minutes, it's easy to imagine the same techniques will at least halve the time of a quality print. With klipper I could already improve the quality of my prints at 200% speed in my very old prusa i3.
Honestly despite being close to the same speeds your benchy looks much better
nice one... wait for it.... =)
Game upped 4:30...
Bravo!!!
Love it, good job.
2 things, wouldn't the hevac help even more with the cooling here (if it's not already on), or would it also affect the unshielded nova hot block too much at those speeds and temps?
Plus can you print in black? It's super difficult to see the details on those white benchies..
But again good job, and congrats on the the record. Waiting impatiently on that 4 min benchy now!!
🖖 👍
Looks really good for this Crasy speed.
Amazing printer and engineering! Really looking forward to see you progress and maybe one day build a HevORT myself! :) can you tell me the specs/model of the servomotors?? I've got some ideas to use a perticular servomotor that has integrated drivers and thinking it could be really neat for this machine.
I wonder what would happen if you edit the gcode to do small "vase mode" prints for the columns (the corners of the cabin). Essentially a simple application of non planar printing. The stringing would be resolved by the head not making those jumps, and probably the overall travel path would be shorter (faster) too. You may have to modify your nozzle slightly to have enough clearance, though I'm not even sure that's the case.
Still turns out better than my 3hr benchy
PLS, could you do a one walled cylinder in Vasemode as fast as possible? :D
I think this would look soooo sick!
You people doing this shit are absolute psychopaths and I love every minute of it. I hate my 3D printer, if I had a HevORT or a Voron, I think I'd do absolutely nothing but 3D print. Welp, time to get one of these!
Sub 5! The race continues...
Damn sub 5 min! I'm always amazed at what is the next hard limit you step into. I'm starting to think the plastic properties themselves are becoming your enemy here. Really pushing the material to the limit.
Mirage did you switch to clips for the glass bed instead of the thermal pads you were using?
Really nice, I love this competition and what you have made with the hevort.:)
It's inspiring.
Some newb/dumb ideas for the flow (as unfortunately I'm not that kind of engineer). Did you consider the following:
- It seems that ABS+ of eSun has a lower fusion temp, and probably glass transition temp It's a kind of blend. And I suppose it's still flowy enough.
- Considering that you don't have retraction, have you considered to lower the fan on the heatsink in order to "preheat" the filament ?
- Or in a more engineered way, have a pre-heat of the filament just below the glass transition temp. It may even be an heated enclosure for the filament which could even be filled with dessicant. Less water in filament, less temperature inertia. I know, probably insignifiant. :)
- Having heat block insulation (silicone and/or ceramic coton).
- Redirecting the flow of cooling air more precisely to avoid as possible cooling the nozzle. Not sure if this is a good idea although, I wonder if the cooling of the nozzle doesn't participate to the cooling of the filament in some sort.
- What about having an additional closer ring (ring to negate the possibly deforming air force by opposing them) really close to the tip of the noze to distribute the cooling air?
- And a crazy one, what about to cool the cooling air? :D Some copper tube with in an icy bucket between the compressor and the hot end. In a more durable way, it could be a (second hand) mini fridge with the bucket in it, no manipulation required. And to more efficiently "load" the cold in the air, maybe a some kind of water nebulization in the air after the compressor and before the icy bucket. If the filament is fully enclosed (roll, ptfe tube, direct drive, hot end), I don't see a real issue to increase a little the room humidity. Depending of the room size, it should be marginal. And to make sure to not spit possible condensed water, have micro hole at the dispenser to renebulize before reaching the hot filament. With all the air flow and temperature is this region, I would guess no water will remain on the model/printing bed.
- If we consider the slicing, maybe having different cooling depending of the location, as some region may need more and other less (eg stack on the boat roof).
Haha, I wonder if someone will read until here. ;)
Don't hesitate to comment. That's just a kind of brainstorming to possibly trigger effective solution(s).
PS: For those who complain about quality, that's not the point here. The idea is to push the state-of-the-art (on speed here) with new creative ideas. In some sort, it's creating problems / finding previously hidden problems to solve them. And from those solutions, applications could be found in a more realistic way where quality matter. :D
Actually, forget what I said about ABS+ of eSun. The melt flow rate is not as good as I though.
- ABS+ - eSun - 12g/10min - 220C°/10Kg
But here some interresting stuffs:
- Technical data sheet --> TitanX™ Black - FormFutura - 41g/10min - 260°C/5Kg
cdn-3d.niceshops.com/upload/file/tds-titanx.pdf
- HF ABS - Extrudr - 44g/10 min - ?°C/?kg
Technical data sheet --> cdn-3d.niceshops.com/upload/file/MaterialDataSheet_ABS.pdf
Those data seem to match your goals as I guess your benchy is around 15g :D
As I understood highspeed head moving requires larger diameter pulleys. Yes?
With every Video i get more impresst. This speed in so cool to see.
My only question is were are the limits?
4D benchy shaped nozzle.
@@RubixB0y I've been thinking similarly. next stop is slicer improvements for semi-non-planar printing and hybridized shape shifting extrusions
Good gawd that's fast! If cooling is the issue have you considered injecting a tiny amount of liquid in the air stream? Like mist coolant on milling machines. What would not affect the adhesion and evaporate fast?
I still have headroom with cooling so far, Melting and extruder speed is the current issue.
Dude I love watching your imPROgresivORT (aggressive and impressive pro progress on the hevort... of course) and has inspired a number of little things on my recent projects. I'm curious what you would think about running x and y (or horizonatal anyway) ballscrews now that you have the servos? and maybe not for you because you're insane and love making two second prints that blind us all but what about conceptually?
Ball screws on XY could be something I explore in the future. One of the main challenge I see is to avoid having the X servo being carried on the Y axis. This would cause in my opinion unbalanced extra weight on the X cross member. There could be ways to avoid that, but the ones I know would involve belts... back to square one...
@@MirageC I've been envisioning a great (imo) design where each x and y are perfectly semetrical and each is powered by a carbon fiber tube coming off the servo shaft going the length of the opposite axis (x along y and y along x) on each side with 45° beveled gears running ballscrews pupendicularly off the carbon fiber shafts on each side of the build plate. No belts. No traveling motors. To help picture think of sort of a ultimaker setup or the dash. I like the idea of Cartesian motion system and no traveling motors or long belts... I really think this would crush it. Only key would be having zero slop 45° gears probably in a little baby gearbox haha
Oh and on the far end of the ballscrew there's another 45° turning another carbon fiber slave tube or else you'd have to have dual servos on each and that would be way Overkill and I think carbon fiber is stiff enough that any play would be imperceptible even in the most intricate prints.
Let the madness continue) Good work. Those extruder and cooling pipes (i mean hoses to them) are made of resin? how do they feel themselves in printing?
Is there rules on nozel size? Couldn't it be much faster with a 1.2 nozzle?
MirageC, what motors are you running for x/y motion?
Is there something like an air chiller to cool down the air before it goes to the part? I imagine sub-zero air would help immensely with part cooling
Unfortunately my air compressor does not have a great capacity, therefore once tank pressure reaches trigger point, compressor will run for the entire benchy print. Tank air might become a bit warmer then due to freshly being compressed and no time for heat dissipation. But as I run a 50 feet hose to the printer, it is not so bad when reaching the air nozzles. And then the final gas expansion at system exit gets the gas temperature even lower. Low enough to create bonding issue if I aim the air jets straight at the nozzle. This is why I aim the cooling jets at the printed part. Cooling molten material too much will create a hard shell at the surface, interfering with layer bonding and the shaping of that material. Sorry for the long response... :)
What about using Co2 instead of compressed air
How are you cooking the parts so quick, I have a machine with a super volcano on it and I can’t for the life of me get enough cooling
YES!!! HAHAHAHA HEVORT HEVORT HEVORT !!! :D
These speed runs reminds me of my pc bench marking obsession and spending silly money to gain a few extra 3d mark points. After all this tuning what can you take from this to print fast, but good quality prints. How long would it take to print a quality benchy from lets say a prusa mk3 ?
About 30min is the fastest ive gotten a benchy on my prusa and still maintain quality
@@tankkiler308 I didn't word it very well, I mean on these super fast printers, how fast can they print, but have the print quality of a normal perfect prusa mk3 benchy.
Honestly the the thing is I really think in a year or two a 5 minute benchy that looks good will be doable. Each step brings more challenges the main two now are better kinematics to prevent vibrations and better part cooling
tes tu du Québec? j'étais pas certain avec l'accent, mais j'ai vue ta minuterie sur ton cellulaire qui étais en francais! Beau travail!
And here I thought the whole point was to be running for our lives for the fastest benchy!!!
Hello
Interesting, how much will cost cheap version with print size 300x300x300?
Perfect results (speed concidered)
0:56 but what did you use, servos or what?
At that print speed, I’m surprised you didn’t burn your hand on molten plastic when you picked up the benchy.
Pretty soon, you’re going to have to have to be spraying liquid nitrogen at the model to cool the filament.
Instead of having the hot end in contact with the model, you should have a gap and have it spraying a jet of molten plastic at the model, with a tiny high velocity jet of liquid nitrogen aimed right at the point on the model where you want the plastic to solidify. Could get some amazing overhangs that way.
On a more serious note, I’d love a camera on the nozzle so I could see what that insane volume of plastic extrusion looks like.
Part of the reason (if not the biggest reason) he is able to get such insane times is that his cooling is... absurd. I'd be interested to see something at these speeds in infrared. I mean, yes, there is this: th-cam.com/video/65FVQ1jArME/w-d-xo.html but the current speeds appear to be much faster.
I wouldn't be surprised if the plastic is almost ambient a couple seconds after it is extruded.
Wow, I'm starting to fear for the belts. :D
That is why they are 12mm belts ;)
@@MirageC Next step: Bikechains. :D
I understand the speed but how do you prevent the filament from curling due to the heat?
The temperature of the filament, on application, is the same as everyone else, isn’t it? I don’t think there’s much flexibility there.
if you all could just stop at 4:20, that would be great
No stop at 4:20.69
Didn't know it was real time (rather than a time-lapse) until the 1 minute mark
clicked almost as fast as this prints.
are those cooling nozzle pipes some factory made pipes or did you do them by self?
made by myself ;)
@@MirageC pipe and m3 hex cap screw? You just glue that pipe to some kind hollow mount stand?
@@TH-wr1dvPipes are 1/8 OD with M2.5 cap screws to block the ends. th-cam.com/video/h-ukClHtai4/w-d-xo.html
Why not make a custom silicone sock for the nova hotend?
Work in progress ;)
@@MirageC I think josh cheated a bit as well. He used some 42v step sticks on the a and b motors
@@imst4722 Well I think that as long as we are being creative to get the chrono down, we are doing the right thing :) Servos, 48V, custom made hot-ends, all of that require significant amount of work.
Nice sweet holly mamma!
_sigh_ hold my beer
What are the rules of this competition? How do you prevent someone from submitting a blob of plastic that roughly has the resembles a benchy?
Your printer does seem quieter than the other two. With less rattling. But I don't know.
Really like your printer :) If it just wasn't sooo expensive to build. :( And it's getting more and more expensive. ^^
I'd really like to know how a benchy looks like in "quality" print speed :) I've actually never seen the print quality of your printer when printing at slow speeds :)
I'd be interested in this too.
I guess it could go a lot faster with g-code optimization there are a lot unnecessary movements
How much did your hervort cost overall and what’s an avarage cost for one ?
you can figure 2000USD for a 315-315-340 sized HevORT
@@MirageC that's with a nova hot end and servos?
i can get behind benchy print speed running
Is there a kit I can build to make this 3d printer
No Kit, but DOM are available here: www.hevort.com
Thanks for the mellow musical background. I don't like it when people play really loud music, especially Heavy Metal type, during their videos. I always have to scramble for the volume control.
This ain't over until someone hits 4.20
Looks more like the Titanic now a days, than like a decent boat.
So, is speed printing a new sport now?
A sport I dont know, but we discovered many techniques ans tricks that will be allow our daily printing to be faster for the same level of quality.
Did you switch back to PLA, or does the Nova just ooze that much with ABS? Also, have you ever thought about enclosing the whole thing and just absolutely blasting it with 100 C air to cool ABS straight down to Tg? Forget about the HevACS or the berd-air, I'm suggesting you just nuke the whole thing straight down to Tg with a violent, constant current of air!
This is ABS. The oozing comes from the small extruder high gear ratio not able to deal with retraction and pressure advance at those speeds.
Air is not dense enough to cool properly if warmed up to 100c. This print is using 60PSI shop air distributed via berd-air pipes. No side duct cooling used.
@@MirageC Gotcha, makes sense.
It's only a matter of time before somebody really starts cheating by keeping the head stationary and moving a very small bed just large enough to fit a benchy at insane accels.
Then, they both move to increase the relative speed
Isn't one of the main limitations the plastic itself? Doubt it would help much.
@@c-djinni Ah but the theory is once you have unlimited mass to spend on the extruder, hotend, cooling system, etc you can in theory work towards solving those problems easier. Assuming you can get the Benchy to stay on the bed at those accelerations....
@@scianiac The problem is that there's a point where the plastic can't distribute the heatengergy within itself fast enough to keep up with the amount of heat that's forced into it. It will burn on the outside while staying cold/"raw" in the middle. Like a badly cooked steak.
Edit: Holy Super Hit, I might have had the best idea of my life so far: Add metal particles to the filament! Copper or silver (maybe even carbon fiber or synthetic diamond?) should help heat transfer a lot and thereby heat the filament more evenly! You'd need a hardened nozzle (which will probably make it a little harder as they tend to heat up a bit worse than brass) to print with metal in the filament, but I'd say it's worth a shot.
Holy shit....
Purely disgusting !
Love it
I can't wait for a legit
Very cool none the less. I think the filament material is holding people back
How could you possibly improve on this?
I have a list... :)
@@MirageC sub 4:00 when? lol
gotta go fast, but at what cost?
At the cost of identifying new technique that will be replicable in our daily printing :)
Gotta start looking into different filaments... That one can't Keep up😂
Swap to a duel feed system not for two colors, but for sufficient flow
Bravo
Vez3D Got 4:35!!
I have a horror movie in my PC called "Benchy in 3:53"... But I am having nightmares watching it! lol!
Man.. do you ever sleep
lol, insane - got below 5min 🍻🍻
No me canso de decirlo: QUEDA COMO EL ORTO.
Don't stop until you get to 4:20.
Why 4:20, been seing this time marker referenced more than once.
hum.
#whysoslow 🤷🏻♂️
Flow is maxed out :P Need shielding from strong compressor and better plastic!
@@MirageC is that hot end flow that you're lacking? Would something like a super volcano help or be too much mass for how fast you're moving?
@@MirageC can you start at a even higher temp? thinking like 380, and then as the hotend cant keep up itll drop in time, but still stay in the temp window you need? mind i dont know what the thermal requirements are to melt at these super high speeds, or the thermal mass that would be needed for that idea to work. but as a motorsports fan too, im loving watching you guys push the limits and go fast, and innovate :)
When 1 sec benchy
That is called injection molding :)
Качество такое себе)
Next time print it in a reasonable color so we can see some surface
can anyone tell me why servo is technically cheating? (except the exorbitant price unless that's the reason)
@Pilot16H so it is basically cost unfairness, so instead of engineering prowess it's just throwing money to the problem?
I still consider my system as an open loop since there is no feedback done to the control board. Servo motors require encoders in order for them to work by nature, all that is done internally to the motor without any return of communication to the MCU.
So in the event of a position error due to mechanical flexion/failure/slippage the system will not know.
The same principle would apply with the current solution offered on 3D printer stepper motors such as the S42Bs.
The only advantage of the servos here is their torque and speed.
@@MirageC how is this cheating?
Speed boating
It resembles a benchy yes. It was fast yes. But...
Lmao it's always HevORT vs race Delta's vs Voron
Yep. Got my voron 2.4 almost ready to go... just some wiring left. Sadly had to go with a nonstandard board due to shortages. And considering I know nearly nothing about electronics, it is rather interesting. (On the bright side, I already have some changes I want to make to the mechanics...)
Dont forget the VzBot =)
Why is it cheating to use servos?
it was in the rules.. dont know why, maybe cost?
Next step to print not shit quality benchy same fast )
i cheated because i played it at 2 speed
and weakly print without defects? all your high-speed videos suck in print quality. someone will print real parts with this quality?no! Show the perfect benchy and say "it's at this speed. if you print higher, the quality will not be perfect"
I think you do realize that we are exploring new territories here. These speed runs had us discovering phenomenons, concepts and machine interactions that we did not see before. We now can develop solutions. You need to see the problem and understand it to fix it. This is pure development, not an aesthetics contest. Of course our goal is aesthetics in the end, but things get dirty when you push the envelop. A 10 minutes Benchy today looks almost perfect compared to what it was 6 months ago.
@@MirageC thanks you for exploring 3d print destination.i have ghost 5 and want to upgrade it on corexy.this videos motivate me to strive for new speed.but i want to see what speed limit have perfect 3d print today.in my mind, it will be very different of this. Example,i have wow effect when see 100mm/s with 10000 acel perfect print-can you amaze me?)
Don't watch speed videos of you look for quality.
At some Point there will be tradeoffs between the 2. Especially with the currently fastest Benchy.