Building a Large Format 3D Printer - Part 2: Motion
HTML-код
- Опубликовано: 21 июл 2024
- Part 2: Constructing a colossal 4’x4’x4’ print volume for my pellet extruder.
Files, Parts, and Info: www.drdflo.com/LF3DP.html
Large Format Printer Playlist: • Large Format 3D Printe...
D-Flo's Amazon Store: www.amazon.com/shop/dr.d-flo
📷 Follow Dr. D-Flo on the Gram: / dr.dflo
Welcome to Part 2 of Dr. D-Flo’s Large Format 3D Printer! This printer has an industrial build envelope of 4’x4’x4’ (1220mm x 1220mm x 1220mm). A typical filament extruder is not capable of pumping enough plastic to cover this massive build volume. Therefore, Massive Dimension’s MDPH2 pellet extruder, which is capable of depositing 1 kg of molten plastic an hour, will be used to generate the layers for some massive objects. However, the MDPH2 weighs nearly 18lbs (~8.5kg), which requires a serious linear motion platform to move around.
Dr. D-Flo constructed 7 of OpenBuilds' C-beam linear actuators to create a rigid and repeatable cartesian system that uses 4 actuators for the Z-axis, 2 for the Y-axis, and 1 for the X-axis. With a couple modifications shown in this video, these actuators that are upgraded for high temperature operation. This video covers the assembly of these linear actuators and their installation into the frame built in part I of this series. By the end of this video, it will be possible to position the MDPH2 at any position in the build envelope.
Subscribe for Part 3!
Table of Contents:
00:00 - Introduction
01:25 - Critical Speed
05:23 - High-Temperature Actuator Design
09:15 - Carriage Assembly
11:23 - Actuator Assembly
13:18 - Quad Gantry Leveling (QGL) Design
17:09 - Machining X-Y Joint
19:25 - Wiring
21:15 - Plasma Cutting Panels
24:20 - Bed Heater Tease
24:20 - Bed Heater Tease
25:55 - Linear Motion Testing
#Massive #DIY #3DPrinter 
Наука
Very nice that you dedicated a full section to the Critical speed. I'm one of those people who talked about it in the previous posts, and I like to think that I could help you at least in a tiny bit. It's a small return for all the advice and information that I've learned from you. Great content
Yes your comments have been very helpful!
@@DrDFlo About the critical speed... Why did you not get leadscrews with a higher pitch? I think igus has some leadscrews with a really high pitch, you could move much faster with the same rpm. The torque of the NEMA 23 motors should be plenty.
@@qutlicopatlixhotutti6552 I was also thinking of the Igus lead screws as well as their high temperature options. I have also seen on some thermwood and I think multicam? cncs that use an hdpe (or at lest it looked like it) stabalizer block that supported the lead screw and rotated out of the way when the gantry passed , that could be a low cost solution
The build is looking amazing. I can't wait to see the world's largest benchy
Oh hell yah
And to see him row around a pond in it😁 (it isn't a very good boat design...)
I hope you have calculated the thermal expansion and a whopping in the aluminium
the difference of 20 degrees Celsius to 100 degrees Celsius What an expansion approximately of 2.2mm at a distance of 1200 mm
8 mm spindle screw has only expansion approximately at the distance of 1.25MM
but by inserting spring washers at the end you may be able to compensate for that
I wish you good luck with the project
I don't think so. Otherwise the Prusa XL wouldn't have that modular bed plate design... And it's only 350mm in size...
@@alejandroperez5368 the prusa did this for heat zones and expansion on the heat zones the build plate will not expand or contract that much due to the heat bed not directly mechanically attached to the heat bed
Those gantry brackets you milled are absolutely gorgeous.
Wow, awesome build project! Looking forward to the next episode!
Very nice build. I'm so envious about the CNC and laser cutter. Also the voron, which I will build myself in the near future.
Keep up the great content
That sound of electric motors being fed square waves!!!
I feel like you could build one beastly battlebot with your shop as well :) Your production and passion you put into your videos is awesome, always look forward to new content !
I'd be curious to see him try this. It's a completely different set of challenges, but I think he could handle it.
Wow, very excited to see the next episode!
I would use brass for the drive nut, I service and build the parts for bucket wheel excavators from RWE in Germany and we use brass for these applications. Love the build, keep up the good work.
jup ptfe probably won't live long under load
Came here to say this
@@TheLordinio I've seen it described as milling bars of soap. It has outstanding chemical and thermal properties, but it's mechanical properties are horrible.
Honestly? If I was to go to the trouble to make new nuts, I'd make them out of oilite.
open builds is great, ordered my aluminum extrusions through them... Great Prices, My kinda website! -Props that thing is a beast!
I'm so hyped for this beast to print a gigantic Benchy!
Incredible video production here. 👏 I'm in the middle of my mechE degree, so I just eat this kind of content up.
Absolutely loved when you touched on the critical speed of the leadscrews and how mass moment of inertia comes into play.
Consider me 'Subscribed'
cant wait to see the first print run
I cannot wait for the first print!
Add an accordian connecting the edge of each rail to the carriage on that rail, and have a smooth bearing attached at segments of the leadscrew along it to act as a moving track to prevent sag
It amazes me that your channel still has sub 100K viewers. Lots of good analysis and information.
I've been so excited for this! Keep up the great work. 😁
Moment of inertia is i^4 and is the measurement of mass from the center of mass. The higher the number the worse the whipping of the lead screw. You really DO an amazing job. I Enjoy your research.
Can't wait for the last video and see how it all comes out.
The Dr. is back at the office!
GE12C for the win! Glad to see those being part of the build! Stop stepping on the bed hahaha
Thanks for your suggestion!!
Looks great - can't wait to see the next part. This'd be great for printing furniture too, for example. "You wouldn't download a car"... why not?
Incredible build
*Ivan Miranda enters chat*
Just found your channel. Love it!
no worries about differences in thermal expansion? looks like the aluminium will expand 1-2mm more than the steel leadscrew when getting to chamber temp. bed could be tricky too at those temps/sizes
Great video, you inspire me with every video!
ohhh i remember the noise my ender 5 made right out of the factory, my wife would love this thing inside the garage, 2 floors below our bedroom^^
Finally is right! Very excited to see this.
Interesting project. Subscribed. One note tho, rater then rotating the lead screw that gets wobbly at longer length. Rotate the nut instead. The draw back is that you have to have the motor mounted on the carrier thus it gets additional weights.
Also I'd look at using support blocks positioned by concertina mechanisms to reduce the unsupported length on your leadscrews.
This looks like it will be super interesting. How cool 😎 👍
Thanks, that has a lot of useful experimental information!
cool machine you made))
in order to make the parts more detailed, I would add a small spindle and cut each layer in layers with a cutter, a cutter with a small diameter) the rigidity of the machine allows you to mill plastic)
Thanks for the video, i want to see you inside a benchy asap haha
This thing is pretty incredible. Excited to see the first layers. What kind of electrical service is run to this beast?
The wiring montage. 👌🏼
Consider taking those lead screws back out and milling a flat at each end for the set screws. I'm picturing the set screw walking right up the thread over time.
Nice job but just be wary that PTFE can cold creep under load. Try to minimize the PTFE thickness and constrain the PTFE by metal where possible i.e. a thin PTFE part in a metal block. Keep up the great work!
Also, free handing the tap is not good when you want it to be square to the part. This alone could induce some of the lead screw whip see in the video, and cause binding and excess friction.
Amazing you unlike most understand much like heated chamber requirements for exotic thermoplastics 140c plus !!! CHEERS finally! the only thing i think you may have some back lash resonance super supporting you here going for high temp resins this chamber will need fire pillows to keep your hydro in the box and maybe a bellow set in my opinion great job selecting Teflon and PC v wheels as back up for now, but you may need the bellows to isolate the heats from thermal expansions on the x,y keep it up all eyes
Part 1 made me really want to try a pellet extruder on a smaller scale. I'm now currently assembling a ratrig vcore 3 500 with a mahor pellet extruder
That sounds like a BEAST! Any online presence to follow that build?
@@DJ-wl5yi not currently. I'll be posting about it as I get it working. Right now I'm still assembling the frame of the kit which has plenty of coverage
hope this works. a 500mm ratrig is on my business road map and a pellet extruder is a great idea on it
I have dreamt of lead screws for my Y axis since 2017. I went looking for M8 trapezoidal replacement and found a supplier at Aliexpress that sells them in much higher pitch than you are using. My idea was to use a closed loop stepper by MKS, and experiment with various pitch values in the range of 8 to 16mm. Your RPM per given feedrate and travel will be dramatically lower thus avoiding the whipping as you exhibited. There will be the matter of anti back lash feature to add in that pre-loads the lead nut..Being that the lead nut is brass as are currently offered, there would be the need for regular lube. I understand that the mass of the components and assemblies are orders of magnitude larger than a little bed slinger. I have no idea if you are dealing with any "stickshun" But maybe it is worth a look. The closed loop option is a small circuit board that plugs in where your drivers go and bolt on to the body of the NEMA23, at least you can rest assured that every pulse will be accounted for. All you have to beyond that is hack the firmware. to suit your choice of pitch. I would be happy to share sources.
Impressive work!
I have been working as a tech in plastics extrusion and injection molding for 20 years. This extruser is cool but they missed something. It should have a suck back function. Wich pulls the screw back just a bit causing whats called suck back. It removes and reverses the pressure so you dont get that ooze when the extruder needs to move. Would be awesome with that function. Then it could move without oozing
Nice work with your CNC milling machine! Just curious if you ever upgraded the spindle bearings for higher speed?
Great work, I am hoping to do this build with my younger brother who is going off to college for engineering this august. Any way we can get the BOM and the CAD for all the additional brackets you manufactured? Thanks Dr. D Flo!
I would definately use slider couplers with this kinda of inertia even though it has thrust washers
Hi. This is an awesome build. I am creating a very similar build myself. I was curious if you had a system figure out yet for your pellet distribution? My build is 1500x3000 and I think I'm moving towards a fan blown process but am curious what your solution is.
Good build. I love the tips.
You can easily CNC rout Aluminum Composite Panels using a V-bit and the rout-and-return method to make custom electrical enclosures. I model them using the Fusion 360 sheet metal workspace to get the flat pattern.
I was looking into this the other day after realizing how expensive enclosures and backplates are currently ($70 for a 21x21" painted backplate?? lol). However, I think for larger boxes, you would be very hard-pressed to save money (and time) building an enclosure or backplate yourself unless you just happened to already have the materials -- most of the cost of the enclosures is just raw material (most good quality electrical enclosures are 14ga powder-coated steel, aluminum and SS are more expensive). I would still be interested to see David (or somebody else) do it, though. lol
Beautiful machine! I would like to ask you to make some runs without a heated chamber or plate to explore the challenges of cold printing and if it is possible at all at such scale. I have this wedge in my brain that tells me 99 ways of printing in a cold environment will not work but there is 1 correct way that can make everything fit together and you can print without the need of heating or cooling.
What a creator, bravo!
Excelente trabajo 👍👍 se ve que sabes mucho de este tema ,felicitaciones
This is an intense project, it will be interesting to see what you can print with it. I shudder to think of how much it will cost to feed it with pellets, even with lower cost materials like ABS. And printing with PEEK.. there's just no room for error, a failed print could cost hundreds along with the time and effort to dry the pellets out
This thing is crazy. However thermal Expansion is crazy and the wobbly X Leadscrew should be adressed.
Thanks for sharing, you taught me a few things! One problem I've noticed though, your Z leveling system with the multiple Z motors is only going to work without breaking things if there is some flex somewhere where there shouldn't be flex. Take a look at the Hevort printer, the short linear rails just under where the bed mounts (I can't find the words to explain it, but you'll see what I mean).
I think he's addressed that with the spherical bearings, they allow a degree of freedom to prevent stress. I think the hevort project uses that decoupled approach on their Z access to work around the runout in cheaper ballbearing lead screws
I like everything about this vid.
how did you match up the threads of the dual teflon blocks for the smaller carriages? With two in series, I would have thought they would require some specific consideration.
Amazing explanations and good way to show how to make things to ensure good results💪💪💪💪
Excellent work!
If leadscrew flex becomes an issue, maybe a mini carriage/bearing combo could work? Very very cool build
This is just awsome !
PEEK parts of this size will be sooooo expensive! Great job and thanks for this amazing build series.
Hence the pellet extruder. $40/kg vs $600/kg filament
@@khaderach19 right the pellets are much cheaper, but did he really build a huge ass machine just to make 1kg prints at a time? Heck that could be made on a foot cubed machine… no his parts will still be extremely expensive for the size they will need to be whatsoever it may be.
Could you add smaller trolleys along the lead screw to limit the span control the whip and increase your critical speed? I had a few ideas that might work to reduce the span. The simplest thing would be to drag them behind the extruder but this would mean managing drag lines which isn't exactly optimal. You could also maybe use a belt system geared at 3:1 and 3:2 off the head to move the sliders but this still sounds pretty complicated to set up. Personally I think the slickest way would be to use the same gearing as you would in the belt system, but instead using a set or planetary gears inside of the support sliders which would then both support as well as be propelled by the main lead screw.
Love the project BTW
Impressive, and well done on the progress thus far. What limit switches are you using? I've been having some trouble with switches, as I've been destroying them on my large scale project, as they aren't triggering reliably, and my gantry just wants to run right through them like Marshawn Lynch.
Have a Idea what if instead of a lead screw drive what about acme thread worm drive that way the long screw doesn’t have to rotate and create the screw wobble.
My custom-built FDM printer is "only" abut 650mm cubed with a 310mm cubed build volume, and it was a pretty crazy project to build. I shudder at the thought of something 9 times larger...
Interesting experiment if you're wanting to increase the diameter of threaded rod or screw size but you don't want the mass to where you can print faster but you don't get the SAG in the rod and the whipping that it causes then have threads turned on a lave in thick wall pipe it should keep its read rigidly but yet you don't have the mass
Where did you get that Metric T8 tap, because I have been looking for one that will fit the openbuilds lead screws
6:10 hold on. I think its important to check the bearing carrier. It might be in plastic. I would tear one down to see.
The data sheet the came with the metal shielded bearing said good for up 250C operation
How come you went with lead screws vs. those larger kind of ball screws? I figured ball screws would be better for the longer travels like this but are they worse for some reason? (other than cost)
I ran into the problem of lead screw sagging many times in the past, and while I always managed to solve it with more or less succes, I had this one idea I never had a chance to test. The idea is pretty simple, just one or more blocks (Teflon in your case) on either side of your carriage which can slide on the same rail, have a hole just a bit bigger than the screw diameter and are tethered to the carriage (and to the next block if there are more) by a string or a thin strip plastic or metal with length equal to a portion of the screw which should be stabilized (approximately 1/4 of total length is two blocks are used, 1/8 if four etc...). This way the carriage always drags the blocks behind and pushes the others. Maybe using elastic string would work even better is fixed at both ends to the frame. Some length is lost, but the blocks can be pretty thin if guided well. Just an idea in case you find out you need greater speeds.
If you wanted to use linear rails, I suppose the wipers could be laser cut from silicone rubber sheet easily enough. Some silicones are rated for 230 oC or even more.
Why not some tight fitting telescoping Teflon
Regarding for lead screws proportional to the fourth power, is that related to the second moment? Like because there''s that extra distance from the centre point?
I may be misremembering it all, I do need to go over all that again
that thumbnail brings up repressed memories
wish i didnt get this reference
What kind of drivers are you using for the steppers? The DM542s are an inexpensive driver for the high torque NEMA32s that are almost silent.
Why didn't you use multiple 'regular size' bed heaters and wired them in paralel or even enable only those you print on currently (with plastics wicht doesn't require a heated chamber)
The leadscrew wobble of the X-Axis is crazy... you realy should swhicht that to metal core timing belts. There are heat resistant versions available.
wouldn't it be possible to make some kind of scizor frame supports for the lead screws like enclosing the frame in a scizor lift the supports would then slide along the lead screw and divide it's length in half or quoters, thirds or wkatever
i know supports on both sides could reduce some travel but it would be like travel reduced by milimeters vs screw resonance frequency multiplyed by the number of supports
in theory 2 times shorter screw should be able to move 2 times faster and accuracy should increase in the low end
what do you think, worth the effort or too much complexiti?
Sounds like you have a product you can sell now with those teflon nut blocks. I'm sure you're not the only one that needed those before
man. a larger leadscrew (or ballscrew) not sure which your using. would have taken so much of that wobble out. love the build thus far though. looks great
Impressive !
I'm not sure of you did it after and just didn't film it, but from the way you applied the idle nut and motor to the lead screws doesn't tension them at all. I would think you would apply the motor side first, then run your idle nut in to pull the desired tension before locking the set screw down.
It would be convenient to have a link to part 1 somewhere in the description :)
This is great :)
Why did you choose flexible motor joints? Don't they introduce play in the axis?
What motherboard and drivers are you using? I am new to building machines and also want to use nema 23 motors, but it seems difficult to find good information.
PTFE is usable at high temperatures but it loves to creep at any temperature. I would think twice and think again about in exactly what capacity I was relying on it. No large static loads, where large is TBD. Teflon coatings on metal structures may be better than teflon blocks IDK. Watch out for thermal expansion problems. Making a bunch of miscellaneous components work with high endurance at high temperatures gives you lots of interesting failure modes. Watch out for anything that needs to be lubricated at all, I would say. Be aware of what you're getting into. Best of luck.
donde conseguiste la cama calentadora
I like it, nice and quiet 😂
Amazing build. Really ambitious that You're planning to use high temperature materials too. Fyi, you can also increase the critical speed of a drive screw by pre-tensioning it.
Another way to go is spin the nut and leave the screw stationary which eliminates the problem. But then the motor is in the heated space and attached to the drive where it's inertia is an issue.
We really need a product that no-one makes. A tube ball screw. Then the stiffness is high, but the mass is low.
@@court2379 spinning the nut is a tricky trade off. While it is a small and light part, it's usually 3-4 times larger diameter so often ends up with a higher rotational inertia than the spinning screw. For the same reason a tubular screw doesn't help with rotational inertia. That said, you can buy tubular screws. They allow you to water cool the screw for high precision applications.
@@martylawson1638 I doubt the nut inertia is going to be higher than the shaft when you are getting into these sizes due to needing to use larger shafts to deal with loads and whipping, and due to the long lengths, but ultimately it would have to be calculated.
I have never seen the hollow ball screws, but if they are for cooling purposes the design is not likely what you would need here. Cooling likely uses a small bore. For lowered inertia, the wall would need to be as thin as possible to keep the inertia close to that of the solid shaft. Mostly likely a hardened alloy, maybe 17-7. Then it would be stiffened by it's cross section and material.
Since the loads are small, I am not sure screw drive is really the best option here. Great for accuracy and staying in place when not being driven, but this printer is huge. It needs to run as fast as possible to get a part done in days instead of weeks.
There is no perfect solution though and just depends on what things the design focus is on as to what is best (apart for just bad engineering, and not saying that is the case here).
I watched your SLA build and it was very involved and detailed. You are producing great content. Do you have your own business? Where did you learn your skills?
Is there a reason you can’t use your mill to make the Teflon wheels?
I dont know, that thing would be printing for a month to make anything the size of its volume right?
7:50 How did you know which Trapezoidal Tap to choose?
Nice project looks similar to what I’m building (600mm x 1200mm) but i went with chinese linear rails and ball screws
Fantastic project. Are you going to be manufacturing with this printer? Looking forward to more videos!
Would it be possible to add bushings/bearings midway between the carriage and the ends of the leadscrew on springs to help reduce the whipping of the leadscrew? As the carriage nears the end of the leadscrew, the springs would compress to keep the bushing centered.
Took a while to figure out your vision here. These bearings will need to move at 1/2 the speed of the carriage then? May work.
Well, something needs to be done.
Perhaps a system where the screw was 51% enclosed, resting in a cup.
I know automotive drive shafts are hollow. Do hollow lead screws even exist?
The solution would be a stationary and tensioned leadscrew an a rotating nut. But keep in mind simple leadscrews aren't nowhere to be meant for high speeds. I'd at least use 1204 cheap ballscrews for that matter. Our DIY large printer will feature 15mm belts for X and Y.
I think there is a reason you couldn't find ptfe drive nuts. It isn't very strong. It will be interesting to see if it holds up.
I would consider belt driven axis and the use of closed loop servos and you'll get a much quitter machine and with much less vibrations from the lead screws. Not going to belts consider ball screws...
Waiting for the linear rail upgrade lmao, I’d also be a little more concerned when standing on the build plate
I can't wait to see the Part3 ! When will it come out?