3D Printed Planetary Gearbox! Resin vs FDM Printing?
Вставка
- Опубліковано 13 січ 2022
- Is resin printing useful for gears??? Results from my experiments show that the parts printed from resin are much more brittle and prone to failures from repeated loading. In this experiment, the PLA I was using was Inland "Tough PLA" and the resin was "ABS-Like" clear resin. This gearbox will be used in some future projects of mine so stay tuned for more!
Free CAD Files: drive.google.com/drive/folder...
3DPrintedLife Video: • My 3D Printed Harmonic...
Thingiverse: www.thingiverse.com/thing:520...
Fusion 360 Gear Plug-In: apps.autodesk.com/FUSION/en/D...
Hardware List (Affiliate Links):
NEMA17 Stepper Motor: amzn.to/3zPRnia
A4988 Stepper Driver: amzn.to/3zRPhy2
30x42x7 Bearing: amzn.to/3felPsR
3x7x3 Bearing: amzn.to/3zROAoq
5x16x5 Bearing: amzn.to/3GlYFN6
M3x10 Cap Screws: amzn.to/3zPQYwa
M3x14 Cap Screws: amzn.to/3GmKSWD
M3 Hex Nut: amzn.to/3rfN3Vm
Keep in mind that there is only 1 tooth in contact at any moment. The bigger the tooth, the stronger it is .The minimum for teeth interference is 13 teeth for small pignons. Depending on the needed gearbox size, you want to choose the biggest possible gear modulus that allow for at least 13 teeth. Bigger modulus = stronger teeth and bigger contact area. Also fit tolerances and surface finish induced by manufacturing process have less impact on bigger teeth. Your pignons look like they have about 20 teeth for 10mm in D, then I assume 0.5 modulus. SO you could choose a bigger modulus for the teeth. In this case 0.75modulus would be the best = 15 teeth would be WAY stronger of course outer ring will also have a reduced number of teeth, but you can keep you diameter and ratio.
I totally agree... for such an application I would use some tougher resin... I would use Liqcreate Tough-X or Composite-X
Either way the point of this test was to check the strength of the two types of 3d printing. If both types were the same design it would be an accurate test. In my personal experience fdm printing is stronger but resin can be better. I personally only use resin with a wax resin so I can easily cast parts in aluminum.
@@Baitrix1 poo
If it is a true envolute, then multiple gears are in contact.
should use helical gear to maximize contact
I'm just impressed how clean the PLA prints look........
Awesome test, Michael!
Would love to see a similar comparison from you as well.
Try this thing out too
Hello CNC kitchen!
Could the WD40 be having an effect on the resin parts? It can make certain plastics brittle. Perhaps try a silicone grease.
Another option could be to use a dry lubricant like graphite, or something more expensive/exotic like molybdenum disulfide. The additional benefit of graphite is that it can fill in surface imperfections while coating every mating surface with a thin layer that should introduce as little added resistance as possible even with tighter tolerances in the version 2 of the resin model.
Probably not it but may be worth testing just in case.
it looked like a wd-40 silicone grease. not the regular wd-40
naa, it's just cheap resin, it works for what it's made for Modeling, you can't expect plain old inexpensive resin to be super resin, for that, one needs a formlabs or something like that.
@@patprop74 true that cheap resin is not going to give anywhere near the same quality as purpose formulated resin from any trustworthy brands. But for the sake of prototyping and testing design variations, it's not a bad idea to use the more inexpensive resins and use whatever trick you can to get most bang for your buck. Because one thing that I have noticed is that it is a rare occasion when printed mechanical pieces work on the first attempt, even if the prints come off the build plate perfectly.
Great video and thanks for the shoutout! I've been dealing with some of the weaknesses of resin prints lately in my own projects, it's great for a lot of things but for functional parts like gears, I definitely prefer FDM. Even if it's just PLA.
I wouldn't say "even if it's just PLA". PLA is pretty damn strong in it's own right. You have to go with high temp plastics to best it in most applications.
what resin did you try? i heard good things about siraya tech blu
So one reason your resin appears to be as brittle as it is, is that you are absolutely baking your parts with UV. At 4:22 in the video you had the UV timer set for 10:00.
That's about 7 to 9 minutes too long, i believe, depending on your resin and the individual part sizes.
I've been printing UV resin for about 3 years now, thousands of parts, and i can tell you that curing time has a BIG impact on brittleness. My initial tabletop game pieces i cooked at about 15 minutes each. During a game session the board was accidentally tipped onto the floor. Of 12 parts, all 12 shattered. Some into seemingly individual resin molecules.
My current go to resin for tabletop durability is Elegoo ABS (80%) mixed with Sirayatech Tenacious (20%). Parts are cured for 30 to 60 seconds per side. This results in an excellent mix of durability and flexibility for tabletop miniatures.
Obviously, for planetary gears, you probably don't want something with much flex. However, you definitely don't want something that will shatter with shear force either.
Recommend you experiment with your resins and curing times (A lot less than you showed) and see if you can dial in something better.
All that said, i think even basic PLA will better in this application than even the best resin, although some of the Sirayatech resins may give that a run. PETG might be even better!
Great feedback. For the next version I have been using “Tough PLA” and it appears to be working very well.
Man you know your stuff! Thanks for sharing all that
Well the curing doesn't just stop when you turn off the light. It will continue crosslinking over the next several months once started until the activator is fully consumed. So if it was going to get brittle, you can't just prevent it.
@@SianaGearz True, but that's the reason that parts are given paint, or a clear UV blocking spray, to prevent this.
@@michaelrechtin i found the 3DMaterials SuperPP to be some very tough resin. they also have a flexible resin that u can mix in like Steven talked about. You should try and reach out to them, maby they will sponsor some resin u can try it out with.
Thanks for sharing, great job and very interesting. I will be waiting your future publications
Did not expect that result, thanks for the video and experiment.
The detailed testing is awesome, and your design is compact and elegant!
Thanks Charlie! Your content is great
It is interesting how often they are coming out with new plastic alloys to make these products more robust. Resin looks great, it's just a matter of time until there are new processes out that bring it to a whole new level.
Watching machines like the markforged get me all kinds of excited for the future tech.
Much respect! I love these gear videos. Just saw the RV video. Looked at your video list and there is so much there I'm interested in. Great work! Sub earned by far
I’m new to your channel, but the last few videos I’ve watched have all been very informative and great quality. Thank you for sharing your work! Subscribed 👍
In Addition to the other suggestions: you UV cured the parts for 10min. My usual curing time lies around 5min. To much exposure will damage the integrity of the material and make it brittle
I also noticed this and thought it was odd..
Yeah, for small parts it's weird to use for than 3-5 min. Imagine how much brittle every small tooth got from 10 minutes of curing
This guy who made the video seems to have some mechanical modelling and electromechanical skills, but either he's trolling or just doesn't have any common sense regarding resin printing.
@@DeathCoreGuitar And he used remarkably brittle resins, and didn't know this would happen?
Glad I wasn't the only one who noticed that.
I'm looking to print a gearbox for my extruder, and am trying to decide between ASA or Resin, and here this video shows up. Then switching to a cyclic test with a cleverly designed setup for testing to stall torque was the icing on the cake! I know ASA is lower modulus than PLAs, but the modified PLAs are usually a little lower modulus. Thanks for this video, this was perfect for what I'm working on!
Great work man and a fun test. Love the style and flow to your video, I can learn a thing or 2 from ya.
I did not expect that. I would have thought that the resin gearbox would be more durable. Great test. Thanks 👍
Resin is more closely related to ceramics. This is maybe not the application for it. Try running those gearboxes under the sun though. FDM won't operate with slightly higher temperatures.
@@Michallote What the hell do ceramics have to do with thermosetting plastics??
@@WhenDoesTheVideoActuallyStart thermosetting clay?
@@Michallote how do you think that would compare with the interior of electronics? I have some Laserdisc player to restore. They need some replacement gears, as well as my Betamax decks.
@@Michallote maybe PLA will have trouble, but with ABS or nylon you`d be good to go
You should try the resin gearbox with a resin that is 1:9 flexible to riding resin. I think it would help with the brittleness.
Hey, great stuff. I've actually designed my own compound planetary gearbox a while back for a 5010 outrunner bldc, but kind of gave up haha. I used lithium silicone grease and it worked great!
Thank you for existing!
I've got a lot to learn still...thanks for this video.
Very nice video! From our experience, basic resins behave more like glass than plastic. The "ABS Like" and "Nylon Like" resins have shown us to have a little more mechanical resistance and flexibility. Although, as in all things, there are different qualities.
Great video and production value! Very happy UA-cam recommended this to me :) subscribed.
Nice project, good design and nicely done video!
You need specific resins for industrial use like this, you could mix 2 resins as well to achieve strong but yet tough
Thank you very much! I am trying to introduce this technology at my job and your video are perfect.
Thank you so much Michael!
well done and thank you for the content!
Great comparisson. Thanks!
Wow looking foward to see difference stronger resin u test
Great video recognising the durability of PLA over resin. Anything where there is any impact or other forces present (other than gravitational) looks like it’s better served by pla. It’s clearly the inelastic nature of resin makes it suitable only in non-stressed applications.
After you have introduced your printers this enthusiastic i hit the sub button!
Good old PLA for the win. Thanks!
Nicely done with both your testing and the overall product! I've found PLA to be surprisingly quite durable in gearing :)
ua-cam.com/video/WCnIJu69aV0/v-deo.html
Great overview ! Excellent work !
Probably already mentioned, but its common to mix flexible resin with standard resin to reduce the brittleness. It opens a whole can of worms though, and it really fair to only test and compare with over the counter products.
I've got some prints out of formfutura though resin. They were printed at different temperatures and illumination times and hardend differently afterwards. One is still elastic aftr a year and the otherone is hard and brittle like tough glas. Even with thr same material, resin can acive vastly different propperties.
For removing the resin prints from the print bed, I recommend getting an oil painters pallette knife. They're strong and flexible yet very thin. Easy to ease underneath a print. No hammering needed!
Very cool test setup! Thank you for the video! Also, in some defense of resins...Which resins were used? Settings? So many variables can affect resin performance. Were they cured underwater? Heating the curing oven that you cure the finished parts in can significantly increase their tensile strength for many resins. It is a complicated sport. Happy printing! :)
I needed this video. Thank you!
very good print quality
Brilliant design
to make the gears last longer, try swapping the WD40 for some grease meant for plastic and when threading screws directly into plastic, tighten them by hand and before tightening, turn the screw counterclockwise until you feel it dropping into the already existing thread, so you don't run the risk of chasing new threades every time you screw it back together.
ua-cam.com/video/WCnIJu69aV0/v-deo.html
Would mobilux ep2 work with plastic?
amazing testing thank you
Unexpected results! Nice video!!
Good stuff. One thing you might try as well is printing the parts using one of the HP MJF (multijet fusion) services out there. I did that with my BLDC compound planetary gearbox and it's impressive. You could probably print these parts for about $10 using JLCPCB's new HP MJF service.
Thanks for you work Bro!
well done! I think the reason for the failure is that the planets have a bearing support only at one end, and the other end is unsupported. this creates a twisting torque.
the resin generally has less tensile strength than PLA, it is expected to fail earlier.
Thanks for the video!
One thing to also take note it is it going to depend on the type of resin, or the type of FDM you use. Some are stronger than others. For example clear resin is known to be brittle. What you want is something like resione tough resin, siraya tech blu strong Resin, eSUN ABS-Like hard tough resin (a personal I use), then there's liqcreate tough-x that is designed for mechanical use. Theres also other brands but these I know and one I have used for more technical builds or things that may be taking a beating.
Also the cure time can have a BIG impact on resin, I saw you are doing them for 9 minutes? That is a bit much and over curing can cause any resin to become britle or form micro cracks. Even with the resin I use if I forget it in its curing box and over cure even the best resin will have issues. 2-5 minutes is normally the go to, though clear resin will cure faster and should be in half that. The fact your print shows yellowing means it cured WAY to long. The cure time also depends on the strenght of the UV light, stronger lights like in those cure stations don't need to be that long while weaker lights may need a few minutes more. For example my manual curing chamber uses a set up with a 405NM uv light and I only cure my prints for 2 minutes at the most. One minute for any clear resin or it starts to yellow and form cracks.
I don't know much about filiment printers as I've yet to get one, but I do know there's some that have metal mixed in that is really strong, and others like silk that are really brittle as well. So it really does depend on what you are using.
Yep, E-SUN ABS Like is the best, a little bit expensive, but it is a product that almost doesn´t require post processing
@@omoralopez @TAiiNE how does print cost compares with filament? Initially I wanted to start with filament (ideally recycling failed prints and making filament from water bottles etc), but it seems like resin might fit my applications better (I would lose the possibility for recycling but I guess resin prints successfully more often).
@@HyperSculptor In our case we have 5 Anycubic printers and 3 Artillery. I think that both technologies are not comparable. IE: One of our pieces printed with resin takes 5 hours, with filament the time goes up to 16.
The cost per gram of both materials is almost the same (ABS Like Resin vs PLA+ Filament), but, at the end of the sum, time is also money.
In addition, the level of detail with the resin is almost as good as if they were injected parts.
We use filament printers to print parts that require less detail, are smaller and, generally, flat.
We have tried many ABS Like resins, and there is a lot of difference between the different brands. I think that E-Sun is the best, but it is still very expensive, so we use one that is of acceptable quality and not that expensive.
@Osvaldo Mora thanks for your reply, it helps making my choice obvious, in term of technology.
@@HyperSculptor You are welcome!
Great video and interesting results. I thought for sure the resin printer would be the winner here.
have you tried this with a tough resin instead of the basic? i know you were only comparing it to pla, but it'd be interesting to what level of resin would = pla's strength and what other fdm filaments would compare with tough resin. AKA abs, asa, pc, peek, etc.
Hi! Any interest in making a video detailing the design process on how to design the gears? These things fascinate me to no end, but i always find it a bit difficult to get into due to the fact that i don't know what tools or processes i should follow to make things such as this where you have teeth count and ratios and all that worked out in a small enclosure. This looks great and professional!
Solid video, really enjoyed it!
Called it! Resin is harder and more brittle. Your results are exactly what I would expect!
I've seen resin used for making near perfect molds for poly or rubber seals though... I'm sure it has its uses that go beyond what other prints could hope to do.
There are more flexible and durable resins, they just tend to be less detailed.
What resin were you using?
The resin used apparently makes a much bigger difference than the material used in FDM (which, to be fair, already has quite big differences).
I've seen resins range from "brittle like glass" to "pretty much slightly weaker ABS".
So it would be really interesting which resins you used for your tests.
Also: I didn't see any lubricant on the resin gearbox when you took it apart. To be fair, the lubricants I usually use are pretty much black, so maybe that gel stuff just can't be seen as easily.
Really well done video.
Man... SO COOL!
Great video. Thank you.
Subscribed. Thank you!
Whenever you start getting near the limits of a material how tough a material is (with toughness being defined in materials engineering term of "area under the stress strain curve") becomes very important and can be the difference in a few hundred cycles failure and how many tens or hundreds of thousands to failure. Gears generally should be tough but with good surface hardness. FDM vs SLA is similar. FDM material selection is generally tougher than SLA, though there are "tough" SLA materials -- they generally have terrible temperature resistance. This is also the PLA - PETG trade off. PETG people rave about it's strength but it's generally more brittle. Good work.
Thats a great video! Congratulations for that. You could test this with a special resin, i know esun has its "hard-tough" one that seems to withstand a lot more.
It's great stuff. Really interesting, it would be a research to do more.
I would also add that if the PLA isn't holding up to move to PETG which in my experience is much more durable and long wearing. PLA tends to be brittle by comparison. As far as printing it isn't much different. Just mind the temps and it's just a smooth if not better.
wonderful! appreciate
Interesting results!
Very good and useful video!!
Finally someone who is using appropiate screws for plastic parts. There are way too many designs out there with machine screws and brass inserts or nuts
_Finally someone who is using appropiate screws for plastic parts._
Not everyone is as smart as you.
@@owowowdhxbxgakwlcybwxsimcwx If you're fastening together something that you don't intend to ever take apart again (how likely is that?...only you can decide), then self-tapping is OK/appropriate. But I suppose if you don't intend to disassemble, then glue the parts together.
As someone who is a novice to 3d printing and who has to overhaul the sturmey archer hub on his bike... this is super interesting! I look forward to the next version
How did it go???
ok, that was surprisingly unexpected!
really nice comparison, I was completely sold on the SLA's strength and this was actually mind-blowing. Thank you very much, here's my like, my sub and why not, my gf as well.
So many resins on the market and new ones coming out all the time. I feel we will soon be looking at FDM as a dinosaur once the better resins are developed. I'm in a group that uses resin molds for DIY plastic injection molding, so this particular use is already something FDM cannot achieve because of the high temperature plastics being injected.
How about ABS?
Great test. I was just recently wondering how PLA would do against resin and this kind of long term test is really interesting.
This is a useless test, on the level of trolling or massive ignorance, because of what resin he used, and the absurd 10 minute curing time to make it even more brittle, etc.
I use both the Creality LD-002H and a Form2 printer. For structural parts like gear, the Formlabs resins out perform Creality by lengths ! I think the MSLA resins contain so much photo catalyst to make them react in the realatively weak UV light source, compared to a spot laser source as found then SLA Formlabs. Photo catalyst seems to make resin more brittle, but having said that, there are third party resin for Creality that come very close to the super expensive Formlabs resins.
Wouldn't that just be a matter of exposure time for an MSLA?
I recently did a huge personal investigation into 3d printed gears and feel like this is a reasonable place to share my knowledge.
I found the smallest you can make an FDM 3d printed gear on an average printer with a normal sized nozzle and expect consistent smooth operation and accurately printed teeth is 1mm. That is, with a modulus of 1mm. You can go smaller, but there will be extra friction and noise and high speed operation will not work very well.
If you are generating the gears in a CAD package, check to make sure the program is actually generating accurate gears and not an approximation. Some programs will just generate an approximation because they expect you to order them off the shelf and only allow you to easily generate them for fancy renders. (SOLIDWORKS)
Herringbones are better to print than helical, and ideally you should be using double herringbone gears because I imagine you are using a 3d printed housing if you are printing gears and can choose your gear parameters, and that will reduce the strain on the housing. The self centering of herringbone gears also helps deal with any warping or tolerance issues.
If you're not using a sealed gearbox, use double helical gears instead because they deal with dirt getting between the teeth better.
Avoid the use of bevel gears if you can help it because they don't tolerate the flexibility of plastic as well, and you will have to take extra care to design everything very stiff.
Planetary gearboxes require way thicker housings than you will intuitively think you need when printed because of the flexibility of plastic.
Don't make the obvious mistake of printing double herringbone or helical planetary gearboxes, assembly is impossible on them. Yeah I made that mistake, felt dumb after.
Lubrication is super important, especially with PLA. "Just a quick test" can quickly warp your gears and you will have to print new ones if doing anything high speed.
definitely worth a sub
Cool video will print the gearhbox my own and trie to build a robotarm with it :)
Thank you. I can choose type of 3D printer.
This test was like comparing a Microscope and a Telescope for looking at the moon, ofcourse one will be better as the other was never designed for this. The resin used in most home printers is for fine detail printing, hense the reason most model makers prefer resin. If you want something to withstand mechanical stresses, the FDM printer will always be the better choice here.
was not expecting for pla to win
Thanks for your practical experienced explanation, but which FDM printer is faster
Great video, thanks. I like to think of resin as able to do what FDM can in optimal situations but clearly that simply is not the case, I don't know what else you could have done to help resin out there and it still lost by a lot.
Amazing
Creative video, thanks :)
If you could adjust your design a little, you could try using helical gears instead of spur gears. This would increase the surface area of the gear tooth interface and might distribute those forces more and reduce the risk of failure. They use helical gears in most automotive drive trains and are also supposed to reduce noise.
как раз большинство производителей используют прямозубые передачи поскольку они проще в производстве. И если речь идет об эффективности, то несомненно лидер это шевронная передача как у Ситроен
Great video! Have you done backlash measurement on this gearbox?
Awesome channel!
Super cool. I’d like to see the modeling process for resizing a exiting model. Seems when I have something that didn’t pan out, I have to mode from square one. Can’t seem to navigate fusion well enough to modify a existing design 🤦♂️. Makes prototyping very tedious
What resins were you using out of curiosity? Really hoping the tough resins get a little stronger for practical use parts I'm playing with some sirayatech build atm might try this print been thinking if ordering some steppers to play with
I strictly use build for all of my functional prints now. Haven't done much testing with it, but housings don't shatter when dropped anymore. I've heard people using blu and build mix (20/80 ratio or similar) have good results as well.
Superb video! Thou would love to see the gearbox done with high end engineering resin with post uv curing.
You proved what I discovered already a while ago. Resin printing is great for toys, for high detail miniatures, and totally unsuitable for any kind of industrial design. On top of that, the so-called "tough" resin is just a bit rubberish so it's not brittle. But also unsuitable for mechanical parts. I tried and tried and tried, but finally gave up.
I print my mechanical parts in ABS, as it is weather, sunlight and stress resistant, 0.1 mm layer hight if I need high detail. Never fails.
Niiiice. Great job
May I suggest using 3D Materials Super PP Resin. That has been the toughest resin I have ever used. It should withstand much more than even the ABS Like resins.
You have to use not only a strong resin, but also an anti friction one. Most resins, even the tough ones like siraya tenatious , are not meant for moving parts. Try syraya mecha resin or resione m58
I am doing something similar.. but i had some of the same experience. I am doing a rack an pinion set up ..but the FDM process takes to long. So i made a mold and casted the resin.. much easier.. but after not to much duty...the resin begin to share of bit, coming to final failure... I change to an epoxy base much softer resin...af it works better. It seems that resin gets to be to brittle, almost glass...but the pla don't do this. I am still working on some solution...but your video shows the same. So if you where thinking that is was due to the process i can tell you even if casted the resin do the same.
Great video, do you have any sources for further reading on this type of gearbox, other than normal planetary gearboxes? Thanks.
cool ！
good to know that the FDM is stronger！
I love the super high gear reduction and self-locking output from a stepped planetary, but mine haven't been giving full torque either. Here your stepper is rated 0.42 N-m torque, and your awesome load cell setup measures about 4 N-m torque, so a healthy 10x torque increase, but an 80x speed decrease. (Why are we losing 8x of our torque? Is it just lost to planet-ring tooth friction?)
@@PeppoMusic I actually tried helical gears in a stepped planetary, and in addition to wanting to eject themselves under torque, they had about the same losses. I'm thinking it's mostly friction, particularly since the tall planets want to tilt under load, every tooth is grinding much more than normal. I should try a planet carrier that holds the planets from both sides so they can't tilt, and see if that has better torque transfer.
@@olawlor Helical gears can transfer more torque for the same width. But will have equally more friction. It's quite easy to see why visually, just take the width of each tooth and rotate it straight in your head and see how it's wider.
Other than that benefit, they are also quieter and sometimes more wear resistant because the transition between each tooth being the load bearing tooth is smoother. This is also easy to see visually, there is slight overlap between each tooth if you draw a line across the width of the gear.
Nice video. The gray filament model looks really smooth, what configuration did you used? it's because the 3D printer resolution, or the material?
Wowwwwwwwwww!
THIS IS THE VIDEO I WAS LOOKING FOR!!!!
I do have 2 FDM printers. I taught that gears would be way better in resin.
At least, with the resin you used it's not a better option. Annnnnd, the FDM is still very capable (so I don't have to buy a sticky printer!!! Great cause I hate cleaning stuff!)
Maybe with a very expensive super strength resin?
Sheer genius. 👍
nice experiment, also the quality of the PLA print is so good. Can you share the settings you use to print PLA?
I been using fdm and many times thought about getting resin option but then I though how lazy I am to deal with curing and all the liquids so for now I’ll stick to fdm, plus fdm serve me well it does what I need and it does it well. When I need a certain part that I cannot replace I just print it. 3D printing is must have in every home.