I feel your struggle. Not because I experienced it myself, but because the cinematography and the editing are communicating so much of those weeks of hardship in 16 minutes.
The secret to becoming a person who finds sanding zen is enthusiastically lying to yourself until you start to believe it. I do a lot of textile work, and i've managed to convince myself that any boring / trivial part (taking apart wrongly sewn seams, finishing a thousand thread ends) is a nice low stakes high reward break with comparatively low screwup chances and freedom from creative thinking or decision making. (All of this obviously goes out of the window when there are tight deadlines. In that case the motto of my artistic practice turns into "why whole-ass it if half-assing will do" )
@@Cinema_Mechanics That's the thing, don't try to convince yourself, just lie like you're 12 and your mum caught you smoking. The being convinced comes after the fact
If you make a full bag and place the mold inside it instead of trying to seal the film to the mold, you'll take leaks in the mold out of the equation. That way you only have to worry about the seal on the bag edges and hoses.
I think the problem is you need an inlet and outlet on opposite sides of the part...and the bags are sold as rolls of sheets to allow flexibility for various size requirements...🤔
My understanding is that you can’t do full bag infusion since the resin would flow all around the mold and not just over the top of it. That said I haven’t tried it.
Bagging the entire mold will work. You may have some resin go the wrong way but most will go where you have the flow media. Composites have a steep learning curve but you are right there.
A 3D-printed mold definitely won’t work here, as the vacuum pressure will collapse the print. The only way it might work is if you print it completely solid without any infill. (Don’t ask how I know this!)
A lot of people think this stuff is easy, and they make a mess. But to be fair, you really aren't that far off. Give it another try and you might surprise yourself. And just FYI, for the first part, the main thing I would change is to use what they call "Wet Preg" technique...which is to lay out the fabric on a plastic sheet and apply resin to each piece individually, squeezing out excess epoxy with a squeegee...then stack them together, wrap over the inner mold, and stick it in the outer mold. The release issue was most likely cause by micro porosity in the mold surface causing mechanical bonding. So use that release film on both sides of the part. Some small surface defects are normal. Perfect shiny carbon parts you see have undergone a lot of finish work, sanding, clear coat, etc.
Love it!!! "I've got this, what could possibly go wrong?" in all of the universe conspiring against order detail. This is what learning looks like and thanks for sharing!
1. It is becoming quickly evident why CF manufacturing is so expensive. 2. I have never watched a video of errors and mistakes and been so entertained. 3. how incredible are the Easy Composite videos in eliciting feelings of "Oh man so easy, even I could do that!" 4. persistence is the key to success. Good job.
The best content on 'machining TH-cam' at the moment - I really appreciate your ingenuity, honesty and beautifully lit/shot/edited pieces. Having done some recent work with resin restoring a dragon boat, this video got me right in the feels as the kids say. Keep it up Steven!
Man, I feel for ya!! When I started making carbon parts for big R/C gliders it was a steep (very steep!) learning curve. I got there eventually though through just dogged persistence. To be fair to Easy Composites (with who I have no personal connection) they are local to me and so often go to pick up materials in person from their trade counter and to pick their brains. I often talk to Paul that does the instructional videos and he is more than helpful in giving advice and inspiration. They do though make it look easier than in real life as you showed in this excellent and very entertaining video. Keep at it fella and you'll get there!
Adding a comment to help boost the video and to please the algorithm Gods I know others have said this, but I absolutely love your videos. They're definitely underrated, and I can't wait to see more of your future videos. I'm sure you'll get the recognition and views you deserve!
Carbon fiber (and composites in general) is one of those things that is so simple in concept, but can be a nightmare in execution to get right. Love your channel
This was a cool video. The perspective of many fails at a project are good to bring to viewers trying at epoxy or other projects its not always an easy road to create things. and your video communicates that as well as sharing with the audience its okay to laugh at yourself on your journey and not to quit cause ultimately your learning and only getting better at what your working on
Your videos are great man, really informative...i feel your pain, they make it look easy, but every new process has such a learning curve....glad you're sticking with it!!
My future self is brimming with confidence after vicariously living through your multiple failures, so I hope you keep up with your warts and all, play-by-play style. The cognitive dissonance of feeling certain that you'll reach your goal and being in awe of how ambitious you are is enthralling!
your failures are inspiring. I had many composite failures before the summer, so many I had to stop it was getting me down. watching this video gives me the perspective that composites are hard and you need to fail a lot before a successful process. drat, now I gotta start it up again!
I love the style of your videos! I had to learn many of the same lessons when working with composites and I was just working on my own robot arm when I saw your video pop up. perfect timing :)
I knew you were gonna say "Easy Composites" as soon as you started talking about the TH-cam rabbit hole lol. Also - I've also been experimenting with a lot of moulding-involved projects lately and if I've learned one thing it's to never use PLA. Ever. It's a shame, because for almost any other application my advice to people would be to always use PLA.
He speaks like the child of Ryan Reynolds and Eeyore who was raised by Mark Rober on the weekends. Another absolute banger of a video. Really hope these videos stay this amazing.
Congratulations on discovering the horrors of composites. Here's some ideas, that may or may not be worthwhile trying. It's up to you to judge that For the U shaped part, your idea was good, but could beneficiate from some improvements. One option is just have an excess of fiber on the top, don't wet it with resin, clamps it in place, then insert your inner mould as it was (maybe use a threaded rod and so nut inserts to pulls together after curing). The idea is to prevent the fibers from being pulled down by the mould. Harder and dumber version could also having inner mold be pulled together during insertion (threaded rod for tension or any other mechanisms could work) so as to minimise contact with the part's walls. Ideally the printed inner mold would be smaller than it is in place. You'd then expand it with a wedge and some clamps for curing. For the bottom radius, have the inner mould be thin enough to "roll" up to the part when under compression, use wedges and clamps when curing to fully round them out at the bottom.
I love your resilience and perserverance. your work shows us all how NOT to do it and gives us a good laugh along the way! but also flex's my empathy, i feel so sorry for you but also know that when you get your arm finalised you will know ALL the ways not to do it! keep making these videos my friend!
Based on my experience, it's always best to infuse resin along the direction of the unidirectional fibers. Another important point is to slow down the infusion rate so the resin can thoroughly penetrate the fibers; otherwise, it can overflow the fibers and trap air bubbles. As you mentioned, infusions can be challenging, and maintaining a clean and precise workflow is essential. For the new mold, I’d definitely recommend using fiberglass. If there’s a leak, you can simply double-bag the entire mold. This approach should make it much easier to achieve a perfect vacuum.
Yep - been there. Particularly failing the drop test and then still thinking that you can infuse a part successfully. When the instructions say 'absolutely no leaks', they really mean that, not 'couple of leaks will be right...'!
Composites are not easy, but the EasyComposites videos make it much easier to learn. These are top notch tutorials but everything is always more complex than a tutorial when you do it for the first time. It is safe to consider that you will fail several times in the begininning..
Composites is the sorta thing where you kinda have to suffer for a bit before you appreciate the medium. Hot resin actually flows much better and its common practice to actually mix and start flowing the resin on the cold side before passing it thru a heating process immeadiatly before it enters the mold cavity. Don't feel too bad. The marine industry has sacrificed an obscene amount of hulls figuring out this process. Some even made it into customer's hands.....
It's fun to watch all the trials! Rang the bell already! Looking forward to the next video! And! The third option can be ordering from pcbway directly if it's in urgent haha.
Am an engineer who turned into making videos and right now am laughing with tears, it's like being a director with no knowledge of turning a story into a audio visual element, yaani its easy to watch a movie and see mistakes than to make the movie itself... love this !!
A few recommandations: 1. Use a full vacuum bag for the U shape part instead of an inner mold. 2. For mold design : avoid double curvatures, minimum radius of curvature of about 20mm for single curvature, double that at least for double curvatures. Also you can do little cuts in the textile at alternate locations to simplify comfortability 3. The West System 25 min is very reactive and only applicable to very small jobs. It is better to have a resin with at least double the amount of time that you expect. 4. Infusion is great but it is one of the most complex process to master. Honestly, I am pretty sure you could do that same job more easily with wet layup and vacuum bag. The easiest vacuum bag technique is to make a fully closed bag and insert the mold I inside. You can even use premade clothing storage vac bags that kind of work ok for DIY.
Given the amount of curing timing mess up you did, the first attempt result was extremely good. We had an idiot do the same thing when we were making our rocket body with a mold he forgot he had to start a timer for the resin curation debut. The waste management was crazy difficult.
As an industrial robotics engineer, I got told many many times by management ppl "Building our own robot shouldn't be that hard, the one we buy are way too expensive." I use to argu against the idea, now I just ignore it.
The bubbles in your vacuum line are likely not due to leaks, they almost certainly bubbles introduced by stirring up the epoxy before you infused it. If the bubbles are also appearing in your molded part, then you should degas the epoxy prior to infusing it using a vacuum pot. I'm enjoying these videos and the droll delivery. Keep it up.
This could totally be the case. From videos examples I thought those mostly appeared along the front “wave” of resin. But I’ll probably have to give that a go.
This is great content! If fiberglass is easier and cheaper, then go for it. At least to get you through the learning curve. For most practical purposes, fiberglass will be as good as carbon fiber, it's just a bit heavier. And on a robot that is full of metal parts anyway, the difference between fiberglass and carbon fiber is probably minimal.
EasyComposites make it look so easy, when in reality it's the most messy, stringy, sticky, oozy thing ever. Still it's carbon fiber, so very futuristic (apart from titanium, which is similarily a PITA to work with)
OMG that first mould design and attempt to build the parts like that... I instantly saw what was wrong when you showed the CAD, but that's second-eyes advantage, because two dumb people together usually spot mistakes that one smart person doesn't, and for other priming advantage (similar to hindsight), since you already told us that this is going to be a trouble vid. In real life i make the same kinds of boneheaded mistakes when working alone, and then wonder what i was thinking and why i thought that was going to go well. The magic of Easy Composites Ltd, UK, is that... they worked out all the kinks and supply high quality, purpose-fit materials, thoroughly designed and tested. And when i'm finding that their materials are a little too expensive... well that just makes me think twice whether i want to use a given method, because odds are, they're expensive for a reason and substituting them with something drastically cheaper won't do me much good. And even then one shouldn't expect success on first try, and maybe not on the second, and in general, yeah having done things a number of times, i'm pretty certain i'm not going to reach that level of success, i mean, it's a skill, just like anything, it needs to be honed. Like, i know i'm trash at designing moulds. I'm a bit lucky that i'm in EU and they have an NL subsidiary, and before Brexit it was super smooth sailing anyway.
Also my suggestion is to not make a mould from the part, but actually 3D print and prep the mould the same way you did previously, because that obviously worked well enough. Because i think otherwise you'll either ruin your good part in a way that neither of us can exactly predict yet, or receive a bad negative at first try, maybe with voids, very likely twisted out of shape. PLA and PETG are cheap, you're not saving money, and you're not saving time because you're not taking trial and error into account.
That first mold should've been three pieces. Solid inner plug and two part outer (2x L shaped pieces.) Lay up on the inner then add the outers and clamp everything together.
I have been there. West Systems Epoxy and 209 extra slow hardener is a much better product for warm weather layups. It takes so long to cure you can mix up a second batch in the middle of a pour or layup with out the part you are working on settings. I don't have a shop or a table inside and I can actually make parts outside in 80+ degrees. With that product. As well even after mold release a epoxy continues to harden for 3 days to a week. You should find your piece much stronger than when you first released it. The pinholes in your infusion are not a problem for structure but they will be for making a mold. You can use an automotive epoxy clear coat (1k) to fill the voids and provide a perfect finish with MUCH less work then surface epoxy.
Not a problem. I have made a lot of mistakes and learned from them. I’m excited to see what you do next. You’re combining my interests in cinematography and carbon fiber fabrication. Great to see
Hey, I love the fact you are making a robot arm. That is one of my dream projects. I just found your channel a few months ago when you started this project and has been awesome to see your progress! By the way, I am a mechanical engineer that works in the composites industry. I have experience in manufacturing in every manufacturing method, designing molds, layup structures, and more! I primarily make military and aerospace components for L3, Cobham, etc. Let me know if you have any questions on how to make better parts! Many times technique and strategy for manufacturing can make all the difference. Please feel free to contact me if you want any advice on anything composites!
Pro tip: upload in 4K, even if your source material is FHD. It uses a different transcoding process and makes the videos look _way_ better. 5K or 8K is even better. Surface defects are somewhat expected in large, flat parts. As long as you get good wetout, you can just fill those in with a gloss sealing pass later. But yeah, composites suck, it just takes personal experience to make it work well.
Maybe for future attempts at a composite material, you could try to collab with Xyla Foxlin. She's done a lot of work with epoxy and carbon fiber/fiberglass. Plus I think it would make for some fun content
Didn't look like you degassed your epoxy before infusion, bubbles are likely just coming out of the epoxy. Lots of gas gets trapped inside it because viscosity. You can improve the appearance by painting a couple thick clear coat layers. Even just rattle can clear with a steady hand can easily turn the worst carbon fiber part into a thing of beauty.
Oof yes been there. Thought I could start a composites company, because I’ve watched all easy composites videos three times. I failed hard. Really hard 😀
I'm only a minute 45 in. I've watched almost all of easy comps videos. I've never done carbon fiber. But Dunning-Kruger 😅😅😅😅 Edit: finished. Wow, if you had a vacuum oven (not worth the price of one or two parts), you could have done prepreg. Thanks for sharing your experience. Easy Composites makes it look easy. But I watch with the sense of "painting is 99% prep". The painting is more fun to watch than prep. I'm sure a month or more is taken to prep everything before his 15 minute video makes it look like a couple hours work. But the end results.....
Please see a doctor about your clubbed nails, as they can be a symptom of certain diseases. Also, nice video, dude! Keep up the great work and don't stop doing this.
How very dare you talk such heresy about Easy Composites! Lol. Been there mate. My first fibreglassing was on a yacht from the '60's. To say it was a disaster would be understatement. got the hang of it, eventually. Ish. Very ish. Fast forward many, many years, I come across EC on YT and eventually jumped in and tried a wee project, for an astrophotography project. My DK readings were high, so I went for it. An absolute, unmitigated disaster and a waste of a chunk of change. After spending way too much time and money, I ended up with a "thing" that looked like it had been made by an alcoholic toddler with Parkinson's. I feel your pain brother...
FIBER GLASS MOLDS ARE WAY WORSE PLS DO NOT WASTE YOUR TIME!!!! Imagine how difficult it is to wet our 10 or more layers of fiber glass whithout any imperfections
At 6:43, you are wrong. Yes, it will increase stiffness, but not strength.The elastic modulus of carbon fiber is wildly different from that of aluminium, so the CF will break before the aluminium gets a chance to take up significant forces. You can adapt the thicknesses accordingly, but that's a nightmare in itself as all of this is heavily dependent on your fiber orientation and the direction of the force applied.
I let out a small 'noooooo' when you bagged around the mold's flange instead of showing the entire thing into one big bag. Also, why not print vertically from ABS at a tiny layer height so that solvent smoothing takes care of everything?
My understanding is that you can’t do infusion in a bag since resin would run all around the mold. But I haven’t tried it. My printers struggle with ABS since they’re not enclosed. Also I don’t trust vertical prints nearly as much for dimensional accuracy. Maybe that’s old fashioned.
@@Cinema_Mechanics Yes, the resin will cover the entire mold and stick if you don't use mould release. But you don't really care about the dimensions of the mould except where the part sits. The bigger issue is that the pressure may crush the mould and/or force resin into it through small gaps - better than letting air into your part. Better coat it beforehand to avoid loosing a kg of epoxy into the void. And yeah, I also avoided ABS and vertical printing like the plague until i got my P1P
Non structural parts, could have easily 3d printed the cover. If you really want the carbon fiber look because you sure don't need it's structural properties, the 3M wrap looks like carbon to anyone who doesn't work with carbon. Not even going to talk about fact that pla covers can be thinner and will probably be lighter. Structural parts, are again not needed. You are not ever going to bend the metal arms swinging a camera around, unless you have an IMAX camera behind your back lol. Now if you made the arms from just carbon (FYI you could easily get china to cut out a full set of arms and plates for the arm for cheap, just ask the FPV drone people.) that would be pretty cool. Keep making fun videos bro :)
The idea was to make them structural. The idea is max rigidity since the arm has to deal with all the acceleration forces. Maybe unnecessary. But I also kind of doubt I could 3d print a cover that’s as light as the CF one.
@@Cinema_Mechanics you have 3d models of the cover so you could always print it out and weigh it. I am specifically talking about the u-shaped cover, not the big flat ones. You could also reduce the thickness of the material on the 3d print because you are not locked into a preset part thickness because of the thickness of your weave. And about the stiffness/strength of the arms you can always just do some material/force simulations. With how thick the aluminum arms are I'm pretty sure you are going to run into issues with the forces applied to your servos/motors/belts before you you run into any issues with the arms stiffness.
@Cinema_Mechanics Start with Glas fiber, dirt cheap, and only about 20% worse mechanical at the same weight. Also, it is a breeze to work with compared to carbon. When you are looking for those stiffening parts, use IMS or UMS fibers, but make sure you are trained. If you thought carbon fiber was expensive, sit down before looking after UMS 🤣 Use glas for your first tries. They always fail. Also, your molds have unnecessary complications, those small indents around the bolts... no... When you do infusion, use spray glue (3M 77) to fix the layers, use an 80gm glas layer top and bottom. That gets rid of pinhole ease. And, if you have complex molds, always use 45deg and twill weave, linnen is for non convex or concave parts. In the inner layers, you can also use non-woven fabrics, uni or bidirectional. And the most important thing you didn't do... Degass your epoxy in a vacuum chamber!!! 😇 Have fun!
Why build the robot from scratch? I think the best thing to do is buy commercial for example ABB robotic arm with broken/obsolete electronics, then you give it some refresh/cleanup and drive the OG motors and encoders with modern compact electronics. I didn't look at prices but I would imagine you can buy $100k robotic arm that is old and broken for like $5k or less. Regarding your first cf part attempt, you made a lot of mistakes totally on your part, I never made a single cf part and I could see that with this mold design the cf would get just shoved down. Using quick cure epoxy in a got rooms is also an obvious mistake. You also gone with wrong technology for this part, with shapes like that and need for smooth flat finish on both sides, you should've gone with forged cf using chopped/tow cf and like 6 part mold.
Please never stop making videos. Most underrated youtube channel right now
Much appreciated!! I’m doing my best to keep it going!
@@Cinema_MechanicsMan, appreciate!
When you start at boss level.
At first I read the exact opposite haha. "Please stop making videos"
sooooo relatable... the stumbling across easy composites videos and thinking... "hang on a minute.... i can DO THIS!"
Gods don't fire pots - we do!
Sure you can!
@@VEC7ORlt Insert raccoon trash can meme here.
This and the superfastmatt videos are the best way to understand everything that can and will go wrong with composites. Great video!
I promise to never include a ham sandwich at least. I hate ham.
I feel your struggle. Not because I experienced it myself, but because the cinematography and the editing are communicating so much of those weeks of hardship in 16 minutes.
Fail videos > success videos, because face it, when we try to do what yt shows, that’s exactly what happens
The secret to becoming a person who finds sanding zen is enthusiastically lying to yourself until you start to believe it.
I do a lot of textile work, and i've managed to convince myself that any boring / trivial part (taking apart wrongly sewn seams, finishing a thousand thread ends) is a nice low stakes high reward break with comparatively low screwup chances and freedom from creative thinking or decision making.
(All of this obviously goes out of the window when there are tight deadlines. In that case the motto of my artistic practice turns into "why whole-ass it if half-assing will do" )
I definitely relate to the half ass it thing. Still working on convincing myself…
@@Cinema_Mechanics That's the thing, don't try to convince yourself, just lie like you're 12 and your mum caught you smoking. The being convinced comes after the fact
If you make a full bag and place the mold inside it instead of trying to seal the film to the mold, you'll take leaks in the mold out of the equation. That way you only have to worry about the seal on the bag edges and hoses.
I think the problem is you need an inlet and outlet on opposite sides of the part...and the bags are sold as rolls of sheets to allow flexibility for various size requirements...🤔
My understanding is that you can’t do full bag infusion since the resin would flow all around the mold and not just over the top of it. That said I haven’t tried it.
@@Cinema_Mechanics Well there's only one way to find out!
Bagging the entire mold will work. You may have some resin go the wrong way but most will go where you have the flow media. Composites have a steep learning curve but you are right there.
A 3D-printed mold definitely won’t work here, as the vacuum pressure will collapse the print. The only way it might work is if you print it completely solid without any infill. (Don’t ask how I know this!)
A lot of people think this stuff is easy, and they make a mess. But to be fair, you really aren't that far off. Give it another try and you might surprise yourself.
And just FYI, for the first part, the main thing I would change is to use what they call "Wet Preg" technique...which is to lay out the fabric on a plastic sheet and apply resin to each piece individually, squeezing out excess epoxy with a squeegee...then stack them together, wrap over the inner mold, and stick it in the outer mold. The release issue was most likely cause by micro porosity in the mold surface causing mechanical bonding. So use that release film on both sides of the part. Some small surface defects are normal. Perfect shiny carbon parts you see have undergone a lot of finish work, sanding, clear coat, etc.
Appreciate the suggestions! I’m gonna totally rethink that first part.
Teachers emphasizing the successes: B+
Teachers emphasizing the failures: AAA+++
Dude, I love the brutal honesty and self reflection. Absolutely brilliant to watch for any maker and/or machinist.
I'm really loving these! You don't hide the mistakes you make, and that makes them all the more educational, believable, and genuinely impressive.
Your videos deserve an Oscar, this is incredible!
So impressed by the editing and commentary of these videos. You should be proud of your work here. Commenting to hopefully they that algo moving
Love it!!! "I've got this, what could possibly go wrong?" in all of the universe conspiring against order detail. This is what learning looks like and thanks for sharing!
1. It is becoming quickly evident why CF manufacturing is so expensive.
2. I have never watched a video of errors and mistakes and been so entertained.
3. how incredible are the Easy Composite videos in eliciting feelings of "Oh man so easy, even I could do that!"
4. persistence is the key to success.
Good job.
@@aarondcmedia9585 thank for watching! Yeah gives you a new appreciation for how cheap CF has become really.
1m camming soon if you keep going this
The best content on 'machining TH-cam' at the moment - I really appreciate your ingenuity, honesty and beautifully lit/shot/edited pieces. Having done some recent work with resin restoring a dragon boat, this video got me right in the feels as the kids say. Keep it up Steven!
Thank you so much! Really appreciate the support!
Man, I feel for ya!!
When I started making carbon parts for big R/C gliders it was a steep (very steep!) learning curve. I got there eventually though through just dogged persistence.
To be fair to Easy Composites (with who I have no personal connection) they are local to me and so often go to pick up materials in person from their trade counter and to pick their brains. I often talk to Paul that does the instructional videos and he is more than helpful in giving advice and inspiration. They do though make it look easier than in real life as you showed in this excellent and very entertaining video.
Keep at it fella and you'll get there!
Adding a comment to help boost the video and to please the algorithm Gods
I know others have said this, but I absolutely love your videos. They're definitely underrated, and I can't wait to see more of your future videos. I'm sure you'll get the recognition and views you deserve!
Thank you!!
Carbon fiber (and composites in general) is one of those things that is so simple in concept, but can be a nightmare in execution to get right. Love your channel
This was a cool video. The perspective of many fails at a project are good to bring to viewers trying at epoxy or other projects its not always an easy road to create things. and your video communicates that as well as sharing with the audience its okay to laugh at yourself on your journey and not to quit cause ultimately your learning and only getting better at what your working on
Your videos are great man, really informative...i feel your pain, they make it look easy, but every new process has such a learning curve....glad you're sticking with it!!
Just found out your channel.... BRO, your voiceover and image quality is stuninng. addictive even for experienced cnc shop owner. Keep pushin
Great video, Stephen! Tough one man... but it's super impressive you're even attempting this.
So funny and educational at the same time 😂. I subscribed and liked.
Thank you for sharing this! I have nothing to helpful or witty add, so this is just feeding the platform's machine spirit.
I just started building my robot arm, damn it is though when you don't want to spend money on bearings.
My future self is brimming with confidence after vicariously living through your multiple failures, so I hope you keep up with your warts and all, play-by-play style. The cognitive dissonance of feeling certain that you'll reach your goal and being in awe of how ambitious you are is enthralling!
your failures are inspiring. I had many composite failures before the summer, so many I had to stop it was getting me down. watching this video gives me the perspective that composites are hard and you need to fail a lot before a successful process. drat, now I gotta start it up again!
I love the style of your videos! I had to learn many of the same lessons when working with composites and I was just working on my own robot arm when I saw your video pop up. perfect timing :)
@@simplerocketscience6222 thanks for watching! Good look on your arm.
I knew you were gonna say "Easy Composites" as soon as you started talking about the TH-cam rabbit hole lol.
Also - I've also been experimenting with a lot of moulding-involved projects lately and if I've learned one thing it's to never use PLA. Ever. It's a shame, because for almost any other application my advice to people would be to always use PLA.
I usually use PETG but wanted to make sure the coating epoxy stuck well. Hence PLA.
I'll be watching this, this.. SuperFilmMat
Love it man. Also. Love your color grading. Can relate to you nonstop
He speaks like the child of Ryan Reynolds and Eeyore who was raised by Mark Rober on the weekends. Another absolute banger of a video. Really hope these videos stay this amazing.
@@blueGate3 lol
I'm so sorry thank you for sharing your pains doing this. I really got a kick out of it 😂😅
Pain is a good teacher. 😅
Congratulations on discovering the horrors of composites.
Here's some ideas, that may or may not be worthwhile trying. It's up to you to judge that
For the U shaped part, your idea was good, but could beneficiate from some improvements.
One option is just have an excess of fiber on the top, don't wet it with resin, clamps it in place, then insert your inner mould as it was (maybe use a threaded rod and so nut inserts to pulls together after curing). The idea is to prevent the fibers from being pulled down by the mould.
Harder and dumber version could also having inner mold be pulled together during insertion (threaded rod for tension or any other mechanisms could work) so as to minimise contact with the part's walls. Ideally the printed inner mold would be smaller than it is in place. You'd then expand it with a wedge and some clamps for curing.
For the bottom radius, have the inner mould be thin enough to "roll" up to the part when under compression, use wedges and clamps when curing to fully round them out at the bottom.
Please don't stop making videos! 🙏
I love your resilience and perserverance. your work shows us all how NOT to do it and gives us a good laugh along the way! but also flex's my empathy, i feel so sorry for you but also know that when you get your arm finalised you will know ALL the ways not to do it! keep making these videos my friend!
Based on my experience, it's always best to infuse resin along the direction of the unidirectional fibers. Another important point is to slow down the infusion rate so the resin can thoroughly penetrate the fibers; otherwise, it can overflow the fibers and trap air bubbles.
As you mentioned, infusions can be challenging, and maintaining a clean and precise workflow is essential.
For the new mold, I’d definitely recommend using fiberglass. If there’s a leak, you can simply double-bag the entire mold. This approach should make it much easier to achieve a perfect vacuum.
Thanks for the tips!
Yep - been there. Particularly failing the drop test and then still thinking that you can infuse a part successfully. When the instructions say 'absolutely no leaks', they really mean that, not 'couple of leaks will be right...'!
This was amazing im not sure how i havent came across your channel before but you best bet im subbin now
Composites are not easy, but the EasyComposites videos make it much easier to learn. These are top notch tutorials but everything is always more complex than a tutorial when you do it for the first time. It is safe to consider that you will fail several times in the begininning..
Yeah hopefully I made it clear how much I appreciate their videos. Just poking some good fun.
Composites is the sorta thing where you kinda have to suffer for a bit before you appreciate the medium. Hot resin actually flows much better and its common practice to actually mix and start flowing the resin on the cold side before passing it thru a heating process immeadiatly before it enters the mold cavity. Don't feel too bad. The marine industry has sacrificed an obscene amount of hulls figuring out this process. Some even made it into customer's hands.....
It's fun to watch all the trials! Rang the bell already! Looking forward to the next video! And! The third option can be ordering from pcbway directly if it's in urgent haha.
+10 for the Dunning Kruger reference.
Ive wanted to buy kits from these guys for a long time :D Love the content my guy!
Am an engineer who turned into making videos and right now am laughing with tears, it's like being a director with no knowledge of turning a story into a audio visual element, yaani its easy to watch a movie and see mistakes than to make the movie itself... love this !!
A few recommandations:
1. Use a full vacuum bag for the U shape part instead of an inner mold.
2. For mold design : avoid double curvatures, minimum radius of curvature of about 20mm for single curvature, double that at least for double curvatures. Also you can do little cuts in the textile at alternate locations to simplify comfortability
3. The West System 25 min is very reactive and only applicable to very small jobs. It is better to have a resin with at least double the amount of time that you expect.
4. Infusion is great but it is one of the most complex process to master. Honestly, I am pretty sure you could do that same job more easily with wet layup and vacuum bag. The easiest vacuum bag technique is to make a fully closed bag and insert the mold I inside. You can even use premade clothing storage vac bags that kind of work ok for DIY.
Given the amount of curing timing mess up you did, the first attempt result was extremely good. We had an idiot do the same thing when we were making our rocket body with a mold he forgot he had to start a timer for the resin curation debut. The waste management was crazy difficult.
As an industrial robotics engineer, I got told many many times by management ppl "Building our own robot shouldn't be that hard, the one we buy are way too expensive."
I use to argu against the idea, now I just ignore it.
I agree. If you earn income from it, you should just buy it. This is all for “fun”.
The bubbles in your vacuum line are likely not due to leaks, they almost certainly bubbles introduced by stirring up the epoxy before you infused it. If the bubbles are also appearing in your molded part, then you should degas the epoxy prior to infusing it using a vacuum pot.
I'm enjoying these videos and the droll delivery. Keep it up.
This could totally be the case. From videos examples I thought those mostly appeared along the front “wave” of resin. But I’ll probably have to give that a go.
This is great content! If fiberglass is easier and cheaper, then go for it. At least to get you through the learning curve. For most practical purposes, fiberglass will be as good as carbon fiber, it's just a bit heavier. And on a robot that is full of metal parts anyway, the difference between fiberglass and carbon fiber is probably minimal.
Great video! Hard experience but you have learned a lot;-)
EasyComposites make it look so easy, when in reality it's the most messy, stringy, sticky, oozy thing ever. Still it's carbon fiber, so very futuristic (apart from titanium, which is similarily a PITA to work with)
This was throughly entertaining!
Times I've said 'oh no' in anticipation, lets say if it was a drinking game...
How’s the hangover?
@@Cinema_Mechanics the liver sends best regards! or is it worst?
This is pretty crazy as I have messing with carbon and having similar results within the last few weeks.
The allure of carbon is just too strong. 😂
OMG that first mould design and attempt to build the parts like that... I instantly saw what was wrong when you showed the CAD, but that's second-eyes advantage, because two dumb people together usually spot mistakes that one smart person doesn't, and for other priming advantage (similar to hindsight), since you already told us that this is going to be a trouble vid. In real life i make the same kinds of boneheaded mistakes when working alone, and then wonder what i was thinking and why i thought that was going to go well.
The magic of Easy Composites Ltd, UK, is that... they worked out all the kinks and supply high quality, purpose-fit materials, thoroughly designed and tested. And when i'm finding that their materials are a little too expensive... well that just makes me think twice whether i want to use a given method, because odds are, they're expensive for a reason and substituting them with something drastically cheaper won't do me much good. And even then one shouldn't expect success on first try, and maybe not on the second, and in general, yeah having done things a number of times, i'm pretty certain i'm not going to reach that level of success, i mean, it's a skill, just like anything, it needs to be honed. Like, i know i'm trash at designing moulds. I'm a bit lucky that i'm in EU and they have an NL subsidiary, and before Brexit it was super smooth sailing anyway.
Also my suggestion is to not make a mould from the part, but actually 3D print and prep the mould the same way you did previously, because that obviously worked well enough. Because i think otherwise you'll either ruin your good part in a way that neither of us can exactly predict yet, or receive a bad negative at first try, maybe with voids, very likely twisted out of shape. PLA and PETG are cheap, you're not saving money, and you're not saving time because you're not taking trial and error into account.
Easy to fall into sunk cost fallacy as well. Once you’ve committed time and energy to an stupid idea, it feels more wasteful to not follow through.
That first mold should've been three pieces. Solid inner plug and two part outer (2x L shaped pieces.) Lay up on the inner then add the outers and clamp everything together.
Awesome video ngl !
Love the content
Underrated
Oh man, if I’m honest this is how the large majority of my projects turn out. Marginal partial success after 10x the work you expected
Haha. The diy dilemma.
For anyone else: Consider trying with fiberglass instead of CF. Materials are much cheaper and the techniques carry over.
You may not be an engineer but you are very good at mimicking other TH-camrs. I hope you fine your own style.
I have been there. West Systems Epoxy and 209 extra slow hardener is a much better product for warm weather layups. It takes so long to cure you can mix up a second batch in the middle of a pour or layup with out the part you are working on settings. I don't have a shop or a table inside and I can actually make parts outside in 80+ degrees. With that product.
As well even after mold release a epoxy continues to harden for 3 days to a week. You should find your piece much stronger than when you first released it.
The pinholes in your infusion are not a problem for structure but they will be for making a mold. You can use an automotive epoxy clear coat (1k) to fill the voids and provide a perfect finish with MUCH less work then surface epoxy.
Thanks so much for the tips!
Not a problem. I have made a lot of mistakes and learned from them. I’m excited to see what you do next. You’re combining my interests in cinematography and carbon fiber fabrication. Great to see
This looked exactly like the last 25years of my live! And yes, it physically hurt one more time!)
Composite manufacturing has always felt more like an art rather than engineering
I am a robotics system engineer. This is great stuff! You could just buy a used robot for a few $G's, but where's the fun in that.
Hey, I love the fact you are making a robot arm. That is one of my dream projects. I just found your channel a few months ago when you started this project and has been awesome to see your progress!
By the way, I am a mechanical engineer that works in the composites industry. I have experience in manufacturing in every manufacturing method, designing molds, layup structures, and more! I primarily make military and aerospace components for L3, Cobham, etc. Let me know if you have any questions on how to make better parts! Many times technique and strategy for manufacturing can make all the difference.
Please feel free to contact me if you want any advice on anything composites!
Much appreciated!
This shows how difficult everything is if you don't have experience!
Pro tip: upload in 4K, even if your source material is FHD. It uses a different transcoding process and makes the videos look _way_ better. 5K or 8K is even better.
Surface defects are somewhat expected in large, flat parts. As long as you get good wetout, you can just fill those in with a gloss sealing pass later. But yeah, composites suck, it just takes personal experience to make it work well.
Much appreciate the tip. I’ll look into that in the future. Source is gonna stay HD for a while but want it to look as good as possible!
Maybe for future attempts at a composite material, you could try to collab with Xyla Foxlin. She's done a lot of work with epoxy and carbon fiber/fiberglass.
Plus I think it would make for some fun content
I think I gotta get my subs up first. 😉
@@Cinema_Mechanics Oh right, your production quality made me forget you're a "small" channel..
Didn't look like you degassed your epoxy before infusion, bubbles are likely just coming out of the epoxy. Lots of gas gets trapped inside it because viscosity.
You can improve the appearance by painting a couple thick clear coat layers. Even just rattle can clear with a steady hand can easily turn the worst carbon fiber part into a thing of beauty.
Yeah I will have to see about degassing. Thanks for the tip! And clear coat is gonna happen for sure.
Oof yes been there. Thought I could start a composites company, because I’ve watched all easy composites videos three times. I failed hard. Really hard 😀
As soon as I noticed flame in the intro I was like "he tried my alcool on CNC suggestion and burned down the machine ouch"😂
Luckily not the case😂
lol. Thankfully no! Haven’t tried the alcohol on the CNC yet though. So can’t guarantee that WON’T happen. ;-)
GREAT VIDEO!
I'm only a minute 45 in. I've watched almost all of easy comps videos. I've never done carbon fiber. But Dunning-Kruger 😅😅😅😅
Edit: finished. Wow, if you had a vacuum oven (not worth the price of one or two parts), you could have done prepreg. Thanks for sharing your experience. Easy Composites makes it look easy. But I watch with the sense of "painting is 99% prep". The painting is more fun to watch than prep. I'm sure a month or more is taken to prep everything before his 15 minute video makes it look like a couple hours work. But the end results.....
Yep. It’s 100% prep. By the time you’re at laminating or infusion it’s too late!
it angers me that we don't have affordable camera robot arms yet
Me too!
Commenting for the algorithm.
You are not only youtuber who has sun ruined mould 😁
Please see a doctor about your clubbed nails, as they can be a symptom of certain diseases. Also, nice video, dude! Keep up the great work and don't stop doing this.
Nice storytelling! and you kinda sound like kevin spacey
How very dare you talk such heresy about Easy Composites! Lol. Been there mate. My first fibreglassing was on a yacht from the '60's. To say it was a disaster would be understatement. got the hang of it, eventually. Ish. Very ish.
Fast forward many, many years, I come across EC on YT and eventually jumped in and tried a wee project, for an astrophotography project. My DK readings were high, so I went for it. An absolute, unmitigated disaster and a waste of a chunk of change. After spending way too much time and money, I ended up with a "thing" that looked like it had been made by an alcoholic toddler with Parkinson's.
I feel your pain brother...
Hahaha. Glad I’m not the only one to feel the struggle.
FIBER GLASS MOLDS ARE WAY WORSE PLS DO NOT WASTE YOUR TIME!!!!
Imagine how difficult it is to wet our 10 or more layers of fiber glass whithout any imperfections
60 seconds in, and I subbed 😅
really nicee
pls tell me all machines you used step by step and price thanks
with that first cover, couldn't you just design it with a draft angle and have the bolts that attach it pull it in to place?
Maybe… recommended draft is 5 degrees. So that’s a fair amount of bend.
At 6:43, you are wrong. Yes, it will increase stiffness, but not strength.The elastic modulus of carbon fiber is wildly different from that of aluminium, so the CF will break before the aluminium gets a chance to take up significant forces. You can adapt the thicknesses accordingly, but that's a nightmare in itself as all of this is heavily dependent on your fiber orientation and the direction of the force applied.
That's an important distinction. I appreciate the feedback!
@@Cinema_Mechanics np, great video!
I let out a small 'noooooo' when you bagged around the mold's flange instead of showing the entire thing into one big bag. Also, why not print vertically from ABS at a tiny layer height so that solvent smoothing takes care of everything?
My understanding is that you can’t do infusion in a bag since resin would run all around the mold. But I haven’t tried it. My printers struggle with ABS since they’re not enclosed. Also I don’t trust vertical prints nearly as much for dimensional accuracy. Maybe that’s old fashioned.
@@Cinema_Mechanics Yes, the resin will cover the entire mold and stick if you don't use mould release. But you don't really care about the dimensions of the mould except where the part sits.
The bigger issue is that the pressure may crush the mould and/or force resin into it through small gaps - better than letting air into your part. Better coat it beforehand to avoid loosing a kg of epoxy into the void.
And yeah, I also avoided ABS and vertical printing like the plague until i got my P1P
Bamboo labs... are you listening?? I'll have to try the whole bag at some point. Maybe it'll help seal my mold!
thanks for reminded rest of us to do reality checks x)
That's so great🎉
why not split that U shape into three flat pieces with a 90degree bend as overlap? way easier than a U shape with 3 equally long sides
6:28 what software is that? Fusion?
Yes, I use Fusion 360 for all the design/sim/cam.
I hope you are using a respirator while cutting the carbon fiber
Yeah, I was. Safety first!
What a pillock
😂
Non structural parts, could have easily 3d printed the cover. If you really want the carbon fiber look because you sure don't need it's structural properties, the 3M wrap looks like carbon to anyone who doesn't work with carbon.
Not even going to talk about fact that pla covers can be thinner and will probably be lighter.
Structural parts, are again not needed. You are not ever going to bend the metal arms swinging a camera around, unless you have an IMAX camera behind your back lol.
Now if you made the arms from just carbon (FYI you could easily get china to cut out a full set of arms and plates for the arm for cheap, just ask the FPV drone people.) that would be pretty cool.
Keep making fun videos bro :)
The idea was to make them structural. The idea is max rigidity since the arm has to deal with all the acceleration forces. Maybe unnecessary. But I also kind of doubt I could 3d print a cover that’s as light as the CF one.
@@Cinema_Mechanics you have 3d models of the cover so you could always print it out and weigh it. I am specifically talking about the u-shaped cover, not the big flat ones. You could also reduce the thickness of the material on the 3d print because you are not locked into a preset part thickness because of the thickness of your weave.
And about the stiffness/strength of the arms you can always just do some material/force simulations. With how thick the aluminum arms are I'm pretty sure you are going to run into issues with the forces applied to your servos/motors/belts before you you run into any issues with the arms stiffness.
Your layup is wrong for torsion. Should be mostly 45°
And you made me laugh a lot. I work with infusion and hand laminating almost 20 years now, I could predict your sentences sometimes 😅
I def didn’t optimize my orientations. Honestly just trying to preserve materials while I learn the process.
@Cinema_Mechanics Start with Glas fiber, dirt cheap, and only about 20% worse mechanical at the same weight.
Also, it is a breeze to work with compared to carbon.
When you are looking for those stiffening parts, use IMS or UMS fibers, but make sure you are trained. If you thought carbon fiber was expensive, sit down before looking after UMS 🤣
Use glas for your first tries. They always fail. Also, your molds have unnecessary complications, those small indents around the bolts... no...
When you do infusion, use spray glue (3M 77) to fix the layers, use an 80gm glas layer top and bottom. That gets rid of pinhole ease.
And, if you have complex molds, always use 45deg and twill weave, linnen is for non convex or concave parts.
In the inner layers, you can also use non-woven fabrics, uni or bidirectional.
And the most important thing you didn't do...
Degass your epoxy in a vacuum chamber!!! 😇
Have fun!
Great tips! Appreciate any and all advice.
You know he is a "cinematographer" because of the vignetting...
The title said Ups and Downs...
I didn't see any Ups😂😭
It built… character?
❤
Why build the robot from scratch?
I think the best thing to do is buy commercial for example ABB robotic arm with broken/obsolete electronics, then you give it some refresh/cleanup and drive the OG motors and encoders with modern compact electronics.
I didn't look at prices but I would imagine you can buy $100k robotic arm that is old and broken for like $5k or less.
Regarding your first cf part attempt, you made a lot of mistakes totally on your part, I never made a single cf part and I could see that with this mold design the cf would get just shoved down.
Using quick cure epoxy in a got rooms is also an obvious mistake.
You also gone with wrong technology for this part, with shapes like that and need for smooth flat finish on both sides, you should've gone with forged cf using chopped/tow cf and like 6 part mold.