commenting again because finishing the surface before decking the mold is genius.. I have been CNC machining carbon molds for years and never thought of this. protecting that edge is painstaking!
Fantastic video, i've spent all day looking around for good solutions to making carbon fiber bend tubes and this is by far one of the cleanest and most efficient ways to achieve this. Thank you alot.
Very cool. Could you cut long thin strips and wrap it around the part at an angle as you wrap then the next layer goes the opposite way and they overlap in different angled directions, sort of similar to how you wrap your handlebar tape but keep laying in different directions?
This is a very innovative technique. In my industry it is often used for convoluted tubing molds so you are right on with the process. One method we also use here is salt core molding. We use a variety of salts to use as cores for molding, although may not expand like silicone, and we can just use water to dissolve the cores and simplify demolding. Tooling require is a bit more, but the initial cost is outweighed by the ease of demold. Very nice!
very nice jaz, and great idea . Many years ago I saw helicopter blade made with air bladders (about 4 bladers) rapted in carbon then inserted in a mold and inflated .
Thanks. I saw a company sell something similar. In that, they would make the insert for you with a material that will expand in heat and cool back to normal size. Looked them up and their price was ridiculous, like more than it would cost me to make a mold. So passed on the idea, glad to see this approach. I didn't bother to try figure out the exact material, but yeah looking back, silicone makes sense. Just get the one with the right thermal properties to handle an oven for carbon fiber.
Dave you are a remarkable engineer. Have you done an update/adjunct to this video? i'd be fascinated to see how you apply this technique to more complex shapes.
Thanks so much! I wish I had an update. We went on to use this technique at a previous employer to produce 4 different parts. I have a great deal more experience and knowledge with this technique now. I just don't have any footage or access to these materials to shoot a video now.
Degassing is a good practice but I can confirm that for this purpose you can get away without doing it. You can also use a shrinking release film like polyolefin if you want. That extends the mandrel life.
To help get the rubber out of the mold, use an air compressor blow guy to shoot air in between the rubber and your part. Don't forget to do both ends and the rubber will push out easily as you shoot air around it.
Cast a silicone rubber mandrel with tubing running through the middle of it. Make the tubing pass through the end caps. Pressurize the tubing to expand the silicone part in the mold. This way the silicone could be a lot smaller than the internal dimensions and even include some internal details that would otherwise lock the mandrel in the part. Air also expands when heated (Boyle's Law) so that gives some additional outward pressure on the part ensuring the carbon fiber layers are well compressed.
The passive pressurization from the expanded solid silicon mandrel is one of the main advantages of this method. Adding gas pressure adds significant complexity to the tooling and additional areas of part molding failure.
I'm right there with you. I made my first cold-air intake about 5-6 years ago by using this method. I was able to create all of the curves that I needed as a bonus. Have you seen the video where they use wood's metal as the mandrel?
Damn so many good tips. Building a sailboat mast section at the moment with two part mold and blender mold but have problems with the tight radius (r7)
Yeah, tight inside corners can be problematic. The composite fabric can "bridge" and lift away from the corner leaving ugly air bubbles and voids in the molded part.
@@designbydave in particular I have the problem during closing that I don't get the layed fibers to stay in place... The are creaping up into the joint.next try today. (Bladder mold)
This is great. If possible could you do a video about how you set up your stick on heaters and controllers? I think many people would like to see that. Thanks
Thanks for checking out the video. That would be great and I had considered going into those details for this video but it was already too long. Unfortunately I do not have access to these tools any more.
Really cool! I'm curious what those complications are with the bladder method and how this internal silicone mold gets around those complications. Have you considered using a higher COE material to provide more compaction (such as paraffin)?
Main problem with a bladder is that it is prone to leaking, which results in a failed part. You can also have compaction issues in "sharp" corners where the bladder has a hard time pressing into. The rubber has PLENTY of compaction. It's actually one of the issue with this method. Too much pressure requires VERY rigid tooling.
@@designbydave i like it.. the outer mould is cast or cnc?? maybe you could use that for the internal ,i.e. split the core at the kink. tap and dye - bolt together, unbolt release??
I wonder if it would work to make the first layer around the rubber out of carbon tape (spiral wrapped) since the interior finish is non-cosmetic. It would also increase strength a little in the other axis.
Pretty much yeah. I would end up making the first layer with 2x2 twill fiber. The twill weave has good compliance and the woven prepregs have a bit more resin which help it stick to the mandrel and also to itself. That allows you to wrap it tightly. Also helps to pre-warm the rubber mandrel.
This is a good, informative video. Thank you Dave for sharing! I read all the comments and wtf is up with these people that have to comment about the music? Amazing that people are so self-important and intolerant that they have to complain about how the music bothered them. They seem like completely miserable people. Also, I will 2nd the comment that McLube makes great release agents. We use it at our shop with great results, plus it's water-soluble.
I know its been a few months, but have you ever tried this with a couple layers of carbon sleeve over the mandrel, then wetted out and placed in the mold? This would make layup much easier and quicker, while giving a uniform pattern.
Woven carbon sleeve would indeed make the layup for this specific geometry much easier but that would require the use of wet resin, instead of prepreg. Part of the purpose of this project was to use prepreg as that's what we would use for future production parts.
also, if you cast your mandrel with a metal tube down the middle, you can pull out the tube to allow easier removal. add the tube before loading up the mandrel.
Tried that and it was the complete opposite of what you'd expect. There is too much friction to insert and remove a metal mandrel "core." Also, the presence of said core prevents the rubber from deforming under tension which makes removal much more difficult. The best solution was the use of a liquid release agent instead of the film shown.
Hi, thank you for this informative video ! I am buildind a folding bike with carbon fiber tube and and the torsional stiffness is nowhere near expectations. With your experience, how much better would a Pull Winded or Braided tube would perform in torsion compared to a Roll Wrapped ? Regards
Thanks for checking out the video. It shouldn't matter what manufacturing method you use as long as you have fibers in the +/-45 direction for tube torsional stiffness. Having said that, filament wound tubes are generally considered superior for torsional stiffness as the fibers are pretty much all going to laid at some angle. Keep in mind that the overall torsional stiffness of your structure isn't necessarily going to be dependent only on the torsional stiffness of your tubes.
Yep, tried that. It made the rubber significantly MORE difficult to remove because the internal solid part prevented the rubber from deforming under tension while extracting.
I had a few sources of prepreg over the years but this stuff was a sample from Rockwest composites. I believe the resin was Newport 301. Curing time depends on the temperature profile.
This seems like it would have some select ideal applications but just doing a regular vacuum bag seems easier overall and you would be able to be much more precise with the fiber layup and orientation
One of the main reasons for this technique is volume production (something not really covered in the video.) Vacuum bagging is very labor intensive and uses a lot of consumables for every part. And a part like this would use a bladder, quite the opposite of a vacuum, but similar principles. The fiber layup demonstrated here is terrible! lol! Literally my first time doing it with a part like this and I definitely wish I had some more recent footage of a proper layup to show that indeed, we can get some pretty intricate laminates together like this.
@@designbydave Oh I know the differences I make custom carbon fiber car parts for a living. I’ve been doing it for about 5 years or so now. All my stuff is vacuum bagged. I’m not disagreeing this is a useful tool at all. It’s something I’d never really thought about. There are 3-4 different viable ways to make a carbon tube. I could probably vacuum bag the same part you made in less time than it would take to do it with the silicone piece. But for mass production on a tube that didn’t need to hold air pressure the silicone might shave off a few minutes long term. The only thing I would suggest to you to do another test is that you should try the layup again. But this time instead of wrapping the UD carbon pre preg around the silicone piece, do all your fiber layups inside the two billet halves, put the silicone piece inside, then clamp the halves together. You’ll get much better compaction and continuity of the fiber matrix that way which will yield more uniform strength and lower weight. Watch the videos from Easy Composites about making a carbon fiber tube and the video where they make a mountain bike frame, you’ll be able to see what I’m talking about. If you stagger lap the carbon layers in the two billet halves together instead of wrapping the silicone piece in the UD then compressing it, you’ll get a much better finish and consistent part every pull you do. You’ll also end up with zero pinholes that way because the silicone piece will be able to serve its purpose more easily with less resistance. It will also be easier to remove that way after the part is cured
@@tylerroe5175 Something critical I didn't cover in this video - One of the main advantages (and purposes) of layup over a male mandrel is that you can wrap fiber continuously around the perimeter of the part. With a two part layup, into the female mold cavity, like you are referring to, there will always be a limiting week spot in your laminate which is the lap joint. The strength across that joint will only be the interlaminar strength of the resin. But like I said before, the main purpose of this project was to test and prove the manufacturing method. We would later refine the laminates and layup process. I wish I had footage of that because its much more impressive then the poor showing you see here!
@@designbydave Definitely understand the logic brother! There are trade-offs between each layup procedure. Benefit of continuous wrapping would be fiber continuity, down side would be equal compressive load during the cure cycle. Downside of lapping joints in the two halves would be fiber continuity/strength distribution, and benefits would be layup speed and more equal compression during the cure cycle. I’m gonna get my hands on some liquid silicone rubber and see if I can find a good application in my manufacturing process for what you’ve shown in the video. Cuz there is definitely a decent argument for it. I’d love to see more carbon videos from y’all in the future! Even though I’ve been manufacturing carbon fiber car parts for years now I’m always on TH-cam looking for improved and more advanced practices.
Vacuum bagging gives you 14.7 psi of compaction force max. This is not sufficient for the elimatiin of porosity and/ or voids throughout the laminate. Trapped rubber acheives pressures much higher than even an autoclave. Once the ends of that tool are capped the pressure exerted on the laminate would be incredible. No porosity and an incredibly low void content.
David, Great video. Fair to say the silicone could have been a bit smaller? Mold closure was a bit tough, you probably used an offset equal to the desired part wall. I'm curious if the production version used more offset and by how much more than tube wall?
Maybe if you get silicon in-between the plies when assembling the layup. Otherwise there is no issue. Silicon rubber tooling is common place in aerospace composite parts manufacturing (for certain parts and processes)
I have been toying with the idea of making similar parts, but with a water soluble solid plug, think of plaster for a cast, but can be dissolved. Same basic process, one mold for the plug, and one for the part. Thoughts?
Yep that's the "hybrid" approach. Both methods have their advantages. A solid rubber mandrel makes for passive compaction and semi-rigid male form to layup on. The main downside is that you have no, or very little control over the pressure. A bladder would allow control over pressure but much more complex tooling.
Hi David, I just found your channel with this video. I see you're working with the prepreg barehanded. Isn't the prepreg toxic? I'm surprised you're not wearing gloves while working with it.
Prepreg resin is just a special kind of epoxy. Probably toxic of you injest it but certainly harmless in small dose contact with the skin. However, after this video I would end up always wearing gloves, but that is to protect the laminate from contamination from the oils in your skin as much as anything.
Yes, Epoxy is toxic when it’s uncured and in rubbery state - prepreg included. Continued exposure (many times) of uncured epoxy will likely cause allergic skin reaction. Always, always use glove when handling uncured epoxies (or other thermosets).
Why didnt you vacuum the mold when you were making the Silicone insert? Seems that would have removed all of the air from the cavities of the mold. Pretty Cool videos, thank you.
I suppose with this mold, since it's small enough that would be a possibility, but it wasn't necessary. The casting process does introduce air bubbles that show up on larger parts but those never presented a major problem. In order to be able to pull a vacuum on a larger mold you would need either a very large chamber (with powerful pump) or a sealed mold which would greatly complicate the tool design.
Yes, but depending on the specific geometry of the bike handle bar (are we talking strait bars, drop bars, etc) the tooling would likely be much more complex.
Camera hides such imperfections ; ) I have a few years of experience now fabricating parts with this technique and it's expected for there to be at least a few pin holes. Experimenting with the resin cure cycle and pulling vacuum on the molds during cure may be what is necessary to completely eliminate such surface imperfections.
You want to cast a length of stainless steel cord into the rubber mandrel, with a little metal tab here and there, that will make it much easier to pull out.
FYI. This precise system was in use in European Formula one teams in 2004 ! I know because I was Laminating the parts. The only difference is that the mould tool was cured in an autoclave
Yes, it has been around for a while and the title is admittedly "clickbait-ie" but that's how you have to work with the TH-cam algorithm these days. The original technical paper I used for reference for this was from 1980.
Very nice but i would recommend adding 2 dowel pins in the moulds so that the two halfes are getting centered. Your part is by the looks of it only centered by the part itself. Other than that great video.
couldn't you just bolt the parts together and then use a flex ball hone to polish the insides of both halves together? or would it not make the curve very well?
I've had the idea of using styrofoam as a mandrel and then pouring gasoline into the finished part to dissolve the styrofoam - I just don't know how gasoline might affect the resin.
Some form of sacrificial mandrel can work but you still would need some form of compaction (mold or shrink tape etc.) I would probably avoid gasoline as it is quite toxic and corrosive
Not necessarily. A gas pressure form of compaction would probably be the ideal process but it comes at the cost of significantly more complex tooling and processing. It is also subject to failure where the entire laminate is ruined (a leaky bladder.) The main advantage with this method is the pressure and compaction is completely passive. The expansion of the rubber mandrel happens as the tool is heated to cure the prepreg resin. No need for a bladder and external pressurization system (air compressor.)
You are pretty much correct but a better term would be bond. Silicon will only really bond to its self but it can stick REALLY good to something like this making it very difficult to remove.
When used in this way- or pretty much any way - micro pores/cracks develop in the silicon surface, retaining particles or filling with resin on a microscopic level. For this reason, a release is used to extend the life of silicon mold and ease part release.
Hi David, what rubber did you use? Do you know the CTE? What tolerance did you run from the final carbon thickness to the rubber OD? Id be really interested in knowing any of these numbers ! Great content. Thanks for putting cool things on the internet.
Unfortunately no, I don't have any footage of the production parts I made. The process is basically the same though. Mostly just a more refined technique for cutting the prepreg (CNC cutting machine) and the layup process.
Surgical Rubber Tubing that starts out with a nominal 1/4-inch inside diameter can be inflated with air or preferably with water to about the size of that solid rubber mandrel, and possibly exert as much or more outward pressure during the hot molding phase. It might be more convenient, and *certainly easier to remove.* Whether it can withstand the temperature of the hot mold would have to be considered, but other types of rubber tubing might be used instead of the natural Surgical Rubber. If doing cold molding, there will obviously be no heat issue. Finally, the exact chemistry of the adhesive will need to be checked for compatibility with the expandable tubing. It might also be possible to simplify the entire process by wrapping the fibers around the inflated tubing while it is straight, then forcing it into a gentle curve for the desired angle at the ends and held that way during curing. A gentle curve is stronger, too.
You are basically just describing bladder molding. This process has several advantages. Passive compaction - mandrel expands with the heating of the tool. The air tight seals, valves and compressor required for a bladder add considerable complexity to the tooling and many sources for failure, ruining the part. Reusable tooling - The mandrel can be reused up to 40 or so times. Bladders are typically one time use. Lastly, the male mandrel allows you to assemble the laminate by wrapping fiber over the mandrel, allowing for continuous fibers across the mold parting line. Also, the rubber makes for SIGNIFICANTLY more pressure then a bladder. We are talking several hundred PSI versus ~120 (typical air compressor.) Having a strait mandrel, wrapping fibers and then bending the assembled laminate to fit into the tool is an idea worth investigating. I just never got around to that.
Okay, yes, a hose that can expand a lot is kinda like a bladder. I just did not think of it quite that way since it expands so much more than 'normal' hoses. As for the kind of pressure which can be exerted, a company where I used to work pumped water into 5/8-inch copper tubes with the aid of a small pneumatic reciprocating pump, and *inflated the copper tubing,* some of which was very strong stuff... hard copper, even cupronickel... not the coiled stuff that is soft. The air supplied to the pump was the same compressed air used by normal pneumatic tools throughout the shop, and the gauge on the pump went up to 10,000 PSI... higher than most hydraulic pumps. So if it is high pressure you want . . . ! Also, you get enough pressure from water freezing to crack outdoor stacks of cast iron cannonballs into pieces just by flooding them with water before winter set in, a method formerly used to dispose of excess ammunition, in the days of black powder cannons. So if your laminate was able to cure in temperatures a bit below 0°C, possibly as low as -40, you could indeed lay up the fibers around a piece of water-filled rubber hose, stick that inside your mold, then freeze the whole thing to get the desired high pressure. You might even be able to skip the end caps of the mold, because the ends of the hose will tend to freeze first, plugging the ends with ice.
I've tried that on a number of different tools without much help. Perhaps if a specific "air channel" was designed into the tooling for this purpose it might work better.
Do you mean a pre-made, fully cured tube? That is not really possible. If you re-heat the matrix to the point that it becomes soft enough to bend, the carbon fibers then become loose and will seperste from the matrix and/or become distorted in such a way that they lose significant strength. Also need to consider the effect that heating and cooling the matrix will have on its structural properties. Having said all that, I seen to remember some manufacture doing the for some of their parts however the bend radius was greater then what you see in this video.
gives me hope for the future... I actually have most of the stuff sitting around here in boxes.. waiting for my health to return... I'm working with an older form of carbon fiber... hemp fiber and silk... and a bioplastic that is tremendously rigid and hard... so .. I'm going to make personal size dirigibles... go up and float... quietly.. all day... solar powered... cheap... and.. they won't fall out of the sky... they fall ^up^ ... slowly
I used to work for a Boeing company doing factory floor automation making the high tech fiber stuff and the aluminum stuff.. for the airframe and powerplant structure... but this is actually an ancient Chinese recipe... I have a 40cm hemp rope.. 13mm diameter... with my bioplastic... weighs a few grams... it's like a piece of rebar at the moment.. slightly flexible if I use a lot of strength... TTD testing will give me shear strength etc... lighter than wood and aluminum... machinable.. ✌️ I'll keep you posted
There is no need to degas the silicone before casting the insert because of the trapped air in the silicone will act like pressure intensifier when you heat the mold. But if you really need degassed silicone it is a good idea to put the whole cast thing in the vacuum chamber again one more time after pouring the silicone. There is no need to use release film on silicone or even release agent, epoxy does not adhere to silicone.
You are basically wrong about everything here. Degassing - The air bubbles don't work as you expect. Excess resin will be squeezed into any bubbles near the surface of the mandrel. This creates sharp, tooth like protrusions of resin on the inside of the part. A non-degassed silicon part will have A LOT of air bubbles. It will look almost sponge-like. Degassing after casting would be ideal yes as the casting process can create bubbles. This gets tricky with large tools though. A release agent is absolutely essential. The epoxy indeed does not bond to the silicon, but it sure sticks like hell, especially when a vacuum is formed during the molding process. Also, the epoxy will, with repeated use, start to tear up the silicon.
Yes, this is not my original idea and its been a standard practice in the composites industry for decades. The title is a bit "click bait-ey" but that's how the TH-cam algorithm works these days. Sorry if that wasn't clear in the video. This is a somewhat novel technique though and there is very little info about it out there, especially on TH-cam..
It would be cool too have a element heater work surface, or even an oil heater style work surface. EG. Leave your aluminium stock on the surface for 1/2 hour or so… good stable success promoting temperature… hahaha… especially in a cold environment. 🇦🇺🤜🏼🤛🏼🍀🍀🍀😂😎🤓🤓🤓
also, free tip. You have voids because there is nothing holding the rubber inside, you should put lids on each end f the mold to in fact Trap the rubber. You see rubber will try to stretch out of the chamber and in the process become "Narrow" that's when you loose pressure, if you trap it, all the pressure stays inside and pushes on your part better. Path of least resistance etc.. Free Tip!
LOL, silicone rubber core molding has been used for over 50 years to make all sorts of shapes including tubes. core pressure is tuned by air gaps so the CTE mismatch creates the desired pressure in the closed mold. Maybe you should read some books on composite fabrication before saying you invented it.
commenting again because finishing the surface before decking the mold is genius.. I have been CNC machining carbon molds for years and never thought of this. protecting that edge is painstaking!
Yeah the parting line was a big issue for some some of the parts my former company produced so we needed a way to improve that. This was my solution.
Fantastic video, i've spent all day looking around for good solutions to making carbon fiber bend tubes and this is by far one of the cleanest and most efficient ways to achieve this. Thank you alot.
Glad it helped!
Fascinating process, simple solution but creative way to exploit material properties. The way the scallop edge is dealt with is neat.
Thanks for sharing the successes and failures during the process. Nice work!
Thanks for checking it out
Very cool. Could you cut long thin strips and wrap it around the part at an angle as you wrap then the next layer goes the opposite way and they overlap in different angled directions, sort of similar to how you wrap your handlebar tape but keep laying in different directions?
how do you have less than 10k subscribers?! this is great and you extremely knowledgable. thanks for sharing!
Thanks for watching
This is a very innovative technique. In my industry it is often used for convoluted tubing molds so you are right on with the process. One method we also use here is salt core molding. We use a variety of salts to use as cores for molding, although may not expand like silicone, and we can just use water to dissolve the cores and simplify demolding. Tooling require is a bit more, but the initial cost is outweighed by the ease of demold. Very nice!
Salt core sounds like a cool idea!
Great video, thanks for sharing. This is my all time favorite process.
Thanks for checking it out
very nice jaz, and great idea . Many years ago I saw helicopter blade made with air bladders (about 4 bladers) rapted in carbon then inserted in a mold and inflated .
Yes, bladder molding is the typical, industry standard of making tubular parts like this.
Cool stuff. Thanks for sharing.
Thanks for watching!
I love the 11 minute camera angle. Awesome Techneque 🇦🇺🤜🏼🤛🏼🍀🍀🍀😎🤓🤓🤓
Thanks
Liked subbed. Need more ppl like mr Aldrich sharing their hacks on yt and elsewhere. Thx DA this one is a keeper.
Glad you like it and thanks for checking it out!
Thanks. I saw a company sell something similar. In that, they would make the insert for you with a material that will expand in heat and cool back to normal size. Looked them up and their price was ridiculous, like more than it would cost me to make a mold. So passed on the idea, glad to see this approach. I didn't bother to try figure out the exact material, but yeah looking back, silicone makes sense. Just get the one with the right thermal properties to handle an oven for carbon fiber.
Yes I'm familiar with those tools. Look great but the cost reserves it for high end manufacturing only.
@@designbydave high end manufacturing is when its so dangerous and poisonous or whatever not just anyone should be doing it i think.
Thanks for the great content.
Thanks for checking it out.
Dave you are a remarkable engineer. Have you done an update/adjunct to this video? i'd be fascinated to see how you apply this technique to more complex shapes.
Thanks so much!
I wish I had an update. We went on to use this technique at a previous employer to produce 4 different parts. I have a great deal more experience and knowledge with this technique now. I just don't have any footage or access to these materials to shoot a video now.
Degassing is a good practice but I can confirm that for this purpose you can get away without doing it.
You can also use a shrinking release film like polyolefin if you want. That extends the mandrel life.
To help get the rubber out of the mold, use an air compressor blow guy to shoot air in between the rubber and your part. Don't forget to do both ends and the rubber will push out easily as you shoot air around it.
In my experience molding several different parts after this video, blowing air into the mold actually is not effective at all.
Water and air works better
Hang Glider Pilots are almost always, very cool and interesting people. And you prove that paradigm in several ways. Stay Awesome Brother 😎🎩♠️
Thanks so much!
Very cool idea. I may give this a try 😁
Thanks for checking it out
Very nice. Earned yourself a sub.
Looking forward to seeing more turorials (sic) as per your channel cover pic.
Thanks for checking it ou
Eeeeeeeexcelent, David.
Thanks!
Your technical vernacular is as admirable as your vocabulary is comprehensive 🎩🎩♠️
That is very kind of you to say.
Cast a silicone rubber mandrel with tubing running through the middle of it. Make the tubing pass through the end caps. Pressurize the tubing to expand the silicone part in the mold. This way the silicone could be a lot smaller than the internal dimensions and even include some internal details that would otherwise lock the mandrel in the part. Air also expands when heated (Boyle's Law) so that gives some additional outward pressure on the part ensuring the carbon fiber layers are well compressed.
The passive pressurization from the expanded solid silicon mandrel is one of the main advantages of this method. Adding gas pressure adds significant complexity to the tooling and additional areas of part molding failure.
Interesting to hear that you developed the method when it's been used by several suppliers of aerospace components for years.
Ok
I'm right there with you. I made my first cold-air intake about 5-6 years ago by using this method. I was able to create all of the curves that I needed as a bonus. Have you seen the video where they use wood's metal as the mandrel?
I haven't seen that video
Damn so many good tips. Building a sailboat mast section at the moment with two part mold and blender mold but have problems with the tight radius (r7)
Yeah, tight inside corners can be problematic. The composite fabric can "bridge" and lift away from the corner leaving ugly air bubbles and voids in the molded part.
@@designbydave in particular I have the problem during closing that I don't get the layed fibers to stay in place... The are creaping up into the joint.next try today. (Bladder mold)
This is great. If possible could you do a video about how you set up your stick on heaters and controllers? I think many people would like to see that. Thanks
Thanks for checking out the video. That would be great and I had considered going into those details for this video but it was already too long. Unfortunately I do not have access to these tools any more.
Awesome man, thank you!!
Thanks for checking it out
very very cool!
Thanks!
Really cool! I'm curious what those complications are with the bladder method and how this internal silicone mold gets around those complications. Have you considered using a higher COE material to provide more compaction (such as paraffin)?
Main problem with a bladder is that it is prone to leaking, which results in a failed part. You can also have compaction issues in "sharp" corners where the bladder has a hard time pressing into.
The rubber has PLENTY of compaction. It's actually one of the issue with this method. Too much pressure requires VERY rigid tooling.
@@designbydave i like it.. the outer mould is cast or cnc?? maybe you could use that for the internal ,i.e. split the core at the kink. tap and dye - bolt together, unbolt release??
Cool.
Does anyone know where to buy CMC (carbon fiber reinforced matrix) used in aerospace (like a rocket's nozzle)?
God bless.
Fantastic, can you share what are the heating elements you use to get and control the mould temperature ? It would be really helpful 👍
They are adhesive backed, silicon sheet heaters. Available from many sources, just Google them.
Hi David; thanks for the full process video! What liquid release agent did you move to to replace the release film?
I can't recall the exact model number but it was a formula specifically for releasing from rubber made by Chemtrend
@@designbydave Try McLube!
I wonder if it would work to make the first layer around the rubber out of carbon tape (spiral wrapped) since the interior finish is non-cosmetic. It would also increase strength a little in the other axis.
Pretty much yeah. I would end up making the first layer with 2x2 twill fiber. The twill weave has good compliance and the woven prepregs have a bit more resin which help it stick to the mandrel and also to itself. That allows you to wrap it tightly. Also helps to pre-warm the rubber mandrel.
This is a good, informative video. Thank you Dave for sharing! I read all the comments and wtf is up with these people that have to comment about the music? Amazing that people are so self-important and intolerant that they have to complain about how the music bothered them. They seem like completely miserable people.
Also, I will 2nd the comment that McLube makes great release agents. We use it at our shop with great results, plus it's water-soluble.
Thanks glad you enjoyed it. Always people will be complaining about music, but I could care less.
I know its been a few months, but have you ever tried this with a couple layers of carbon sleeve over the mandrel, then wetted out and placed in the mold?
This would make layup much easier and quicker, while giving a uniform pattern.
Woven carbon sleeve would indeed make the layup for this specific geometry much easier but that would require the use of wet resin, instead of prepreg. Part of the purpose of this project was to use prepreg as that's what we would use for future production parts.
Hi David, thanks for sharing the process! May I ask which sheet wax you used for the offset?
It's from Freeman Supply Company. I think Easy Composites in the UK also sells a similar product
New sub here, thnks for sharing your work.
Thanks for checking it out!
also, if you cast your mandrel with a metal tube down the middle, you can pull out the tube to allow easier removal. add the tube before loading up the mandrel.
Tried that and it was the complete opposite of what you'd expect. There is too much friction to insert and remove a metal mandrel "core." Also, the presence of said core prevents the rubber from deforming under tension which makes removal much more difficult. The best solution was the use of a liquid release agent instead of the film shown.
Hi, thank you for this informative video ! I am buildind a folding bike with carbon fiber tube and and the torsional stiffness is nowhere near expectations. With your experience, how much better would a Pull Winded or Braided tube would perform in torsion compared to a Roll Wrapped ? Regards
Thanks for checking out the video. It shouldn't matter what manufacturing method you use as long as you have fibers in the +/-45 direction for tube torsional stiffness. Having said that, filament wound tubes are generally considered superior for torsional stiffness as the fibers are pretty much all going to laid at some angle. Keep in mind that the overall torsional stiffness of your structure isn't necessarily going to be dependent only on the torsional stiffness of your tubes.
I can see casting the silicon over an armature for support while carbon wrapping and to aid extraction from the part.
Yep, tried that. It made the rubber significantly MORE difficult to remove because the internal solid part prevented the rubber from deforming under tension while extracting.
Dave do you use the heating elements on larger parts as well? Or do you have to resort to an oven of some sort? Awesome video!!!!!
Yes, just larger heating elements
Hey Great Video! Which kind of Prepreg do you use? It seems to cure much faster?
I had a few sources of prepreg over the years but this stuff was a sample from Rockwest composites. I believe the resin was Newport 301. Curing time depends on the temperature profile.
This seems like it would have some select ideal applications but just doing a regular vacuum bag seems easier overall and you would be able to be much more precise with the fiber layup and orientation
One of the main reasons for this technique is volume production (something not really covered in the video.) Vacuum bagging is very labor intensive and uses a lot of consumables for every part. And a part like this would use a bladder, quite the opposite of a vacuum, but similar principles.
The fiber layup demonstrated here is terrible! lol! Literally my first time doing it with a part like this and I definitely wish I had some more recent footage of a proper layup to show that indeed, we can get some pretty intricate laminates together like this.
@@designbydave Oh I know the differences I make custom carbon fiber car parts for a living. I’ve been doing it for about 5 years or so now. All my stuff is vacuum bagged.
I’m not disagreeing this is a useful tool at all. It’s something I’d never really thought about. There are 3-4 different viable ways to make a carbon tube. I could probably vacuum bag the same part you made in less time than it would take to do it with the silicone piece. But for mass production on a tube that didn’t need to hold air pressure the silicone might shave off a few minutes long term.
The only thing I would suggest to you to do another test is that you should try the layup again. But this time instead of wrapping the UD carbon pre preg around the silicone piece, do all your fiber layups inside the two billet halves, put the silicone piece inside, then clamp the halves together. You’ll get much better compaction and continuity of the fiber matrix that way which will yield more uniform strength and lower weight.
Watch the videos from Easy Composites about making a carbon fiber tube and the video where they make a mountain bike frame, you’ll be able to see what I’m talking about. If you stagger lap the carbon layers in the two billet halves together instead of wrapping the silicone piece in the UD then compressing it, you’ll get a much better finish and consistent part every pull you do. You’ll also end up with zero pinholes that way because the silicone piece will be able to serve its purpose more easily with less resistance. It will also be easier to remove that way after the part is cured
@@tylerroe5175 Something critical I didn't cover in this video - One of the main advantages (and purposes) of layup over a male mandrel is that you can wrap fiber continuously around the perimeter of the part. With a two part layup, into the female mold cavity, like you are referring to, there will always be a limiting week spot in your laminate which is the lap joint. The strength across that joint will only be the interlaminar strength of the resin.
But like I said before, the main purpose of this project was to test and prove the manufacturing method. We would later refine the laminates and layup process. I wish I had footage of that because its much more impressive then the poor showing you see here!
@@designbydave Definitely understand the logic brother! There are trade-offs between each layup procedure. Benefit of continuous wrapping would be fiber continuity, down side would be equal compressive load during the cure cycle. Downside of lapping joints in the two halves would be fiber continuity/strength distribution, and benefits would be layup speed and more equal compression during the cure cycle. I’m gonna get my hands on some liquid silicone rubber and see if I can find a good application in my manufacturing process for what you’ve shown in the video. Cuz there is definitely a decent argument for it.
I’d love to see more carbon videos from y’all in the future! Even though I’ve been manufacturing carbon fiber car parts for years now I’m always on TH-cam looking for improved and more advanced practices.
Vacuum bagging gives you 14.7 psi of compaction force max. This is not sufficient for the elimatiin of porosity and/ or voids throughout the laminate. Trapped rubber acheives pressures much higher than even an autoclave. Once the ends of that tool are capped the pressure exerted on the laminate would be incredible. No porosity and an incredibly low void content.
What do you use for your heating elements?
Adhesive backed silicon sheet heaters
David, Great video. Fair to say the silicone could have been a bit smaller? Mold closure was a bit tough, you probably used an offset equal to the desired part wall. I'm curious if the production version used more offset and by how much more than tube wall?
Fine tuning the silicon mandrel size is a very tricky task that requires some level of trial and error.
@@designbydave We use to have a saying in my way back tech days, "That's proprietary! I can't tell you that!"
Put a hole down the middle of the rubber and blow it up like a ballon with compressed air during curing it will be super easy to take out
One of the main points of a trapped rubber tool is not needing to use air pressure which adds significant complexity to the tooling and process.
Presence of silicone may cause delamination in composites!
AFAIK such methods are not allowed for structural parts in the aircraft industry.
Maybe if you get silicon in-between the plies when assembling the layup. Otherwise there is no issue. Silicon rubber tooling is common place in aerospace composite parts manufacturing (for certain parts and processes)
I have been toying with the idea of making similar parts, but with a water soluble solid plug, think of plaster for a cast, but can be dissolved. Same basic process, one mold for the plug, and one for the part.
Thoughts?
Oh my dude, I did exactly that. Check my channel.
@@designbydave Awesome will do! Keep at it, BTW both monkeys and footballs need love to... 🤣
I still have problems with fusion cam but it's so strong
I’m curious to know what you’re using as your liquid release agent. I’ve been looking for a good liquid release agent.
For the rubber or the mold surface? Both are from Chemlease, but I can't recall the exact model numbers.
@@designbydave The release that you used instead of release film.
Very cool what product did you use to makec silicone rubber
V-340 from Freeman Supply Company
Where do you purchase the sheet wax? Awesome video💪🏼
Freeman supply company. Thanks for watching!
can we do this at home?
Suggestion you should have made a soft hollow plug thear holds that shape... ie a silicone bladder mold...
Yep that's the "hybrid" approach. Both methods have their advantages. A solid rubber mandrel makes for passive compaction and semi-rigid male form to layup on. The main downside is that you have no, or very little control over the pressure. A bladder would allow control over pressure but much more complex tooling.
Olá David. Como é feito quando o tubo tem mais que uma curva ?
Hi David, I just found your channel with this video. I see you're working with the prepreg barehanded. Isn't the prepreg toxic? I'm surprised you're not wearing gloves while working with it.
Prepreg resin is just a special kind of epoxy. Probably toxic of you injest it but certainly harmless in small dose contact with the skin. However, after this video I would end up always wearing gloves, but that is to protect the laminate from contamination from the oils in your skin as much as anything.
Yes, Epoxy is toxic when it’s uncured and in rubbery state - prepreg included. Continued exposure (many times) of uncured epoxy will likely cause allergic skin reaction. Always, always use glove when handling uncured epoxies (or other thermosets).
@@designbydave Also, dry cutting and sanding of carbon fibres shouldn't been done without a proper respirator. A simple dust mask won't cut it.
Why didnt you vacuum the mold when you were making the Silicone insert? Seems that would have removed all of the air from the cavities of the mold. Pretty Cool videos, thank you.
I suppose with this mold, since it's small enough that would be a possibility, but it wasn't necessary. The casting process does introduce air bubbles that show up on larger parts but those never presented a major problem. In order to be able to pull a vacuum on a larger mold you would need either a very large chamber (with powerful pump) or a sealed mold which would greatly complicate the tool design.
Can you do multiple angles and vary clocking to make 3D tube? I think I know how it can be done.
Multiple angles, yes to a certain degree at least. Need to be able to extract the rubber mandrel.
Not sure what you mean by "vary clocking"
Do you think this process could be used to fabricate a bike handlebar?
Yes, but depending on the specific geometry of the bike handle bar (are we talking strait bars, drop bars, etc) the tooling would likely be much more complex.
What type of Wax sheets did you use? It looked thin but when you held it up to the camera it looked to be thick.
They are sold by the Freeman supply company. Available in many thicknesses
Hello Sir where did get that thin sheet of wax also what was the final
weight thank you?
Freeman supply company.
Don't recall the final weight
Hi Dave, what is the wax sheets that you use and do you have a link to where to buy it ?
Freeman supply company
How did u manage not to get any pin holes even without a vacuum bag to contain the resin in the fabric?
Camera hides such imperfections ; ) I have a few years of experience now fabricating parts with this technique and it's expected for there to be at least a few pin holes. Experimenting with the resin cure cycle and pulling vacuum on the molds during cure may be what is necessary to completely eliminate such surface imperfections.
Would you be able to share what silicon you’ve used and what is the thermal expansion quorficient of the silicone you’ve used?
Sorry, I don't have that info any more. @@robertocoelho441
You want to cast a length of stainless steel cord into the rubber mandrel, with a little metal tab here and there, that will make it much easier to pull out.
Tried that and its the exact opposite of what you expect. The solid metal core prevents the rubber from deforming (as much) while under tension.
FYI. This precise system was in use in European Formula one teams in 2004 ! I know because I was Laminating the parts. The only difference is that the mould tool was cured in an autoclave
Yes, it has been around for a while and the title is admittedly "clickbait-ie" but that's how you have to work with the TH-cam algorithm these days. The original technical paper I used for reference for this was from 1980.
Very nice but i would recommend adding 2 dowel pins in the moulds so that the two halfes are getting centered. Your part is by the looks of it only centered by the part itself. Other than that great video.
Yes, alignment pins are necessary and would be added to later, production molds.
If i want to order something of my desire somethign simple do you take orders ?
I don't think that's something I can do. Thanks for checking out the video.
couldn't you just bolt the parts together and then use a flex ball hone to polish the insides of both halves together? or would it not make the curve very well?
Are you referring to the mold halves? I'm not familiar with a "flex ball hone"
And they are available in a wide variety of grits, sizes, abrasive compound, etc. These are just an example.
I want you to build a mold like that for me how can I get in touch with you
I've had the idea of using styrofoam as a mandrel and then pouring gasoline into the finished part to dissolve the styrofoam - I just don't know how gasoline might affect the resin.
Some form of sacrificial mandrel can work but you still would need some form of compaction (mold or shrink tape etc.)
I would probably avoid gasoline as it is quite toxic and corrosive
This video was really cool, and it could be titled "why carbon fiber parts are expensive!" Haha
Exactly. Manufacturing with composite materials is still quite expensive
again would an inflatable liner still not be ideal?
Not necessarily. A gas pressure form of compaction would probably be the ideal process but it comes at the cost of significantly more complex tooling and processing. It is also subject to failure where the entire laminate is ruined (a leaky bladder.) The main advantage with this method is the pressure and compaction is completely passive. The expansion of the rubber mandrel happens as the tool is heated to cure the prepreg resin. No need for a bladder and external pressurization system (air compressor.)
also.. that thermal sheet is rather fancy... whats that called please..
sheetwax
Great effort but why not use trapped air rubber? I have done exactly the same thing with inflating a bicycle inner tube.
A solid mandrel offers several advantages over a bladder. I've addresses this several times in other comments.
I thought silicone did not stick to anything but it self. So why the release agent?
You are pretty much correct but a better term would be bond. Silicon will only really bond to its self but it can stick REALLY good to something like this making it very difficult to remove.
When used in this way- or pretty much any way - micro pores/cracks develop in the silicon surface, retaining particles or filling with resin on a microscopic level. For this reason, a release is used to extend the life of silicon mold and ease part release.
Hi David, what rubber did you use? Do you know the CTE? What tolerance did you run from the final carbon thickness to the rubber OD? Id be really interested in knowing any of these numbers ! Great content. Thanks for putting cool things on the internet.
I used Bluesil V-340 from Freeman Supply company. 45 Shore A hardness, CTE= 2.5X10-4 in/in/degree C. All tech specs on the Freeman supply web site
Have a link to the more updated process?
Unfortunately no, I don't have any footage of the production parts I made. The process is basically the same though. Mostly just a more refined technique for cutting the prepreg (CNC cutting machine) and the layup process.
Surgical Rubber Tubing that starts out with a nominal 1/4-inch inside diameter can be inflated with air or preferably with water to about the size of that solid rubber mandrel, and possibly exert as much or more outward pressure during the hot molding phase. It might be more convenient, and *certainly easier to remove.* Whether it can withstand the temperature of the hot mold would have to be considered, but other types of rubber tubing might be used instead of the natural Surgical Rubber. If doing cold molding, there will obviously be no heat issue. Finally, the exact chemistry of the adhesive will need to be checked for compatibility with the expandable tubing. It might also be possible to simplify the entire process by wrapping the fibers around the inflated tubing while it is straight, then forcing it into a gentle curve for the desired angle at the ends and held that way during curing. A gentle curve is stronger, too.
You are basically just describing bladder molding. This process has several advantages.
Passive compaction - mandrel expands with the heating of the tool. The air tight seals, valves and compressor required for a bladder add considerable complexity to the tooling and many sources for failure, ruining the part.
Reusable tooling - The mandrel can be reused up to 40 or so times. Bladders are typically one time use.
Lastly, the male mandrel allows you to assemble the laminate by wrapping fiber over the mandrel, allowing for continuous fibers across the mold parting line.
Also, the rubber makes for SIGNIFICANTLY more pressure then a bladder. We are talking several hundred PSI versus ~120 (typical air compressor.)
Having a strait mandrel, wrapping fibers and then bending the assembled laminate to fit into the tool is an idea worth investigating. I just never got around to that.
Okay, yes, a hose that can expand a lot is kinda like a bladder. I just did not think of it quite that way since it expands so much more than 'normal' hoses. As for the kind of pressure which can be exerted, a company where I used to work pumped water into 5/8-inch copper tubes with the aid of a small pneumatic reciprocating pump, and *inflated the copper tubing,* some of which was very strong stuff... hard copper, even cupronickel... not the coiled stuff that is soft. The air supplied to the pump was the same compressed air used by normal pneumatic tools throughout the shop, and the gauge on the pump went up to 10,000 PSI... higher than most hydraulic pumps. So if it is high pressure you want . . . !
Also, you get enough pressure from water freezing to crack outdoor stacks of cast iron cannonballs into pieces just by flooding them with water before winter set in, a method formerly used to dispose of excess ammunition, in the days of black powder cannons. So if your laminate was able to cure in temperatures a bit below 0°C, possibly as low as -40, you could indeed lay up the fibers around a piece of water-filled rubber hose, stick that inside your mold, then freeze the whole thing to get the desired high pressure. You might even be able to skip the end caps of the mold, because the ends of the hose will tend to freeze first, plugging the ends with ice.
Using pressured air might ease the removal just like they do with the handlebars of the motorcycles
I've tried that on a number of different tools without much help. Perhaps if a specific "air channel" was designed into the tooling for this purpose it might work better.
most people in germany they use silicone balloon tube for casting
Thanks for checking out the video. Lots of reasons to not use a gas pressurized bladder which I've already commented on.
what type of silicone should i use?
V-340 from Freeman Supply Company
I wonder if a sealed plastic bag filled with expanding foam would work?
An interesting idea. Would need to investigate how much pressure that foam expansion can provide.
Hi, can you tell me which liquid rubber did u use?
V-340 from Freeman Supply Company
can a rollcage be made this way ?
Yes
Have you ever tried bending carbon fibre tube by filling with sand and heating and bending? 👍
Do you mean a pre-made, fully cured tube? That is not really possible. If you re-heat the matrix to the point that it becomes soft enough to bend, the carbon fibers then become loose and will seperste from the matrix and/or become distorted in such a way that they lose significant strength. Also need to consider the effect that heating and cooling the matrix will have on its structural properties.
Having said all that, I seen to remember some manufacture doing the for some of their parts however the bend radius was greater then what you see in this video.
You work with carbon without mask. You are crazy.
I definitely am wearing a respirator when cutting it with the saw. No need any of the other times.
this looks like a good one, bet this pipe could take a beating.
Thanks, it's pretty strong yes.
way cool shop, bro... can I come over and play at your house please
gives me hope for the future... I actually have most of the stuff sitting around here in boxes.. waiting for my health to return...
I'm working with an older form of carbon fiber... hemp fiber and silk... and a bioplastic that is tremendously rigid and hard... so
.. I'm going to make personal size dirigibles... go up and float... quietly.. all day... solar powered... cheap... and.. they won't fall out of the sky... they fall ^up^ ... slowly
I used to work for a Boeing company doing factory floor automation making the high tech fiber stuff and the aluminum stuff.. for the airframe and powerplant structure... but this is actually an ancient Chinese recipe... I have a 40cm hemp rope.. 13mm diameter... with my bioplastic... weighs a few grams... it's like a piece of rebar at the moment.. slightly flexible if I use a lot of strength... TTD testing will give me shear strength etc... lighter than wood and aluminum... machinable.. ✌️
I'll keep you posted
Thanks for checking it out. Sadly this shop no longer exists.
There is no need to degas the silicone before casting the insert because of the trapped air in the silicone will act like pressure intensifier when you heat the mold. But if you really need degassed silicone it is a good idea to put the whole cast thing in the vacuum chamber again one more time after pouring the silicone. There is no need to use release film on silicone or even release agent, epoxy does not adhere to silicone.
You are basically wrong about everything here.
Degassing - The air bubbles don't work as you expect. Excess resin will be squeezed into any bubbles near the surface of the mandrel. This creates sharp, tooth like protrusions of resin on the inside of the part. A non-degassed silicon part will have A LOT of air bubbles. It will look almost sponge-like. Degassing after casting would be ideal yes as the casting process can create bubbles. This gets tricky with large tools though.
A release agent is absolutely essential. The epoxy indeed does not bond to the silicon, but it sure sticks like hell, especially when a vacuum is formed during the molding process. Also, the epoxy will, with repeated use, start to tear up the silicon.
Anyone know the music about 13 mins in?
It's just some generic royalty free music. Can't recall the source now.
Maybe some pressurized air or liquid can help with removing the rubber mandrel
I tried that but didn't seem to help. The liquid release agent I would end up using later made the madrel much easier to remove.
Well your not first, BST in South Africa has done this for years..
Yes, this is not my original idea and its been a standard practice in the composites industry for decades. The title is a bit "click bait-ey" but that's how the TH-cam algorithm works these days. Sorry if that wasn't clear in the video. This is a somewhat novel technique though and there is very little info about it out there, especially on TH-cam..
It would be cool too have a element heater work surface, or even an oil heater style work surface.
EG. Leave your aluminium stock on the surface for 1/2 hour or so… good stable success promoting temperature… hahaha… especially in a cold environment.
🇦🇺🤜🏼🤛🏼🍀🍀🍀😂😎🤓🤓🤓
That would work but the solution I had here was perfectly adaquate.
And when you're done making comosites, you can give that core to your SO. ;)
Yes
You might be able to mask that edge with aluminium tape and shot blast the piece
Sounds like that might be worth a try
lol been doing it for 12+ years but yeah, new to some I guess.
also, free tip. You have voids because there is nothing holding the rubber inside, you should put lids on each end f the mold to in fact Trap the rubber. You see rubber will try to stretch out of the chamber and in the process become "Narrow" that's when you loose pressure, if you trap it, all the pressure stays inside and pushes on your part better. Path of least resistance etc.. Free Tip!
Alt title: "why are carbon fiber parts so expensive"
Yep. The material, the tooling and the labor, all quite expensive.
no links :(
What links are you looking for? Sorry might have forgot to include them and its been a while since I made the video.
LOL, silicone rubber core molding has been used for over 50 years to make all sorts of shapes including tubes. core pressure is tuned by air gaps so the CTE mismatch creates the desired pressure in the closed mold. Maybe you should read some books on composite fabrication before saying you invented it.
Thanks for checking out the video!