I'm sure you've seen it, but in case you haven't, Time bicycles makes frames with this process. Not sure if it would help you here, but instead of a 3d printed core and bladder, they mold a wax core for the inside of the part, and once the part is cured, they just melt the wax out.
I just seen a super neat trick where a guy wrapped his layup and inner core wih fishing line, then upon closing the mold (as much as possible), he snipped each loop of fishing line and pulled it out allowing it to expand into the opposite side mold. You could potentially wrap the whole mold and layup with fishing line which would move the fishing line to the outside for removal...
great video, working on something similar at the moment :) Here's an answer to your question for the "composites tape instead of blue tape" to stick the fiber to the inner core: Use some strips of carbonfiber UD, add some composites spray tack and use these as a tape to hold it to the 3d core (look for hi-tack 71 from 3M ) never used it from 3m so you might have to do some research (I'm using Fusion Fix from easycomposites) that way you can make a piece of carbonfiber tape similar to prepreg
I've tried that on a previous project and it didn't work out too well. The 3d printed soluble material we used wasn't quite as soluble you would want and removing the mandrel was very difficult.
Fascinating watch see the science and magic behind beautiful carbon parts. After watching this whole thing, I have only one question… will you be taking orders for this finished ‘product’?
Thanks for checking it out Ryan. Definitely will not be taking orders. I don't "do" production any more and these 3D printed molds aren't really durable enough to make more then a couple parts each anyhow. Also the finished product will be extremely cost prohibitive. Having said that. We have materials to make a set or two extra of the corner brackets.
there are very reliable plastic clamps that you can on your hoses. Also it's best practice not to use OIL based products on your dry cloth to mark your preform patterns. :)
I'd print that black part in TPU.. its like a hard rubber.. I think you could get a seal if you go the tolerances tight. It can print glass smooth even on my entry-level printer.
It's been a while since I did this project but I want to say that I had a reason back then for not using TPU... but I can't recall so you may be on to something!
I am not an expert, but I havea general knowledge of how these work and I realize I'm missing something--shouldn't the bag need to go all the way through?
To make the parts completely hollow, yes the bag would need to go all the way through. I tried to emphasize "mostly" hollow although now I realize I omitted that detail in the title. The airfoil section is only 15mm at its thickest, so there just isn't much room for the bag to pass trough, especially with the resin feed line. So, we just oped to keep that part solid. With better planning of the tooling, it should be possible to divert the resin feed line to the side of the part allowing space for the bag to pass though. I didn't think of this when designing the tools as the process was being co-developed with the tooling. I am hoping to explore this further on a future project.
I think it depends on the part geometry. With a part of this size, it doesn't seem the extra pressure is necessary (especially if we were using a proper, low viscosity infusion epoxy.) However if the part was much larger, or had a thicker laminate, then the extra pressure could help push the epoxy into the part.
great work, but my question is..why do you use a core? You don't need a smooth surface inside and with this geometry you can do a resin infusion without a 3dprinted core
@@designbydave maybe it's different, I usually work with prepreg, but I think it's a geometry that can be do without core. Or using a little core in the smaller section . You can insert the bag before close the mold, so it's easier to control
I think that your void problem may be due to your use of a pressure pot. You're blowing air into the mold along with the epoxy, then you're not allowing enough time under vacuum for the air to work its way out. Your pressure pot could be a real asset if used in a different way. Put the epoxy in it, attach a vacuum source and degas the epoxy! Less air in means a smaller void risk.
From our various tests, I can be pretty confident that this is not the case. You'll notice in this video that we purged the air from the feed line in the pressure pot. Also, I made a part without it. I don't have enough experience (any experience) with the resin infusion process to say, but the one thing for sure we are not doing is using an infusion specific resin. They must make that stuff for a reason. haha
@@designbydave No, what I noticed was that I could literally see the air bubbles in the line @13:55 , and a few seconds later someone said, "Bubbles let us know it's flowing!". I beg to differ - bubbles let you know that you're pushing air-entrained epoxy into the mold. Each bubble followed the path of least resistance, and some of them led to the mold surface, where they arrived after their path to the vacuum feed was blocked by resin. Hence, voids. Remember that air is compressible and liquids are not, so it's very difficult to push air out of a fluid. When hydraulic systems (e.g. car brakes) are bled, the fluid that contains air bubble has to be pushed completely out of they system by clean fluid that doesn't contain any air. You could have done the same thing and kept the flow through the mold until you saw airless epoxy at the vacuum end, but that would be wasteful of epoxy (and you'd need a capture pot to protect the vacuum pump). Another thing you could try is to block off the feed once the epoxy is half-way along the tube, and allow vacuum to run for a few minutes to degas in the tube. That won't help if bubbles are entrained throughout the entire epoxy mix though - it'll only degas that initial feed. (Sometimes, that might be enough.) You can also try warming your epoxy components before mixing by placing them in a warm water bath. A lower viscosity will entrain less air to begin with, and will also degas faster. You can also warm the mold to prevent a cold mold from increasing the viscosity. Just beware the decrease in pot-life. It might be an idea to mix a small test batch at your intended temperature to verify viscosity and pot-life before committing to it for a part. If you watch the Exploring Composites channel, you'll see that he uses a variety of heat and vacuum techniques to control the epoxy flow speed and path to ensure that any remaining air in the system can escape before the epoxy locks it in.
I follow you on this one. The problem is you have a build up of pressure on the start of the tube entering the part, reducing air bubbles in volume you see in the tube, then go over to to the vacuumside causing the airbubbles to expand, meaning every bubble going in, will expand up to 5 times or more in size under vacuum. @david : like you mentions below, using infusion resin would help a lot on this process. Another point that you might work on is closing down pressure side and let the vacuum do its work more to the end, if needed repeat a few time Anyways good job and good results coming out of the mould!
@@designbydave Using pressurized resin with a soft bag inside looks a recipe to excess resin. Did you compare the weight of the finished pieces? I'm pretty sure from previous experience that leaving the vacuum on in the final stage is not enough to drive out the excess resin. On the other hand it look like you can easily modify your setup to slightly pressurize the bag, enhancing compactation and fiber ratio. As a tackyfier you can also use a VERY SMALL amount of unmixed part A epoxy. The mixed resin you are going to infuse will move and mix with it, and all will harden. This leaves you plenty of time to work.
I'm sure you've seen it, but in case you haven't, Time bicycles makes frames with this process. Not sure if it would help you here, but instead of a 3d printed core and bladder, they mold a wax core for the inside of the part, and once the part is cured, they just melt the wax out.
Thanks. Yeah I'm pretty sure 3T does the same.
I just seen a super neat trick where a guy wrapped his layup and inner core wih fishing line, then upon closing the mold (as much as possible), he snipped each loop of fishing line and pulled it out allowing it to expand into the opposite side mold. You could potentially wrap the whole mold and layup with fishing line which would move the fishing line to the outside for removal...
Interesting trick, I'll have to keep that in mind
great video, working on something similar at the moment :) Here's an answer to your question for the "composites tape instead of blue tape" to stick the fiber to the inner core: Use some strips of carbonfiber UD, add some composites spray tack and use these as a tape to hold it to the 3d core (look for hi-tack 71 from 3M ) never used it from 3m so you might have to do some research (I'm using Fusion Fix from easycomposites) that way you can make a piece of carbonfiber tape similar to prepreg
Great idea, thanks so much! Love your channel btw! Looking forward to the next project.
Just earned a new subscriber. I’m impressed.
Awesome, thank you!
for the lap ply you could use supper77 adhesive as a tackyfier . Spray the printed insert before installing.
Oh spray the insert. That is a great idea, thanks! I had though to use a spray adhesive, but not to spray the core.
You can purchase a spray can adhesive that is applied to the cloth so it will be tacky during layup.
Yeah that would be good thanks
is it possible to use solubile materials?
I've tried that on a previous project and it didn't work out too well. The 3d printed soluble material we used wasn't quite as soluble you would want and removing the mandrel was very difficult.
Prepreg and a bladder mold seems like a much simpler way to go. Is there any benefit to doing it this way outside of maybe cost?
Fascinating watch see the science and magic behind beautiful carbon parts. After watching this whole thing, I have only one question… will you be taking orders for this finished ‘product’?
Thanks for checking it out Ryan. Definitely will not be taking orders. I don't "do" production any more and these 3D printed molds aren't really durable enough to make more then a couple parts each anyhow. Also the finished product will be extremely cost prohibitive. Having said that. We have materials to make a set or two extra of the corner brackets.
there are very reliable plastic clamps that you can on your hoses. Also it's best practice not to use OIL based products on your dry cloth to mark your preform patterns. :)
I'd print that black part in TPU.. its like a hard rubber.. I think you could get a seal if you go the tolerances tight. It can print glass smooth even on my entry-level printer.
It's been a while since I did this project but I want to say that I had a reason back then for not using TPU... but I can't recall so you may be on to something!
I am not an expert, but I havea general knowledge of how these work and I realize I'm missing something--shouldn't the bag need to go all the way through?
To make the parts completely hollow, yes the bag would need to go all the way through. I tried to emphasize "mostly" hollow although now I realize I omitted that detail in the title. The airfoil section is only 15mm at its thickest, so there just isn't much room for the bag to pass trough, especially with the resin feed line. So, we just oped to keep that part solid. With better planning of the tooling, it should be possible to divert the resin feed line to the side of the part allowing space for the bag to pass though. I didn't think of this when designing the tools as the process was being co-developed with the tooling. I am hoping to explore this further on a future project.
So the last proses is the best ?
The infusion without the pressure?
I think it depends on the part geometry. With a part of this size, it doesn't seem the extra pressure is necessary (especially if we were using a proper, low viscosity infusion epoxy.) However if the part was much larger, or had a thicker laminate, then the extra pressure could help push the epoxy into the part.
great work, but my question is..why do you use a core? You don't need a smooth surface inside and with this geometry you can do a resin infusion without a 3dprinted core
The purpose of the core, in this final from of the process is mainly to support the lap joint during the layup process.
@@designbydave maybe it's different, I usually work with prepreg, but I think it's a geometry that can be do without core. Or using a little core in the smaller section . You can insert the bag before close the mold, so it's easier to control
Try to spray the inner bag (touching area with interior) with release wax to extract it easily
Good idea thanks! Removing all the bag isn't important for these parts but it might be on future projects.
Probably won't help much. These tube bags usually are pretty much self-releasing, the catch is when a pleat gets trapped in resin.
You almost need a balloon animal balloon for your bladder. It would start small and expand to fit.
Great thanks!
Thanks for checking it out.
I think that your void problem may be due to your use of a pressure pot. You're blowing air into the mold along with the epoxy, then you're not allowing enough time under vacuum for the air to work its way out. Your pressure pot could be a real asset if used in a different way. Put the epoxy in it, attach a vacuum source and degas the epoxy! Less air in means a smaller void risk.
From our various tests, I can be pretty confident that this is not the case. You'll notice in this video that we purged the air from the feed line in the pressure pot. Also, I made a part without it. I don't have enough experience (any experience) with the resin infusion process to say, but the one thing for sure we are not doing is using an infusion specific resin. They must make that stuff for a reason. haha
@@designbydave No, what I noticed was that I could literally see the air bubbles in the line @13:55 , and a few seconds later someone said, "Bubbles let us know it's flowing!".
I beg to differ - bubbles let you know that you're pushing air-entrained epoxy into the mold. Each bubble followed the path of least resistance, and some of them led to the mold surface, where they arrived after their path to the vacuum feed was blocked by resin. Hence, voids.
Remember that air is compressible and liquids are not, so it's very difficult to push air out of a fluid. When hydraulic systems (e.g. car brakes) are bled, the fluid that contains air bubble has to be pushed completely out of they system by clean fluid that doesn't contain any air. You could have done the same thing and kept the flow through the mold until you saw airless epoxy at the vacuum end, but that would be wasteful of epoxy (and you'd need a capture pot to protect the vacuum pump).
Another thing you could try is to block off the feed once the epoxy is half-way along the tube, and allow vacuum to run for a few minutes to degas in the tube. That won't help if bubbles are entrained throughout the entire epoxy mix though - it'll only degas that initial feed. (Sometimes, that might be enough.)
You can also try warming your epoxy components before mixing by placing them in a warm water bath. A lower viscosity will entrain less air to begin with, and will also degas faster. You can also warm the mold to prevent a cold mold from increasing the viscosity. Just beware the decrease in pot-life. It might be an idea to mix a small test batch at your intended temperature to verify viscosity and pot-life before committing to it for a part.
If you watch the Exploring Composites channel, you'll see that he uses a variety of heat and vacuum techniques to control the epoxy flow speed and path to ensure that any remaining air in the system can escape before the epoxy locks it in.
I follow you on this one. The problem is you have a build up of pressure on the start of the tube entering the part, reducing air bubbles in volume you see in the tube, then go over to to the vacuumside causing the airbubbles to expand, meaning every bubble going in, will expand up to 5 times or more in size under vacuum. @david : like you mentions below, using infusion resin would help a lot on this process. Another point that you might work on is closing down pressure side and let the vacuum do its work more to the end, if needed repeat a few time Anyways good job and good results coming out of the mould!
@@MatthieuLibeert Thanks
@@designbydave Using pressurized resin with a soft bag inside looks a recipe to excess resin. Did you compare the weight of the finished pieces? I'm pretty sure from previous experience that leaving the vacuum on in the final stage is not enough to drive out the excess resin.
On the other hand it look like you can easily modify your setup to slightly pressurize the bag, enhancing compactation and fiber ratio.
As a tackyfier you can also use a VERY SMALL amount of unmixed part A epoxy. The mixed resin you are going to infuse will move and mix with it, and all will harden. This leaves you plenty of time to work.
make bicycle lugs with this process! CF Bicycle 2.0!
Yeah that would be awesome! Maybe someday
Top
Thanks
Subbed!
Thanks!
Many bike manufacturers use bladder and wet lam or prepreg.
To add infusion really complicates and adds more labour