As somebody more familiar with gravity casting and silicone molds, this definitely seems like a way to get into hobby injection molding without cnc machining experience.
If you add a radius to the thru hole for filling it should help a lot. You can see that the mold fractured along 45 degrees/the shear plane and this is encouraged by that corner creating a stress riser.
While this would certainly remove the stress raiser, overfilling the resin and sanding off flat at the mould half backs will mean that the epoxy will be better supported in the injection moulding machine platens, so it may not be necessary to machine a radius in the aluminium.
Agree with the suggestion for radii (radiuses?) on those sharp corners in the aluminum carrier. It’s pretty much an article of faith when designing plastic parts - which your epoxy inserts also are, despite the aluminum filler - to use radii rather than sharp right angles on all shoulders, ribs, gussets and bosses to remove creation of stress points that build in weakness and may fracture - as this one has. Radii there will also help to reduce air entrapment when casting the epoxy. Spent about 15 years casting liquid urethane elastomers for engineering use, and the other article of faith I’d suggest is to slowly run the epoxy down the side of the aluminum and allow it to flow over the 3D printed pattern pushing the air in the bottom of the cavity before it. Maybe tilt the mold to about 25 degrees and slowly bring it back to level as it’s filling. Dumping the material straight onto the top of the pattern entraps air in the pattern detail right from the start. Cheers from Melbourne, Australia - thanks for a great video, love your work!
I have experienced similar defects when printing molds flat on the build plate. I attribute both the rough edges and compressed Z height on the top features to the same issue. The printer is not rigid enough to apply the required force to squeeze all of the resin out from under the large base section before the exposure starts. The resin continues to squeeze out during the exposure, causing the partially cured rough edge. The Z column deflects slightly so the layers with large cross sectional area print slightly too thick, but layers with small cross sectional area do not. I have been able to eliminate the rough edge and improve dimensional tolerance on my prints by adding long pre-exposure wait times and heating the vat to decrease viscosity. Printing at an angle solves the problem also but of course requires supports.
Bravo John- I am using 3D prints for masters with molds- appreciate the failure analysis and your approach to this method. As always- be safe and look forward to your next bit of education. Thanks for letting us watch over your shoulder.
While I'm just a beginner to this, I have been making resin mold masters for epoxy casting and I have a few tips from my experience over the last couple of years. I think that "traditional" top-down laser printers have fewer forces to contend with since their parts are in compression, not tension, since they sit on a perforated plate and descend into a vat of resin. That resin acts as a sort of support for the whole thing. If you need parts of a more geometric design that require tight tolerances that might be the way to go, but it is pricey, I was quoted about $1000 for a part the size of a softball. Printing things flat on a consumer printer can make your parts deform in the z axis. I always get stretching, sometimes as bad as 100%. It depends on the resin and the temperature of your print area. Using a heated enclosure at around 30 degrees celcius can help, when the resin is thinner there's less forces. If you print at an angle, round things won't be round. I've had a heck of a time trying to make parts with gasketed fitted lids. I've had to print the part over sized and machine it on a mill and lathe to get the precision I need. If you print hollow, which reduces the weight and thereby the distortion, you will need drain holes and an ultrasonic cleaner to get all the uncured resin out or your parts will swell and crack over time.
I don't remember the UV resin brand/model being mentioned, but it looks like standard everyday resin. When I am trying to attain more accurate forms, I've had good luck with Sirayatech Build resin in smokey black. I would also try to print the master with the flat parts directly on the flat FEP, and hollow the whole thing to about 1-2 mm, with lots of ribs for structure. You could then back fill the hollow reverse side with resin and pressure cure it for strength.
Thank you for this very informative testing video. This will save so much time and effort for others. The smaller hole for epoxy entry you mentioned while pouring might be all you need for reducing the blowout. I wonder if the hole could be so small that a syringe could be used to add the epoxy. Not sure if that would introduce bubbles. The smaller the hole, the harder it will be to remove the epoxy mold to be replaced by another mold, but each mold may need its own aluminum holder due to the sanding as the aluminum will be affected in the sanding process that the molds will need to be paired with their aluminum counterpart.
I used this same method (also with Epoxacast HT) to prototype molds and I had a lot of trouble with uneven shrinkage/deformation on mine so in the end, the mold halves were misaligned by 0.050" or more. Also had issues like you with bubbles not popping due to the high viscosity, and found that much higher time sitting in the vacuum chamber helped, but don't have a pressure pot to use like you. The end product can also be quite brittle, so features that stick out and small details are more vulnerable. I think this works well with smaller molds with less detail, but probably won't be pursuing it for any larger or more intricate parts again.
Agreed. I think this is a potentially better over pure 3D printed molds. But clearly not a replacement for CNC machined molds. For the next version, I'm hoping that Rulon can do a production run to find out how long it lasts.
This is really neat. I've used high-temperature 3D printer resin to make moulds, experimenting with a home-made Buster Beagle 3D injection machine at the same time. I've heard of people getting a hundred parts out of a run with direct printed parts which was good enough for my use case because printing another mould is easy enough. Usually I ended up getting plastic stuck in some lettering after a few injections on my own machine and now I'm planning to upgrade my machine and try mould release. It will be interesting to see how durable the epoxy is--if it's over 3x as good as directly printing a mould it would be worth the effort. Thanks for the video!
I would suggest you fill the square opening on the one side completely and make sure it is flat wrt the rest of the mold. This will do two things. First, it will support the injected material while under pressure and prevent catastrophic cracking. Second, it will create a thermal contact with the pressure plate so mold can cool faster. If you have a cool aluminum plate to set the mold on after injection that would also help to cool things down more quickly through the large sides of the mold.
Instead of pouring the epoxy into a large hole, I would try 2 small 1/8" holes. One to inject the epoxy into with a syringe and the second is for air to escape. I think 2 small, round 1/8" holes would be less compromised than a larger square. More pressure against a square edge, than round experiences. Just a thought.
Nice work. You may want to apply heat grease or a thermal compound in the interface between the inner epoxy mold and the outer aluminum case. It would surely be messy, but it is likely to help with the thermal performance.
From my understanding of the warping, it's due to the print getting stuck on the vat because of a vacuum/suction formed as it prints. That's why when you printed it at an angle, you had better results. The surface area of the print was smaller, so there was less of a vacuum being formed during the print.
Great video. I would add 4 3/8" -16 flat headed bolts thru the back side of the mold. One at each corner. Put mold release on the bolts before threading them in then tighten them well. Before you add the epoxy making sure you put more mold release on the whole cavity. They should protrude 2/3 to 3/4 of the way into the cavity area. . Putting a small relief angle on the sides will also help releasing the mold knowing the bolts will hold it in place. Without the relief angle you may want to make the bolts captive that way helping to push the mold out when not needed. Hope that helps.
I haven't taken the time to read smooth on's specs for this epoxy in regards to shrinkage over time. I've worked with resins and epoxies for years and one thing that might be helpful, I don't know, but the epoxy like glass and concrete is great in compression and terrible under tension. You could reinforce that epoxy with fiber reinforcement. If you took some fiberglass and removed 50% of the material so the weave was opened up and you layered that through the mold my thought is that would be near indestructible. I've done this before but never with the epoxy that you are using. You would have to use a putty knife to impregnate the glass cloth and then hand build the layers stopping about an eighth of an inch of the surface. Another thing might be to electroform the surface of the mold and keep the epoxy as a backing. Electroforming would be plating the surface whatever metal you choose and let it build up to a certain thickness. Adding air passes in the thicker parts of the mold for air circulation to aid cooling might be beneficial. Like anything, you need to R&D and figure out the process and come to solutions. There are many ways to go about making these molds without complex machining.
Same here. Thought this was meant to be a single use item... but if/when he fixes the design, I think it could be used dozens, if not 100s of times. It could be 1000s, but I'm not really sure of that. Would be cool to see though... like a future video after long-term use.
I’ve had problems with filled lettering if I didn’t thoroughly wash the excess resin out. The IPA didn’t penetrate the recesses - acetone helped a LOT with making crisp relief features release resin but damage the resin if left too long.
Great vid! Very interesting to watch, and I wanted to say thanks for your channel it is very helpful. I think you might be able to drill a through whole/slot in through the corner of the metal mold holder in place of the cutout at the back to pour the epoxy in. The corner because all the air would naturally bleed out there, and then it wouldn't allow for the epoxy to blow out
Had the same idea as well. Did not know they made an epoxy mix.. Was going to try graphite powder as well as aluminum powder, for heat dissipation and maybe a mix of the 2. Was also going to try monster clay, to make a base mold. But not sure of how hot the heat off gassing of the epoxy would be?(Temperature Gun of the epoxy would help) Testing.. will see PS.. would buy those blanks if they were available to buy..
One thing that helps minimize distortions/warping of the 3D printed masters is to use a resin with a bit of flex. I print engineering prototypes and had a horrible time printing thin straight edges on parts until I saw a few TH-cam videos comparing various resins and started using Sunlu ABS-Like resin. It's fairly cheap but has little warping compared with most resins.
3:41 Your model is showing typical signs that you need more time on the "rest after retract" setting or whatever the equivalent is in your slicer. Basically your light is turning on and curing resin while resin is still flowing/squishing out of the gap between the film/screen and the last layer of the model. This is especially a problem with more viscous resins and is often referred to as "bloom." Another possibility is that the PWM or light intensity setting needs to be turned down a little bit, but I'd experiment with the rest after retract first. Some printers like my Mega 8K have either really intense light sources that are capable of overwhelming the screen's masking, or the masking is too weak. Turning this setting down can help prevent a different type of bloom or light bleed. This is typically more of a problem with larger parts with lots of straight edges.
that failure following the shape of the hole in the frame is amazing, I wonder if the sheer edge had something to do with it super cool, thanks for sharing Ɛ>
Can I make a suggestion. Make the hole round. If you look at the cracks, the seemed to crack exactly like old planes with square windows used to crack. In the corners. That is why you don’t get planes with square windows anymore. It causes too many weak points.
Nicely done. I wonder if the resin shrank uniformly enough to adjust the mold size to account for shrinkage (no tape). Something I did to overcome the flatness issue for 3D printed parts was to print the part flat with a chamfer on the platten side. The chamfer lets you get a metal spatula under it. I had printed parts and made silicone molds without curing the printed parts. Curing introduces stress and curvature so I just cleaned and blow-dried the 3D printed parts and then cast it.
I haven't tried it myself, but a while back I was looking into which resins had the highest temperature ratings and found the Stone Coat countertop resin to seemingly have the best numbers on paper. I had thought about using that resin along with mixing in either copper or aluminum powder to help with thermal conductivity. That resin is technically a coating resin and not a casting resin, so it may not work as desired when used this way, but could still be worth experimenting with.
@JohnSL I want to make a suggestion that builds on some of the other suggestions below: Instead of filing the epoxy from the working face of your mold through the aluminum frame, add fill and vent holes to corners the 3D printed part. Use a large syringe to force the epoxy through the fill hole while the mold is upside down (the printer part is below the metal frame) This will allow the epoxy to flow over and away from your details, carrying any surface bubbles with it. When the epoxy starts to come out of the vent hole, you will rotate to mold so that the vent hole is the highest point and continue to inject the rest of the epoxy slowly. This way you can use the weight and viscosity of the epoxy to push out the bubbles and you can keep a solid back on the mold to avoid stress concentrations that will eventually shatter your mold. The face of the mold will need to be sanded anyway, so removing the injection and vent sprue will not add much extra effort. If you still have issues with air cavities, contact Robert Tolone (another TH-cam channel)who has an extensive experience in molding and casting parts.
Thanks. That's an interesting idea. By using the pressure pot, I didn't have any issues with air bubbles. My plan is a little different, but somewhat similar. I'm going to put pouring/vent holes into two corners of the frame back and pour into the "lower" hole. That will ensure that if there are any bubbles, they will remain at the back. But I think if I put the mold at the right angle, I won't event get those bubbles.
Particle reinforced epoxy is quite fragile... The aluminum frame should completelly support the back and sides of the molded epoxy block (as you mentionned in the end).
Instead of a rounded rect cutout in the aluminum frame, you could do a round threaded bunghole. Then you could screw a brass or aluminum insert into the hole to have full even support of the final mold. Just sand off the back of the epoxy mold as well so that it can sit flat with the cover in place.
Great video John thanks for sharing! The resin method is really interesting. The heat curing step of the resin does look like it poses some challenges. I wonder if using less volume of resin overall would aid in reducing the amount of deformation that could occur? Although that might cause warpage if the resin isn't thick enough. What are your thoughts of the possibility of the aluminium housing expanding during the heat treatment and possibly causing the resin to pull away from the walls when cooled? Reducing the size of the inlet when pouring the resin could make it more difficult for trapped air to escape as you fill the mold. Several/many small holes around the entire area might help if you think that's a possiblity. anyway thanks again for sharing and looking forward to the next one!
You need to angle the bolster inner side walls and the sides of the epoxy inserts at 1.8%, and plastic/plaster gauge the back faces of the inserts to the bolsters to work out the correct depths. Support your epoxy sides 360° and you will have the maximum advantage from the epoxy. Nice video. I think this would work. If you're using PVC, watch out it is explosive if left in a sealed heated chamber.
The sagging of the print is the print sticking to the PET film between layers, silicone spray emigrate it a bit but the trick is the 45 degree tilt is making the contact area of the part between layers to become smalle this less release stress from the PET film
John I have mostly given up building supports in the printing apps like Lychee et al and place them directly in my CAD. For me the biggest problem with all these resin slicers is that they cater to the "toy solder" rather than engineering community where absolutely dead flat isn't post-print fixable and critical to print correctly. I solved a similar issue to your problematic manually supported edge by building a "v-shaped" chisel support into the model edge itself. The chisel base is large enough to use as part of a raft and the chisel "point" hits the model edge with about 0.3mm thickness along it's entire length, rather than tryng to use arbitrary "point" supports in app. The chisel support then just snaps off cleanly along the length of the part and so allows you to take advantage of the 45 deg print angle for resolution. Depending on the model/angle this can allow one face of the chisel to be co-planer with one of the model surfaces and so produce a super clean parting line as well.
I believe if you have more supports that are thinner at the bottom it will warp less. Warping is caused by resin shrinkage and very heavy supports at the bottom make part shrink unevenly
Could be possible to machine directly into a epoxy brick so you dont have to create a positive? great content! My grandfather had a big old injection machine and some clients brought him bronze molds. Less durability but way cheaper too.
What if you mixed the SLA resin with aluminum powder to combine the physical properties and help the heat spreads across the mold? This way you’d skip the entire epoxy+remolding part, and still have the thermal benefits of the aluminum mix
If I may, I observed other resin casters pouring resin into the hardener prior to hand stirring. It was said to give a better mix. Resin adhering to the wall of the mixing cup was then not a problem. It is something I will try in the future.
Interesting. Would it be worth having a small pressure release hole/path opposite the sprue where the injected plastic could vent and release ? I know nothing about injection molding. Just a thought. Great documentation and fun seeing what you were up to.
Wow, that crack was crazy! The video was interesting all the way through, I like the little challenges but that crack and how the plastic filled the broken part was really cool, I was not expecting that!
I was surprised it didn't crack sooner. I thought it was being built as a single use item, but was glad to see he got several uses from it... and if/when he improves the design I think he might get 100s of uses from it. I'm not sure if 1000s of uses will be possible though... but that would be great to see. Would open the doors for a lot of DIYers and inventive people.
Interesting experiment. Detailed analysis. Actually, the opening could be filleted on the inside, and additional plate can cover the hole with silicone(any possible bubbles removed with vacuum) to takeup extra space, with more even spread on pressure. Just some 💭
Crafsman has a video called "3D Printed Injection Molds (actually work)!", the frame he has in that video has proper support all around the 3D printed part. That looks kinda important, even he mentions that he was worried about the pressure and his 3D printed mould probably held up because it had a proper backing part too.
Hee, hee. I was wondering if someone would notice that. It's not the best practice, but I figured the pressure pot would cover that sin, and it did. The only bubbles I saw where on the back where the air wasn't able to escape.
@@JohnSL haha it was the first thing I thought when I saw it, but didn’t want to be “that guy” 😆 and when I saw you use the pressure pot afterwards I figured it would be be fine and it totally was. just now that pouring from the side becomes automatic I thought I would throw it in as a nod to the master 😊
Really interesting stuff. The granate plate - was that a cheap option you picked up? I've used glass in the past. You can see if glass is flat by it's aberations - or so someone told me.
Does the epoxy shrink when cooling? If so, could that cause the mold to be a lot weaker and more likely to crack? Like, if the epoxy shrinks, then the pressure between the epoxy and aluminum might not be as strong and might give it a bit of clearance to "stretch"... but if it's a hard/brittle epoxy, then it will crack instead of flexing. It's just a thought and a question. I really don't know. I just thought about it when the mold was being made, when the shrinkage was being shown. If it shrank that much there, then I thought it might be possible to shrink (to a much lesser extent) on the inside after it starts to cure. I'm sure being thinner didn't help much either. I also wondered about the pressure used for the bubbles and was wondering why pressure and not a vacuum to remove the bubbles? (or vacuum then pressure).
Can you sinter it? I'm looking into various metal resin products that I'd be able to make a master mold of Stainless steel, sinter it and use that master as a repeatable mold for lower fired projects such as glass fusing and bronze. I'm told as long as the metal fill is 80-85% or higher sintering could work.
Thanks for this... I am gearing up to do a series of injection molds and I was going to try 3D printed molds and hadn't considered epoxy. I have a Morgan press and will have the sprue in the top of the mold, not the side... Any thoughts on how that would change the process... thanks again, this is great stuff!
Thanks so much for you video, i Want see the part 2 ... i have some question, firts im from mexico, second the Epoxy can be change for resin of 3D Print?, i can see you have a problem whit the plastic, maybe your problem is the time cold and no the presure, i tried make a simple inyection machine with 3D print (ENDER 3) and the problem was my Mold and no the press... and last question which one the same epoxy injection similar?
Epoxy properties fall down with temperature, some can take more heat than others but you should really take a look at TDS for some numbers. Usually "high temp" resins have only a small fraction of their room temperature performance at the claimed max operating temperature, even some aerospace grade adhesives. About voids, you should take the mold in the vacuum chamber before the pressure pot. Vacuum will expand the bubbles, hopefully they will surface and burst (may need some help with mechanical action / vibration if resin is too viscous). The pressure pot just reduce the size of the bubbles proportionally to the pressure applied, it will not pop them. Double the pressure, half the volume, and so on. If you can arrange a good master/aluminium layout with a suitable resin inlet and vacuum outlet, I would try something like a vacuum infusion/VARTM process, it will ensure a nearly void free mold.
I tried using a vacuum chamber. The bubbles roles to the top, but didn't break. I tried to vibrate it a little, but didn't have a good solution. Even so, I didn't have issues with bubbles in the mold.
Do the plastic beads get hard if they're left on the heat too long? With PLA plastic (3d printing) if you leave the machine on with the hotend at printing temperature for a long time (several hours) without printing it will cause the plastic to harden like a ROCK and it's so hard you have to melt it out or burn it out. It can't be removed with physical force and it's probably not a good idea to try to Dremel it out.
Great video. I've noticed the same Z axis errors when printing a similar part "flat" on the build plate of a resin printer. I'm curous to know if anyone has an explaination.
You're supposed to put the model at 45° in two axis not just one. Also 45° may not be the correct angle for your printer and print. You should use the arctan for the correct angle. Some resins are better for more dimensionally accurate prints as well. I also recommend keeping the resin and print environment at 77°F for best results.
As an armchair expert in mold making, I'm wondering, can you add something like nails to the epoxy resin to act like rebar? Though it might do more harm than good if its coefficient of thermal expansion doesn't match.
Put the hole to fill in the epoxy to a corner or perhaps even drill a hole from a side surface or two, one to fill in the epoxy and one to let the air escape.
Id say cnc cooling channels in the aluminum before you permabond the 3d plastic and then add a regular pc cooling radiator with reg fittings etc. Kind of in the same way what nasa did back in the day with their rocket nozzles.
Vacuum bubbles 2-3x out of epoxy before pouring.. Add 1/32 milled fiber or micro carbon fibers, if available, to Epoxy. Definitely fully support the sandwich.. Let us know your success!
Non ho ben capito come mai non è stato fatto interamente in alluminio lo stampo a cnc ma potresti provare a rifarlo usando, invece che del gelcoat caricato per stampi, della resina mescolata con polvere di carbonio.
Kind of making an educated guess here, but it looks like your resin 3d print is suffering from over exposure bloat. Not sure what printer you use and what settings you have and all that.
![PASULOL รามเกียรติ์ ตอนที่ 4 นนทกพบรัก [Ramakien Ep.4: Nonthok in love]](http://i.ytimg.com/vi/tMUQgmZzf_A/mqdefault.jpg)
As somebody more familiar with gravity casting and silicone molds, this definitely seems like a way to get into hobby injection molding without cnc machining experience.
If you add a radius to the thru hole for filling it should help a lot. You can see that the mold fractured along 45 degrees/the shear plane and this is encouraged by that corner creating a stress riser.
While this would certainly remove the stress raiser, overfilling the resin and sanding off flat at the mould half backs will mean that the epoxy will be better supported in the injection moulding machine platens, so it may not be necessary to machine a radius in the aluminium.
@@dewexdewex true, I guess it's more of an improvement rather than a solution
& smaller filling hole 👍🏻
Agree with the suggestion for radii (radiuses?) on those sharp corners in the aluminum carrier. It’s pretty much an article of faith when designing plastic parts - which your epoxy inserts also are, despite the aluminum filler - to use radii rather than sharp right angles on all shoulders, ribs, gussets and bosses to remove creation of stress points that build in weakness and may fracture - as this one has.
Radii there will also help to reduce air entrapment when casting the epoxy. Spent about 15 years casting liquid urethane elastomers for engineering use, and the other article of faith I’d suggest is to slowly run the epoxy down the side of the aluminum and allow it to flow over the 3D printed pattern pushing the air in the bottom of the cavity before it. Maybe tilt the mold to about 25 degrees and slowly bring it back to level as it’s filling. Dumping the material straight onto the top of the pattern entraps air in the pattern detail right from the start.
Cheers from Melbourne, Australia - thanks for a great video, love your work!
@@nongbloke agreed, fellow Melbournian :)
I enjoyed the video, fun to see the thought process and iteration, looking forward to part 2
I have experienced similar defects when printing molds flat on the build plate. I attribute both the rough edges and compressed Z height on the top features to the same issue. The printer is not rigid enough to apply the required force to squeeze all of the resin out from under the large base section before the exposure starts. The resin continues to squeeze out during the exposure, causing the partially cured rough edge. The Z column deflects slightly so the layers with large cross sectional area print slightly too thick, but layers with small cross sectional area do not. I have been able to eliminate the rough edge and improve dimensional tolerance on my prints by adding long pre-exposure wait times and heating the vat to decrease viscosity. Printing at an angle solves the problem also but of course requires supports.
Thank you. That all makes sense.
Bravo John- I am using 3D prints for masters with molds- appreciate the failure analysis and your approach to this method. As always- be safe and look forward to your next bit of education. Thanks for letting us watch over your shoulder.
While I'm just a beginner to this, I have been making resin mold masters for epoxy casting and I have a few tips from my experience over the last couple of years. I think that "traditional" top-down laser printers have fewer forces to contend with since their parts are in compression, not tension, since they sit on a perforated plate and descend into a vat of resin. That resin acts as a sort of support for the whole thing. If you need parts of a more geometric design that require tight tolerances that might be the way to go, but it is pricey, I was quoted about $1000 for a part the size of a softball. Printing things flat on a consumer printer can make your parts deform in the z axis. I always get stretching, sometimes as bad as 100%. It depends on the resin and the temperature of your print area. Using a heated enclosure at around 30 degrees celcius can help, when the resin is thinner there's less forces. If you print at an angle, round things won't be round. I've had a heck of a time trying to make parts with gasketed fitted lids. I've had to print the part over sized and machine it on a mill and lathe to get the precision I need. If you print hollow, which reduces the weight and thereby the distortion, you will need drain holes and an ultrasonic cleaner to get all the uncured resin out or your parts will swell and crack over time.
Injection molding at home
I don't remember the UV resin brand/model being mentioned, but it looks like standard everyday resin. When I am trying to attain more accurate forms, I've had good luck with Sirayatech Build resin in smokey black. I would also try to print the master with the flat parts directly on the flat FEP, and hollow the whole thing to about 1-2 mm, with lots of ribs for structure. You could then back fill the hollow reverse side with resin and pressure cure it for strength.
I also recommend a 24 hour post-cure in a UV oven, submerged in water. This will ensure the part doesnt warp as it ages.
Thank you for this very informative testing video. This will save so much time and effort for others. The smaller hole for epoxy entry you mentioned while pouring might be all you need for reducing the blowout. I wonder if the hole could be so small that a syringe could be used to add the epoxy. Not sure if that would introduce bubbles. The smaller the hole, the harder it will be to remove the epoxy mold to be replaced by another mold, but each mold may need its own aluminum holder due to the sanding as the aluminum will be affected in the sanding process that the molds will need to be paired with their aluminum counterpart.
I used this same method (also with Epoxacast HT) to prototype molds and I had a lot of trouble with uneven shrinkage/deformation on mine so in the end, the mold halves were misaligned by 0.050" or more. Also had issues like you with bubbles not popping due to the high viscosity, and found that much higher time sitting in the vacuum chamber helped, but don't have a pressure pot to use like you. The end product can also be quite brittle, so features that stick out and small details are more vulnerable. I think this works well with smaller molds with less detail, but probably won't be pursuing it for any larger or more intricate parts again.
Agreed. I think this is a potentially better over pure 3D printed molds. But clearly not a replacement for CNC machined molds. For the next version, I'm hoping that Rulon can do a production run to find out how long it lasts.
This is really neat. I've used high-temperature 3D printer resin to make moulds, experimenting with a home-made Buster Beagle 3D injection machine at the same time. I've heard of people getting a hundred parts out of a run with direct printed parts which was good enough for my use case because printing another mould is easy enough.
Usually I ended up getting plastic stuck in some lettering after a few injections on my own machine and now I'm planning to upgrade my machine and try mould release. It will be interesting to see how durable the epoxy is--if it's over 3x as good as directly printing a mould it would be worth the effort. Thanks for the video!
I think it will be a lot more durable, but we won't find out until I make a new set of molds.
Maybe add a removable back panel, for casting and after solidifies, sand down and close with a clamping screw.
I would suggest you fill the square opening on the one side completely and make sure it is flat wrt the rest of the mold. This will do two things. First, it will support the injected material while under pressure and prevent catastrophic cracking. Second, it will create a thermal contact with the pressure plate so mold can cool faster. If you have a cool aluminum plate to set the mold on after injection that would also help to cool things down more quickly through the large sides of the mold.
Instead of pouring the epoxy into a large hole, I would try 2 small 1/8" holes. One to inject the epoxy into with a syringe and the second is for air to escape. I think 2 small, round 1/8" holes would be less compromised than a larger square. More pressure against a square edge, than round experiences. Just a thought.
My plan for next time is pretty much what you're suggesting, but with perhaps 1/4" or a little larger holes at two corners rather than in the center.
Nice work. You may want to apply heat grease or a thermal compound in the interface between the inner epoxy mold and the outer aluminum case. It would surely be messy, but it is likely to help with the thermal performance.
From my understanding of the warping, it's due to the print getting stuck on the vat because of a vacuum/suction formed as it prints. That's why when you printed it at an angle, you had better results. The surface area of the print was smaller, so there was less of a vacuum being formed during the print.
Great video. I would add 4 3/8" -16 flat headed bolts thru the back side of the mold. One at each corner. Put mold release on the bolts before threading them in then tighten them well. Before you add the epoxy making sure you put more mold release on the whole cavity. They should protrude 2/3 to 3/4 of the way into the cavity area. . Putting a small relief angle on the sides will also help releasing the mold knowing the bolts will hold it in place. Without the relief angle you may want to make the bolts captive that way helping to push the mold out when not needed. Hope that helps.
I haven't taken the time to read smooth on's specs for this epoxy in regards to shrinkage over time. I've worked with resins and epoxies for years and one thing that might be helpful, I don't know, but the epoxy like glass and concrete is great in compression and terrible under tension.
You could reinforce that epoxy with fiber reinforcement. If you took some fiberglass and removed 50% of the material so the weave was opened up and you layered that through the mold my thought is that would be near indestructible. I've done this before but never with the epoxy that you are using. You would have to use a putty knife to impregnate the glass cloth and then hand build the layers stopping about an eighth of an inch of the surface.
Another thing might be to electroform the surface of the mold and keep the epoxy as a backing. Electroforming would be plating the surface whatever metal you choose and let it build up to a certain thickness. Adding air passes in the thicker parts of the mold for air circulation to aid cooling might be beneficial.
Like anything, you need to R&D and figure out the process and come to solutions. There are many ways to go about making these molds without complex machining.
This was very informative. Can't wait to see round 2.
Same here. Thought this was meant to be a single use item... but if/when he fixes the design, I think it could be used dozens, if not 100s of times. It could be 1000s, but I'm not really sure of that. Would be cool to see though... like a future video after long-term use.
I’ve had problems with filled lettering if I didn’t thoroughly wash the excess resin out. The IPA didn’t penetrate the recesses - acetone helped a LOT with making crisp relief features release resin but damage the resin if left too long.
Great vid! Very interesting to watch, and I wanted to say thanks for your channel it is very helpful. I think you might be able to drill a through whole/slot in through the corner of the metal mold holder in place of the cutout at the back to pour the epoxy in. The corner because all the air would naturally bleed out there, and then it wouldn't allow for the epoxy to blow out
Had the same idea as well. Did not know they made an epoxy mix.. Was going to try graphite powder as well as aluminum powder, for heat dissipation and maybe a mix of the 2. Was also going to try monster clay, to make a base mold. But not sure of how hot the heat off gassing of the epoxy would be?(Temperature Gun of the epoxy would help) Testing.. will see
PS.. would buy those blanks if they were available to buy..
One thing that helps minimize distortions/warping of the 3D printed masters is to use a resin with a bit of flex. I print engineering prototypes and had a horrible time printing thin straight edges on parts until I saw a few TH-cam videos comparing various resins and started using Sunlu ABS-Like resin. It's fairly cheap but has little warping compared with most resins.
Pretty darn cool! I'm excited to see Part 2.
I haven't tried this myself yet.. but if you pull vacuum on the epoxy, you may be able to reduce bubbles
3:41 Your model is showing typical signs that you need more time on the "rest after retract" setting or whatever the equivalent is in your slicer. Basically your light is turning on and curing resin while resin is still flowing/squishing out of the gap between the film/screen and the last layer of the model. This is especially a problem with more viscous resins and is often referred to as "bloom."
Another possibility is that the PWM or light intensity setting needs to be turned down a little bit, but I'd experiment with the rest after retract first. Some printers like my Mega 8K have either really intense light sources that are capable of overwhelming the screen's masking, or the masking is too weak. Turning this setting down can help prevent a different type of bloom or light bleed. This is typically more of a problem with larger parts with lots of straight edges.
that failure following the shape of the hole in the frame is amazing, I wonder if the sheer edge had something to do with it
super cool, thanks for sharing Ɛ>
Can I make a suggestion. Make the hole round. If you look at the cracks, the seemed to crack exactly like old planes with square windows used to crack. In the corners. That is why you don’t get planes with square windows anymore. It causes too many weak points.
Nicely done.
I wonder if the resin shrank uniformly enough to adjust the mold size to account for shrinkage (no tape).
Something I did to overcome the flatness issue for 3D printed parts was to print the part flat with a chamfer on the platten side. The chamfer lets you get a metal spatula under it.
I had printed parts and made silicone molds without curing the printed parts. Curing introduces stress and curvature so I just cleaned and blow-dried the 3D printed parts and then cast it.
I haven't tried it myself, but a while back I was looking into which resins had the highest temperature ratings and found the Stone Coat countertop resin to seemingly have the best numbers on paper. I had thought about using that resin along with mixing in either copper or aluminum powder to help with thermal conductivity. That resin is technically a coating resin and not a casting resin, so it may not work as desired when used this way, but could still be worth experimenting with.
That is amazing. I'm looking forward to part 2
@JohnSL I want to make a suggestion that builds on some of the other suggestions below:
Instead of filing the epoxy from the working face of your mold through the aluminum frame, add fill and vent holes to corners the 3D printed part. Use a large syringe to force the epoxy through the fill hole while the mold is upside down (the printer part is below the metal frame) This will allow the epoxy to flow over and away from your details, carrying any surface bubbles with it. When the epoxy starts to come out of the vent hole, you will rotate to mold so that the vent hole is the highest point and continue to inject the rest of the epoxy slowly. This way you can use the weight and viscosity of the epoxy to push out the bubbles and you can keep a solid back on the mold to avoid stress concentrations that will eventually shatter your mold. The face of the mold will need to be sanded anyway, so removing the injection and vent sprue will not add much extra effort. If you still have issues with air cavities, contact Robert Tolone (another TH-cam channel)who has an extensive experience in molding and casting parts.
Thanks. That's an interesting idea. By using the pressure pot, I didn't have any issues with air bubbles. My plan is a little different, but somewhat similar. I'm going to put pouring/vent holes into two corners of the frame back and pour into the "lower" hole. That will ensure that if there are any bubbles, they will remain at the back. But I think if I put the mold at the right angle, I won't event get those bubbles.
Particle reinforced epoxy is quite fragile... The aluminum frame should completelly support the back and sides of the molded epoxy block (as you mentionned in the end).
Very interesting video, thank you for sharing!. I just wonder why you didnt consider 3D printing the mold itself rather than using epoxy? Thank you!
Instead of a rounded rect cutout in the aluminum frame, you could do a round threaded bunghole. Then you could screw a brass or aluminum insert into the hole to have full even support of the final mold. Just sand off the back of the epoxy mold as well so that it can sit flat with the cover in place.
Great video John thanks for sharing!
The resin method is really interesting. The heat curing step of the resin does look like it poses some challenges. I wonder if using less volume of resin overall would aid in reducing the amount of deformation that could occur? Although that might cause warpage if the resin isn't thick enough. What are your thoughts of the possibility of the aluminium housing expanding during the heat treatment and possibly causing the resin to pull away from the walls when cooled?
Reducing the size of the inlet when pouring the resin could make it more difficult for trapped air to escape as you fill the mold. Several/many small holes around the entire area might help if you think that's a possiblity.
anyway thanks again for sharing and looking forward to the next one!
You need to angle the bolster inner side walls and the sides of the epoxy inserts at 1.8%, and plastic/plaster gauge the back faces of the inserts to the bolsters to work out the correct depths. Support your epoxy sides 360° and you will have the maximum advantage from the epoxy. Nice video. I think this would work. If you're using PVC, watch out it is explosive if left in a sealed heated chamber.
The sagging of the print is the print sticking to the PET film between layers, silicone spray emigrate it a bit but the trick is the 45 degree tilt is making the contact area of the part between layers to become smalle this less release stress from the PET film
John I have mostly given up building supports in the printing apps like Lychee et al and place them directly in my CAD. For me the biggest problem with all these resin slicers is that they cater to the "toy solder" rather than engineering community where absolutely dead flat isn't post-print fixable and critical to print correctly.
I solved a similar issue to your problematic manually supported edge by building a "v-shaped" chisel support into the model edge itself. The chisel base is large enough to use as part of a raft and the chisel "point" hits the model edge with about 0.3mm thickness along it's entire length, rather than tryng to use arbitrary "point" supports in app. The chisel support then just snaps off cleanly along the length of the part and so allows you to take advantage of the 45 deg print angle for resolution. Depending on the model/angle this can allow one face of the chisel to be co-planer with one of the model surfaces and so produce a super clean parting line as well.
Great video
I have to try this in the future.
Any expirience with coolig channels in the Epoxy to reduce cycle time?
I believe if you have more supports that are thinner at the bottom it will warp less. Warping is caused by resin shrinkage and very heavy supports at the bottom make part shrink unevenly
Thanks. I'll have to try that.
Could be possible to machine directly into a epoxy brick so you dont have to create a positive? great content! My grandfather had a big old injection machine and some clients brought him bronze molds. Less durability but way cheaper too.
great video. How about adding kevlar fibers at 90 degree angles inside the mold before casting the expoxy.
What if you mixed the SLA resin with aluminum powder to combine the physical properties and help the heat spreads across the mold?
This way you’d skip the entire epoxy+remolding part, and still have the thermal benefits of the aluminum mix
If I may, I observed other resin casters pouring resin into the hardener prior to hand stirring. It was said to give a better mix. Resin adhering to the wall of the mixing cup was then not a problem. It is something I will try in the future.
Interesting. Would it be worth having a small pressure release hole/path opposite the sprue where the injected plastic could vent and release ? I know nothing about injection molding. Just a thought. Great documentation and fun seeing what you were up to.
Out of curiosity, what’s the end product for ?
Silicone = Vacuum pot, Epoxy = Pressure pot. Vac pot will pull volatiles from the epoxy and make it have more gas.
Wow, that crack was crazy! The video was interesting all the way through, I like the little challenges but that crack and how the plastic filled the broken part was really cool, I was not expecting that!
I was surprised it didn't crack sooner. I thought it was being built as a single use item, but was glad to see he got several uses from it... and if/when he improves the design I think he might get 100s of uses from it. I'm not sure if 1000s of uses will be possible though... but that would be great to see. Would open the doors for a lot of DIYers and inventive people.
Interesting experiment. Detailed analysis. Actually, the opening could be filleted on the inside, and additional plate can cover the hole with silicone(any possible bubbles removed with vacuum) to takeup extra space, with more even spread on pressure. Just some 💭
Enjoyed the video, the process and looking fwd to Part 2
enjoyed the video m8 well put together , liked the bit about troubleshooting the print's and the mold too
Интересный эксперимент. Есть как плюсы так и минусы. В принципе имеет право на жизнь такой подход
Great video. Got any info on your table top injection machine?
Crafsman has a video called "3D Printed Injection Molds (actually work)!", the frame he has in that video has proper support all around the 3D printed part. That looks kinda important, even he mentions that he was worried about the pressure and his 3D printed mould probably held up because it had a proper backing part too.
Great video. Very interesting results. Would shifting the pouring hole to the side help, away from the actual part?
Yes, that's one of the changes I plan to make, which is to make the pouring hole smaller, and put it in a corner.
@@JohnSL less chance to anger TH-cam mold & casting god Robert Tolone as well 😉 I saw you pour straight over the details 🙂
Hee, hee. I was wondering if someone would notice that. It's not the best practice, but I figured the pressure pot would cover that sin, and it did. The only bubbles I saw where on the back where the air wasn't able to escape.
@@JohnSL haha it was the first thing I thought when I saw it, but didn’t want to be “that guy” 😆 and when I saw you use the pressure pot afterwards I figured it would be be fine and it totally was. just now that pouring from the side becomes automatic I thought I would throw it in as a nod to the master 😊
From the specs it looks like JB weld has higher temperature resistance. But with aluminum filling this resin probably has higher thermal conductivity.
Now I wonder if adding alumininium powder to the resin would improve the thermal characteristics.
Amazing video. Where to buy the aluminum housings for which you make the parts? Thank you
I guess your pour hole was too big. That's very promising.
Really interesting stuff. The granate plate - was that a cheap option you picked up? I've used glass in the past. You can see if glass is flat by it's aberations - or so someone told me.
Does the epoxy shrink when cooling? If so, could that cause the mold to be a lot weaker and more likely to crack? Like, if the epoxy shrinks, then the pressure between the epoxy and aluminum might not be as strong and might give it a bit of clearance to "stretch"... but if it's a hard/brittle epoxy, then it will crack instead of flexing. It's just a thought and a question. I really don't know. I just thought about it when the mold was being made, when the shrinkage was being shown. If it shrank that much there, then I thought it might be possible to shrink (to a much lesser extent) on the inside after it starts to cure. I'm sure being thinner didn't help much either. I also wondered about the pressure used for the bubbles and was wondering why pressure and not a vacuum to remove the bubbles? (or vacuum then pressure).
Can you sinter it? I'm looking into various metal resin products that I'd be able to make a master mold of Stainless steel, sinter it and use that master as a repeatable mold for lower fired projects such as glass fusing and bronze. I'm told as long as the metal fill is 80-85% or higher sintering could work.
Thanks for this... I am gearing up to do a series of injection molds and I was going to try 3D printed molds and hadn't considered epoxy.
I have a Morgan press and will have the sprue in the top of the mold, not the side... Any thoughts on how that would change the process... thanks again, this is great stuff!
It should also work for a Morgan. You would just need frames where one frame as the sprue in the back.
Thank you! Nice and useful investigations!
Thanks so much for you video, i Want see the part 2 ... i have some question, firts im from mexico, second the Epoxy can be change for resin of 3D Print?, i can see you have a problem whit the plastic, maybe your problem is the time cold and no the presure, i tried make a simple inyection machine with 3D print (ENDER 3) and the problem was my Mold and no the press... and last question which one the same epoxy injection similar?
Another really good video, very educational!
Epoxy properties fall down with temperature, some can take more heat than others but you should really take a look at TDS for some numbers. Usually "high temp" resins have only a small fraction of their room temperature performance at the claimed max operating temperature, even some aerospace grade adhesives.
About voids, you should take the mold in the vacuum chamber before the pressure pot. Vacuum will expand the bubbles, hopefully they will surface and burst (may need some help with mechanical action / vibration if resin is too viscous). The pressure pot just reduce the size of the bubbles proportionally to the pressure applied, it will not pop them. Double the pressure, half the volume, and so on.
If you can arrange a good master/aluminium layout with a suitable resin inlet and vacuum outlet, I would try something like a vacuum infusion/VARTM process, it will ensure a nearly void free mold.
I tried using a vacuum chamber. The bubbles roles to the top, but didn't break. I tried to vibrate it a little, but didn't have a good solution. Even so, I didn't have issues with bubbles in the mold.
Sweet wingnuts! Thanks for sharing!
Do the plastic beads get hard if they're left on the heat too long? With PLA plastic (3d printing) if you leave the machine on with the hotend at printing temperature for a long time (several hours) without printing it will cause the plastic to harden like a ROCK and it's so hard you have to melt it out or burn it out. It can't be removed with physical force and it's probably not a good idea to try to Dremel it out.
Great video. I've noticed the same Z axis errors when printing a similar part "flat" on the build plate of a resin printer. I'm curous to know if anyone has an explaination.
You're supposed to put the model at 45° in two axis not just one. Also 45° may not be the correct angle for your printer and print. You should use the arctan for the correct angle. Some resins are better for more dimensionally accurate prints as well. I also recommend keeping the resin and print environment at 77°F for best results.
That was such a beautiful failure mode...
The issue you having with the 3d printed parts (the one printed flat or without supports) maybe due to the longer exposure time of the first layers.
As an armchair expert in mold making, I'm wondering, can you add something like nails to the epoxy resin to act like rebar? Though it might do more harm than good if its coefficient of thermal expansion doesn't match.
Put the hole to fill in the epoxy to a corner or perhaps even drill a hole from a side surface or two, one to fill in the epoxy and one to let the air escape.
this is cool.. the injection machine ... does it have a name ?
Great videos as always
Aren't bubbles in epoxy usually dealt with vacuum instead of high pressure? 🤔 I don't understand how could both work.
Id say cnc cooling channels in the aluminum before you permabond the 3d plastic and then add a regular pc cooling radiator with reg fittings etc. Kind of in the same way what nasa did back in the day with their rocket nozzles.
Vacuum bubbles 2-3x out of epoxy before pouring.. Add 1/32 milled fiber or micro carbon fibers, if available, to Epoxy. Definitely fully support the sandwich.. Let us know your success!
Good info! Thanks John!
is part 2 planned or will it be infinite part 1?
Try before vacuum spray surface of resin with mold release.
That sounds interesting. I'll have to try that.
Would it be useful to have some long mineral fibers in the aluminum epoxy to hold the mold together under pressure?
Very interesting. Ciao, Marco.
Non ho ben capito come mai non è stato fatto interamente in alluminio lo stampo a cnc ma potresti provare a rifarlo usando, invece che del gelcoat caricato per stampi, della resina mescolata con polvere di carbonio.
Hi I am planning to make 3D printed mold of for my injection machine with low volume. Can I refer the 3D resin used to make these mold ?
You might want to chamfer the edge i that hole. Sharp corner creates a stress point where the epoxy gave up
Why don't you fasten the wire rings with resistance welding?
Total newb here what are the brush things for? They make the parts with holes inside?
I discovered an issue with the channel listing. Part 2 is missing. It has been like that for quite some time. :)
I've been quite busy with other projects. But should be back with part 2 soon. Sorry for the long delay.
Hello. What is the name of the gray paste you used?
Thank you for the lessons!!
Lol oh maybe the huge hole in the back let the pressure out. Just a guess. Cheers for the vid!
The failure is at 45 degrees indication of shear stress
would a vacuum chamber help remove the epoxy alum air bubbles? works with silicone
what printer equipment and resin are you using?
Please recommend me an injection machine or CNC that are affordable, I want to start doing this.
I can do modeling well.
Thank you!
i wonder about concrete molds?
Hi what ratio is the epoxy to the metal powder?Thanks in advance!
Kind of making an educated guess here, but it looks like your resin 3d print is suffering from over exposure bloat. Not sure what printer you use and what settings you have and all that.
Very interesting