STRONGER 3D Prints with GLUE? (and CARBON FIBERS)

[@CNCKitchen]

Don't forget to share this video on Facebook, Reddit, Twitter and other social media!

[@allonyllc1879]

Do carbon fiber videos please.

[@Passco666]

n Hello Stefan.. Can you doing similar test? I would like to know what happend if you use infill let say 15% and into the space you inject epoxy?!

[@jamesbrown99991]

If you're doing CF tests, could you also please try reinforcing with woven cotton/polyester fabric/cloth/textile? It's much cheaper than CF, and literally everyone has some available

[@RomanoPRODUCTION]

Happy 2020 ❤

[@lucasvanbeers9906]

I have experience with using carbonfiber+epoxy and fiberglass+epoxy on 3D prints

[@andrewbiskup]

Single handedly the best 3D Printing KZbin channel. Why you're not at a million subs is mind blowing like the glue not strengthening anything. As a mechanical engineer, your contributions to the STEM community are incredible.

[@CNCKitchen]

Thank you!

[@fufun4me]

I never looked at his subs and just assumed he was massive bc of the use of his testing and how no brainer it is to watch his videos if you want to learn. Luck plays too much of a roll. Sad

[@andrewbiskup]

You must not be an "actual engineer" because we do use these for more than prototyping. We use them final parts often and are always trying to better engineer our designs in all industries so they can function properly whilst being intended to be solely 3D printed. GE metal 3D prints dozens of FINAL PRODUCTION jet engine parts and nozzles, not just prototypes and they're not DIYers. Being an engineer includes proving your design. People for years claimed putting glue into your part made it stronger, but never showed anything to prove such claims like Stefan did with simple tensile tests. Stefan made the effort to test a hypothesis the community itself was unsure of, and gave us very simple results so I'm not sure what other "raw data" you're looking for. He gave a load to failure for each instance. That's about as raw as it gets.

[@andrewbiskup]

Did you even watch the video? He posted data graphs several times with the exact raw data you speak of. No need to act like to a child when someone points out you making obviously false and factless statements. If you don't like a video, move on. Simple.

[@andrewbiskup]

So all his work and effort is useless because you can't read a graph and didn't get a table. Go find something else to complain about. Everyone else found this helpful bro

[@yucannthahvitt251]

When you further investigate your carbon fiber reinforcement, you should use the proper type of resin for CF work instead of hobbyking epoxy.

[@CNCKitchen]

Definitely. The working time of 15 min alone isn't optimal.

[@akinz3079]

I wonder if shoe goo/amazing goop etc. would help. Flex seal could also be decent

[@newdeathscope]

@@CNCKitchen west systems is my epoxy of choice, it works really well.

[@anonony9081]

@@CNCKitchen also try some sort of carbon fibre wrap so the entire print is reinforced

[@sneaky_tiki]

Outside of working time, I don't see it mattering much (if at all) what kind of epoxy you use for this sort of test. You're not trying to minimize the weight of the epoxy, you don't care about wicking properties. You basically don't care about anything, because the plastic will always break, and the carbon fiber never will (in his test jig). It all comes down to whether the carbon fiber is positioned correctly for the loading scenario. Oh, the amount of heat generated by the epoxy reaction might matter, since it can deform the plastic, but I don't know if that's something you can easily account for when looking at epoxies.

[@3DPrintingNerd]

Wow, didn't expect those results. This is fantastic - thank you Stefan! Also, was thinking, I wonder how something like 3DGloop would stack up against CA and 2-part epoxy?

[@rileyfenley522]

3D Printing Nerd I was just about to suggest PLA 3D Gloop.

[@SkansgardCNC]

I have used fiberglass weave and ca-glue to greatly increase strength of thin PLA parts on a mask. Works very well! :)

[@Bronze-td8jq]

Can you please give more detail on how you accomplish this? Am very interested. Thanks!

[@SkansgardCNC]

@@Bronze-td8jq Pretty simple; I wet the part with ca-glue and push on fine fiberglass weave and push it on with a small brush so the CA can soak the fibreglass before it hardens. If I need more layers, I use activator between layers. This is probably not as strong as epoxy-infused glass fiber, but a lot better than plain 3d-print. Also might be a good idea to sand between layers if you want it to look good. I have used it inside masks etc so looks have not been important :)

[@jcugnoni]

3d printing + composite is really the way to go for high strength parts. I would love to see more of this.

[@gadjetsvideo]

Ever thought of printing the hooks hollow and filling the inside with resin, maybe through a hole left in the top. I've often wondered if it would be possible to add some resin filler during the printing process with a very low amount of infill.

[@g6qwerty]

LOL I just thought of the same thing now.

[@frankbauerful]

I use woven glass fibers with epoxy. Unlike carbon fibers they are transparent and almost invisible. Because the fibers are woven they do not have a weaker orientation.

[@AntiVaganza]

Sorry, just saying "woven" doesn't give us enough info. Perhaps you are mixing up chopped strand fiber mat with woven which does have less specific fiber orientation but have other issues. Woven for sure will have weaker orientation...

[@Flav_M]

And are more affordable at comparable results (for us mere mortals is good enough), and very easily acquired locally at auto parts stores. That's defn bang for the buck results. Makes PLA not such a crappy spotter at GYM afterall...

[@truantray]

But CF is much, much stronger than glass. A cheaper alternative is aramid fiber, the generic name for Kevlar.

[@jakegarrett8109]

okleydokley technically there are some fiberglass nearing or equal too average carbon fiber (higher stiffness and ultimate load strength), I’m sure it’s probably more expensive than CF and CF just has a cool aesthetics factor! I’m not sure of any industry besides aerospace that uses the stronger ranges of fiberglass. Kevlar like you said is an excellent material and well suited for some other tasks as it happens to make fantastic flexible hinges for composite planes (when making a UAV jet plane, we used a semi-monocoque design so the outside was carbon fiber and then we used a little strip of Kevlar as the hinges for flaps). Fiberglass is so much cheaper though so I understand why people want to use it, definitely better than nothing! It was like $3 for a decent piece of extremely light (like 100 grams per square meter, normal checker pattern stitch) fiberglass when I bought some recently, but CF you have to pay out the nose for that stuff (the closest I have on hand is some almost ghost weight, it’s like 20 grams per square meter and it’s like a random squiggle mesh that’s so thin and hollow you can see through it, basically min weight possible that’s useful, so you can make incredibly light weight slow flyer planes, but the cost is insane). I’d say it’s about 20-50x more for CF cloth than conventional fiberglass cloth, but sometimes the project is mostly about man hours and mold times and it may end up being only like 20% extra cost in the end, so like for this video it was so small of a patch like what’s the difference between 1 vs 2 pennies? Nothing basically, might as well use the best one. Big projects you really gotta analyze if it’s needed!

[@weareallbeingwatched4602]

@@jakegarrett8109 glass fiber can be proper tough as nails.

[@dipt_tpid]

I love the effort you put in these videos. Well done

[@juststeve5542]

Very interesting Stefan, especially the epoxy actually reducing strength.

[@glennleader8880]

It might be worthwhile putting the parts into a pressure chamber and a slower curing time for the epoxy.

[@martinpirringer8055]

How I strengthen is to use the part as a skeleton and wrap it in fiberglass

[@martinpirringer8055]

@@RazgrizDuTTA Currently brush and squeegy (spelling?) Still experimenting to find the best way. Parts are 12to 16inches in some direction. so pretty big. The final product will be 130-150 lb 30x30x48 in thereabout robot.

[@martinpirringer8055]

@@RazgrizDuTTA That is what the experiments are for. Been looking for something but epoxy to use. EPOXY on PLA is not good as its exothermic and can cause distortion. Its a pain too as you got to mix it. On ABS you can use ABS slurry make it as thick as possible where you can still work it through the glass and soften the part up with acetone then acetone weld the ABS slurry soaked glass to it. now Gorilla glue adheres well to HIPS and ABS but makes the glass brittle as its extremely hard and seems to crush some of the fibers. On foam core Polycrylic works well also on cardboard and wood. On cardboard and wood Elmers glue all works well too. But 2 part poly ethylene or epoxy is best but you got to work quick. I will still run some tests with certain jb-weld products that have been specially formulated for HIPS and PETG. So atm its still work in progress to find the best solution possible

[@pepe6666]

you got me thinking about merely inserting strong stuff into the print. and i think m3 bolts are strong. we often think of using bolts to join parts together, but merely having inserts for a long bolt might be worthwhile to make strong segments where parts take a lot of punishment.

[@martinpirringer8055]

@@pepe6666 We insert all kind of things sometimes even aluminum tubing. You can see an early prototype move here. www.chiefdelphi.com/uploads/default/original/3X/3/6/36e8fbdd064daf1db337ffb20366218e0d2115d2.mp4 I think at that point it was about 9kg of filament reinforced at some places with 1/2 in 16 gauge aluminum square tube

[@mauriziocastrocastro4538]

@@martinpirringer8055 Does adding the aluminum made the prototype significantly stronger? If yes, do you have any documents or links i can read? I´ve been looking for ideas to make my prototypes stronger so if you have any information, let me know.

[@ajhartmanaero]

I own a company that manufactures carbon fiber car parts and would be extra interested in any carbon fiber tests you do. I just picked up my first 3D printer and have some ideas of how I want to marry the two techniques.

[@FilmFactry]

I've printed a PLA spline for a gearbox hollow. Then poured epoxy and some metal nuts as filler. The epoxy becomes super hard, and there is only 2 plastic shells.

[@Flav_M]

cut up some glass fibers and mix in with the resin

[@shenqiangshou]

Would love to see more tests with the carbon fibre, both the continuous strand and the normally available chopped CF infused PLA. Thanks!

[@DanteYewToob]

A cool technique I recently saw on reddit, was a guy who prints large figures and stuff with only two walls and zero infill and then fills them with epoxy by slush casting the inside with a few layers of resin. He drills holes in the bottom, injects thin resin and moves the part around to coat the inside and let's it settle in the smallest or weakest part so they become solid. I'm very curious how well this works and if it's worth it! Faster and stronger prints, potentially?! I need to know.

[@bigbearnelson]

I would love to see you test different methods of gluing pla parts together. A practical application being a full size Marvel mjolnir replica that's printed in several parts. CA glue, epoxy, JB Weld, PVC cement, stuff like that.

[@nightfighter78]

Excellent wideo again! I was sooo waiting for extra hook with longer carbon fiber reinforcements! That seems to have good potential.

[@ammerudgrenda]

Yes, I was waiting for that too.

[@Andreas-gh6is]

For some applications it might be a good idea to design 3d printed shapes that are optimized to be reinforced with carbon fiber material. Maybe spiral grooves. Maybe a tunnel. Also, the horizontal orientation fibers still confer some strength in the vertical direction, because they have a thickness and aren't perfectly horizontal everywhere.

[@patrickfle9172]

Ich benutze gelegentlich cfk verstärkte Teile aus dem 3d Drucker oder gedruckte Formen für die Verarbeitung von Carbon/Epoxy Für Funktionsteile mußt Du für beide Verfahren ein Bisschen umdenken. Roving verstärkte Teile profitieren von Nuten, in die Du die Fasern einlegen oder einwickeln kannst. Hierbei kannst Du auch Haken und Pins andrucken, die beim Wickeln helfen. Zu enge Umlenkradien lassen die Fasern brechen (!) besonders bei HM-Fasern. Ich habe auch schon Faserträger gedruckt, bei denen ich den Kern anschließend entfernt habe. Beim Formenbau must Du, anders als beim Verstärken, darauf achten, daß das gewünschte Teil _nicht_ fest mit dem Druck verklebt. Hier ist dann ggf. ein Trennmittel erforderlich, dessen Lösemittel die gedruckte Form nicht anlöst. Deinen Haken könntest Du praktisch im CAD von einem Formblock subtrahieren und diesen dann geeignet splitten. Diese Negativform kannst Du dann mit harzgetränkten Fasern füllen, die Formhälften schließen und alles härten lassen. Nach Entformen und Besäumen hast Du dann einen massiven Haken aus CFK. Kannst mir auch gern die Datei für den haken schicken und Ich mach Dir die Form.

[@Saanq]

You can also try a glue which has ethyl acetate as solvent, since it also dissolves PLA and should make an inter-linking structure.

[@MaplePanda04]

Painting the part with clear nail polish would work. Acetone-free nail polish remover also contains the stuff if I remember correctly.

[@ddkessner]

Things to try: 1. Same hooks, but physically thinner. Printed at 100% infill. Ideally they should have about the same amount of filament as the hooks in this video, but packed into a thinner hook. 2. Take the hooks from #1 and submerge them in a slow and runny epoxy. Put that into a vacuum chamber so you can pull the air out of the crevices in the hook and replace the air with epoxy. Remove from the chamber and hang the hooks up to cure, while letting any excess epoxy drip off. 3. Have some holes going through the hook. Fill the holes with epoxy and carbon fiber threads. 4. Same as #3, but use normal fiberglass fiber/threads. Thanks and good luck!

[@MacroAggressor]

I really like your first idea, testing different cross section geometry using the same amount of filament.

[@dougingraham5807]

Your results were as I expected. The breakage at the end of the CF reinforcement was perfect. I started using CA in 1973 with my RC models. I stopped using it when I became sensitized to it in the early 1990's. A single whiff of it will give me flu like symptoms for several days. I miss using such a useful material. Thanks for all the testing.

[@fabioarrua]

P100 respirator with organic vapor cartridge can get you back in the game

[@Dustmadeout]

Could you also try making a channel inside of your part and fill it throught the hole with epoxy resin?

[@ammerudgrenda]

That would probably make it heavier without increasing the strength.

[@aronderksen7055]

Or you could run some carbon fiber thru to hole with resin.

[@Daniel-nm8vi]

Stefan YES, Please YES! Yes to More tests of increased strength, stiffness and impact resistance created by surface or buried (pausing the print) epoxied carbon/glass fiber roving (& fabric) reinforcements. I think this is a quick, simple & cheap method to GREATLY improve the real life functionality of 3D printed objects.

[@pioneerAv]

I wonder if you could impregnate a print with resin similar to wood with stabilizing resin. Perhaps you print in such a way to make voids that are conducive to this.

[@mn1816]

You should try with a solvent glue, like Weld-on #3 (Methylene Chloride based). It works on PLA and ABS just as well. Due to its really low viscosity, it does get in between layers and gaps dissolving and welding the plastic.

[@kurtlindner]

10:35 had me rolling, that delivery. :D :D Please do some more with the Carbon Fiber reinforcement.

[@frollard]

to add another data point - I had some prints that I rushed (making a prop rifle for a cosplay)...had very very bad layer adhesion because of the print speed I used. I had to reglue them together with CA and epoxy after since they wanted to split constantly. You are trying to improve a well calibrated print and seeing minor results. I don't want to recommend it as a crutch, but to a bad print, the glues are absolutely a life saver.

[@skyak4493]

This vid did a fine job of showing there is little improvement from a glue coating. Stress risers are critical to real world strength. I would like to see more with composites but it is a very complicated topic -maybe not for general audience. For example: -PLA might not be compatible with common composites because it shrinks after printing and is uniquely amorphous. Or maybe the one flaw smooths over the problem the other creates. A good test would be to print, then glass the exterior, then "anneal". If the composite does not survive this basic test the components are incompatible. -ABS may make a more compatible core -the adhesion to the core is the big challenge and difference in stiffness creates a stress riser at the junction. Carbon fiber is least likely to work. Glass is much closer and even polyester or nylon fabric might work better. The question on every test should be where did it fail? When the sandwich stops failing at the junction you can start optimizing -shell for strength and stiffness, core for shear strength.

[@archeebookingagent1909]

i did similar test but not in 3d printed world.. the epoxy resin will heat up while curing and therefore will affect the PLA bond but epoxy itself is not a structural component.. it is sort of keeping the fibers/ filament together .. CF will will perform however it twists (fibers have flex) depends on the type of epoxy you use.. If you could get a hand on vacuum / release / pump system (foam and core with PVA and make the fibers woven and fully impregnated it would probably damaged your jig.. same might happen if you use a layer of woven or unwoven fiberglass or kevlar or hybrid.. .. Carbon fibers themselves work as unit .. if there is a hairy crack it all goes... and also repair of CF is very difficult and usually 1 cm crack requires 10 cm repair... good job Stefan.. thank you for all these.. I like to watch this as I have done same things just using weights and springs 25 years ago ... mostly with wood during boat building

[@IscuAndrei]

I've made a few parts with carbon fiber reinforcement for my airsoft replicas (variations of ergonomic pistol grips). Since I printed them standing I was worried about layer separation. The method I used was coating the part with resin, wrap the twill on the pistol grip then use peel-ply and a vacuum bag. The first grip had some torture tests involved hammering on all 3 axes, loading the part with 10kg and heating it to ~90 degrees centigrade, etc. It survived so well that I sanded it down and applied a thin coat of resin and after a polish it was mounted on my replica. I presume the vacuum helps the resin flow in the small spaces between the layers while the carbon shell gives it good impact and shape retaining performance. However, without vacuum, I cannot see how the resin could get in between the layers considering its viscosity and surface tension.

[@adriFernandez]

It would be awesome to know more about the effect of the carbon fiber reinforcement. Very nice video. You are one of the main information sources for 3D printing. Keep pushing.

[@Flav_M]

The timing is so perfect for me, just picked up an ender 3 on 11:11 sale ... Told myself I wasn't allowed one until I learned how to get started with Fusion360(since 99% of what I'll prob print is functional stuff and tests of them lol). Having never used ANY design or altering software(windows paint doesn't count for dots and lines, although...nevermind) I still have a huge mountain ahead, but I got as far as designing with threads and most of the basics, I'd love to get into assembly and fitting parts together, seeing stress points...basically the good stuff that you DO :D And that brings me to what I started to say.... AWESOME content, exactly what I needed when I needed it. I was just bitching last night in chat during Joel's giveaway that PLA is such a disappointment when you need functional stuff because of layer separation and then you give us this. I'm assuming resin filled is next? If not, It Should be ... that's my next trial for my functional junk, PETG filled with warmed up resin to flow nicely into infill. I've got some two ton stuff laying around somewhere from a project a while back, that should do nicely lol Jjust ordered a roll of petg, since now after a month of testing I've gotten things really close to dialed(just did a complete thorough check over the WHOLE printer and Glad I did, just about everything needed an additional turn with the wrench and the bed rollers were actually never aligned 100% and started wearing funny, but I detected it early and greased it all. HUGE improvement since my first print where I scratched the aluminum bed...I guess I'll forever have my very first imPRINT :( so i'm ready to try PETG and if that goes well maybe some tougher stuff . Once again, awesome stuff, will be watching it all. Can't wait to see what else I get to learn. Happy Holidays and a Super New YEAR!!

[@iteerrex8166]

Hey Stefan, I have had this idea for a while. For prints that are thick enough, one could print a cavity in side the object and insert a piece of metal. It could be an AL rod or some spring steel and such. It would be hidden and add tons of strength. I haven't tried it but it should work good. Thanks for the video.

[@Flav_M]

or fill with epoxy the entire cavity ;)

[@Anyone700]

You would not fill with epoxy because it is an exothermic reaction and would melt the print out of spec. Also it is too brittle. What you would do is model a thread into the 3D print use matching size threaded rod. Run the rod through to get the hole to spec then put a glob of epoxy on the hole and a line up the rod. Thread it all the way in, then trim flush with a cut-off wheel.

[@iteerrex8166]

@@Anyone700 Yes, that's the idea, of course the details would depend on the situation and application. Good point about the heat. And brittle not bridle.

[@Flav_M]

@@Anyone700 That is very very size and shape specific. Using slow curing(or low heat generating) epoxy it'll be just fine.

[@davidstonier-gibson5852]

I bought a load of cheap 3, 4, 5mm drill bits once off eBay, and have used them for exactly that.

[@lucas3820]

Very happy that this channel has English subtitles, I can "hear".

[@TheAndonen]

Stefan, watching your videos makes me feel I was switched to the Discovery channel (or similar) and watching high production informative and interesting scientific episode. Appreciate the ton of work you did to make it happen.

[@cefoltran]

I abandoned super glue several months ago when I did an unpretentious test. I was making a structure out of PVC pipes for a monitor stand (www.thingiverse.com/thing:3679144) and I decided to test if the glue I used on the pipes would work on PLA. For my surprise, I found this PVC glue to be far superior to super glue. Not only it gives a strong bond, but also left no white spots on the parts. Moreover, because it drys slowly, parts can be repositioned. This type of glue is particularly useful when one needs to join parts that fit inside one another, like pipes do! While liquid, the glue acts as a lubricant and can facilitate the fitting of tite pieces. I also reinforced some parts by coating it with PVC glue and them wrapping a layer of PVC film over it, analogously what you did with the carbon fiber. I suggest you try this kind of glue.

[@papamidnightfpv]

Hey, here's something I picked up from drone pilots. Super glue can be hardened instantly using baking soda. Cyanoacrylate and sodium bicarbonate. They say it is harder than glue itself. It's for fixing broken blades. A broken propeller blade is similar to a layer adhesion problem on a thin part. If it is actually that strong it might make for a good video. Love your work

[@UNVIRUSLETALE]

Top quality videos, you definitely deserve millions of subs

[@grahamdwells]

Good job with the squarespace promotion quick and to the point. Progress bar was a very nice touch too.

[@19mitch54]

I grew up on a beach in Southern California and surfed since I was ten. My best friend's dad tried to make him a surfboard by applying fiberglass and resin to a cheap styrofoam board. It was a disaster. The resin dissolved the styrofoam. (Polyurethane is use to make surfboards.) I think anything that dissolves in acetone will also dissolve in epoxy resin. That may be why your samples were weakened. Maybe, after the resin hardened, the sample would continue to get stronger as the solvents evaporate. I have seen in industry where vacuum is used in a process called "wet out." Maybe you could try a filament that doesn't dissolve in acetone. You may try to approach your designs as 3D printed forms to wrap resin impregnated carbon or fiberglass cloth over where the finished fiberglass/carbon would bear the bulk of the load and the 3D printed part makes most of the shape.

[@Mike01Hu]

Another classy video! The failure mode that hits me most is delamination so using carbon fibre is a probable solution. The CF filled polycarbonate filament is OK but still weak in this plane. I had considered putting channels or holes through the high stress points and filling with potting resin with CF laid in or pushed through. I do use holes with welding wire pushed through with epoxy and that is very successful. Time is my enemy!

[@CAPTAINCAPSLOCK111]

Worth revisiting? YES! I would even go so far and think about a technique that is somwhat similar to making magnet or metal inlays, but with CF. I.e. Stopping the print, insertig the fiber and then print on. But you would need a method of wetting the fiber then...hmmm..

[@Falconguygaming]

I've been using super glue even before 3d printing as a crude additive manufacturing technique. Even if it doesn't add much strength it makes for an easy smoother before paint

[@МУЅТ]

I’d love to see this done again with a .6 or .8 nozzle with significantly larger layer heights. The larger heights would leave larger grooves or ruts inbetween the layers (which is what attributes to more visible layers/layer height) thus giving a substantially increased amount of surface area for the glue. I don’t think you’d see a huge difference in CA glue (maybe the thin) but I think you’d see a lot in the epoxy. Also would be nice to test multiple purpose driven layers of epoxy on a a single print. For instance Smooth On has a huge variety. Their TASK™ 9 ColorMatch has a tensile strength of 7,800 psi but their Free Form AIR HT which is an epoxy putty with a post cured HDT rating of 204• C. So potentially being able to add specific traits to the print via their respective coats of epoxy. Maybe making the outer most coat be the hdt high temp resistant idk. But even just the putty bc you can work it into all the layer line grooves and scrape off the excess flush with the print. So you wouldn’t even be changing the models dimensions really. If anything making them closer to the original model.

[@belenhedderich3330]

Thank you Stephan to keep on doing research on this subject Merry Christmas!

[@armoth6958]

Hi Stefan, great video!! I’m wondering what would happen if the hooks are printed with a pre-determined hole in which you can inject epoxy or resin to “fill” the stress point(s).. maybe a stupid idea but I share it anyway

[@aus109]

Dont forget coating your prints will also add heat resistance

[@yonutz333]

Yes, i would be interested to see how carbon fiber reinforced parts behave and how strong they are

[@WH1T3_No1SE]

how about printing a hook, leaving hole on top, and filling it with glue or epoxy afterwards

[@e_g4239]

Like a faster way to print 100 % infill maybe it's better

[@samuela6271]

Solid epoxy is very brittle and heavy.

[@BennXdesign]

@@samuela6271 and kind of expensive, but it may worth a try for certain parts...

[@BloodSprite-tan]

if you're going through the effort of doing that you're better off making silicone mold and making an fiber glass epoxy composite hook in the first place. epoxy is very strong, but it's not mention for structural stability, and as such you need some sort of reinforcement.

[@RevanPlloyd]

I was thinking something along those line as well - Epoxy - Epoxy with fibers (the short fibers carbon/glass) - cement - plaster (All off witch have the draw backs do to the way there cure. Exothermic , Oxygen, water evaporation) Like cast In place, but you keep the mold on. Also it wold make for a good exploration of infill and infill modification, seeing as what ever material would have to flow all the way through the part or to specific locations. I have had some luck use epoxy to increase compression strength on a print.

[@mitupadhyay1451]

I think you should try with the change of temperature effects on the prints with epoxy. Because even if it doesn't improve strength it must affect the temperature behaviour and also hardness. About CF it is very interesting the failure mode where it fails at the end of the fibers. But there are also law density mesh which will add a little material but the strength can be much better in both lateral and transverse direction. Thanks for the video, loved it.

[@court2379]

I think simply sanding and polishing a part will show there are no voids to fill with glue. Try an abs part. Dipped or vapor softened in acetone. I am wondering if you can eliminate the difference in properties in each direction.

[@dominiksamol9068]

I love this series and think it can be a very good basis for actual construction with 3D parts involved. For the fibers. You already did simulations where virtual fibers or reinforcement should be added. If you leave a actual a canal or whole where you can pull the wetened fibers through (almost like sewing or like the steel enforcements in concrete, you will get an improvement and still have the clean look. Of course for perfect results you need to make sure the channel is full and also that there are no air bubbles. For a project I used already carbon, resin, and 3D printing for a wind turbine with 2m diameter. Iff applied correctly you can have the quick results and "perfect form" of printing with a strong and usable result

[@wordreet]

I'm going to suggest that glass fibre roving would have a similar effect to the carbon fibre. CF is used primarily for it's linear stiffness once it's epoxied into a component. That's why I use carbon fibre tubes, and have also used CF roving, in the wings of the foam gliders I build. But woven glass fibre cloth, or even GF tissue, (which is omni-directional), will make a BIG difference to the layer strength test results. Though, as you discovered, you would need to cover the hook past the point of suspension with the fibre. GF adds strength cheaply, and Kevlar is even stronger, though waaaay more expensive. People! Please be aware of the safety issues involved with glass/carbon/kevlar fibres. These materials release microscopic fibres when handled, so please wear some safety gloves, a mask and eye protection! Also ventilate the area if you are using epoxy or polyester resin! Oh yes, that reminds me Stephan, polyester resin is never as perfectly rigid as epoxy, so it may not suffer from that fine cracking we saw under load. So it may last a little bit longer in your tests. Heh, it's also cheaper.

[@jamesanderson2381]

Another great video, thanks Stefan! As you correctly noted the epoxy has a lower failure strain than the filament you used which has a more plastic failure mechanism, I wonder if regular PLA would be a better match to the failure strain of this epoxy? Alternatively there might be a toughened epoxy with a higher failure strain that could do a better job also.

[@iceblu3710]

A local startup is using 3D prints with the bottom shell print missing. They flip the part and fill with UV resin to make a solid part. Its only half as messy as just making a resin mould but you could easily inlay a strip of carbon fiber while adding layers to cure.

[@tronskifpv]

thank you for doing all the technical videos. These are the types of videos I really appreciate. I want to make the strongest prints with the most appropriate materials I can.

[@elijahsimmons2900]

Would LOVE to see more composite reinforced 3d printing. That's what I'm working on now for my robotics team.

[@pirobot668beta]

I played around with a material call tulle; an open mesh fabric used in garments. About 10% fabric, 90% holes. Anyway, it can be layered into a print to make a support mesh for 'chain mail', without the links! Pause the print, slip in a piece of tulle, resume print. The holes in the fabric let the plastic 'print through', incorporating the fabric into the print. A search for 'carbon fiber netting' pulls up a few promising leads: mosquito netting with high carbon % and bound in nylon...might be a thicker than normal layer, but should be strong. Y'all not afraid to edit gcode... The Holy Grail would a netting form of carbon/kevlar.

[@AS-ug2vq]

Fill the hooks with epoxy, superglue, cement, PVA glue. Create channels in infill for the filler to fill inside in different patterns and one in solid.

[@BladeWalker91]

I once repaired a scissors with locktite super glue and a mesh of thin shielding wiring, that I took from some coaxial cable lying around. I embedded wiring into a plastic using a soldering iron. Scissors are still working to this day for about year and a half. Maybe you can use similar method to add more strength to your 3D prints?

[@WVRetreat]

I use CA to strengthen wood parts and one thing that is important to make sure the CA can wick between the wood fibers. In your experiment, I don't think the CA is able to appropriately wick in between the layers. I would suggest printing gaps into the layer to ensure maximum penetration of the CA. You'll also need to use thin CA for the best penetration.

[@foxxo-dev]

I love how you are so intrested in these topics. You deserve to have more subscibers!

[@KiR_3d]

Hi, Stephan! Wrap all stress zones (or over the whole part) with 2-4 thin layers of carbon fibers placed diagonally: 45, 135 deg. Similar method is usual for composite construction parts in the aviation (they use not carbon fibers but more sturdy materials as it's vital). I think you'll get about 200-300% increased strength (with my very modest opinion). IMHO it's better to reinforce your testing equipment before this test as well (especially if you don't know how much it can stand with).

[@dylanvadakumchery6747]

Wouldn't solvent cement be more effective as it helps "fuse" the layers together?

[@goatgoatgoat1234]

I've been using XTC3D from Smooth-on to reinforce parts for a few years. That combined with proper print orientation seems to have helped increase load. I'd be interested in how that performs vs this epoxy. Great vid! Always enjoy seeing the results of your experiments.

[@JSCB-365]

I see a lot of people asking you to do a full video on The carbon fiber as well as Woven fiberglass. Even one recommendation to add fibers inside the print, while the printer is paused. This could lead to prints with optimized geometry where fiberglass or Carbon fiber could be added to the internal structure of the print

[@mayoropl1]

Once I've printed PLA lever, it consisted of 2 parts. I used leftovers of epoxy - it was super liquid and super runny, and full cure time was something like three days. If I remember correctly - it was clear epoxy for floors coating. All epoxy become absorbed inside of my printed parts (not outside as in your case) and all gaps between PLA "fibers" were filled with epoxy - capillary forces did their job. Guess different epoxy will give you different results.

[@AlvaroQF]

This video is really really educational, congratulations! You clearly show the importance of printing direction and how much do carbon fibers improve part toughness and the importance of the fibers positioning. I would have loved if you had tested 1, 2 and 3 layers in the inside side of the parts printed in XY direction and Z direction, this would have been superb!! Maybe an idea for a future video.

[@JanTec3D]

very good video, Gut gemacht Stefan!

[@nobocks]

what you can try : Pause half print, dump epoxy and wait to its harden then resume print. You can also try with others things like sands, cement, ect :)

[@Andreas-gh6is]

People have done similar things with filling the infill cavity with cement and the like. Usually after the printing finished. I don't think it's a good idea during the print.

[@syber-space]

Pretty sure he made a video on that already...

[@chloemcholoe3280]

People have tried leaving holes in their prints to fill with epoxy later and it worked very well

[@CAPTAINCAPSLOCK111]

Crazy. I just had the same idea.

[@ajmtranz]

I would be curious on how 100% infill holds up. I print functional automotive motorsports parts for race teams. I print all of these with CF-PETG at 100% infill.

[@danlaskowski5494]

One more thought about layer strength. In the wood industry, we impregnate wood with chemicals to provide insect resistance, etc. You do that by putting the wood in a vacuum and then adding the chemical and it gets sucked into the center of the wood. I've wondered what would happen if you tried that with ABS and acetone? Suggestion: put parts in a vacuum. Close off the vacuum and open a valve to allow in some acetone which will immediately boil into a gas and be sucked into every crook and cranny of the ABS part. Would that improve layer adhesion?

[@Runoratsu]

I usually coat strength-critical parts in a little CA glue, then add low stretch cotton ribbons while the base coat hasn't set yet, and add another coat of thin CA glue on top (soaking the ribbons). Pretty quick, cheap and uncomplicated, and strong enough for my projects so far.

[@alunmo]

have you tried filling the space in the infill with resin? maybe even squirt it in and then pour it back out.

[@asterchades]

It might seem a bit like I'm taking the mickey, but just hear me out on this one: give it a try with good ol' PVA/wood glue. My experience has CA/super glue with PLA being... not a good mix. I've been able to make it work but it just doesn't like to set properly, to the point where I can still move pieces around days after initial application. PVA glue, however, gives a very strong hold in short order, the same as you'd expect when using it for wood, quite possibly due to the organic nature of PLA. I'd also be curious about the "CA cement" made when combining super glue with sodium bicarbonate, since it seems like an easy and accessible way to potentially increase the strength of the super glue part.

[@Polymate3D]

Have you considered printing your hook hollow and injection resin / concrete etc inside and compare as it should remove the Z axis weakness in a way that we can easily be done. Possibly you could mix the filling with your variable infill method?

[@MikaelJSandersson]

Yes, please give us more carbon fiber improved stuff! :)

[@ewildgoose]

Try using a vacuum chamber. It seems to draw the glue into the voids in the 3D print. Another video here used it with just PLA I think to make a print airtight. With a low viscosity ca or epoxy it would then be bonding the actual layers?

[@laser162003]

HIGHLY reccomend GFlex epoxy. it bonds really well to the plastic when I tested bonding an 8mm solid aluminum shaft and carbon tube into a hex part. the shaft and tube failed before the bond. GFlex is not as brittle as most epoxy and works so well, that it has become my go-to

[@vonscherfarms925]

i know this is an old video but ive reinfocred my pla prints using countertop epoxy since it has a high ipact and heat resistance. i cant speak towards layer adhesion but on my glock frames it makes them last significantly longer than a standard PLA glock frame

[@Wezix1]

Hello Stefan. PLA would not be the material of choice for me to test glue strengthening. I usually use CA glue on pure ABS prints where i know a certain part will endure extra stress. I would choose to experiment on more plastic based materials like ABS and PET.

[@mocmaniac1571]

Nice video, to complete the test, could you compare a thicker printed part with equal outside dimensions as the one covered with CA glue ? And further more one with continuous carbon fiber all around the hook ?

[@TheLinuxBust]

Thank you for the video! I really appreciate the animated graphs next to the test videos. By the way, at 5:56, was the text supposed to say "Thin CA Glue" instead of "Refer[e]nce"?

[@TeskesLab]

I know this video is old now, but I wonder if the same test could be repeated for ABS instead of PLA. Mainly because both are reactive with Ciano-acrylate compounds. I usually use CA glue for making my ABS parts shiny and to vanish with the layer marks. I notice that just a drop of CA glue is enough to melt the surface a bit (the remaining CA glues that drips from the part, drips in the filament color). So I wonder if that might actually increase layer adhesion / resistance for the parts since it actually melts the part (although just on surface)

[@ginger404]

Have you tried fine fiberglass mesh and glue/resin that's intended for artificial fingernails? It's strong, flexible, thin, lightweight, and with proper application crystal clear and smooth. You can sand and shape it, and buff it to a high polish. The mesh itself can be purchased in uncut form (not preshaped for nails) and the best has a light adhesion so it stays in place as you apply the glue.

[@ge2719]

What about designing the inner support structure to be hollow with a channel throughout so you can fill it with resin? A hole at the top for pouring and a hole or two for air to flow out like on a mould. Would need to be something that doesnt get too hot.

[@Atimar01]

will internal support help? if you put diagonal structure instead of infill in the straight part of the hook, it should provide extra strength and since thers no infill there, it should use almost no extra pla. internal support should go from half way top to half way bottom side, as center support. 2 or 3 above and below that 1 with 2mm distance should do. obviously for flat printing....

[@EasyThere]

Larger gaps on the inside and pour the glue/resins on the inside to fill in the gaps. if you want to cap the ends, print the end caps separately, then fill the gaps with the glue and dry. To attach the end caps use a light dusting of baking powder.

[@xSensu]

GEEEZ, i'm in awe of how nice that printer is printing. I have 3 printers and cannot seem to tune them in to look even close to that nice.

[@ipccleaning2542]

Have you thought about using minimal infill and adding a resin or glue into the shell through a hole?

[@MaximilianonMars]

A commenter who does restoration work and repairs on plastics commented that different mix ratios of the 2 parts in the epoxy result in different finished properties, example: they said to match the less brittle ABS you need 1:1.2 binder to hardener so the material properties match. Using more hardener in your tests may remove the cracks and stress risers you talked about as it increases the flexibility of the finished epoxy. I am taking this on faith, I haven't tested it myself yet.

[@MaximilianonMars]

Found the comment on the video "Easy Plastic Reconstruction and Repair" by Tech Tangents By AkBkukU. Siana Gearz commented: "...I do a lot of plastic repairs. Thin reinforcement on the inside with a superglue and blotter paper composite, or same with epoxy. You can also get "plastic glue" which is bog standard CA glue and a felt tip pen to use beforehand which does something to the surface of ABS and other Styrene based plastics, not sure what! For epoxy i use UHU 1:1 epoxies personally, both 30-minute type and super durable 24-hour type, and usually i heat the parts with a hair dryer while the epoxy is setting or throw it on the hot bed of the 3d printer - curing it hot (up to about 80°C) makes it much more durable for some reason and speeds up curing, also helps kick off a reluctant mixture. The magical thing about good epoxy is that you don't necessarily get a super hard brittle material, you can vary the proportion of binder and hardener to achieve a range of mechanical properties. Adding about 20% more hardener than binder (which is actually hardener-epoxy-plastifier mix balanced such that it won't cure on its own) will match ABS plastic flexibility remarkably well, which can prevent breakage in the future. But adding more hardener makes it more difficult to cure, which is where heat comes in. You can also tint it in a million ways for seamless repairs..."

[@justinwebster667]

Please further investigate the fiber reinforcement, I would love to know more about it!

[@elvinhaak]

I wonder: what would the strenght do when you make a small opening in the part and fill it up with a syringe with epoxy? I guess it will 'glue' the infill together ?

[@spinnetti]

I was thinking to paint PLA parts with SLA resin both for strength improvement and also to fill the surface to get a better sanded surface.. haven't tried it yet but plan to do on a R/C airplane spinner maybe tomorrow.

[@microdesigns2000]

Yes, you should revisit, wrapping the fiber from top to bottom.

[@cleverbast594]

You should try out different fiber reinforcement. Carbon, glass, kevlar, and cotton are what I can think of off the top of my head. There is also woven vs mat vs single direction. If you want to go even deeper, there are resins that are designed for making composites from different materials. Anyway, thank you for your wonderful channel! It's really nice to see these tests done scientifically!

[@Piccyman1]

How about making a small hole into the support material and fill with resin

[@Buchsbaumschere]

I've had great success printing parts without top layers and completely filling it with resin. Works best with gyroid infill when there are no trapped areas inside the infill.

[@bigbearnelson]

Not hating but aren't you effectively making a 3d printed mold for a resin part?

[@theartandscienceofflight9239]

Nice test to see done. I have used multiple layers of thin CA to strengthen parts printed in PETG. I would be interested to see if the glue coating affects different filament types differently. I was really hoping to see the carbon fiber test as a clear winner. Would you consider repeating the test with a longer roving? Basically to cover the distance between the attachment points in your test jig? Thanks!!