Several of you have pointed out that I was quoting numbers in newtons, but the scale was set to pounds. That was my mistake. I set up the scale to read newtons when I assembled my test setup, but later checked the accuracy of the scale using my gym weights, which are marked in pounds. Apparently I changed it and did not set it back. I did scan through the footage to verify it was set to pounds for all of the materials, so the comparisons are valid. Sorry for the confusion.
The biggest advantage of the Carbon Fiber filled filaments is the strength to weight ratio, not it's raw strength. A part made from fiber filled will be lighter, with a similar strength.
Freudian slip? Hungry and thinking about Fig Newtons cookies. I print a lot of PETG... so more research into the wide ranging results would be of interest. I mostly use it because it gives a lot before breaking. Thanks for the testing.
You should try those tests with HTN CF25, which is high temp nylon with 25% carbon fiber, it's also known as black aluminum. I just started using it to test the tools in the 3d printable tool set that I'm selling. I tested a 7/8ths socket made of CF25 on a wheel lug nut and it was way stronger than PLA. The socket printed in PLA got up to 14.4 foot pounds of torque but the CF25 got up to 50.1 ft lbs and that was printing it without the suggested temp settings. I printed it on a CR10 v3 which has a max hotend temp of 270c and the suggested print temp for CF25 is 280c - 310c. I'm currently waiting for an all metal/bi-metal heat break to be delivered so I can upgrade my printer and try printing the CF25 at higher temps to see if that makes it stronger. Even printing the CF25 at 270c though, it still had a metal sounding ring to it when you hit it with your fingernail.
There is a variable that you missed here. We use 3d printed plastic parts for combat robotics and do a lot of testing of different materials. Our tests have shown that the pigments in the plastics make a huge difference. For example, the white ABS likely uses Zinc Oxide as a pigment which greatly reduces the strength of the plastics... while the carbon that is generally used to pigment black filament doesn't affect the strength much. We've been in contact with several of the filament manufacturers, and their engineers have backed up our findings as well. If you want good consistent results, you should probably test all of the parts using the same color... preferably "natural" undyed, but that isn't available for things like cf filled or some of the better flavors of PLA+ Our testing so far (for impact resistance) has two colors of Duramic 3d PLA+ being the top runners... (Black and Red). These filaments print best at a bit higher temp (I use 225c) but it is well within the capabilities of even an inexpensive hobby grade printer like an ender 3.
I have seen the same in material testing of injection molded parts. Especially nylons, the lighter the color (gray, blue, white) generally the weaker and lower surface energy they have. Same reason if you are having bed adhesion problems, sometimes switching to a natural or black color will fix the issue.
Interesting that you mentioned zinc oxide. What's your source? To my knowledge titanium dioxide is the most common white pigment, when opacity is desired.
for strength id say pla+ for short powerful repetitive impacts pa6 cf nylon hoffman tactical did a whole video on it considering what he builds with it for impact testing i trust his word, polymaker often hops on his streams and also agree
We did similar testing for our products and got mostly the same results. PLA is the best all arounder. But we have a temperature requirement and that ruled out PLA for us. We tried it just to make sure and just a little heat in an oven and oddly enough tossing it in the freezer made the parts deform to the point that they stopped working. During our testing we did discover a few key things that may be affecting your results. 1) the rate at witch you pull matters, that rate is based on the sampling rate of the scale. If you move to quickly the force will fail the part before the scale updates the reading. Giving you a bad reading. Consider using a Load cell and the good old Oscilloscope to recode the loads. 2) the angle of applied force matters, that angle must be consistent. If the forces shift slightly it can lead to the part being loaded favorably or unfavorably depending on the shape of the parts. Try a more rigid linear slide and add an electric actuator to slowly and consistently apply the load. We got ours off amazon. 3) print settings matter, they effect layer adhesion and that effects the ultimate performance. Your results are hinting at possible issues with controlling the printing conditions. Bubbles in the material and to cold of print settings will lead to huge swings in performance. Dry everything in the oven at 100F for 8 hrs and run it all on hot settings. 4)sample more than 2 or 3. these materials fluctuate naturally and any sample less than 10 is asking for misleading (biased) results. We have a saying. The first data set is the worst data set. Old filament may have issues, new is preferred, the temperature of the room matters. The more brittle the material the more it fails statistically and less predictably, and cold temperatures will cause this specifically for thermal plastics. The temperatures should be around 75F Or the operating temperature that the part will experience during it's life. 5) load the part realistically, if your part is going to hang something then make extra sure that as you load it the part is free to deform and move in the way it would in real life. Over restricting the movement of the part can lead to unnatural twist or bending that could cause early or late failure. This can often times be hard to do with things that hang. Consider a sand bucket free floating body test. Hang a metal bucket and put your load cell between the bucket and part. Use a funnel and slowly allow sand to fill the bucket till the part fails. This allows the load to move with the part, which can not be done with a constrained loading machine. put something under the bucket to catch it. like a stool or box so it doesn't fall far or tip over and spill. Good luck I look forward to seeing a follow up video.
Excellent comment, thank you for posting ! PETG must be printed properly to achieve the best results, and each printer must be setup to print it. As pointed out layer adhesion in PETG needs careful attention. The conditions of each filament roll are very important, age, how stored, dryness, etc. Nylon requires careful attention, I;m doubtful of switching material to do testing can achive that without careful tuning. I believe we ask too much of these printers, expecting the same machine to easily adapt to very differnet materials. Thank you for posting an excellent video !
Good points. I'd also add that there are also Tough PLA blends out there that may improve some qualities and some that have CF in them. For me, most of the CF materials I have used add no strength to the part and frequently make it harder to print. However, that matte finish is very very nice
Yep, I thought it was kind of obvious that the materials weren't all printed adequately. Delaminating layers was the give away imo. It's also expected - most of us have a favored material and use it way more often. Fresh filament, honed settings... yeah the prints weren't done with the same quality. So you're going to publish the results you tested right?
You can also easily anneal PLA to make it even stronger. I usually just anneal PLA on the printer bed at 120c, and put a plastic container over the part so that the heat stays enclosed close to the part. I’ve found doing that for PLA brackets make them soooo much stronger and also increases the heat deflection point from 60c to 90 or 100c.
Both my old cheap Chinese printers can do that, one with the smaller 15cm x 15cm bed gets there quickly and holds it without a problem, the other printer with a ender 3 sized bed I’d take to 110 at most. Any printer that says it can print abs should do at least 110c as usually you need the bed at that temp because of warping.
No surprises for those how are familiar whit engineering! For the rest look at technical data sheets the important numbers are: "TENSILE STRENGTH" will let you compare which will break under higher load for the same section aria. PLA is ~100N/mm^2, Nylon ~60-70N/mm^2, PETG ~50N/mm^2, ABS ~40N/mm^2 (you see the order matches whit the test results). The "Young modulus or Elastic modulus" will tell you know how rigid a material will be, higher the number higher the rigidity. Carbon reinforced Nylon >3500N/mm^2, PLA 2800-3500N/mm^2, ABS 1800-2400N/mm^2, PETG 1400-2000N/mm^2. Again no surprise. Whit other words READ THE TECHNICAL DATA SHEET! As for temperature, materials maintain these properties under, 45 °C for PLA, 140°C for Nylon, 105°C for ABS, 80°C for PETG of course these warry according to blend and manufacturer. In terms chemical resistance PETG is resistant to far more chemicals than any other while ABS is probably the worst in this group (depending on blend). If failure mod is a factor look at elongation at break, higher the percentage the more ductile the material and will fail les violently.
The issue is the difference between what is stated in the datasheets and what the 3d printed material actuslly can do. Often, there 's a huge difference and most datasheets don't even include layer adhesion.
My biggest gripe with PLA is creep over time even at low temperatures. That means if you use it as a bracket, you shelf is going to sag and fall down eventually. If you threat it for a fastener it will lose clamping force over time. if you use it for anything structural with a load on it it wont hold its shape. Makes it pretty much useless for me in any practical application.
I recently started using Creality CR-PLA. My 5 S1 printer came with it and I gotta say, this stuff is amazing! Very clean, accurate and has the best hardness I have ever seen in a printed part.
i did not use mine for over a year and am now doing stress tests to determine which filament to print my race bike aero winglets. so far CF-PA6 and creality CF-PLA come out very strong. I was very surprised about the quality of the creality CF-PLA
Nice testing. Harder to test, but I think that consistency between copies is perhaps a more important measure than pure strength for a few parts. You can design for a specific strength if you know the behaviour of a material, but if the material shows large variations then the only way is to always overbuild them, which may not be what you have room for.
Great job I like the test. Consider new and aged testing. Many plastics perform drastically different even with just a year of aging. Annealing will also affect plastic strength. It has been my experience that nylon rules given a couple years.
Good sciencing. PETG and ABS need a well heated enclosure for the cross layer adhesion to work well, I generally print them at a slightly higher temperature than the recommended max and they're usually stronger over a longer time interval. PLA's strength deteriorates over time, especially if they're exposed to direct / indirect sunlight. One thing to test / consider is post processing the prints in the oven, tightly packed in some super fine salt. It heat cures the layers to form a very strong part.
@@Clough42 in my experience, many white materials have a heavy pigment ratio. White plastisol screen print ink can be like treacle compared to black and other colours, for example! Not sure if it's the same for the plastics we print with, but I wouldn't be shocked if it turned out that white filament ended up being measurably weaker than other colours. Not that it would change your ABS results.... But it would be interesting to test, either way!
@@20vK I've noticed this in some parts as well. Colors definitely play an impact on the structural strength of the plastic around them, due to the size of their pigments. Clear is usually best for the greatest impact resistance and flexibility, but it also tends to be just a little bit easier to scratch as well. 😊🌎❤️
@@OJesusX3 ive noticed even in resin printing that clear resins (not filaments) like superPP, Nylon Green, and Siraya Blue have impressive properties over every other resin. they are like glass but can withstand the tremeandous forces demanded of suppressors and firearm frames you may be onto something with pigment particles playing a role in overall chemical properties of the print black carbon products have all behaved like charcoal for me when broken opaque white parts are more elastic and break like pvc
You all are some knowledgeable folks.. Love this video and the comments. I'm a simple CNC Programmer I remove materials to make parts. I'm just starting with this 3d Printing endeavor.. I'll be watching more of these type of videos to find the best and not break the bank 3d Printer.. Thanks again.
Interesting test! One thing I would change for a future test is to use a different colour of ABS, as white ABS needs a lot of colourant that reduces the part strength.
I would love to see you testing 3DX Tech’s ASA (or ABS) reinforced with Carbon Fiber. It’s currently my favorite strong and stiff filament. I’d be glad to pay for the roll and support your content anyway I can. Keep up the great vids!!
Hi James, I do print quite a bit of PETG, it is my go-to for all the reasons (or people's conceptions) of why someone would choose to use it. I do find the print characteristics can vary easily by print settings and even ambient conditions when printed, let's call this "finicky". For me this is a consideration in material choice. I.E. your choice of the carbon fiber reinforced may have lower values, but its less finicky and you get far more consistent results. As this anomaly in PETG is worth further investigation, I would love to see you take your scientific approach to it. As always, great testing and fair reporting of the results.
I fully agree with that. Part has to be designed with PETG in mind. I have found PETG to be quite brittle (almost glass-like) if designed or printed in such a way to not let it flex - eg more walls, less in-fill. Also, I always try print at highest temperature possible, in an enclosure.
I also use PETG a lot, but one purpose it doesn't seem to work well at is as shelf brackets. The creep doesn't seem to stop over time. In my office I have a wall of filament made of shelf brackets made of black petg. About a year later, one of the rows just collapsed, dropping all the filament. The layers on the bracket looked like they had slowly pulled apart over time. PETG is known for very good layer adhesion, so this was a bit of a surprise, but ironically this same thing happened to "3D Printing Nerd" about a year after he set up a filament rack with brackets. For being under a load for a long time, PETG needs CF. I decided to use PA-CF and that has worked very well. A couple months later, another row collapsed in the same way. I then replaced all the brackets with PC-CF and they've held since (2 years now) with no deflection or creep.
PETG strength can vary _significantly_ based on the temperature of the extrusion and the temperature of the build volume during the print. For best results, keep these factors steady during printing and bias toward hotter temps for stronger parts. That first value (in lbs) is repeatable and typical of a hot PETG print. PETG is _by far_ my favorite printing material for parts that need to be strong, but can also be somewhat flexible.
@@alejandroperez5368 Both valid points, but if the finish isn't as important as the structural integrity, PETG holds up best, especially considering it's inter-layer adhesion is top-notch. I haven't worked with any of the super hot stuff, like PEEK, but for regular ranges, it beats out the competition for me.
If y'all can, try to build or buy an enclosure as they can greatly help with the efficiency and layer adhesion of our parts. Even a thin sheet or a plastic bag helps. 😊🌎❤️
i dont understand how some people print firearms with carbon petg, but i can bend the parts with my fingers and they explode like frag grenades under any stress. parts printed at 250 on a 90 bed at moderate flow rate
One dimension I think which is also important, is how good one can glue them. Nylon is hard to glue and ABS can be glued with acetone by dissolving the surface and getting a very good bond. Many dimensions to the properties. Also not all materials are the same when it comes to ease of printing.
@@paulgallagher2937 What can they glue? PP? PE? Nylon? Low surface energy plastics? Nice. Yes, most of these solutions which actually work are rather expensive. Thanks for your reply, wasn't aware of this product.
Great test, thanks. Like you, I am surprised by the outcome. I'm certainly interested to see a bit more PETG vs PLA testing to see if it's possible to get that super-strength result in a consistent way.
Due to the peculiar design of the bracket, the breaking test appeared to be primarily over-stressing only the least-well-reinforced mounting hole. The angled portion seemed only to serve to change the vector of the stress on the hole. This seems difficult to interpret meaningfully.
It always is a tensile test. Thats what stretching is. The point where it begins to bend is called the yield point. Point they all snap is called the ultimate tensile strength.
It's like when you see those strength tests on wrenchs and the snapon wrench breaks while the cheap crap deforms so much it just spins. Is it useful for a tool to survive more at the expense of its ability to actually perform its function? Nope.
This is interesting and poses the question "Does PLA have the higher tensile strength because it's a purer material?". By that I mean it has fewer modifiers and additives. You could further ask whether the additions to PA-CF are actually bonded well to the carrier material and are just "in suspension"? and not adding to tensile strength.
@@relativisticvel pound is a unit mass bro, just like gram - pound-force and gram-force are force units on each due to gravity. I’m no dam physicist but that’s at least the most common way to use those words
This is what I've have always heard too, but I haven't seen any good long term testing that actually demonstrates that PLA creeps more than most of the other common 3d printed plastics. What little info I have seen demonstrates that PLA performs similarly to if not slightly better than PETG, ABS and ASA at room temp. The outliers appear to be PC which creeps the least, nylon 6 the most, and nylon 12 is more or less in the middle.
@@chrislosada6701 Personally I haven't had any issues with my PLA prints that are kept at a relatively dry and constant 74F environment. I'd imagine not all PLAs are the same either. Really I'm looking for a controlled experimental data, with a large sample of many different filaments and many other factors such age of prints, age of the filament, whether or not it has been exposed to light for any length of time, moisture content, if it was printed with a heated chamber or a heated bed and of course different environmental conditions. I know it's a lot of work, but that's what is needed to get truly useful data, not just anecdotes from random people on the internet.
PLA is actually one of the best choices for making (low speed) gears out of, because it has such great toughness/bond strength, stiffness and is incredibly resistant to wear. However, over the last few years, they have been making more subtle introductions into PLA which can make it more rubber-like. This does make it more resistant to cracking, but I wouldn't be surprised if some of you are not experiencing the same results. It is usually a very dense, and very hard wearing material. 🙂🌎❤️
@Zakir Siddiqui (wikipedia) In materials science, creep (sometimes called cold flow) is the tendency of a solid material to undergo slow deformation while subject to persistent mechanical stresses. It can occur as a result of long-term exposure to high levels of stress that are still below the yield strength of the material. In other words: Let's say this TV stand can tolerate 40kg without breaking. Even though, it will slowly and permanently sag (withing several days / weeks) under a load of just 10kg... PLA is clearly not the right material of choice for applications where a permanent load is at play. I've experience even more creep with "PLA+" (a small amount of different polymers such as TPU are added to turn PLA less brittle and improve printability)
That was surprising! Thanks for doing this. For my personal prints, some of which I have products made, I do try different materials (well, just pla and petg) to see what material is better suites for the particular application. Thanks for doing this
hello, i am gonna print a vacuum cleaner wheel, which material do you recommend, petg or pla? the wheel design has small clips to plug in to vacuum cleaner
Thanks for the testing, James. I'd be very interested in the cost/unit of the various material. So many things are engineered to a price these days. For anything done at scale, it has to be a consideration.
A valid question. However, I will just point out that "for anything done at scale", one probably isn't using fused filament fabrication, and instead using something like injection molding or such... which kinda changes the whole game. Then again, there are degrees of "at scale", so, again, a valid question!
I only use carbon filaments for that matte look. Carbon filaments just look really good, but it just means there is less plastic.. it essentially a filler material, it cant really do much. Good PLA is just really good, its cheap and the parts are strong and rigid. I think a lot of these special filaments, are kinda hype.. there is just way higher margins in them.. and everyone starts looking for special stuff at some point. To feel all fancy.
I was wondering about that. It's the fiber that matters to carbon fiber. Otherwise it's coal dust. I'd be interested to see the manufacturing process to understand if they have any way of keeping the characteristics of long fibers in these filaments. Hell, Kingsford's cheap. I'll start making that black stuff by the weekend. Just need a comic book name for it. The wife's probably not going to be happy with me.
@@Ideasite there are very different grades of carbon fiber filled materials ranging from ground cf (dust as you mention), milled, aerospace grade (see 3dxtechs website), and continuous fiber (like markforged). CF can definitely improve material characteristics - IIRC flexural modulus of ASA is almost 2x that with CF (based on 3dxtechs TDS)
@@Ideasite also the fibers need to reach from layer to layer to actually make them useful. I could see how they might help reduce warping but i dont really think they make the parts stronger.
Nice job keeping it simple and applying the test to the use case in question. The trouble with 3d printing materials is that the geometry of any given part seems to have more effect than the mechanical properties of the polymer. Stefan at CNC kitchen has done a lot of work characterising different filaments mostly testing ultimate tensile strength with the layers and across the layers which shows the disparity in strength between the different axes. I have found when using PLA the problem is temperature, even in the UK a part left in direct sunlight can easily reach 40C and start to deform. I printed a nice drawer to store bits and bobs under my Ender 3, 2 days later it was banana shaped and I had to destroy it to get hold of my spare nozzles and cutters.
That’s why it’s important to understand the characteristics of each material for your project. Pla is more rigid but generally where it falls short is on heat deflection and impact strength. Of course there are alloys that try to minimize those. If you don’t need heat deflection and impact strength 99.9% of the time you can buy the pla. Would love for you to test taulman gf and cf nylons, there finish is injection mold quality.
Good analysis. This was the first video of yours I watched. It was a very good idea to point out that the application should dictate which material parameters are the prime concern (and therefore which material to use), and that there are many more properties of these materials that were not explored and could be prime factors in selecting the best material for your application. Well done.
The biggest advantage to CF filaments is dimensional accuracy. PLA also has a slightly higher density than the other filaments, so other than just having a lower strength/weight ratio, a 1kg spool of it will print fewer parts. Those are all minor issues though compared to PLA's poor creep characteristics and relatively low glass transition temp (parts will just turn into blobs over here in Australia if they're left in a hot car).
PLA does that all around the world and will never stand up to a hot car test anywhere. in my opinion PLA is a hobbyist filament to teach people how to use a 3d printer and it has no other use than printing ornamental things that you don't count on lasting.
I thought carbon fibre filaments are just to make parts lighter and prints easier for things like abs and nylon. They aren't continuous fibres so they don't add any strength and typically increase brittleness. Adding CF to nylon basically removes most of the benefits to printing nylon in the first place, I don't understand the appeal
Quite. I'm designing some posts that need to survive hot air for hours, and neither PLA nor PETG will cope with the temperatures for long. ABS or ASA or better, and nylon can handle even higher temperatures.
This is a great clip. One of the best parts about it is the thought provoking subject and the fantastic group of followers. I learn so much from James, but a ton more from the rest of you. It's amazing~! Thanks, James!
CNCKitchen has a whole series on testing different filaments. It's a pretty thorough testing regime, too. You may want to check out Stefan's videos and maybe see about getting him some Carbon Fiber nylon to test. This was interesting and nice to know that I'm not going to be a total idiot for making furniture parts out of PLA.
Stefan's testing is more of a 'laboratory' style parts testing while the testing here is more 'real world' as the part put under load is for a real application. Stefan's testing can give you a good 'base line' while in this video this is more the way the engineers will test actual parts and materials for a given application now there are other things to take into consideration as well as layer orientation when designing and printing a part. Those are the infill density, printing temp (both at the printhead and the environment), part cooling amount, RH of the filament and environment during printing, the way the part is sliced for printing (i.e. weaker areas of the print more walls/infill). All these factors and more can affect the resulting parts total strength and weak points.
This is only the 3rd video of yours I have watched. I would just like to say that my experience has been really good. Your videos just seem to have this perfect balance of information to go along with what you are doing or explaining. It’s rare these days to get to the end of a video and be like “oh, it’s over”. You definitely seem to capture my attention very well and I enjoy all the bits of information you add in. You are very good at explaining things. Love your work. Thank you
Thank you James! I am starting to print tools for my work shop. Mostly 90% for woodworking. 90* for clamping, corners for belt clamps and so on. I have been very happy using eSun PLA +, even as a beginner a year ago. Until I watched your video I was thinking I should use a material other than PLA.
@@fluffypinkpandas plus or pro can be whatever you want and it different between different filament makers, so both things does not means something specific but used to differentiate "modified" product
I just printed eSun PLA+ which is a copolymer with TPU as far as I know, seems a bit more flexible, so probably less brittle and stronger. The filament itself feels much tougher to break in hand. It's supposed to be almost 2x as strong as normal PLA. I need to print it over 210*C, because at that temp layer adhesion was slightly too weak and I got a small separation from pulling off bed.
PLA and ABS aren't UV resistant, and PLA has a very low glass transition temperature. The PA12-CF is used for high temperature engineering applications, and I imagine the NylonX can also fit that role. For general purpose applications, PLA is good, but if you need UV resistance or some amount of heat resistance (such as a part in the interior of your car), PETG is the better choice. The other materials have limited, niche cases where they're preferred, but are vastly more expensive (usually 100-200% more). I personally print mostly in PLA, but have recently started using PETG for applications around heat sources. I also have PA12-CF for making car parts around the engine bay, such as end caps for headlights/fog lights, small brackets, etc.
What stands out is that the dial indicator doesn’t return to zero on the carbon fibre samples, so it distorts in a short period of time and doesn’t return to its original shape. The biggest advantage of the PA/CF is temperature and chemical resistance, like for automotive use 👍
PETG has pretty good chemical resistant too. I've actually used PETG to print functional automotive parts that see gasoline and that live in the hot engine bay of a vehicle and it has held up perfectly without issue.
Impact strength is one of the weaknesses of regular PLA, it's quite brittle. Some improved PLAs trade off some ultimate strength for significantly toughness e.g. e-Sun PLA-ST.
As someone who does this stuff professionally, I really loved your approach on the PTEG. That high outlier is dangerous to leave in place for your data, especially with an n of 3. I'd hire you if that was an interview question.
Great video James, I'd like to add one important thing to the PLA 1st place. As others pointed out the problem for PLA are elements and temperature, therefore these lovely brackets could have a serious issues outside. But what I haven't seen in the comments, is the PLA has tendency to flow under load. It could not be a problem for such a small part vs weight of soundbar, but... Few weeks ago I talked to the guys from MyCello (3d printed tailor made chellos). We have agreed on this PLA behaviour as they had to switch from PLA bodies to the PETG, which seems to be OK for they use. On the other hand I have printed many holders in my garrage from PLA, after few years they're still fine. Even parts on my motorbike where they got beaten by sun and rain. I love CF filaments for plastics which are hard to print due to warping...
As others have pointed out, it appears your scale was set incorrectly to lbs - doesn't invalidate the results, just changes the units :) very interesting video, I've been looking into buying the X1 Carbon, but after this video (and other, similar videos from CNC kitchen), I'm finding it harder and harder to justify the purchase.
One recommendation I would make for petg is make sure you're using as minimal cooling as possible for strength. Petg has poor crystalization properties if cooled too quickly. Any chance between the one part and the other two you had changed the cooling percentage? Great video as always.
Eric, I have not done any sort of printing myself but my first thought was the printing or cooling cycles were not the same. Clearly to me anyway this was evident in the results, but also makes me wonder if the others were printed with the best/optimal settings
The problem is, PETG is hygroscopic. So if you decrease cooling, the final part will be full of strings. That's why it needs to be dried out first which is extra work. And that's why PETG is not worth it for me. PLA is the king.
Most filaments do absorb moisture to some extent, and that's why it helps to create a heated filament dry box. (Most of us end up making ours out of clear storage totes, printed roller guides, and using whatever insulation is most available to us. Foam, cardboard, etc) PETG is a great medium! They're all great mediums, but they do have different properties in which each excels. I generally print with PETG myself, (for its ease of use and relatively high thermal resistance, (it doesn't *typically* ever melt in a car)) although when it comes to surface durability, (scratch resistance) stiffness, and bond strength, PLA is unique in that it's characteristics (and density) is like that of a cast resin part than other plastics. While it doesn't take much heat, it can form some Incredibly tough surfaces and it does make for some good (low speed) gears and high load bearing surfaces and rollers which will last a long time. Even in real world scenarios. (I've made lasting parts out of PETG as well. 😊🌎❤️) I've heard that annealing can greatly improve PLA's temperature resistance, (and overall parts strength as well,) but I've yet to try this. 🙂 Keep in mind that PLA is only 1 of the many bio-plastics we have designed and I'm sure that there will be more that may replace many of the more common ones that we use today, (it all in the formulations and the bonds which gives plastics and resins their unique properties) perhaps making several, easy to use, low cost plastics that can rival most of the things that we use today. Fingers crossed gentleman! 😊🌎❤️
Sure would be interesting to see a finite-element model of stress to see how it distributes during the pull. I'm thinking that failure is precipitated by the weakest area, after which other areas fail having been let down by the first area. It looks to me like the design is too dependent on the very thin region around the recessed screw adjacent to the "diagonal" side. So if might turn out this is really a test of that specific kind of feature. Not that it's an issue for the actual application, just for the testing results. Regardless -- good on James for doing some tests and entertaining us!
Wow, this was an eye opener for me, and like you, I was very surprised at the results. The PETG was disappointing, as I expected more from it. Great video, and another sub. Thanks!
@@gareth4168 lb is a confusing measurement as it can be both a force and a mass depending. I’m an American, so imperial makes more sense visually, but when I went to school and imperial was used it was a real pain for engineering. Imperial mass: Slugs is the SI unit lbm is another (but it’s commonly just written as “lb” for added confusion” Then we have force: lbf (usually written, once again, as “lb”).
@@dreamrestore213 because if he reports the strength ratings in Newtons on some website, people will think they are around 4.5 times weaker than they really are since it’s actually in lb not N.
Great real material testing. In other section i would like to see similar study for temperature resistance because this is another critical property in order to select it. I’m a beginner in 3D printing and i have much to learn. Thanks for your work and your channel.
I'd be really interested to see bending and tensile tests of regular nylon against nylon CF. My guess is it might be marginally better in bending but no better in tension. In the world of fiber-reinforced composites, your strength is very dependent on fiber length and load transfer. Unfortunately, these chopped CF filaments have such short fibers that loads don't really transfer to the fibers and instead they just shear around in the plastic. They might improve stiffness in bending though. Again, these are just my assumptions.
A big factor in the strength of composites is the adhesion between the filler and the matrix. In this case CF is the filler and Nylon is the matrix. Unless the stresses in the material can bridge the filler to the matrix material effectively you essentially just made a part with a bunch of voids. If the Nylon or CF were modified to have better adhesion to each other the strength could be drastically improved.
My experience with PETG (and looking at videos here) is what you saw: very high inconsistency. When I have weird part failures of usable parts, it's usually PETG. It's inconsistent, it's picky with printing, and susceptible to moisture. It's also the filament most likely to damage your printer: I've lost a hot end because the print failed and the filament started sticking to the nozzle, then built up and completely encased the hot end and was in the cooling fans, warped the ducts, etc. On top of that, PETG either doesn't stick to the bed, or it sticks far too well, and I damage the bed removing it. I put my printer in a heated enclosure, and because of this, ABS/ASA have been a little easier to print than most PETGs.
These tests went almost exactly as I expected. Except the PETG break test. The biggest issue with PLA is heat for sure and it moves at MUCH lower temperatures. IF you haven't already check out CNCKitchen
@@alejandroperez5368 Yes. A car in summer is hot, but not unusually so. The point was that it doesn't require being in the engine bay, or a component of an oven to deflect. There are many real-world scenarios that are simply without the useful temperature range of PLA. That said, it's still my primary filament for functional prints.
I don’t recall if you did a modifier around the screw holes to beef up the part at the load point . If you just had the 4 perimeters that could impact. Be interesting if you did a solid 5-10mm around bolt in all test materials as that top screw kept cracking. It would also eliminate that Petg riser suspicion. Cost vs strength the pla would probably dominate but if your preference is the pa12 Then use it. I prefer petg as it easy to print and is suitable for all my cases so far. Just on parts ductile failure maybe better than catastrophic failure depending on use case.. Unless you add a higher factor of safety to avoid the catastrophic failure - which in fusion you can test for. Interesting dealing with materials, but preference and ease after win out. Why stuff around for half a day trying to get PC to plot or eliminate warping issues, when you can just drop on petg (or pla or ….) and go off do something else.
Your last point is why I fell in love with carbon-filled polycarbonate. I got a roll just to test, and while it took work to get my printer up to temp (280C hotend, 105C bed), it has always just worked. It's strong, rigid, and has a lovely finish. It's much more expensive than PLA, of course, so it won't be used for everything, but I know I can just drop it in and have it work. I wouldn't buy another roll if I had to babysit it just to get a single successful print. Regardless of any special properties, the most important is often that it prints reliably.
Care is needed when using PLA for engineering because it will creep over time if it is under constant load. ABS is much more stable over time and under load and that's why it is commonly used for engineering.
Nice video. Consider having a larger sample size as well as maybe adding a couple of other materials such as ASA or PC. Personally I stick with PLA just because of how darn easy it is to print, but I'm always open to new materials if they can prove having better material properties for certain applications.
Very interesting set of tests. Testing materials is hard, and you did some easy ones, and I don't blame you. But I am a bit unconvinced of the value of UTS tests, especially for a plastic material that can creep. Really, for real world design, it's the yield strength you would work to, surely - OK, after yielding it may hang on for a bit, and that may save someones life or stop something important getting broken, but you'd never design to that standard. The problem is that finding the yield point is hard, especially on a relatively elastic material like you are using. In support of my comment, you yourself were looking closely for signs of yield on a couple of the tests. But yield strength might well give very different answers - the ones that were strongest did seem to bend a long way, no way to tell if they would have gone back, or how far, but quite possibly worse than the low UTS materials.
I print a lot of PETG because a lot of the parts I make should be tough, but for hardness PLA is hard to beat. Two and a half years ago I designed a prototype speed handle for my vice. I wasn't sure if it would survive long enough for me to decide whether I liked the design and make it from metal, so I printed two of my final design that passed minimal use testing. I used PLA. I'm still on the first of those two PLA prints, and I rarely even need to use the long steel handle that came with the vice. My speed handle has a 6-point and a 12-point socket on it, and the socket I normally use is the 12-point socket. It doesn't have a lot of engagement, yet it has not stripped out. I didn't anneal it; I didn't expect it to last this long and wasn't trying for longevity. It just really exceeded my expectations.
Unless temperature resistance is a concern a good quality PLA+ (I really like polymakers PLA pro good balance between stiffness and impact resistance) will outperform basically all filaments until you get up into filaments that require heated chambers to get maximum layer adhesion.
been printing with PLa for 7 yrs. Never tried anything else , I always felt it was decent for most applications. And this testing corroborates it . Not to say other filaments arent better for certain applications , im just too lazy to experiment for small gains.
I really love this type of videos and it helps tremendously people to decide what material to use in their project. Only problem is that, it sometimes can inadvertently provide impressions that may not necessarily be true in all circumstances. I am saying this because I was quite surprised with PA-CF performance. My experience is that regardless of number of walls, or infill density, PA-CF strength can vary wildly depending on temperature and speed they were with. It also has very low tolerance to imperfections when laying down the layers, whilst PLA just seem to fuse together with imperfections very easily. Because PA-CF requires much higher temperature and the CF reduces its viscosity, it is very easy to print weak parts due to low temperature settings (not to say humidity, which is obvious). I'd experiment printing at higher temperatures and test it's strength again. I used to print it at 260/270, but was only when I printed it 280+ that it started to show true strength. It's also interesting that in klipper, the pressure advance value has a wildly different behavior at higher temperatures, which really indicates how its viscosity can vary depending on the temperature, which can indicate it has not fully undergone glass transition and will affect layer adhesion. Both of which are crucial to print strong parts. By the way. Try annealing PLA. That is something really impressive. It becomes more temperature resistant than nylon and feels like is as strong as aluminium. Only problem is that it shrinks so, makes it tricky with precision assembly.
i recently started using the carbon fibre stuff and loved the surface finish of it, i did suspect it wasn't giving any extra strength. Glad you also came to the same conclusion. overall it looks good but not worth the extra money. thanks.
For PETG i had similar inconsistencies and did some research on it.. It is realy important to print PETG fully dried to have consistent results.. Because the explanaiton is that the moisture inside does microbubbles during printing and you get part that is similar to foam.. after that I religiosly dried and even printed out of a dryer and it made significant improvments specialy the flex before it breaks is bigger...
Super informative, thanks! I think with a different part geometry you could have a lot more strength. Each part seemed to fail at the screw holes, beefing that area up or even having that spline continue in between those two screw holes could make a big difference. But I'm sure it was designed around certain requirements and you made it as strong as possible for its application. Cool video.
This has been my experience as well. I love the finish of a carbon fiber reinforced filament. The finished part just looks so much better. But it has never been as strong as its PLA counterpart.
yeah, i say, the strenght of a material can differ even from the same manufacturer. Color can also affect the strenght... soo, yeah. I got told that abs was soo good and strong, i usually print in petg. It just works best for me, strenght, doesent react to chemicals, uv and so on. Anyhow, got a abs roll from amaxon, printed, alot of stinking, yeah, and not that strong, not enought for me tu use it instead of petg anyway. Used up that roll, and got a customer that wanted some stuff i abs printed. Bought another roll from another manufacturer, and that plastic was rock hard! the print looked soo good, but it got all warped up and cracked between the layers in the end sadly enought... soo, yeam, amazon basic petg is what i use.
I have been using PLA and PLA pro for almost all my prints, and they are very strong. I have recently started printing with more wood infused PLA and also Metal Infused PLA, Both are very strong and make some unique prints.
Have been designing and printing functional parts this last year and for absolute strength the way you print is almost as important as the material. Also even with pla you want to keep it as dry as possible, even a week in humid air and you will start getting air coming out of your nozzle leading to under extrusion. Personally with pla+ for a balance of max strength and print time I keep to 2mm minimum for walls with a 0.6 nozzle and layers perpendicular to where force will be applied. Also you can cheat a little with PLA, if the part touching the bed will experience the most force. Lastly you can heat the bed up after the print is done and before it unsticks, to harden that surface. I’ll usually heat the bed slowly to 80C for a while then let it cool and make sure nothing warps as it cools. Part becomes much harder and more heat resistant.
I didn't *set out* to do a test of PETG, but I accidentally did and it mirrored your results. IE, inconsistent strength. My application specifically needed some degree of heat resistance (I printed holders for parts that were spray coated and then put into a curing oven). Some of the holders were plenty strong for what I needed, but several just busted during assembly. It was the same eSun PETG you used (though a different color). Gonna try turning down cooling as is suggested here and see if that makes a difference in strength.
Very interesting. I haven't had any real issues with it in the past, but I think the geometry of this part and the resulting overhang created some imperfections in the print that gave the fracture a place to start.
@@Clough42 In my case it was absolutely because it was a long, thin part that had threading on both ends. The only orientation I could print it was the worst one for durability, and the ones that broke all broke in the threads, even with 5 perimeters. Definitely a weak spot. That said, the one printed with drastically reduced part cooling seems to be much more tougher. Much better surface finish too! So I'm changing *that* filament profile permanently.
At least I did not catch you mentioning.... 1: what the infill amount is for each part. 2: what type pattern for the infill you used, etc. All this does make a difference. Also in my designs. I design my part to be an even multiple of the nozzle printing width. I use a REVO HL .6 nozzle and a .4 layer height and a .5 layer width. So for walls I might for example make my walls 4mm thick. In Simplify3D my slicer I then will use 2 Walls called Perimeters in simplify3D. Which is actually 4 walls 2 interior and 2 exterior. So that adds up to 2mm and then 2mm of infill I print either Cubic or Honeycomb infill pattern. At 8-10% infill @ 120% infill width my parts are really strong. For only 8-10% infill. I used to use 50-90% infill thinking that would make for a stronger part. Then I discovered that going by what I stated above I get just as strong of a part with saving a lot of printing time and filament. So without taking the above into account when you or others are doing these types of tests does not tell the whole story. I also discovered that as you go down in infill percentage and maybe increasing the infill width I get a bigger wider infill pattern. I have always wondered if this is actually more rigid than printing a smaller pattern and more infill. I never did a test. Which someone should do such a test comparing the above mentioned setting. Take PLA and use different infill patterns and infill % amount and also vary the Infill Width. This test would show a lot and I think help a lot of users. And you would have to vary the number of Walls to have a true test.
One thing to note with PETG is that it's hygroscopic and if you print it when it's too moist it will actually become brittle due to the interactions with the steam created when printing. To get the best results with PETG it has to be very dry, I always run mine for a few hours in the dehydrator before printing and I print from a dry box to minimize absorption during a print.
Exactly my thoughts, it looks like first one was printed fresh out of the package, but other ones caught some humidity from the air which resulted in poor structural strength. Dry box is a must with petg.
I don't have a dry box (just haven't gotten around to it) but I absolutely run PETG in the dehydrator for a few hours prior to use. I certainly get more consistent prints that way.
My only experience with a strong 3d print material was "tough-pla" i don't remember what brand, but the name was no joke, that stuff was really frickin strong. I tried tapping screws into it like you would with normal pla and ho boy was that an adventure, it took so much force to work into that material, and I don't think I ever had a single part break, because it wasn't just stiffer, it was also more ductile too, none of the brittleness of normal pla.
I'm not surprised by the PETG results - I've had the same experience. Sometimes it bends and sometimes it just snaps - even if it's the same part and the same settings.
Great test! I think there is one additional property relevant for your application. PLA tends to flow more due to its low glass transition temperature. Would be interesting to see how PLA performs in the initial deflection test when loaded for a week or so.
@@alejandroperez5368 sure 🤦♂️ PLA is perfectly fine for mechanical and structural parts. I use PLA for 10 years now and I don't consider most of my parts "Toys".
Awesome job. In the breaking strength test most to all cracked right at the screw cap hole mounts, meaning how tight you had those on each piece would affect the results. In further tests the weakest spot that will break should not be the screw area nor the tied pull area, eliminating a good amount of variability.
Very interesting and educational for a newbie to 3D printing. I have taken screen shots of both of your graphs and I will use them when choosing a material type under those needs. Yes, heat and UV ect. must be considered for different circumstances but for most applications, unless you are Machining critical parts, I think your testing covers most needs. Thank you.
Have a look at CNC kitchen, he tests loads of materials in quite a lot of tests, including impact strength and failure temperature and also tests different techniques like annealing or remelting in salt.
Thanks for sharing. I use PLA for all types of prints, it can be over and under cooked. Color can affect the temperature for the best prints. Printer quality has tripled therefore the prints are fantastic and stronger. Some of my slower prints are mind blowing and look like injection molded parts with PLA, I have painted some to make them more UV protected. Thanks for the testing. There is a video on PLA outboard propeller testing this video made me try harder for better prints with PLA, faster, slower, temperature control, infill.
I'm so glad you made this video. Thank you! I've been printing various parts for a while now and, without doing any rigorous testing, I was under a strong impression that PLA is an under-appreciated material. I also suspected that mixing in carbon fibers doesn't really do much for the strength of the final part, but couldn't really confirm it. Now we know! I still like the *look* of PA-CF parts, though. It's funny how various myths take hold and persist - PETG vs PLA is another one. You will often hear from 3d printing afficionados "for strength, use PETG". This is nowhere as clear-cut! These days I use PLA for pretty much everything, except when I need those additional 8-10C of temperature resistance, then I'll go for PETG.
Thank you for this video it's always nice to see realtime results. I appreciate the repeatability of this test. Obviously with any repeatable statistical analysis the higher the number of tests samples the closer to an actual result, that is my only suggestion. I know that time, effort, and money make it hard to have hundreds or thousands of samples. But for any of these types of results to be reliable, the number of test samples should be in the thousands. Maybe there could be a cross channel collab, where each channel studies many samples?
Another consideration is that PLA has issues with sustained loads. I had a headphone stand with a desk clamp for example and the PLA version loosened every few weeks, until it broke, while the petg version still holds strong after over a year.
Great video, my new favorite material is PC. It prints great for me and love the properties, it's extremely hard. I would love to see how it does in the tests.
Interesting test. PLA is great, but the lack of heat resistance does limit its usefulness in a lot of cases (although I still try using it whenever I can get away with it). The PETG result was surprising -- I normally print mine pretty hot (255 to 260C) with a very slight extrusion multiplier (~101%) and 85C bed, and the result is usually excellent with the Matterhackers PETG
Several of you have pointed out that I was quoting numbers in newtons, but the scale was set to pounds. That was my mistake. I set up the scale to read newtons when I assembled my test setup, but later checked the accuracy of the scale using my gym weights, which are marked in pounds. Apparently I changed it and did not set it back. I did scan through the footage to verify it was set to pounds for all of the materials, so the comparisons are valid. Sorry for the confusion.
Units don't matter among relative results.
The biggest advantage of the Carbon Fiber filled filaments is the strength to weight ratio, not it's raw strength. A part made from fiber filled will be lighter, with a similar strength.
Freudian slip? Hungry and thinking about Fig Newtons cookies.
I print a lot of PETG... so more research into the wide ranging results would be of interest. I mostly use it because it gives a lot before breaking.
Thanks for the testing.
How much did each sample weight?
You should try those tests with HTN CF25, which is high temp nylon with 25% carbon fiber, it's also known as black aluminum. I just started using it to test the tools in the 3d printable tool set that I'm selling. I tested a 7/8ths socket made of CF25 on a wheel lug nut and it was way stronger than PLA. The socket printed in PLA got up to 14.4 foot pounds of torque but the CF25 got up to 50.1 ft lbs and that was printing it without the suggested temp settings. I printed it on a CR10 v3 which has a max hotend temp of 270c and the suggested print temp for CF25 is 280c - 310c.
I'm currently waiting for an all metal/bi-metal heat break to be delivered so I can upgrade my printer and try printing the CF25 at higher temps to see if that makes it stronger. Even printing the CF25 at 270c though, it still had a metal sounding ring to it when you hit it with your fingernail.
There is a variable that you missed here. We use 3d printed plastic parts for combat robotics and do a lot of testing of different materials. Our tests have shown that the pigments in the plastics make a huge difference. For example, the white ABS likely uses Zinc Oxide as a pigment which greatly reduces the strength of the plastics... while the carbon that is generally used to pigment black filament doesn't affect the strength much. We've been in contact with several of the filament manufacturers, and their engineers have backed up our findings as well. If you want good consistent results, you should probably test all of the parts using the same color... preferably "natural" undyed, but that isn't available for things like cf filled or some of the better flavors of PLA+
Our testing so far (for impact resistance) has two colors of Duramic 3d PLA+ being the top runners... (Black and Red). These filaments print best at a bit higher temp (I use 225c) but it is well within the capabilities of even an inexpensive hobby grade printer like an ender 3.
I have seen the same in material testing of injection molded parts. Especially nylons, the lighter the color (gray, blue, white) generally the weaker and lower surface energy they have. Same reason if you are having bed adhesion problems, sometimes switching to a natural or black color will fix the issue.
Interesting that you mentioned zinc oxide. What's your source? To my knowledge titanium dioxide is the most common white pigment, when opacity is desired.
Wow. Duramic PLA+ black is my go to for durability! Seems you came to the same conclusion.
for strength id say pla+ for short powerful repetitive impacts pa6 cf nylon hoffman tactical did a whole video on it considering what he builds with it for impact testing i trust his word, polymaker often hops on his streams and also agree
Funny, this is true of fabrics too. Often lighter colors undergo harsher chemical treatment.
We did similar testing for our products and got mostly the same results. PLA is the best all arounder. But we have a temperature requirement and that ruled out PLA for us. We tried it just to make sure and just a little heat in an oven and oddly enough tossing it in the freezer made the parts deform to the point that they stopped working. During our testing we did discover a few key things that may be affecting your results.
1) the rate at witch you pull matters, that rate is based on the sampling rate of the scale. If you move to quickly the force will fail the part before the scale updates the reading. Giving you a bad reading. Consider using a Load cell and the good old Oscilloscope to recode the loads.
2) the angle of applied force matters, that angle must be consistent. If the forces shift slightly it can lead to the part being loaded favorably or unfavorably depending on the shape of the parts. Try a more rigid linear slide and add an electric actuator to slowly and consistently apply the load. We got ours off amazon.
3) print settings matter, they effect layer adhesion and that effects the ultimate performance. Your results are hinting at possible issues with controlling the printing conditions. Bubbles in the material and to cold of print settings will lead to huge swings in performance. Dry everything in the oven at 100F for 8 hrs and run it all on hot settings.
4)sample more than 2 or 3. these materials fluctuate naturally and any sample less than 10 is asking for misleading (biased) results. We have a saying. The first data set is the worst data set. Old filament may have issues, new is preferred, the temperature of the room matters. The more brittle the material the more it fails statistically and less predictably, and cold temperatures will cause this specifically for thermal plastics. The temperatures should be around 75F Or the operating temperature that the part will experience during it's life.
5) load the part realistically, if your part is going to hang something then make extra sure that as you load it the part is free to deform and move in the way it would in real life. Over restricting the movement of the part can lead to unnatural twist or bending that could cause early or late failure. This can often times be hard to do with things that hang. Consider a sand bucket free floating body test. Hang a metal bucket and put your load cell between the bucket and part. Use a funnel and slowly allow sand to fill the bucket till the part fails. This allows the load to move with the part, which can not be done with a constrained loading machine. put something under the bucket to catch it. like a stool or box so it doesn't fall far or tip over and spill.
Good luck I look forward to seeing a follow up video.
Excellent comment, thank you for posting !
PETG must be printed properly to achieve the best results, and each printer must be setup to print it.
As pointed out layer adhesion in PETG needs careful attention.
The conditions of each filament roll are very important, age, how stored, dryness, etc.
Nylon requires careful attention, I;m doubtful of switching material to do testing can achive that without careful tuning.
I believe we ask too much of these printers, expecting the same machine to easily adapt to very differnet materials.
Thank you for posting an excellent video !
Good points. I'd also add that there are also Tough PLA blends out there that may improve some qualities and some that have CF in them. For me, most of the CF materials I have used add no strength to the part and frequently make it harder to print. However, that matte finish is very very nice
Very true about the sample size. Really though, you don't get any statistical significance until you hit 23 data points at the absolute bare minimum.
Awesome comment! 👏👏👏
Yep, I thought it was kind of obvious that the materials weren't all printed adequately. Delaminating layers was the give away imo. It's also expected - most of us have a favored material and use it way more often. Fresh filament, honed settings... yeah the prints weren't done with the same quality.
So you're going to publish the results you tested right?
You can also easily anneal PLA to make it even stronger. I usually just anneal PLA on the printer bed at 120c, and put a plastic container over the part so that the heat stays enclosed close to the part. I’ve found doing that for PLA brackets make them soooo much stronger and also increases the heat deflection point from 60c to 90 or 100c.
What printer are you using that has a print bed with 120°C+?
@@nils1953 most are capable of that lol
@@johnandrews9433 lol. Okay then it's just my oldtimer one that struggles to get up to 90.
Both my old cheap Chinese printers can do that, one with the smaller 15cm x 15cm bed gets there quickly and holds it without a problem, the other printer with a ender 3 sized bed I’d take to 110 at most. Any printer that says it can print abs should do at least 110c as usually you need the bed at that temp because of warping.
@@rvent3605 hmm. My ultimaker2 only goes up to 100°
No surprises for those how are familiar whit engineering! For the rest look at technical data sheets the important numbers are:
"TENSILE STRENGTH" will let you compare which will break under higher load for the same section aria.
PLA is ~100N/mm^2, Nylon ~60-70N/mm^2, PETG ~50N/mm^2, ABS ~40N/mm^2 (you see the order matches whit the test results).
The "Young modulus or Elastic modulus" will tell you know how rigid a material will be, higher the number higher the rigidity.
Carbon reinforced Nylon >3500N/mm^2, PLA 2800-3500N/mm^2, ABS 1800-2400N/mm^2, PETG 1400-2000N/mm^2. Again no surprise.
Whit other words READ THE TECHNICAL DATA SHEET!
As for temperature, materials maintain these properties under, 45 °C for PLA, 140°C for Nylon, 105°C for ABS, 80°C for PETG of course these warry according to blend and manufacturer.
In terms chemical resistance PETG is resistant to far more chemicals than any other while ABS is probably the worst in this group (depending on blend).
If failure mod is a factor look at elongation at break, higher the percentage the more ductile the material and will fail les violently.
Thank you, that was useful.
The issue is the difference between what is stated in the datasheets and what the 3d printed material actuslly can do. Often, there 's a huge difference and most datasheets don't even include layer adhesion.
My biggest gripe with PLA is creep over time even at low temperatures. That means if you use it as a bracket, you shelf is going to sag and fall down eventually. If you threat it for a fastener it will lose clamping force over time. if you use it for anything structural with a load on it it wont hold its shape. Makes it pretty much useless for me in any practical application.
I recently started using Creality CR-PLA. My 5 S1 printer came with it and I gotta say, this stuff is amazing! Very clean, accurate and has the best hardness I have ever seen in a printed part.
i did not use mine for over a year and am now doing stress tests to determine which filament to print my race bike aero winglets. so far CF-PA6 and creality CF-PLA come out very strong. I was very surprised about the quality of the creality CF-PLA
@Clough42 watching the video 8:38 i noticed that the scale is set at LB but you call it newtons , its actually half the show value !
Nice testing. Harder to test, but I think that consistency between copies is perhaps a more important measure than pure strength for a few parts. You can design for a specific strength if you know the behaviour of a material, but if the material shows large variations then the only way is to always overbuild them, which may not be what you have room for.
Great job I like the test. Consider new and aged testing. Many plastics perform drastically different even with just a year of aging. Annealing will also affect plastic strength. It has been my experience that nylon rules given a couple years.
Good sciencing. PETG and ABS need a well heated enclosure for the cross layer adhesion to work well, I generally print them at a slightly higher temperature than the recommended max and they're usually stronger over a longer time interval. PLA's strength deteriorates over time, especially if they're exposed to direct / indirect sunlight. One thing to test / consider is post processing the prints in the oven, tightly packed in some super fine salt. It heat cures the layers to form a very strong part.
The ABS was printed at 240C in an 80C heated print chamber.
@@Clough42 in my experience, many white materials have a heavy pigment ratio. White plastisol screen print ink can be like treacle compared to black and other colours, for example! Not sure if it's the same for the plastics we print with, but I wouldn't be shocked if it turned out that white filament ended up being measurably weaker than other colours. Not that it would change your ABS results.... But it would be interesting to test, either way!
@@20vK I've noticed this in some parts as well.
Colors definitely play an impact on the structural strength of the plastic around them, due to the size of their pigments.
Clear is usually best for the greatest impact resistance and flexibility, but it also tends to be just a little bit easier to scratch as well. 😊🌎❤️
@@OJesusX3 ive noticed even in resin printing that clear resins (not filaments) like superPP, Nylon Green, and Siraya Blue have impressive properties over every other resin. they are like glass but can withstand the tremeandous forces demanded of suppressors and firearm frames
you may be onto something with pigment particles playing a role in overall chemical properties of the print
black carbon products have all behaved like charcoal for me when broken
opaque white parts are more elastic and break like pvc
@@fluffypinkpandas Incredible. And thank you very much for adding your experiences my friend. This helps 😊🌎✨
You all are some knowledgeable folks.. Love this video and the comments. I'm a simple CNC Programmer I remove materials to make parts. I'm just starting with this 3d Printing endeavor.. I'll be watching more of these type of videos to find the best and not break the bank 3d Printer.. Thanks again.
Interesting test! One thing I would change for a future test is to use a different colour of ABS, as white ABS needs a lot of colourant that reduces the part strength.
Yeah probably the fairest way to do it would be with all filaments being their natural color with no colorants.
I would love to see you testing 3DX Tech’s ASA (or ABS) reinforced with Carbon Fiber. It’s currently my favorite strong and stiff filament. I’d be glad to pay for the roll and support your content anyway I can. Keep up the great vids!!
Hi James, I do print quite a bit of PETG, it is my go-to for all the reasons (or people's conceptions) of why someone would choose to use it. I do find the print characteristics can vary easily by print settings and even ambient conditions when printed, let's call this "finicky". For me this is a consideration in material choice. I.E. your choice of the carbon fiber reinforced may have lower values, but its less finicky and you get far more consistent results. As this anomaly in PETG is worth further investigation, I would love to see you take your scientific approach to it. As always, great testing and fair reporting of the results.
I fully agree with that. Part has to be designed with PETG in mind. I have found PETG to be quite brittle (almost glass-like) if designed or printed in such a way to not let it flex - eg more walls, less in-fill. Also, I always try print at highest temperature possible, in an enclosure.
I also use PETG a lot, but one purpose it doesn't seem to work well at is as shelf brackets. The creep doesn't seem to stop over time. In my office I have a wall of filament made of shelf brackets made of black petg. About a year later, one of the rows just collapsed, dropping all the filament. The layers on the bracket looked like they had slowly pulled apart over time. PETG is known for very good layer adhesion, so this was a bit of a surprise, but ironically this same thing happened to "3D Printing Nerd" about a year after he set up a filament rack with brackets. For being under a load for a long time, PETG needs CF. I decided to use PA-CF and that has worked very well. A couple months later, another row collapsed in the same way. I then replaced all the brackets with PC-CF and they've held since (2 years now) with no deflection or creep.
PETG strength can vary _significantly_ based on the temperature of the extrusion and the temperature of the build volume during the print. For best results, keep these factors steady during printing and bias toward hotter temps for stronger parts. That first value (in lbs) is repeatable and typical of a hot PETG print. PETG is _by far_ my favorite printing material for parts that need to be strong, but can also be somewhat flexible.
PETG loves to stick to the nozzle. And you won't find matte finish filaments. Not so nice looking.
@@alejandroperez5368 Both valid points, but if the finish isn't as important as the structural integrity, PETG holds up best, especially considering it's inter-layer adhesion is top-notch. I haven't worked with any of the super hot stuff, like PEEK, but for regular ranges, it beats out the competition for me.
@@alejandroperez5368 Try the carbon infused petg for a really clean Matt finish
If y'all can, try to build or buy an enclosure as they can greatly help with the efficiency and layer adhesion of our parts. Even a thin sheet or a plastic bag helps. 😊🌎❤️
i dont understand how some people print firearms with carbon petg, but i can bend the parts with my fingers and they explode like frag grenades under any stress. parts printed at 250 on a 90 bed at moderate flow rate
One dimension I think which is also important, is how good one can glue them. Nylon is hard to glue and ABS can be glued with acetone by dissolving the surface and getting a very good bond. Many dimensions to the properties. Also not all materials are the same when it comes to ease of printing.
You can use Loctite 770 primer and instant adhesive. Very expensive but works great.
@@paulgallagher2937 What can they glue? PP? PE? Nylon? Low surface energy plastics? Nice.
Yes, most of these solutions which actually work are rather expensive.
Thanks for your reply, wasn't aware of this product.
@@AlJay0032 Yes it will work on PP, PE and even silicone rubber
I did try using acetone because its way cheaper, but not even close to same results. They got some magic chemicals in the primers and activators.
Great test, thanks. Like you, I am surprised by the outcome. I'm certainly interested to see a bit more PETG vs PLA testing to see if it's possible to get that super-strength result in a consistent way.
Due to the peculiar design of the bracket, the breaking test appeared to be primarily over-stressing only the least-well-reinforced mounting hole. The angled portion seemed only to serve to change the vector of the stress on the hole. This seems difficult to interpret meaningfully.
A possible explanation for the high load PETG result is that it deflected so much it kind of turned into a tensile test.
It always is a tensile test. Thats what stretching is. The point where it begins to bend is called the yield point. Point they all snap is called the ultimate tensile strength.
It's like when you see those strength tests on wrenchs and the snapon wrench breaks while the cheap crap deforms so much it just spins. Is it useful for a tool to survive more at the expense of its ability to actually perform its function? Nope.
This is interesting and poses the question "Does PLA have the higher tensile strength because it's a purer material?".
By that I mean it has fewer modifiers and additives. You could further ask whether the additions to PA-CF are actually bonded well to the carrier material and are just "in suspension"? and not adding to tensile strength.
Your crane scale appears to be on "lb" not "N". What unit of force were you actually measuring?
I noticed that too. Presumably that's "pounds force", about 4.45N.
@@GenericAnimeBoy pounds is a force not a mass.
Definitely look like lbf...if those values were Newtons, those things would be incredibly weak.
Units don't matter - dial provides a percentage comparison.
@@relativisticvel pound is a unit mass bro, just like gram - pound-force and gram-force are force units on each due to gravity. I’m no dam physicist but that’s at least the most common way to use those words
Hi James. PLA is very prone to creep, even under very light loads. PLA brackets would continuously creep under the sole weight of the sound bar IMHO
This is what I've have always heard too, but I haven't seen any good long term testing that actually demonstrates that PLA creeps more than most of the other common 3d printed plastics. What little info I have seen demonstrates that PLA performs similarly to if not slightly better than PETG, ABS and ASA at room temp. The outliers appear to be PC which creeps the least, nylon 6 the most, and nylon 12 is more or less in the middle.
@@creamofbotulismsoup9900 I got a spool holder I designed on PLA. Thing started creeping a couple days later.
@@chrislosada6701 Personally I haven't had any issues with my PLA prints that are kept at a relatively dry and constant 74F environment. I'd imagine not all PLAs are the same either. Really I'm looking for a controlled experimental data, with a large sample of many different filaments and many other factors such age of prints, age of the filament, whether or not it has been exposed to light for any length of time, moisture content, if it was printed with a heated chamber or a heated bed and of course different environmental conditions. I know it's a lot of work, but that's what is needed to get truly useful data, not just anecdotes from random people on the internet.
PLA is actually one of the best choices for making (low speed) gears out of, because it has such great toughness/bond strength, stiffness and is incredibly resistant to wear.
However, over the last few years, they have been making more subtle introductions into PLA which can make it more rubber-like. This does make it more resistant to cracking, but I wouldn't be surprised if some of you are not experiencing the same results. It is usually a very dense, and very hard wearing material. 🙂🌎❤️
@Zakir Siddiqui (wikipedia) In materials science, creep (sometimes called cold flow) is the tendency of a solid material to undergo slow deformation while subject to persistent mechanical stresses. It can occur as a result of long-term exposure to high levels of stress that are still below the yield strength of the material.
In other words:
Let's say this TV stand can tolerate 40kg without breaking.
Even though, it will slowly and permanently sag (withing several days / weeks) under a load of just 10kg...
PLA is clearly not the right material of choice for applications where a permanent load is at play.
I've experience even more creep with "PLA+" (a small amount of different polymers such as TPU are added to turn PLA less brittle and improve printability)
The nylong is probably printed wrong? is it printed in a heated chamber? PA12 requires that if not it will be very brittle.
That was surprising! Thanks for doing this. For my personal prints, some of which I have products made, I do try different materials (well, just pla and petg) to see what material is better suites for the particular application. Thanks for doing this
hello, i am gonna print a vacuum cleaner wheel, which material do you recommend, petg or pla? the wheel design has small clips to plug in to vacuum cleaner
Well done. I switched to PLA+ quite a long time ago. It's reliable, prints well, and has extraordinarily good layer adhesion.
Thanks for the testing, James. I'd be very interested in the cost/unit of the various material. So many things are engineered to a price these days. For anything done at scale, it has to be a consideration.
A valid question. However, I will just point out that "for anything done at scale", one probably isn't using fused filament fabrication, and instead using something like injection molding or such... which kinda changes the whole game. Then again, there are degrees of "at scale", so, again, a valid question!
I work at a metallurgical lab as a machinist. I would like to pull some filaments on our tensile tester at different temps.
I only use carbon filaments for that matte look. Carbon filaments just look really good, but it just means there is less plastic.. it essentially a filler material, it cant really do much.
Good PLA is just really good, its cheap and the parts are strong and rigid.
I think a lot of these special filaments, are kinda hype.. there is just way higher margins in them.. and everyone starts looking for special stuff at some point. To feel all fancy.
I was wondering about that. It's the fiber that matters to carbon fiber. Otherwise it's coal dust. I'd be interested to see the manufacturing process to understand if they have any way of keeping the characteristics of long fibers in these filaments. Hell, Kingsford's cheap. I'll start making that black stuff by the weekend. Just need a comic book name for it. The wife's probably not going to be happy with me.
PLA doesn't hold up to heat and it's brittle. This may or may not matter to your application.
@@Ideasite there are very different grades of carbon fiber filled materials ranging from ground cf (dust as you mention), milled, aerospace grade (see 3dxtechs website), and continuous fiber (like markforged). CF can definitely improve material characteristics - IIRC flexural modulus of ASA is almost 2x that with CF (based on 3dxtechs TDS)
@@YEETPRIME No. Carbon fiber is only there to make the filament printable (reduce warping). Some carbon fiber dust won't give it any strength.
@@Ideasite also the fibers need to reach from layer to layer to actually make them useful. I could see how they might help reduce warping but i dont really think they make the parts stronger.
Nice job keeping it simple and applying the test to the use case in question.
The trouble with 3d printing materials is that the geometry of any given part seems to have more effect than the mechanical properties of the polymer. Stefan at CNC kitchen has done a lot of work characterising different filaments mostly testing ultimate tensile strength with the layers and across the layers which shows the disparity in strength between the different axes.
I have found when using PLA the problem is temperature, even in the UK a part left in direct sunlight can easily reach 40C and start to deform. I printed a nice drawer to store bits and bobs under my Ender 3, 2 days later it was banana shaped and I had to destroy it to get hold of my spare nozzles and cutters.
That’s why it’s important to understand the characteristics of each material for your project. Pla is more rigid but generally where it falls short is on heat deflection and impact strength. Of course there are alloys that try to minimize those. If you don’t need heat deflection and impact strength 99.9% of the time you can buy the pla. Would love for you to test taulman gf and cf nylons, there finish is injection mold quality.
Good analysis. This was the first video of yours I watched. It was a very good idea to point out that the application should dictate which material parameters are the prime concern (and therefore which material to use), and that there are many more properties of these materials that were not explored and could be prime factors in selecting the best material for your application. Well done.
The biggest advantage to CF filaments is dimensional accuracy. PLA also has a slightly higher density than the other filaments, so other than just having a lower strength/weight ratio, a 1kg spool of it will print fewer parts. Those are all minor issues though compared to PLA's poor creep characteristics and relatively low glass transition temp (parts will just turn into blobs over here in Australia if they're left in a hot car).
i hear having annealed edges helps
@@fluffypinkpandaswhy bother annealing your prints and adding another post processing step when you can just use a better filament
@@alksmdlaks i agree just use better filament. but also anneal the better filament
PLA does that all around the world and will never stand up to a hot car test anywhere. in my opinion PLA is a hobbyist filament to teach people how to use a 3d printer and it has no other use than printing ornamental things that you don't count on lasting.
I thought carbon fibre filaments are just to make parts lighter and prints easier for things like abs and nylon. They aren't continuous fibres so they don't add any strength and typically increase brittleness. Adding CF to nylon basically removes most of the benefits to printing nylon in the first place, I don't understand the appeal
Yes, but also, heat and chemical resistance can be very important for some specific applications!
Quite. I'm designing some posts that need to survive hot air for hours, and neither PLA nor PETG will cope with the temperatures for long. ABS or ASA or better, and nylon can handle even higher temperatures.
This is a great clip. One of the best parts about it is the thought provoking subject and the fantastic group of followers. I learn so much from James, but a ton more from the rest of you. It's amazing~! Thanks, James!
chat gpt ass response
@@MegaChickenPunch I wish I was that smart
Thanks for the results James. Very nice job. I appreciate your dedication.
CNCKitchen has a whole series on testing different filaments. It's a pretty thorough testing regime, too. You may want to check out Stefan's videos and maybe see about getting him some Carbon Fiber nylon to test.
This was interesting and nice to know that I'm not going to be a total idiot for making furniture parts out of PLA.
Stefan's testing is more of a 'laboratory' style parts testing while the testing here is more 'real world' as the part put under load is for a real application. Stefan's testing can give you a good 'base line' while in this video this is more the way the engineers will test actual parts and materials for a given application now there are other things to take into consideration as well as layer orientation when designing and printing a part. Those are the infill density, printing temp (both at the printhead and the environment), part cooling amount, RH of the filament and environment during printing, the way the part is sliced for printing (i.e. weaker areas of the print more walls/infill). All these factors and more can affect the resulting parts total strength and weak points.
We appreciate your honesty, and your fantastic videos sir! Keep them coming if you would be so kind.
This is only the 3rd video of yours I have watched. I would just like to say that my experience has been really good. Your videos just seem to have this perfect balance of information to go along with what you are doing or explaining. It’s rare these days to get to the end of a video and be like “oh, it’s over”. You definitely seem to capture my attention very well and I enjoy all the bits of information you add in. You are very good at explaining things. Love your work. Thank you
chat gpt ass response
Thank you James! I am starting to print tools for my work shop. Mostly 90% for woodworking. 90* for clamping, corners for belt clamps and so on. I have been very happy using eSun PLA +, even as a beginner a year ago. Until I watched your video I was thinking I should use a material other than PLA.
It would be interesting to see how PLA+ compares to standard PLA in this test.
in my experience PLA+/PRO layer adheason is much much better
i heard pla plus and pro are just PLA ABS cross cantamination ratios
@@fluffypinkpandas plus or pro can be whatever you want and it different between different filament makers, so both things does not means something specific but used to differentiate "modified" product
@@SergeiSugaroverdoseShuykov modified pla and modified abs huh.
seems about right
I just printed eSun PLA+ which is a copolymer with TPU as far as I know, seems a bit more flexible, so probably less brittle and stronger. The filament itself feels much tougher to break in hand. It's supposed to be almost 2x as strong as normal PLA. I need to print it over 210*C, because at that temp layer adhesion was slightly too weak and I got a small separation from pulling off bed.
PLA and ABS aren't UV resistant, and PLA has a very low glass transition temperature. The PA12-CF is used for high temperature engineering applications, and I imagine the NylonX can also fit that role.
For general purpose applications, PLA is good, but if you need UV resistance or some amount of heat resistance (such as a part in the interior of your car), PETG is the better choice. The other materials have limited, niche cases where they're preferred, but are vastly more expensive (usually 100-200% more).
I personally print mostly in PLA, but have recently started using PETG for applications around heat sources. I also have PA12-CF for making car parts around the engine bay, such as end caps for headlights/fog lights, small brackets, etc.
What stands out is that the dial indicator doesn’t return to zero on the carbon fibre samples, so it distorts in a short period of time and doesn’t return to its original shape.
The biggest advantage of the PA/CF is temperature and chemical resistance, like for automotive use 👍
PETG has pretty good chemical resistant too. I've actually used PETG to print functional automotive parts that see gasoline and that live in the hot engine bay of a vehicle and it has held up perfectly without issue.
@@bluegizmo1983 agreed but then they don’t have the rigidity of the PA/CF.
I like where this series is going. Keep up the great work!
Deflection over time is something I'd really like to see with the same part. Thanks for the vid!
Impact strength is one of the weaknesses of regular PLA, it's quite brittle. Some improved PLAs trade off some ultimate strength for significantly toughness e.g. e-Sun PLA-ST.
As someone who does this stuff professionally, I really loved your approach on the PTEG. That high outlier is dangerous to leave in place for your data, especially with an n of 3. I'd hire you if that was an interview question.
Great video James, I'd like to add one important thing to the PLA 1st place. As others pointed out the problem for PLA are elements and temperature, therefore these lovely brackets could have a serious issues outside. But what I haven't seen in the comments, is the PLA has tendency to flow under load. It could not be a problem for such a small part vs weight of soundbar, but... Few weeks ago I talked to the guys from MyCello (3d printed tailor made chellos). We have agreed on this PLA behaviour as they had to switch from PLA bodies to the PETG, which seems to be OK for they use. On the other hand I have printed many holders in my garrage from PLA, after few years they're still fine. Even parts on my motorbike where they got beaten by sun and rain. I love CF filaments for plastics which are hard to print due to warping...
I can confirm that PLA flows under load. I built a belt tensioner for a printer with it and it didn't survive a year.
This was a very informative video, thanks! Keep up the great work
As others have pointed out, it appears your scale was set incorrectly to lbs - doesn't invalidate the results, just changes the units :) very interesting video, I've been looking into buying the X1 Carbon, but after this video (and other, similar videos from CNC kitchen), I'm finding it harder and harder to justify the purchase.
Why? You have found more reliable printer, bambulab bannana bed problem made you rethink?
Well done. Concise and straightforward.
One recommendation I would make for petg is make sure you're using as minimal cooling as possible for strength. Petg has poor crystalization properties if cooled too quickly.
Any chance between the one part and the other two you had changed the cooling percentage?
Great video as always.
Yup, i run petg at 10% fan cooling and pla is nowhere near the amount of bending force that petg can take.
Agreed I run PETG at 0% with 10% on overhangs I think there was some layer adhesion issues or something. My PETG parts have been really strong
Eric, I have not done any sort of printing myself but my first thought was the printing or cooling cycles were not the same. Clearly to me anyway this was evident in the results, but also makes me wonder if the others were printed with the best/optimal settings
The problem is, PETG is hygroscopic. So if you decrease cooling, the final part will be full of strings. That's why it needs to be dried out first which is extra work. And that's why PETG is not worth it for me. PLA is the king.
Most filaments do absorb moisture to some extent, and that's why it helps to create a heated filament dry box. (Most of us end up making ours out of clear storage totes, printed roller guides, and using whatever insulation is most available to us. Foam, cardboard, etc)
PETG is a great medium! They're all great mediums, but they do have different properties in which each excels.
I generally print with PETG myself, (for its ease of use and relatively high thermal resistance, (it doesn't *typically* ever melt in a car)) although when it comes to surface durability, (scratch resistance) stiffness, and bond strength, PLA is unique in that it's characteristics (and density) is like that of a cast resin part than other plastics.
While it doesn't take much heat, it can form some Incredibly tough surfaces and it does make for some good (low speed) gears and high load bearing surfaces and rollers which will last a long time. Even in real world scenarios. (I've made lasting parts out of PETG as well. 😊🌎❤️)
I've heard that annealing can greatly improve PLA's temperature resistance, (and overall parts strength as well,) but I've yet to try this. 🙂
Keep in mind that PLA is only 1 of the many bio-plastics we have designed and I'm sure that there will be more that may replace many of the more common ones that we use today, (it all in the formulations and the bonds which gives plastics and resins their unique properties) perhaps making several, easy to use, low cost plastics that can rival most of the things that we use today. Fingers crossed gentleman! 😊🌎❤️
Thank you for making this video. I was surprised with the results.
Sure would be interesting to see a finite-element model of stress to see how it distributes during the pull. I'm thinking that failure is precipitated by the weakest area, after which other areas fail having been let down by the first area. It looks to me like the design is too dependent on the very thin region around the recessed screw adjacent to the "diagonal" side. So if might turn out this is really a test of that specific kind of feature. Not that it's an issue for the actual application, just for the testing results. Regardless -- good on James for doing some tests and entertaining us!
You hit the key way before me! That thin spot queers the whole deal. (Can you use that expression any more?)
He should've tested on a "coupon" of material and not a specific element, this is not the correct way to test materials
Wow, this was an eye opener for me, and like you, I was very surprised at the results. The PETG was disappointing, as I expected more from it. Great video, and another sub. Thanks!
You’re saying newtons of force but your digital scale is set to lbs not newtons
I saw the same - 1 lb is about 4.5 N - assuming James is on the surface of earth as lb are mass and N is force :)
What does that matter for this test genius..??
@@gareth4168
lb is a confusing measurement as it can be both a force and a mass depending.
I’m an American, so imperial makes more sense visually, but when I went to school and imperial was used it was a real pain for engineering.
Imperial mass:
Slugs is the SI unit
lbm is another (but it’s commonly just written as “lb” for added confusion”
Then we have force:
lbf (usually written, once again, as “lb”).
@@dreamrestore213 because if he reports the strength ratings in Newtons on some website, people will think they are around 4.5 times weaker than they really are since it’s actually in lb not N.
But it's not a cookie, it's fruit and cake
Great real material testing. In other section i would like to see similar study for temperature resistance because this is another critical property in order to select it.
I’m a beginner in 3D printing and i have much to learn. Thanks for your work and your channel.
I'd be really interested to see bending and tensile tests of regular nylon against nylon CF. My guess is it might be marginally better in bending but no better in tension. In the world of fiber-reinforced composites, your strength is very dependent on fiber length and load transfer. Unfortunately, these chopped CF filaments have such short fibers that loads don't really transfer to the fibers and instead they just shear around in the plastic. They might improve stiffness in bending though.
Again, these are just my assumptions.
A big factor in the strength of composites is the adhesion between the filler and the matrix. In this case CF is the filler and Nylon is the matrix. Unless the stresses in the material can bridge the filler to the matrix material effectively you essentially just made a part with a bunch of voids. If the Nylon or CF were modified to have better adhesion to each other the strength could be drastically improved.
My experience with PETG (and looking at videos here) is what you saw: very high inconsistency. When I have weird part failures of usable parts, it's usually PETG. It's inconsistent, it's picky with printing, and susceptible to moisture. It's also the filament most likely to damage your printer: I've lost a hot end because the print failed and the filament started sticking to the nozzle, then built up and completely encased the hot end and was in the cooling fans, warped the ducts, etc. On top of that, PETG either doesn't stick to the bed, or it sticks far too well, and I damage the bed removing it. I put my printer in a heated enclosure, and because of this, ABS/ASA have been a little easier to print than most PETGs.
These tests went almost exactly as I expected. Except the PETG break test. The biggest issue with PLA is heat for sure and it moves at MUCH lower temperatures. IF you haven't already check out CNCKitchen
@@derschwartzadder "even a car interior in summer can destroy it". Even? That's an actually hot environment. If you said "even in winter" then yeah.
@@alejandroperez5368 Yes. A car in summer is hot, but not unusually so. The point was that it doesn't require being in the engine bay, or a component of an oven to deflect. There are many real-world scenarios that are simply without the useful temperature range of PLA.
That said, it's still my primary filament for functional prints.
I don’t recall if you did a modifier around the screw holes to beef up the part at the load point . If you just had the 4 perimeters that could impact. Be interesting if you did a solid 5-10mm around bolt in all test materials as that top screw kept cracking. It would also eliminate that Petg riser suspicion.
Cost vs strength the pla would probably dominate but if your preference is the pa12 Then use it. I prefer petg as it easy to print and is suitable for all my cases so far.
Just on parts ductile failure maybe better than catastrophic failure depending on use case.. Unless you add a higher factor of safety to avoid the catastrophic failure - which in fusion you can test for.
Interesting dealing with materials, but preference and ease after win out. Why stuff around for half a day trying to get PC to plot or eliminate warping issues, when you can just drop on petg (or pla or ….) and go off do something else.
Your last point is why I fell in love with carbon-filled polycarbonate. I got a roll just to test, and while it took work to get my printer up to temp (280C hotend, 105C bed), it has always just worked. It's strong, rigid, and has a lovely finish.
It's much more expensive than PLA, of course, so it won't be used for everything, but I know I can just drop it in and have it work. I wouldn't buy another roll if I had to babysit it just to get a single successful print. Regardless of any special properties, the most important is often that it prints reliably.
Care is needed when using PLA for engineering because it will creep over time if it is under constant load. ABS is much more stable over time and under load and that's why it is commonly used for engineering.
I really enjoyed your presentation style. Subscribed.
Nice video. Consider having a larger sample size as well as maybe adding a couple of other materials such as ASA or PC. Personally I stick with PLA just because of how darn easy it is to print, but I'm always open to new materials if they can prove having better material properties for certain applications.
I am looking at ASA for it's outdoor uses. I see all these great things to print that in in Arizona melt
WOW, I'm more surprised then you are. I would have thought PLA would have been weaker. Thank you for the very informative informational video.
Very interesting set of tests. Testing materials is hard, and you did some easy ones, and I don't blame you. But I am a bit unconvinced of the value of UTS tests, especially for a plastic material that can creep. Really, for real world design, it's the yield strength you would work to, surely - OK, after yielding it may hang on for a bit, and that may save someones life or stop something important getting broken, but you'd never design to that standard. The problem is that finding the yield point is hard, especially on a relatively elastic material like you are using. In support of my comment, you yourself were looking closely for signs of yield on a couple of the tests. But yield strength might well give very different answers - the ones that were strongest did seem to bend a long way, no way to tell if they would have gone back, or how far, but quite possibly worse than the low UTS materials.
I print a lot of PETG because a lot of the parts I make should be tough, but for hardness PLA is hard to beat. Two and a half years ago I designed a prototype speed handle for my vice. I wasn't sure if it would survive long enough for me to decide whether I liked the design and make it from metal, so I printed two of my final design that passed minimal use testing. I used PLA.
I'm still on the first of those two PLA prints, and I rarely even need to use the long steel handle that came with the vice. My speed handle has a 6-point and a 12-point socket on it, and the socket I normally use is the 12-point socket. It doesn't have a lot of engagement, yet it has not stripped out.
I didn't anneal it; I didn't expect it to last this long and wasn't trying for longevity. It just really exceeded my expectations.
you say newton but the scale clearly points to lb.
He clearly doesn't give a fig (newton).
Unless temperature resistance is a concern a good quality PLA+ (I really like polymakers PLA pro good balance between stiffness and impact resistance) will outperform basically all filaments until you get up into filaments that require heated chambers to get maximum layer adhesion.
Sure, except if your part is under constant load. Then the creep that PLA experiences will likely make you regret using it.
@@supergiantbubbles Nylons have significantly more creep than PLA does.
I'd be interested to see how the nylon performs after absorbing some moisture (from ambient humidity)
Same here. Please print some for the same test after 2 months or maybe just a day in the sauna.
Awesome video - tests like these are so very helpful - appreciate the work
been printing with PLa for 7 yrs. Never tried anything else , I always felt it was decent for most applications. And this testing corroborates it . Not to say other filaments arent better for certain applications , im just too lazy to experiment for small gains.
What are you printing? Toys? Try printing something that will be inside a car, or in direct sunlight with some weight on it.
@@alejandroperez5368 almost all pellet gun parts, things that arent exposed to sun for long periods.
What a great vice you have there $$$$$$. And the rest of your shop tools. Very nice setups.
"Than" not "then"
Me likes the way your thanking.
I really love this type of videos and it helps tremendously people to decide what material to use in their project.
Only problem is that, it sometimes can inadvertently provide impressions that may not necessarily be true in all circumstances.
I am saying this because I was quite surprised with PA-CF performance. My experience is that regardless of number of walls, or infill density, PA-CF strength can vary wildly depending on temperature and speed they were with.
It also has very low tolerance to imperfections when laying down the layers, whilst PLA just seem to fuse together with imperfections very easily.
Because PA-CF requires much higher temperature and the CF reduces its viscosity, it is very easy to print weak parts due to low temperature settings (not to say humidity, which is obvious).
I'd experiment printing at higher temperatures and test it's strength again. I used to print it at 260/270, but was only when I printed it 280+ that it started to show true strength. It's also interesting that in klipper, the pressure advance value has a wildly different behavior at higher temperatures, which really indicates how its viscosity can vary depending on the temperature, which can indicate it has not fully undergone glass transition and will affect layer adhesion. Both of which are crucial to print strong parts.
By the way. Try annealing PLA. That is something really impressive. It becomes more temperature resistant than nylon and feels like is as strong as aluminium. Only problem is that it shrinks so, makes it tricky with precision assembly.
i recently started using the carbon fibre stuff and loved the surface finish of it, i did suspect it wasn't giving any extra strength. Glad you also came to the same conclusion. overall it looks good but not worth the extra money. thanks.
For PETG i had similar inconsistencies and did some research on it.. It is realy important to print PETG fully dried to have consistent results.. Because the explanaiton is that the moisture inside does microbubbles during printing and you get part that is similar to foam.. after that I religiosly dried and even printed out of a dryer and it made significant improvments specialy the flex before it breaks is bigger...
Super informative, thanks! I think with a different part geometry you could have a lot more strength. Each part seemed to fail at the screw holes, beefing that area up or even having that spline continue in between those two screw holes could make a big difference. But I'm sure it was designed around certain requirements and you made it as strong as possible for its application. Cool video.
This was not an experiment in how good the design was but to compared different filaments.
This has been my experience as well. I love the finish of a carbon fiber reinforced filament. The finished part just looks so much better. But it has never been as strong as its PLA counterpart.
Great video! I would love to see a video testing the creep over time being tested using the same test brackets...good part choice.
yeah, i say, the strenght of a material can differ even from the same manufacturer. Color can also affect the strenght... soo, yeah. I got told that abs was soo good and strong, i usually print in petg. It just works best for me, strenght, doesent react to chemicals, uv and so on. Anyhow, got a abs roll from amaxon, printed, alot of stinking, yeah, and not that strong, not enought for me tu use it instead of petg anyway. Used up that roll, and got a customer that wanted some stuff i abs printed. Bought another roll from another manufacturer, and that plastic was rock hard! the print looked soo good, but it got all warped up and cracked between the layers in the end sadly enought... soo, yeam, amazon basic petg is what i use.
I have been using PLA and PLA pro for almost all my prints, and they are very strong. I have recently started printing with more wood infused PLA and also Metal Infused PLA, Both are very strong and make some unique prints.
Have been designing and printing functional parts this last year and for absolute strength the way you print is almost as important as the material. Also even with pla you want to keep it as dry as possible, even a week in humid air and you will start getting air coming out of your nozzle leading to under extrusion. Personally with pla+ for a balance of max strength and print time I keep to 2mm minimum for walls with a 0.6 nozzle and layers perpendicular to where force will be applied. Also you can cheat a little with PLA, if the part touching the bed will experience the most force. Lastly you can heat the bed up after the print is done and before it unsticks, to harden that surface. I’ll usually heat the bed slowly to 80C for a while then let it cool and make sure nothing warps as it cools. Part becomes much harder and more heat resistant.
I didn't *set out* to do a test of PETG, but I accidentally did and it mirrored your results. IE, inconsistent strength. My application specifically needed some degree of heat resistance (I printed holders for parts that were spray coated and then put into a curing oven). Some of the holders were plenty strong for what I needed, but several just busted during assembly. It was the same eSun PETG you used (though a different color). Gonna try turning down cooling as is suggested here and see if that makes a difference in strength.
Very interesting. I haven't had any real issues with it in the past, but I think the geometry of this part and the resulting overhang created some imperfections in the print that gave the fracture a place to start.
@@Clough42 In my case it was absolutely because it was a long, thin part that had threading on both ends. The only orientation I could print it was the worst one for durability, and the ones that broke all broke in the threads, even with 5 perimeters. Definitely a weak spot. That said, the one printed with drastically reduced part cooling seems to be much more tougher. Much better surface finish too! So I'm changing *that* filament profile permanently.
Well thought out and intelligent conversation. Earned the sub!
At least I did not catch you mentioning.... 1: what the infill amount is for each part. 2: what type pattern for the infill you used, etc. All this does make a difference. Also in my designs. I design my part to be an even multiple of the nozzle printing width. I use a REVO HL .6 nozzle and a .4 layer height and a .5 layer width. So for walls I might for example make my walls 4mm thick. In Simplify3D my slicer I then will use 2 Walls called Perimeters in simplify3D. Which is actually 4 walls 2 interior and 2 exterior. So that adds up to 2mm and then 2mm of infill I print either Cubic or Honeycomb infill pattern. At 8-10% infill @ 120% infill width my parts are really strong. For only 8-10% infill. I used to use 50-90% infill thinking that would make for a stronger part. Then I discovered that going by what I stated above I get just as strong of a part with saving a lot of printing time and filament. So without taking the above into account when you or others are doing these types of tests does not tell the whole story. I also discovered that as you go down in infill percentage and maybe increasing the infill width I get a bigger wider infill pattern. I have always wondered if this is actually more rigid than printing a smaller pattern and more infill. I never did a test. Which someone should do such a test comparing the above mentioned setting. Take PLA and use different infill patterns and infill % amount and also vary the Infill Width. This test would show a lot and I think help a lot of users. And you would have to vary the number of Walls to have a true test.
One thing to note with PETG is that it's hygroscopic and if you print it when it's too moist it will actually become brittle due to the interactions with the steam created when printing. To get the best results with PETG it has to be very dry, I always run mine for a few hours in the dehydrator before printing and I print from a dry box to minimize absorption during a print.
Exactly my thoughts, it looks like first one was printed fresh out of the package, but other ones caught some humidity from the air which resulted in poor structural strength. Dry box is a must with petg.
I don't have a dry box (just haven't gotten around to it) but I absolutely run PETG in the dehydrator for a few hours prior to use. I certainly get more consistent prints that way.
My only experience with a strong 3d print material was "tough-pla" i don't remember what brand, but the name was no joke, that stuff was really frickin strong. I tried tapping screws into it like you would with normal pla and ho boy was that an adventure, it took so much force to work into that material, and I don't think I ever had a single part break, because it wasn't just stiffer, it was also more ductile too, none of the brittleness of normal pla.
I'm not surprised by the PETG results - I've had the same experience. Sometimes it bends and sometimes it just snaps - even if it's the same part and the same settings.
Another great video, thanks for your time making these.
Great test! I think there is one additional property relevant for your application. PLA tends to flow more due to its low glass transition temperature. Would be interesting to see how PLA performs in the initial deflection test when loaded for a week or so.
PLA is only for decorative parts. It's the toy of filaments.
@@alejandroperez5368 sure 🤦♂️ PLA is perfectly fine for mechanical and structural parts. I use PLA for 10 years now and I don't consider most of my parts "Toys".
Awesome job. In the breaking strength test most to all cracked right at the screw cap hole mounts, meaning how tight you had those on each piece would affect the results. In further tests the weakest spot that will break should not be the screw area nor the tied pull area, eliminating a good amount of variability.
Very interesting and educational for a newbie to 3D printing.
I have taken screen shots of both of your graphs and I will use them when choosing a material type under those needs.
Yes, heat and UV ect. must be considered for different circumstances but for most applications, unless you are Machining critical parts, I think your testing covers most needs.
Thank you.
Have a look at CNC kitchen, he tests loads of materials in quite a lot of tests, including impact strength and failure temperature and also tests different techniques like annealing or remelting in salt.
Thanks for sharing. I use PLA for all types of prints, it can be over and under cooked. Color can affect the temperature for the best prints. Printer quality has tripled therefore the prints are fantastic and stronger. Some of my slower prints are mind blowing and look like injection molded parts with PLA, I have painted some to make them more UV protected. Thanks for the testing. There is a video on PLA outboard propeller testing this video made me try harder for better prints with PLA, faster, slower, temperature control, infill.
I'm so glad you made this video. Thank you! I've been printing various parts for a while now and, without doing any rigorous testing, I was under a strong impression that PLA is an under-appreciated material. I also suspected that mixing in carbon fibers doesn't really do much for the strength of the final part, but couldn't really confirm it. Now we know! I still like the *look* of PA-CF parts, though.
It's funny how various myths take hold and persist - PETG vs PLA is another one. You will often hear from 3d printing afficionados "for strength, use PETG". This is nowhere as clear-cut! These days I use PLA for pretty much everything, except when I need those additional 8-10C of temperature resistance, then I'll go for PETG.
From what I've seen, the carbon adds stiffness, not strength.
Thank you for this video it's always nice to see realtime results. I appreciate the repeatability of this test. Obviously with any repeatable statistical analysis the higher the number of tests samples the closer to an actual result, that is my only suggestion. I know that time, effort, and money make it hard to have hundreds or thousands of samples. But for any of these types of results to be reliable, the number of test samples should be in the thousands. Maybe there could be a cross channel collab, where each channel studies many samples?
Another consideration is that PLA has issues with sustained loads. I had a headphone stand with a desk clamp for example and the PLA version loosened every few weeks, until it broke, while the petg version still holds strong after over a year.
Great video, my new favorite material is PC. It prints great for me and love the properties, it's extremely hard. I would love to see how it does in the tests.
Interesting test. PLA is great, but the lack of heat resistance does limit its usefulness in a lot of cases (although I still try using it whenever I can get away with it). The PETG result was surprising -- I normally print mine pretty hot (255 to 260C) with a very slight extrusion multiplier (~101%) and 85C bed, and the result is usually excellent with the Matterhackers PETG
Would love to see some heat testing. Its pertinent to my application and probably the most significant factor.