This is brilliant! As other people have pointed out, the pigments used can greatly affect the strength. It's widely known that transparent/translucent PLA is terribly brittle, and also filaments with a bit of moisture (aka ones that have been owned for a few years without a drybox) will behave quite differently compared to brand-new ones!

It's refreshing seeing actual mechanical engineering commentary in this field. Many channels have broached this subject but very few actually do it right.

My biggest issue with that kind of testing is that it completely forget that the pigment can have a huge impact on strength and printability. The impact on strength is not necessarily direct though, some pigment are more keen on absorbing moisture. I have two good quality PLA filament rolls from the same source one is coral pink and prints wonderfully the other is white and is a chore to print with always foaming and leaves a harsh almost gritty surface. I think the issue with the white filament is that they put too much white pigment in it as it was noticeably more opaque than the pink one. So my suggestion for your testing is to test different colors for each manufacturer and to test with "soaking" your filament in high moisture environment to compare moisture absorption and performance degradation. For pigment testing a white and black should be easy to find for any manufacturer and deep blue and red are also probably available for most of them. You could try to ask them directly to have samples for all of the colors from a manufacturer and make filament review video on their products ? For moisture, having 3 points of data a high and low with a driest strongest and a low weak with the filament after days underwater and one after something like 5 hours in high humidity (a long print without dry box feeding). That way we could see which filament are the most affected by moisture as well as how fast they are affected thanks to where the 5h point would be between the two extremes. Even if there is a lot of variable you will probably be able to reduce the number of tests if for example one filament only loses 10% of strength from being underwater, testing after 5h in high humidity is probably useless. For pigment you might see trends that some colors are problematic while others never cause any changes, or some manufacturer have big discrepancy as they focus on looks while other test their pigments to assure similar results for more engineered applications.

So good. Minor thing on the website. Cost/Strength should be the cost of the printed part. The 2 walls vs 4 walls of the same material should have different cost to print. This will give insight into if the additional material usage is worth it.

Visited the website. It is a good start. I wish the filament list was persistent when you changed tests. I have to reload new filaments everytime I switch tests. It doesn't remember the previous test setups either, which also could be fixed. Next, it would help if they all defaulted to the same perimeter/infill just to eliminate a step when adding data. Last, need a "Download result" or image, or something. Not complaining, just giving some feedback. Nice job!

That looks really good, I never knew digital calipers have SPI.

this guy is high on my list of ¨favorite youtubers that post infrequently¨

I think if you did a slight redesign to the sample hold down clamps (standardized insert shape, a foot/shoe on one side, with a knob screw on the other), you could probably reduce it to 1 knob each side. 8 screws seems like a pain in the butt. It's one of those convenience things you don't really consider when designing, but you notice right away when you go to use it. Another area you could test (would require a different module) would be to test the filament diameter consistency. I find if there's a high variance, it'll affect print quality a lot.

CNC Kitchen has gone through this very extensively, you might could collaborate on obtaining this data and making it available all in one place.

This is incredibly good, I've been looking for the hardest material (I mean, the one with the highest posible young module). But when you look into google, people never use the correct terminology. Having a web page with the tensile strength graphs is super helpful. That's a pure act of philanthropy. I really like your content. Keep doing videos like this. And BTW, I did like the part where you explained the testing rig, and i was not boring at all. I've would have loved if it were longer. Again, keep doing this amazing work.

The sheer dedication to do all of this testing, chappau!

Please test strength with multiple samples from the same filament printed at different temperatures. I noticed when I print petg too hot it gets fragile. Please verify hot end temp when printing.

Very impressive! Your onshape CAD modeling is outstanding aswell!

Fantastic job, Michael! Thanks a bunch for all the testing! 😃 Stay safe there with your family! 🖖😊

I would love to have the printer settings used for each filament, along with color, finish, and humidity. The Carbon X1 sacrifices some strength for print speed because the melt zone in the hot end is smallish. To counter this, they use higher stock temps. The problem comes in the fact that thermoplastic is a poor conductor of heat. Therefore, the outside of the filament might get to 230C, the center might only hit 190C. At work, we use an X1 Carbon. At home I use a highly modified Ender 3 V2 Neo. The same print at work might take 30 minutes to print, but 5 hours at home. But the home print is much stronger with the exact same material, infill, and perimeters.

I wish I had time for these kind of projects, thanks for doing the leg work for us and making it free.

excellent video cant wait to see more test, i am doing a bit of material testing myself but for one engineering material only. It quite interesting to see what kinda result you get for other filament. also a note make sure you dry all your filamnet as i found when it take on water it weaken significantly

That website looks awesome! Nice video :)

When you do durability testing it is best to compare the same color filaments or filaments without any dye in them. CNC Kitchen has a video showcasing how different color materials perform differently.

This is extremely useful, thank you!

Suggestion for Materials: BASF PA, PAHT CF15 and PP. Besides this, very well done, good job with the machine and all the programming and stuff. Impressive 👌

excellent and intensive work thank you

Some time back I was working on a website to help a small group improve data management. I used a library called Streamlit and was really happy with how good is was for getting a website up and running quickly, especially where there's data involved. I think it will make it easier for you to scale the current website in future! Also, I came across another project called pynecone that looked really interesting but I haven't had the chance to try it out.

It never ceases to amaze me how expensive PLA is abroad. In Argentina the prices are between 6 and 12 dollars the most expensive.

That machine you built is most impressive! Excellent engineering

Honestly thanks bro, probably going to use this site a lot.

I swear by Hatchbox and Polymaker, but I am also graced with a Microcenter very near by. I thought the Inland brand was cheap junk until I tried a spool while on sale. Man, best PLA I've used yet (regular PLA so far, need to try PLA+, Matte, Pro, and Silk). Thanks for the data!

Good info, would it be possible to print the test pieces in different orientations arent isentropic materials, be interesting to see the failure modes if there printed vertically and how that affects tensile strength :)

Good stuff. Thank you.

Great job.

DATA! MORE DATA! Love this video. Now I want to build my own tensile tester.... my own inner engineer needs more data!

Amazing work

Great video. The electronics isn’t boring. I’d love to see how it works

I hope this website is going to get more attention

Subscribed! Just found your channel yesterday. Good stuff. This is also exactly what I have been looking for. It would be amazing if you could test out Bambu Lab PAHT-CF. I'd like to see if their tensile measurements are accurate.

Damn dude really cool video!!

Great!!! I'm interrested by your traction machine :)

Oh c'mon - don't act like you've never seen a video of CNC Kitchen (@CNCKitchen). But I must admit, your machine looks way nicer. Thumbs up. Because I really like your videos. 10/10 presentation and execution.

Congratulations on the project! Im new on the maker world and can only imagine the work you put in designing the strenth machine (mechanical, electrical and programing) and programing that awesome website. For curiosity how old are you and what did you study? You learned all this self-taught? Thanks and congratulations again, you are really inspiring person.

very cool, wish i knew how to read/interpret the data

Hi! This is a cool project! Please test the Geeetech brand too! Thanx

Awesome!

If you leave lots of little holes in the surface then paint your part with resin, when it soaks in you will have an insanely strong print.

Replace the 8 hold down bolts with two single action clamping mechanism Also, recording video of the breakage, automatically, would probably yield extra qualitative data

I have a test idea I don’t think I’ve seen anyone do. Establish the limits of material strength, than take the weakest material and build up the design in CAD until it’s as strong as the strongest. Then compare size (how much extra material you need to use), weight, cost, etc… I think it would be interesting to see how much more PLA it takes to match CF or something in basic strength test. I generally design parts in CAD for the application AND the material so generally I can make PLA work. I just use more infill, more exterior layers, or just make the part larger. Obviously this doesn’t work in all applications but for the simple parts I make it generally does.

Fantastic work! Polymaker Pla Pro, gotta test that one! Would also love to see nylons tested, specifically Bambulab HT-PA-CF, and Polymaker PA6CF and PA12CF.

I haven't heard a consensus of TPU being in the PLA mixture. If you look at the PolyMax SDS, it has PMMA in it (and it does tend to stand out as an easy print, so maybe there's something there), but most of the others I've found are about 97% PLA/3% CaCO3.

I'd love to see you break test the filament by itself to see if the raw "String" break rates scale up to the creation and determine how much % is gained or loss from the added density and melting. *I guess what I'm wondering is can I use a fish scale and figure out the raw PLA break strength using a bucket of water as weight and then scale up with math how much PLA will be melted into a given space and know what it's hypothetical break strength is without having to actually break the finished product

Interesting. I suppose that the printer and print settings would affect the strength though? Some material that is weaker here might bond better at +20 C extra etc? Or the other way around. But overall it's probably valid tests.

Awesome video! Very impressive machine. I've used step counting in a similar machine and found it very accurate, so it would be cool if you plotted step vs. caliper and see how close they track. Also, would you consider testing other people's samples? We could provide pictures, samples, and maybe a $ donation :)

Bro I needed that Allen wrench once when changing my bike wheel. I didn’t have one so I used a long m10 connecting nut and a wrench😂

You had me at Stack Overflow. 😂

How is the stress calculated ? For solid filaments ita straughtforward as you divide the load bt the cross sectional area. What about about the other sections with different infills ? What area did you use ? Is it the same as a solid cross secrion ?

I love bambu labs paht-cf! Its my most used filent for those parts that need to be super tough. I am super excired to see some numbers on it

Thermal resistance is not something I considered at all to test. Same with impact. Very interesting. Definitely want to see more...

Yes please use other materials and testing method, one method that I would love to see is torque test to failure and permanent deformation with gears to see the strenght of material in mechanical parts

U could have used smaller sample as per ASTM D638 to reduce wastage and printing time.

Did you learn control theory? I think yes, and I want to learn it too please any tips like getting minimum knowledge for quickly getting started with hands on?

Hacking the caliper encoder via its SPI and a pico is genius. Is there a github for the code? I'd love to implement this for closed loop feedback in a servo stage

It woud be cool to get more data with different ways of stress on the material so ppl woud know wich material is best for which job on the project

Please do overture PLA and PLA+

Elegoo is even cheaper, they sell 10kg (1kg spools) for 100 euros. including tax. I spend 7 euros on 1kg of black pla which is insane. Was going to produce my own pla but I dont really have a reason to do so anymore

the caliper outputs measurement data over SPI ?

can you please do this with updated resins tough ones, ive found a custom blend of any cubic tough and siraya tech build to offer a super pla like result and is throw at the wall tough!! ill pay for a bottle if you can get us the number my man!!!!

Website is good but not responsive (on phone), btw great work tell me if you need any help regarding website

You should have been my materials science professor.

Could you try carbon fiber filament

Can you add IIID MAX PLA.

I don't see what temperatures you used, which is going to make a significant difference to part strength by way of whether the extrusions just mechanically interlock like hot glue or fully fuse.

Can I have the 3d models for the 3d printed quad copter please.

compression tests soon???

I think for science you need to try making your own filaments now. Recycled, sure, but also see if you can make filament out of milk.

What engineering degree did you take in college? Was it Aeronautical, Aerospace, Electrical engineering, Mechatronic, Mechanical, or Computer engineering?. The reason I am asking is because I am currently thinking about applying to one of these programs and do not really know what would be the best bachelors but also masters to take. Got any ideas?

Why in the world would you use a Pi Pico instead of just using one of the SPI busses directly on the Pi? That's the whole damn point of them.

Hey, can you make a 3D printed V8 diesel motor I think is gonna be amazing. can you please do it. 🙂

wait no protopasta htpla? :( and you need to do a color comparison of just one brand. Ive always heard white flows worse and that pink is the toughest color, like the pigment changes the plastics properties

I swear by 3dfuel pla+ is $40 a kg but it is worth it in my opinion with print quality and strength

I think Cost vs Strength is misleading. It should be material cost spend on print relative to resulting strength, not just 1kg plastic cost.

Can you do a disc spinning to see what one is the strongest. I am lead to believe that Petg is stronger that Pla.

What, no Amazon Basic????

Nice job but you spent way to much time on the build (albeit a very nice job) and not nearly enough on the plot curves. You need to eli5 which point of the curve you would select for what application and which filament you would go with for that in a clear concise manner.

Does PLA stronger than resin?

you don't use tough PLA? that stuff is strong!

bro using HTML to create the software 💀

run a car over it

Polytera is absolute garbage.