This is great info thanks for this. I have always added a couple walls to your settings 😅😅 because I though I was making it stronger. Guess ill stop lmao
Hey I just wanted to say that your videos are really awesome! Im a young teen who is really interested to widen my knowledge in 3D printing and material science as I'm having a little project of my own and your videos are helping a lot:)
You altered the parts which invalidated your tests. And do not use only walls, use a mix of the two which was found by CNC Kitchen to be the strongest. I use 8 walls and 98% infill, the very very very tiny voids in the infill actually cause it to be stronger on crush forces which again was tested by CNC Kitchen using his scientific methods.
@@PatrickGunderson I just watched all of CNC Kitchens test video's. Really good stuff. In fact, I just ordered all of the parts to make a "real" tension test machine. Should be fun.
@@HoffmanTactical I would love if you gave some details on your tensile testing setup. I'm getting into using pretty expensive materials for functional prints and I'd really like to be able to do tensile tests to make sure I'm getting the most out of my materials.
too bad you're wrong considering they are all altered in the same fasion.... he simply changed the variables, not individually, but as a whole. Science isn't hard, guy...
@@Shep01 "hurrrrr cant stop da signullll" lmao stop, you arent a freedom fighter. You pay taxes to pedophiles and allow your gun "rights" to be chipped away year after year. Youll do fucking nothing.
So I used the AR lower model but I printed with ABS besides the issues that come with printing ABS the moddle had shrunk enuff to be un usable. Should I try scaleing up the model or just print with PLA?
Check the distance between the take down pin holes, if it matches the upper, then your scaling is fine. If that is the case, try adjusting your XY size compensation. I'd also go with PLA or PETG if you have some on hand.
Increasing infill flow to 130% or even 150% will give you a very good infill layer adhesion while keeping printing time the same. You might want to set your extruder temp a bit higher though, like 10°C more.
Do you mean extrusion width? I wish I could adjust the infill flow separately from the perimeters ;) With a 0.4 MM nozzle and 0.2MM layers I recommend 0.56 MM extrusion widths for infill and 0.45 MM for perimeters.
Good work, a couple of problems I see however. This is only testing on 1 axis & gives you nothing with respect to torsion or deflection not to mention impact. I'm also dubious as to the rigidity when considering the broken print lines in the milled sections. Can't help but think of 2 welded steel plate sections vs 1 solid steel plate section when looking at flexibility vs rigidity. A great start though👍
Great testing. Have you experimented with using larger print lines even without changing the nozzle? Apparently about 120% to 140% extrusion on a .4 nozzle makes for stronger parts. An idea for another test anyway.
cut up a sanding sponge into cubes and put them under the feet and frame along the x axis including under the steppers(At the frame) and it will eliminate most of that printer noise. I just did it because I have 4 printers on a table and my prusa was making the other 3 print with vibrations. A couple sanding sponges later and I have never ever heard the printers print this quiet.
I think people have things a little mixed up... The idea of using more perimeters is for strength on the XY directions because under bending loads the internal tension and compression will be parallel to the perimeters, and then the polymer strength will bear the load, not the adhesion between lines. (The same goes for tension on the XY.. and compression.. but it is not likely that it fails under compression). Now, for layer adhesion, I believe that 100% rectilinear infill could be better because the next layer interlocks with the valleys and ridges of the previous layer that was printed at a 90° angle respect to the new layer
Just what I was trying to show in this video ;) This was a pretty bad way of doing it though. After building my tensile tester I did repeat the experiment and found that rectilinear is only a little bit better then walls, as far as layer adhesion goes.
I would change your infill to perimeter overlap percentage to around 30-35%. There is also a feature in CURA (and I'm sure prusa slicer) called infill wall multiplier. Basically you can double up, triple up, etc your infill walls running in parallel. This means instead of say 50% infill and single infill walls you can run 25% infill with double walls. It will give you very similar strength but cut your filament usage by 40% or so. By also increasing the infill perimeter overlap you get better strength throughout.
I forgot to mention that by increasing your perimeters from 2-4 (or whatever you choose) AND upping your infill / perimeter overlap percentage, instead of getting a default 20% infill overlap of your default wall count you will be getting a 30+ percent overlap of your new wall count. This should be much stronger in bonding the walls to infill and give you stronger infill.
Yes. I've repeated this test with a proper tensile tester and found infill to be about ten percent stronger with PLA. I'll have to do a video at some point on it!
@@MikeJones-mf2rt I've tried a couple Nylons (Matter Hackers and Taulman). Will be doing some videos on them. I don't care for Nylon because of the difficulty to print. I think a good PLA based filament is the best all around option. Every thing else is still "experimental" so to speak.
Interesting test. The issue for me is that you printed the test parts standing up so that your test is essentially loading them normal to the layer lines instead in the plane of the layers. They are likely all breaking due to delamination at the layer lines because of this test orientation. So what this test is telling you is that the infill pattern doesn't significantly change layer adhesion properties. Still good to know. You might want to repeat the test with a part printed lying on it's side. Then you'll be loading the part in the plane of the layers so that the infill pattern has a greater chance to effect strength and stiffness. The 45 degree infill might take shear load better than 100% wall in this case and help with bending stresses.
Absolutely. The goal of the test was to find which has better layer adhesion (don't know why I didn't mention that in the video :). That is important because on my AR-15 lower design, a lot of the stress is normal to the layers rather then in plane with them.
I find your results interesting. CNC Kitchen has shown multiple times that more walls beats infill amount for return on investment. I'm also curious as to how much the infill can be reduced before you see a change in performance. As once again, I've seen lots of tests showing that 100% infill isn't as strong or at least is minimal strength added over 70-80% infill. But I'm also guessing that could change depending on the type of infill. Lots of variables... ha. And of course the closest you can get injection molded solidarity the better! But that would require overextrusion etc. But then I wonder how much creating all those nuclei air pockets contributes to weak spots. FEA required I suppose. I'd like to see more of this type of testing with actual lowers instead of simply using test pieces as the test pieces can get a true idea of the material, but the changes in geometry stresses on a real part are what really matters in the end. Either way keep up the good work!
I have done a bit of testing with a real testing set up like CNC kitchens (his test videos are great BTW). Walls are only stronger in the horizontal direction. Infill is stronger in the vertical, layer adhesion direction. So if you expect the part to fail at the layers, use 100% infill with three walls. If the part is going to fail across the lines, then pack as many walls as you can in. Basically, there is not a one size fits all answer. It depends what you are doing.
I think both are important, and a balance needs to be struck. You want to have a strong print, so you need at least few top/bottom layers and walls (at least more than 3), but you also want your print to be light, especially if it's a big one, and don't use all your filament all at once. So 30-50% infill also need to be used.
When you machined a slot wouldn't that have weakened the ones with more perimeters more than the ones with solid fill? You machined off the perimeters on one side. I liked your video I would love to see another like it.
You should really use carbon/nylon filament with 20-25% carbon fibre content, its way stronger than PLA, its probably the strongest filament using plastics
Since doing this video I've tested a bunch of different plastics. There are a lot of trade offs. Some PLA's are pretty near the top. a carbon fiber Nylon I tested was the strongest at about 13 KPSI. Unfortunately it had poor layer adhesion. Another thing to consider is how much the plastic warps. All the Nylons I've tested warp too much to print a lower with. PCTG I think is a really good option. Prints nice too. Much better impact reistants then Nylon and better layer adhesion and HDT then PLA+. It's not quite as strong horizontally though. About the same as plain Nylon.
Why not use a honeycomb infill pattern instead of rectilinear for increased strength or the triangular hex pattern? Both those patterns should have increased strength over the rectilinear pattern
@@HoffmanTactical i have seen alot of files that are 100% infill but are recommended to use a honeycomb or triangular hex infills due to it providing better strength in the printed frames i get better strength in my prints using those patterns over the rectilinear. I am no expert but that has been my experience
Gonna have to learn the primates and infills. And like to leave multiple comments on the same video. Does it help you. Let me know Do handful for each.
ik im a year late but milling that slot completely invalidates the test, the reason upping perimeters or 100% concentric infill over rectilinear infill makes ur part stronger is the print lines are in the direction of the stresses the part would likely receive, ie along the walls. Milling a slot doesn't preserve the direction of the print, it would have an entirely different internal structure if printed with the slot already 3d modeled into the part.
aligned rectilinear likely has different strengths in the layer direction depending on if it is bent in the X or Y direction; prints will always be strongest perpendicular to the majority of the print lines.
Love that your doing testing like this! But i do want to point out that this test may have been flawed. when milling out the samples it is possible that the friction from the tool may have heated that material to its glass transition temperature. If this were the case than you had unknowingly increased the fusion between the layers which may explain why you seen similar performance between the samples.
@@HoffmanTactical Thats cool man! i am very curious with this topic myself.. Oh btw just printed your receiver! Got it assembled and should be test firing it next weekend! So thank you for the work on that!!!!
Props for putting in the work, but that's a bad method. If the print isn't untouched it becomes unstable. Suddenly, after the milling, the cross section is not something the printer created
LOL. I know, this is an old video. Though to be fair, if done slowly the milling will have only small effect as the interior of the part is not hot enough to change significantly.
I watched a lot of your videos now and my question is why do you keep using PLA? Have you tried any of the high-strength polymers? I'm printing in carbon fiber polycarbonate blend, and in 75/25 nylon carbon fiber, both of these materials are far superior to PLA! Why do you continue to use an inferior material? And your scientific tests were anything but. I'm not trying to be overly critical, but this is not the way to test materials, and in light of the fact you altered those materials after you created them. For all you know you could have heated the parts through friction when you drilled through their cross-section them and change their mechanical properties. Also using a lever and fulcrum to stack your weights means you did not exert the full force on the test piece, but in actuality exerted a percentage of the force by the very nature of the width of the metal plates over the distance of the lever. And then there is the problem of stacking and re-stacking the weights which causes material fatigue which again would not allow you to achieve empirical results. And then there's the problem of testing the materials on the layer lines which really isn't going to matter which method you print your going to be constrained by the coefficient of adhesion. At the very least you should have done two sets of coupons (that's what test pieces are called) print one with the layers vertically the way you did and the other one horizontal cut your printed thickness in half for consistency's sake, and then test them to see if there is a real difference. Anyway good luck
I love the detailed feedback! I'll try to cover some of these points in the next video. Layer adhesion is what I was trying to test, if you look at how I print my lowers you will see why this matters. Milling the parts was definitely not best. Since this is strictly a comparative test, as long as the stacking process is the same for every coupon, I should still get results I can compare. Anyway, I have used PETG for lowers in the past, that is my standard plastic I use for almost all of my printing, along with ABS. If I can get the lower to hold up with PLA, it can only get better with more exotic materials.
Face palm. Take this vid down and redo it. Your new ones are better. You altered the piece/cut the walls and ruined the test. 8 walls 98% rectlin infill
Yes. I know ;) This was before I started any real testing, but I've maintained it here for posterity. I can tell you though that 100 % infill will serve you better then 98%. Some state that the slightly lower infill helps with impact, but that is not so. It hardly saves any weight and compromises your layer adhesion some what. I personally use three walls and 100% aligned rectilinear infill. Emphasis on the aligned infill.
Could imagine how long the video would be if he slowed down. He give out lots of information in as concise and quick as possible. I'm alright with that
