Which Infill Pattern is the STRONGEST? | 3D Printing Testing Lab

A little while ago, we released a video where we observed the dynamics of different infill patterns and how they broke down under stress. The problem was, we didn't have a force gauge to provide any numbers. Well, now we do.
Witness the strength of 3d printed parts as they bear the brunt, showcasing the unique behavior of each infill pattern under extreme stress. We explore the resilience of 8 commonly used infill patterns, unraveling the strength and complexity hidden behind them. We also discuss some additional considerations that should be weighed against the strength of the infill pattern you choose, in order to ensure a balanced and reliable part. We hope this video helps you understand the practical implications of different infill choices in 3D printing and can help you decide what the best infill pattern is for your next project.
Don't forget to LIKE and SUBSCRIBE to Slant 3D for more great Mass Production 3D Printing content!
#3dprinting #3dprint #3dprintingideas #massproduction #manufacturing #infill #3dprintingcommunity
🔗 IMPORTANT LINKS 🔗
Get a Quote for Your Production Project: www.slant3d.com/
Slant 3D Etsy Plugin: www.slant3d.com/slant3d-etsy-...
Get Our STL's: www.angled.xyz/
Get Affordable High-Quality Filament: www.tangledfilament.com
Try Shapr3D (Use Code: Slant3d): www.shapr3d.com/download?utm_...
Our Favorite Products: www.amazon.com/shop/slant3d
About Slant 3D
🏭 High-Volume 3D Printing: Scalability Meets Flexibility
Slant 3D's Large-Scale 3D Print Farms utilize 1000's of FDM 3D printers working 24/7 to offer limitless scalability and unparalleled flexibility. Whether it's 100 or 100,000 parts, our system can handle it reliably, while still allowing for real-time design updates, ensuring products evolve with the times. This adaptability is key in today's fast-paced world.
🌿 Sustainable Manufacturing: Eco-Friendly Efficiency
Embrace a system that drastically reduces carbon emissions by eliminating carbon-intensive steps in the supply chain, such as global shipping and warehousing. Our approach minimizes this footprint, offering a more sustainable manufacturing option.
⚙️ Digital Warehouses: Parts On-Demand
Think of print farms as a "Digital Warehouse", meaning we can store your parts digitally on a server rather than physically on a shelf. parts are available on-demand, reducing the need for extensive physical inventory.
Produced by Slant Media
As an Amazon Associate, I earn commission from qualifying purchases.
Produced by Slant Media

[@slant3d]

You're stuck on a deserted island with only a 3D printer. What's the first thing you're printing?

[@no-page]

I don't know, but it will be on a raft...

[@taon2325]

A Benchy 🙃

[@ThatGuy-ou4ev]

Nothing, because there is no power.

[@VektorVirus]

TPU pocket pushy. STL coming soon.

[@justandy7963]

a fully functional helicopter

[@triandot]

most people don't consider parts that were squished like 30% to still be usable so i think this video needs an additional graph for peak force before visible (or a specific low percent of) deformation

[@nyrixx]

immediately thought the exact same thing, every test looks to start with each infill already showing some level of deformation.

[@SUBtrauma]

Also, I usually orient my parts so that any load sits vertically on the infil. Gyro when load/impacts come from all sides.

[@grekiki]

You mean yield strength?

[@jameslmorehead]

I'd really like to see your press get some horizontal stabilization and plates larger than the test sample. The lateral motion will cause the data to be invalid since the actual force vector is not being used, but rather an amalgamation of forces within the press and the sample. To fix a similar problem with our press so we could use it for die stamping, I used sheets of UHMW to fill the gap on either side of the moving fixture of the press. This needs to be done equally on both sides so you are not pushing the moving fixture to one side or the other. Also, with the plates being smaller than the samples, stress risers are being formed in the sample. Given this is a qualitive test setup rather than a quantitive, allowing the walls to be fully within the test should not skew the data, but rather just shift it. It might also allow the testing of the incorporation of overlapping walls with the infill. Also, testing in the XY plane looks nice, testing in the Z axis would also be very helpful data.

[@Maleko48]

thank you

[@FrankOnline007]

Second that

[@Liberty4Ever]

It's possible to buy guided cylinders that do not allow lateral slip.

[@Tonyo_oteli]

Agree on this one The test is invalid beside stupid 😂

[@BibboRacing96]

just a square indent should be fine

[@noahkatz9616]

As stated below by @jameslmorehead and myself in your video 9 days ago on infill strength, the results are suspect due to the load not being uniformly applied with end plates, and variability in the location of the round ram relativeto the samples' edges.

[@802Garage]

He has also stated in videos now that these tests are the "for fun" version of the much more legitimate and scientific testing they plan to do later with more advanced methods using more expensive machines.

[@johnlombardo]

I’d really like to see each infill compression tested at different axes, and then move on to torsion, shearing, bending and tension as your test lab expands.

[@volnas1665]

Same, I want to print something to support some weight and I came here to see how much can it handle. Obviously I wouldn't load the prints from the side like this so this video is kinda useless.

[@nickrobertson9533]

I would think crushing the boxes on the "camera" side would show a totally different result from the very different geometry!!! I would love to see that as a side by side...

[@evanbasnaw]

This is very useful to me. I need to print for compressive strength on most of my items so these last 3 videos have given me some important insight into slicing for my applications.

[@Tinker_Balambao]

Looking forward to sheer, tension, and torsion testing. I would love to see this same test and the above tests done on samples that have spherical and circular forms.

[@SliderSS]

Love the video, id love to see one where you show how strong some of the patterns are on their strongest axis, versus the weakest axis, could be useful when deciding how to design the product and infill pattern, Like using grid infill but pushing from the top not the sides

[@oasntet]

You should be graphing peak force vs weight. Different infills use different amounts of material at the same infill %. How much did the cubic print weigh vs the stars print?

[@MichaelBabcock]

This exactly. Also vs time. Strength per minute of printing, strength per gram of material are both relativistic measures people will care about in varying circumstances. Imagine one infill is 25% stronger but takes twice as long to print, that might not be worth it in decision making.

[@aledstarkey7523]

Exactly what I was thinking

[@willofthemaker]

Another variation you should do is the angle of the infill. your lines are at 45° but what if they were perpendicular? Generally, id expect them to be stronger in line with the infill. I bet a pretty large difference between gyroid and 90° grid infill for example.

[@chrisdixon5241]

Came to make the same comment that triandot did below. I'd be interested to see how much force a part can sustain before deformation begins and how much deformation a part could withstand and "recover" from (spring back to its original uncrushed state or very close to). If I'm building a shelf, a slight bowing could be ok but if it starts to buckle then my infill wasn't up to the task - how much weight can it take? Love what you're doing, keep it up!

[@TheReal-MFer]

Thank you so much for making this video!!! (Everyone listen up...I actualyl spoke with these guys...they are Legit as hell and very helpful). I mentioned this as a great idea for a video and WHAM here it is... They REALLY are here to help us out!! We need MORE collaborative creators like this!! Just my .02!

[@slant3d]

Thanks

[@Liberty4Ever]

Great video! Thanks! I design and 3D print functional parts so the engineering parameters are very important to me. Hopefully you can get a better compression test fixture that keeps the ram centered on the part. The videos show the ram sliding sideways as the part deforms by a significant amount, which could cause significant error in the measurement for materials that fail by diagonal shearing. It would also be good to have a small hydraulic pump to continuously advance the ram rather than a car jack to advance the ram a few millimeters at a time. PrusaSlicer allows the infill angle to be changed. It would be interesting and probably informative to see directional fill patterns tested in different orientations. I was gratified to see the print times listed near the end, but it would be good to graph the infill time (not total print time) and the grams of infill on the same bar chart as the compressive strength, or the specific infill time and mass relative to the compressive strength (seconds/newton and grams/newton) which are probably the most useful metrics when optimizing infill pattern selection for print time or material minimization. Force displacement curves are very useful because ultimate part failure isn't the limiting factor. More often, much less part deformation is the failure point so infill with more initial rigidity would be a better choice rather than a compressible infill. It'd be great to see slow motion of the part with a superimposed numerical readout or bar graph showing the force. I'm currently printing 250 ABS parts with gyroid infill. I used the TLAR infill selection method (That Looks About Right), then print one, flex it in several directions and declare, "That ain't goin' nowhere." I'd like to have some better data to inform my engineering opinions and subsequent choices to optimize parameters for additive manufacturing.

[@Phazaar]

This is useful, but misses three key points: 1) Not capping the ends reduces the efficacy of some infills since free expansion into the Y axis allows catastrophic failure earlier. 2) Using a test part smaller than the contact patch on the press disadvantages some infills more than others. 3) Not capturing the moments visible deformation + unrecoverable deformations occur is a shame - getting the measurement visible in the footage would help :)

[@Mash229]

For this to be actually useful information it needs to be a uniform load applied to the top of the brick against a uniform base and the parts need to have equal weight (infill of the same % doesn't mean same weight). Maybe additional part orientation tests too and we also need to see force to deformation mapped out so its clear when a part becomes "broken" for different applications. Without that info these tests are more "nice to know" rather than actionable data. Number of wall loops as another variable would also be extremely helpful for extra points.

[@AwestrikeFearofGods]

Gyroid is a two-domain open-cell pattern, so it fills fairly well. I could see smaller air bubbles getting trapped in local high spots, so it would take some agitation or vibration to fill it 100% without voids.

[@stew675]

I printed out a grid infill cube for fun with no walls. I was able to see through it in every dimension (all faces), and also when views from the corners. It's actually a remarkably open pattern in that sense.

[@conorstewart2214]

Yeah that was what I was thinking, gyroid is not closed cell like cubic is.

[@enterrr]

Yes, it was a mistake in the video to say gyroid is closed-cell like cubic. One of the things gyroid is famous for is open cell in all directions - so if need be can be filled with epoxy, sand, cement, what have you

[@SpatialGuy77]

Yeah, you should vibrate the base plate the entire printing time… 🫨🤦🏻‍♂️ Gr8 Idea…NOT 🤣😂 Michael 🇦🇺

[@AwestrikeFearofGods]

@@SpatialGuy77 Or... Fill after printing, Dick 🇺🇸

[@Enjoymentboy]

I've seen a few videos like this and i don't see the print orientation ever taken into account. Grid might be wrap 6 when crushed from the side but what about when the pressure comes from the top? If i want extra vertical strength I typically use grid. The direction the pressure and shear is going to be coming from us hugely important when picking the infill pattern.

[@kinanhloubi1935]

Thanks for these tips! We produce 3D printed objects, and we print them hollow. We used sand to give them weight. But this process takes a lot of time. Would you recommend other material as fill materials instead of sand?

[@_droid]

The thing about cubic is that not only is it one of the strongest, it also prints fast and easy because it's a linear pattern. The non-linear patterns are definitely rough on the printer.

[@klaidasrunele]

Great video! Short and sweet

[@MaxWithTheSax]

Good to know. I always use cubic for rigid prints and gyroid for TPU.

[@christianerhard8875]

you should run the same test oriented with the vertical axis (pressing as if from the current POV). I would love to see the comparison. Which is best from both axis'?

[@ahmedekin]

Thank u for these useful information.

[@2bakeww]

This was very helpful for me, am looking to 3D print model rocket fins. I'm going to conduct specific test on the parts I print myself, what force Gage/press did you use?

[@TheSEMOMU]

I use the option "infill line multiplier" in Cura Slicer this make the lines thicker improving strength.

[@enyoc3d]

0:45 the ram is moved left by the print - is the print that strong or is that slop in the press? I used to have a harbor freight H-press that I made HDPE shims for to take up the slop - would be super simple to 3d print some great vid as always, thanks

[@raonipr]

i think a good test on this series would be about balancing infill vs perimeters within a same-ish total weight to look for that sweet spot, but maybe with a slightly bigger cube to be a bit more translatable to average structural prints

[@gaveintothedarkness]

Id like to see this too, weight for weight, which infill has the best strength and characteristics.

[@jplummer]

It'd be nice to see how these perform when pressed along the z axis.

[@phillarcher8104]

thanks for that update ;p it would be cool to see how the outside shape work with infill , maybe a version where it's a sphere maybe behaves different than square or a cylinder . Thanks for sharing the knowledge ;p

[@MyTree]

Hi. Thanks for the video. Could you make exactly the same video testing the Z axis strength? The stress was only in the side, which is irrelevant in my project.

[@gj4769]

Why not make filament from hdpe plastic ? I know there is a issue with warping and bed adhesion problem , but both of these can be solved I've read research paper to solve both these problems.

[@Challenge-oo1tb]

This is amazing. cam we get the same test but for twisting force please!

[@atomicsmith]

I think you need some wider mandrels. This really just tested how you aligned the patterns with the edges of the two cylinders. The failure in every case comes from the edge where the sample overhangs.

[@marshallb5210]

Worth mentioning which patterns are non-overlapping, even the X1C can scrape across overlapping infill and knock over parts, gyroid avoids that issue for example

[@GreenFox1505]

I'm very confused by your test parameters. It seems the the force you recorded is just what the press was at when you decided to record the force. Maybe it was when it "failed", or "catastrophically failed", but there is no definition of what that means. It seems like it was just "yeah, that's enough" and you wrote down a force number. Many of these, I felt "failed" well before you stopped pressing them and others I felt could take more but you stopped. You could have done how much the object compressed (total deflection) at a given weight/force applied. You could have done how much force was required to compress to a certain level. But instead it seem to be you just stopped when you felt like it.

[@breakflight]

I assumed he compressed until the pressure decreased instead of increased. Then he recorded the highest pressure it achieved. He didn't say that though, so I could be wrong. It would help if it was more explicit. And maybe also record the pressure when it deforms by a threshold amount, like 3mm.

[@Michael-Archonaeus]

Very useful. Thank you!

[@aayush212]

Can you please make a video on how the gradual infill affect the strength of the 3d printed part ? And what about using it for printing gears ?

[@naasking]

What are the failure loads if you normalize by part weight? Cubic prints a lot more material than lines at the same percentage, and more material typically translates to more strength, so you have to remove that factor if you want to compare the contribution of the infill pattern alone to strength.

[@levmans8865]

Great comparison. It would be interesting also to compare them with 100% infill strenght.

[@BrilliantDesignOnline]

Why were the samples not fully supported at top and bottom? Also, why were the rams permitted to yaw sideways? It seems to me the best test for consistency would be two plates larger than the samples by 50% and no lateral yielding of either plate to give a true result across the tested infills, limiting to ONLY compressive force.

[@ethancohen6005]

Can you make a video of infill for impact resistance, or "springiness"? How likely it can be to regain the shape after deformed

[@TheMrMused]

odd testing in that there are steps on both the top and bottom, which likely set the initial point of failure. Good if that's what you're trying for, but not so good if you're testing straight up compression of the infilled part and trying to isolate the infill rather than how the infill responds to offset failure.

[@conorstewart2214]

It would be interesting to see the same tests but loaded in the Z axis to see how much stronger cubic and gyroid are than other infills like grid or lines in the Z direction.

[@romainlmml3852]

I just have a question, what is the matherial used ?

[@goury]

A test of multiline infills would be nice to have

[@Caldoric]

To be fair, as neat as it is to see the open "cells" etc as everything gets crushed, it would be more useful to do these tests with all 6 sides of the cube covered, for maximum stress distribution/reinforcement, as well as crushing from the various perpendicular axes with said fully-enclosed cubes, because that's a whole other thing to consider. _Also_ also, as a few others have said, getting a bigger press surface would be better so it more evenly crushes the samples would be better, as well as reinforcing the linear-moving parts so they don't experience side-slippage as it tries to crush samples. Lots of variables to tweak.

[@jessicaalvis7063]

I can't seem to find those other videos about tensile strength. Do you have a link?

[@802Garage]

Would be great when you get into heavier data gathering to know how much material each infill adds to a test piece. If the part is 10g without infill and 20g with infill for example, the infill was 10g. Then we can get strength per material usage ratings as well. :D

[@thec-m]

As someone without a 3D printer, do all of these use similar amounts of material for infill or would that vary too? Material waste is a big factor in which infill type to use. Or do they all take up 20% off the space?

[@How_Interesting.]

in the videos you should also show the stress strain graph as the parts are stressed and in general as much information as can possibly be packed on the screen just knowing the peik force used may not always be the most helpful Maybe try get a materials engineer as a mentor to give suggestions on what to include so that it is most useful and you are making the best use of your time.

[@gage5650]

As someone that has done construction materials testing in lab, compressing the entire object between larger surfaces would yield a more scientifically accurate result. There are ASTM standards for compression strength testing using 2" cubes and cylinders (3"x 6" and 6"x 12") that would be worth looking up and following. I also think you could further test and compare the direction of compression, in this test you are essentially compressing across the grain of the infil as opposed to along the grain. It is most likely that like wood, 3D prints along the grain infil would be stronger than across the grain of its infil. Look up beam testing for shear strength testing.

[@malloot9224]

Stefan made a variable infill years ago that was thick against the walls and thinner as it spanned spaces. Still think it's a shame it's not any in any slicer yet.

[@slant3d]

Very true

[@1_HighDuke]

It'd be interesting if you put the weight of each cube on screen since some infills use more material than others at the same infill density.

[@XYZ-print]

Very interesting video!

[@slant3d]

Glad you enjoyed it

[@atomictyler]

it seems like this would be better if the cube bottoms were fully supported. this test was kind of just showing which direction they compress.

[@Fractured_Scholar]

Energy absorption during large impacts? Could be good for cheap car parts considering collisions?

[@davehanna1540]

is this peak force when it deforms or being squished.. also your only compressing in one direction.

[@kokodin5895]

gyroid is not closed cell infill, it is the one you can fill in with liquid the eatiest because each line of "cells" is connected to oposing direction line few layers deeper making it the only infill with direct path from each space to another . it looks from the top like it is closed off but it is not. it could have just crossing sines but i think it would be harder to print

[@Gambiarte]

I'm newbie in 3D printing, but I see cubic as one of the best, having cubic which is strong and fast to print I cant see why using honeycomb as it is very slow to print. Honeycomb has several directions changes, which makes the print head a lot of accelerations and decelerations, cubic is basically a lot of straight lines side to side. Also, cubic has a good relation/distribution between X x Y x Z axis strength. My default is CUBIC!

[@TS_Mind_Swept]

Results didn't go as expected? I was expecting cubic to be the strongest (esp off the print orientation, which is the strongest for most infills) and it was, the others ranked kind of as expected as well; though the numbers themselves might have surprised me. Butt good points about different infills for different applications, I feel like that was the most beneficial part of this SeemsGood

[@bridgetclinch3678]

Why squish across the X/Y axis rather than Z?

[@yddishmcsquidish3904]

A numerical value for force would've been very informative. If only load cells costed an amount within a farm owners budget!

[@catscratchfever1473]

What about in the other direction? Keep it in the same orientation it printed in, not as fun to watch but useful to know.

[@viv3d]

Please do an episode on 3D printing Threads.

[@mpaczkow]

While interesting, I am not sure of the utility of this study as, in use, is it the initial deformation or the peak force that is the parameter to measure - when will the object lose its utility? Also, material and design is so important in the decision - the results would be different if the object was round instead of square (and probably isotropic related to infill direction). The test done here is more 2D (XY ) than 3D (XYZ). Finite element analysis would guide the multi parametric decision without building anything - there are free add-ons for CAD programs online.

[@FileTh1rt3en]

Compressive strength in the other dimensions with the same settings you used here.

[@artineogda]

i use cubic for from beginning. For TPU i use gyroid

[@thewatcher9778]

that ram started drifting causing a different type of load geometry on the parts instead of a stright linear loading

[@uiopuiop3472]

i use the rectalinear

[@butre.]

you have a very different idea of "failure" than I do

[@sleepib]

I think this is a bit misleading, in most situations the main way infill adds strength is by keeping the outer walls from buckling. If you want more strength you're usually better off adding perimeters instead of increasing infill density. Maybe even reduce infill density, because thicker walls need less stabilization. There are some exceptions where infill helps more, like printing with TPU, or when you have a very chunky part.

[@Tex_actual]

Just started watching and I’m putting money in the fact that the octagon is the bestagon.

[@breakflight]

Don't most of these have their strongest axis perpendicular to the one compressed in this video?

[@dhruvmistry6623]

What about snap fit parts ?

[@justandy7963]

maybe a 3d printer that could use one type of polymer for the walls and another type for the infill?!

[@davidsteinkraus]

What about lightning infill on Bambu?

[@uran238fr]

Just a number is not enough. We need graphs. We need to see when elasticitiy ends. We need to see when plasticity ends and it breaks. We need the full spectrum. One number says nothing. Also we need pull measurements. :)

[@MrHeHim]

Every infill has different amount of filament that it uses despite having the same percentage. It's usually only about a 5-10% difference. And there's certain infills the suffer from weak connections caused by part cooling or simply an untuned printer that otherwise would have been as strong or stronger. Anywho, my two cents

[@edfx]

Cubic is the strongest, I wonder if its the weakest when pressed on another axis

[@blalesnail3098]

The masses of these samples are pretty obviously wildly different regardless of what the infill percent says. We need an equal-mass comparison.

[@PrograError]

IMO for strength test this is a bit inaccurate since the only "support" is the X and Y axis where actual prints would have another "support" along the Z axis. it might be a bit more accurate to do a I beam inside the infill area. tho might just talking out of my arse...

[@rodesdara]

In most cases material failed in tensile strength. As for any material - usage must be adapted to forces applied and best strength of material.

[@engine0991]

Is that really 20% infill on each? The Triangle looks so much less dense than the Grid

[@frazier6612]

gyroid is NOT "celled off," it has continuous airflow often greater than that of rectilinear

[@StasonKalbason]

Молодцом!

[@jerbear7952]

You should have tested all of these under a full moon. I also would have like to have seen you test every single filament while wearing different color hats. God the comment sections of KZbin testing videos

[@toto_chanel6952]

Total destruction is not what we use when designing a device. What we are more interesting at is the elasticity. So should you be able to define the maximum applied force before going out of the elasticity zone will be a strong asset for designers.

[@tonykitching5810]

Cant help but wonder if the fact that the press dia is misleading this test. Why not have the test pieces contacting on a flat area BIGGER than the piece ? The press is almost acting like a spear.

[@givemeanameman1]

I am grateful for the knowledge you share, and most of your videos I have found extremely helpful, however I feel this video was not even close to as useful as it could have been and not have taken much if any more time to expand. Testing Elastic deformation/Plastic Deformation/Full Failure on each piece would have been much more informative. For most instances this is not very useful outside of border cases such as creating sacrificial parts such as a plastic cog designed to fail at X force to save other parts of the machine. For most 3d printed parts, past Elastic deformation can be considered full failure and time to replace.

[@hypertectonics7009]

the patterns are anisotropic so the orientation of the patterns and direction of the force have a huge influence which these tests don't take into account. Some patterns will be stronger than others in particular orientations only.

[@slant3d]

1 variable per test

[@chinook9785]

I predict something with staitwalls and closed 3d cell is the strongest, maybe cubic infill. Let's see.

[@chinook9785]

I was right all along.

[@chinook9785]

Even though you can't see through it, a gyroid is an open cell lattice that divides space into two regions throughout the part. The space of every unit cell of the gyroid is connected with every other cell. It is possible to create two holes anywhere, and they are connected as long as a full cell is opened by the two holes.

[@RC-jf6rj]

your upper shaft keeps sliding to the left and right...if it's not pushing straight down, it's really not a valid test...it's putting a false sheer on the parts...

[@H4HDJD]

You also didn’t bring up the amount of filament used for each, like time something to consider

[@SandyWalsh]

You need a press that doesn't move horizontally

[@Vlow52]

That test seems wrong. You need to locate the net towards the strongest force applied to an object. Try breaking honeycomb pushing perpendicular to the hexagons, it will be many times harder. The best one theoretically is not a gyroid but a net of diamond structure made of the single special shape similar to tetrahedron.

[@toto_chanel6952]

Hi thanks for effort to provide us data for better design. However, your testing is not a compression test the way it is setup. It is in fact a peneration test from the top and bottom. In addition a compression test means that the force vector is always in the same diection, Unofrtunatelly your piston being shifting laterraly means the angle of applied force is shifted. This unfortunately leads to inconsistent results to dray conclusions.

[@MaxThor10TT]

Results are void : the orientation matter, like for honeycomb, weak on the side, strong on the front/back. And as already said, pressure is not homogenous.

[@milolc]

I'm sorry, this testing isn't useful. It would be more useful to know how much they can withstand before any significant deformation.

[@aone9050]

@@milolc surprised you never got comments, maybe a correct criticism in a sea of also correct criticisms? I literally have no reason to finish the video, this is useless lol.

[@klausbrinck2137]

u forgot to put some metal plates above/below the specimen, leading to unfair comparisons. The way u choose, cannot guarantee that the placement is always the same anyways.