Extrusion Width - The magic parameter for strong 3D prints?

Start your Squarespace free trial today at www.squarespace... and use code CNCKITCHEN to get 10% off your first purchase.
Extrusion width defines how wide the lines of extruded material are. Higher values require more pressure to squeeze the molten filament out of the nozzle that also presses the layers together. I investigated how different extrusion widths impact the quality and the strength of our 3D prints and discovered that this might be a way to produce strong prints in a short amount of time.
💚 Support me 💚
Patreon: / cnckitchen
Join as a KZbin member!
Original CNC Kitchen Threaded Inserts: geni.us/CNCKIn...
Merch: teespring.com/...
Buy an Original Prusa i3 printer: geni.us/CNCKPrusa
PayPal: www.paypal.me/...
Shop at Matterhackers(US): www.matterhacke...
Shop at 3DJake(EU): geni.us/zHvnB
🎙Check out my PODCAST with Tom Sanladerer
/ @themeltzone
🛠Tested Materials
Spoolworks PLA: e3d-online.com...
📚Papers and further information
Slic3r Manual - Flow Math: manual.slic3r....
⚙ My gear (Affiliate Links):
🎥 CAMERAS & LENSES
Panasonic GH5 - Professional 4k60 camera: geni.us/LMN0CmS
Panasonic GX80/GX85 - Great value system camera: geni.us/M2Sm
10-25mm f1.4 - Awesome Lense: geni.us/ZTBH
30mm f2.8 macro - Great Macro Lense: geni.us/vEwqD
12-35mm f2.6 - Great allround lense: geni.us/S9GOsr
14-140mm f3.5-5.6 - My go-to travel lense: geni.us/fSAyKo
25mm f1.4 - Nice prime for photoraphy: geni.us/mqWM
🎙AUDIO
Rode Video Mic Pro - Shotgun mic: geni.us/6JFRdJ
Rode Film Maker Kit - Wireless mic: geni.us/XMD2N
Rode NT-USB - Studio Mic: geni.us/YVONvy
🔴 LIVE STREAMING
Elgatoo Stream Deck: geni.us/ppIiAL
Elgatoo HDMI USB Capture Card: geni.us/imhD
Logitech C920 - Overhead camera: geni.us/ViVgB
🏆 Do you want to help me cover my running costs? Send me a dollar or two over PayPal, it helps me a lot!
www.paypal.me/...
🌼 Even watching the ads before my videos helps me a lot!
Follow me on Twitter: / cnc_kitchen
Follow me on Instagram: / cnckitchenyt
DISCLAIMER: This video was sponsored by Squarespace.

[@CNCKitchen]

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

[@martinpirringer8055]

If you actually limited the speed to 2mm3/sec then the question is did it become stronger due to the increased layer width or the slower print speed. at 100% - .4 layer width 1mm of extruded material contains .4 * .16 = .064mm3 material so 2/.064 = 31.25 so that was printed at 31.25 mm/sec whereas the 250 * was printed at about 12.5mm/sec so the question is does the slower speed or increased layer width make a difference. Also from my experience the protocol would be to properly tune in each setting. So i would suggest this procedure 1.) establish max extrusion rate for your material and machine like indicated here grabcad.com/tutorials/dialing-in-a-filament-and-specifying-the-max-volumetric-e-xtrusion-value 2.) maybe deduct 20% to allow for the wide range of pressure difference At this time lets assume you use PLA and 1.) established 13 mm3/sec deduct 20 % gives about 10 (lets use ten math is probably easier 3.) calc minimum speed of test which would be at 250 % which would be 10/.16 = 62.5 so 60mm/sec would be a good speed with a .4 nozzle. So in Prusa slicer the max speed and max volumetric E could be set to 60mm/sec and 10mm3/sec. With that the nozzle move speed will be constant as per my research (calling filament manufacturers) the print speed at which the nozzle moves makes a difference 4.) Dial in each extrusion width to establish a proper extrusion multiplier. to achieve similar (very close) component thickness the extrusion multiplier can vary as much as .1 (i.e .95 to 1.05) especially at the extremes. to do that I would print in your case a 1, 1,5, 2 and 3mm wall. here is a link to a print that might be useful grabcad.com/library/frc1989-filament-dial-in-1/details?folder_id=7076739 This particular one has a 1mm. 1.25mm 1.5mm 2mm and 3mm wall. Unless you are using linear advance measure close to the middle with a caliper. This also might teach you a thing or 2 about your slicer as you can also check for horizontal layer adhesion (tip print it with .96mm layer and you can peal the 2mm wall into 2) Then print a flat block at least 20x20 in size to inspect the surface you do not want the extrusion multiplier that high that you get a "rough ride" = have a wavy surface. Interesting different materials on the same setup in different conditions can require an extrusion multiplier as low as .9 and as high as 1.1 Now you can run your tests and should get good data on the influence of layer adhesion Then you can re-run the test at 40, 20, 10mm/sec as max speed to see the difference the speed makes. My testing rig are my hands with which I break test pieces so not that accurate. But it seems between layer width speed plays a role too. Now i don't think it makes that much difference with a .4 nozzle but with a .8 it is much more noticable and to get good prints I usually use 110-125% for outside perimeter and 150-225 for the inside perimeters with a layer height of .4 (50%) as with a .8 a print with .4mm layer height will finish in the same time as one with a .2 due to the limiting factor being the max Volumetric E which for HIPS is about 18 mm3/sec the material we print most often and with above procedure you get not only prints strong enough but also accurate enough - look at this test www.chiefdelphi.com/t/rack2020-test-with-updated-video/363607 That is all printed with a .8 all the yellow parts on an anycubic chiron you have 10 gears meshing and another 8 with 2 double. the bearings are also 3DP but on a prusa clone out of Nylon also with a .8 nozzle. You are looking at about 3kg of hips and 1/2kg of nylon in this video as everything is printed solid that means enough perimeters (6-16 depending on the part) to make everything a solid piece of plastic. Layer width for .88 to 1.6 on those prints. Looking fwd to a message as to what exactly was tested a mixture of print speed and layer width or one or the other. Would like to see all of these tested so I stay tuned and a patron

[@VincentGroenewold]

LoganDark yeah we should always get these quality videos for free forever right!? Sooo weird he asks for a bit of money to do all this, but you’re right I’ll be unsubscribing to miss all the info he has just to avoid an ad. ;)

[@martinpirringer8055]

@@LoganDark4357 Have you any idea how much work goes into that - not to speak about the filaments and printer(s) etc. Just curious - how much have you contributed to his channel?

[@AntiVaganza]

@@martinpirringer8055 Which is fine, but no need to "beg". If people feel like sharing, they will.

[@martinpirringer8055]

@@AntiVaganza The way it came across to me was ask nicely

[@flymypg]

The re-watchability of these videos is huge: There's so much here to try and apply!

[@Nabikko]

True!!!

[@BitsOfInterest]

I keep coming back to this series everytime I have a problem, like now when I'm trying 0.5 line width to make design of parts easier 😉

[@NaterNorris]

I've been watching these over and over and learning loads. My first gear fix for a mixer worked right off the bat because of these videos. Nice to see a scientific analysis of the 3d printing tech.

[@alexandredeoliveira8075]

re-watching right now

[@erikwithaknotac]

A year after I first saw this vid, here I am again...

[@jimhize]

I've had good results by setting external perimeters to 110%, and internal ones to 160%. Gives a mix of strength and details.

[@rusername]

thanks

[@danijel124]

You should make a video where you make the ultimate strong print with all settings combined

[@bassbeats92]

Is there a video like that? Or not yet?

[@zodd67]

I was thinking the same then put it up against the most standard print settings

[@lukerickert5203]

Looking at the footage of loading of the hooks I think you can reduce the data scatter if you center the load every time (with some sort of added part). It appears the load placement can change a bit from sample to sample which is going to change the bending moment in the part. Nice work by the way.

[@bersK00]

Maybe he can add small dips or ridges that hold/center the hooks in place on the part.

[@panicraptor2837]

You should make a video on non-printed fill material, e.g. filling the inside by injecting epoxy and other glue types. Parts come out much stronger, but there are many different options for injected infills, and also different infill percentage for printed infills.

[@kirbo1619]

so, molds.

[@WhenDoesTheVideoActuallyStart]

@@kirbo1619 Yeah, I'd just make it out of Epoxy. Bonus for carefully mixing in some 13mm chopped glass fibre. Just don't breathe the fibres nor leave them around nor sand/cut parts with glass/carbon fibres in them. They're an arse to work with.

[@sub-sonicwavescaraudio1814]

This is GENIUS!!! I am SO glad KZbin exists!!! I never would've thought about filling the inside with epoxy. I actually have epoxy for the outside of some of my models though on the outside it's not for strength it's more for the finish of the product but there are some things that I make that could definitely benefit by having a boxy resin fill the inside rather than in the build

[@forivall]

@@sub-sonicwavescaraudio1814 oh heck, I'm also thinking of using this for better transparent parts by using a clear epoxy resin.

[@sub-sonicwavescaraudio1814]

@@forivall That is also a great idea!! I enjoy your analytical approach to problem and solution the information always useful and helpful. I have learned quite a bit from watching your content and I am definitely thankful for all your videos and teachings!!!!

[@Rushmere3D]

When I print vase's I use an extrusion width of 1mm on a .4 nozzle, it creates a really strong print.

[@cameronknowles6267]

Rushmere3D I think that’s a 250% extrusion width

[@surronzak8154]

You are the kind of poeple who slightly change the world, thank you very much for your work. Du bist wahnsinn

[@curtisgauthier9205]

I have been printing a lot of parts that are single wall without infill from one of 3D LabPrint's RC planes and this video helped me solve a huge problem that I have been having where the wall lines don't stick together. I didn't realize before that increasing the line thickness above the nozzle diameter was possible but I just tried it and it's turning out great now! Thank you!

[@prxrb]

Interesting stuff as always! I've been printing with e3d style nozzles and extrusion widths of up to ~150% nozzle diameter regularly as well ever since I forgot to do a nozzle swap at some point and noticed that things generally worked fine anyway. It's so convenient this way, as using a 0.6mm nozzle covers a huge range of usable extrusion widths and layer heights. As for bonding strength, also consider that increasing extrusion width while holding layer height constant increases the contact surface area between layers. The cross section of an extrusion is basically a rectangle, with the sides capped by semi-circles that have a radius equal to half the layer height. These semi-circles do not touch the above and below layers. For example, assuming 0.1mm layer height (25% nozzle diameter), the percentage of each extrusion that is actually bonded to above and below layers for a 150% extrusion width is 150% - 25% = 125% nozzle diameter. And for a print done with layer height = 25% nozzle dia, and extrusion width = 100% nozzle dia, the bonded width is only 100% - 25% = 75% nozzle dia. Thus printing something with two perimeters at 150% extrusion width would result in total bonded width = 125% * 2 = 250% nozzle diameter, while doing three perimeters at 100% extrusion would result in the same wall thickness, but with a total bonded width of only 75% * 3 = 225% nozzle diameter. I wouldn't be surprised if the increase in layer bonding is more a result of the increased adhesion area than pressure effects. Practically, I dont think it really makes a difference as the general advice of using wider extrusions for stronger parts doesnt change, but it might be interesting to test if there's a difference in layer bonding between two prints done at 0.248mm layer height and 0.56mm extrusion width, one done with a 0.4mm nozzle, and one done with a 0.6mm nozzle, to try to isolate out the pressure effects.

[@nprsem]

Looks like a materal for another video

[@indigo1417]

Great explanation

[@Numenor7]

This is great. Looks like 150% might be the magic number. Would be nice to see the hooks tested at 150% and 2 or 3 layers. I don't imagine many people will prefer to use 200% due to the surface quality.

[@theheadone]

I generally use 150% extrusion width for mechanical parts, although mostly it was because I print most mechanical parts with PC-CF and the larger extrusion width helps with the first layer sticking to the bed, great video!

[@Kilgorio]

I gave you money

[@Kilgorio]

And I commented to myself

[@Kilgorio]

What a madlad

[@CNCKitchen]

Thanks for the support, man!

[@daviddunn1766]

I've used wide extrusion width for making vase mode more successful on steeper angles in parts. I've used up to 0.8 on a 0.4 nozzle with decent results.

[@joyjacob7463]

So what should be the initial layer line width in such a case?

[@sergeantseven4240]

You aren't really supposed to increase or decrease the extrusion width past 25% of the nozzle diameter. Anything over or under will cause inconsistent extrusions. If you really want to increase your extrusion width and decrease print time or increase print quality its better to use different size nozzle diameters. I have a set of nozzles ranging from 0.15mm to 1mm, You have to do a lot of settings adjustments for each nozzle size but its much better than pushing extrusion width past the +- 0.1mm of nozzle diameter rule.

[@250_fpv7]

The last Test with the different perimeter settings brings me to an idea: I can imagine why 2 thicker extrutions lines holds together stronger than 4 smaller ones. But as we learned as a kid from playing Lego, it would be better to have the lines overlapping each other. So, would it be possible to have one layer printed at 2 perimeter 200% and the next one at 133% 3 perimeter and then alternate between these two? Is there a way to set these settings per layer? Because that could cause the layers to interact with each other even better. The Material of the middle line from the 3 * 133% layer could float into a potential gap between the two lines from the 2 * 200% layer.

[@MrGreenRunning]

Wonder if PETG would have the same properties.

[@SeishukuS12]

Yep

[@CNCKitchen]

I assume that there should be the same tendency in terms of layer adhesion.

[@martinpirringer8055]

And so does HIPS, ABS and Nylon and probably evreything else I have not printed yet

[@EvilGTV]

PETG definitely benefits (in my case anyway) from at the very least having the infill extrusion width bumped up by a good bit, makes the infill print cleaner and faster, without the normal snotballing behaviour on the infill.

[@BelowAboveAverage]

@@EvilGTV Same here. I definitely get better infill with PETG by bumping it up to 150%.

[@hugogijzen3852]

I whould like a overall tutorial on how to get stronger prints 😊.

[@danielhenaff6630]

Hello Thank's for the translate in french because my english is not enough so good. Daniel from Belgium

[@daliborfarny]

Amazing, your videos really helps me with my own projects. Keep them coming!

[@noelwade]

Stefan - thanks as always for the tests and wonderful data! However, one thing has struck me about your infill tests and comments: Your tests load the parts in tension; never in compression. An internal web or "honeycomb-like" structure (even if its a triangle, rectilinear, or other pattern) is mostly contributing to compression resistance, and perhaps helping with shear loads, correct? It seems like - even disregarding layer adhesion issues with FDM/FFF - this would always be true because you [almost] always have a joint or bonding point between the internal structure and the outer shell (for example, between the ribs and the skin of a wing). That area will likely be the weakest point or the area with the greatest chance of imperfections, stress-risers, discontinuities, etc. Please let me know if I'm overlooking something here - I dropped out of Aerospace Engineering after only 1 year (to start a career in IT). :-)

[@strictnonconformist7369]

I believe compression strength is not tested because it’s not the most important aspect most of the time in 3D printed parts, and you can have absolutely horrible layer adhesion without it being very fatal to compression strength, so long as the layers are properly aligned. It would still be interesting to see that, including testing both the same tests he’s doing here for layer adhesion compared with compression tests, and also deliberately see bad prints for layer adhesion for compression compared to clean ones.

[@avexoidavex3577]

Please more infografik (animated, if you have possibility) and textual information. There are many watchers that dont understand English well when somebody talks. The matters of your videos are very interesting for amateurs and professionals. Fiele Danke aus dem UdSSR. Auf wiedersehen!

[@Sneakydecoy]

Can you do a test on Fillamentums CPE? I would like to see their claim "improved over PETG in every way" tested.

[@MaplePanda04]

Printing with something containing chlorine sounds like a terrible idea. I’m not sure about everyone else, but I certainly don’t have ventilation setup I’d be confident in.

[@Sneakydecoy]

@@MaplePanda04 I don't think Copolyester releases any chlorine. There is no mention of that anywhere. It also doesn't make sense to have a filament that releases it, as chlorine is very corrosive and would destroy the metal parts of the printer. Many plastics does however release nasty stuff if you set it on fire, but if your print is burning you have bigger problems then fumes..

[@MaplePanda04]

@@Sneakydecoy Even at printing temperatures, the filament can release funky stuff. PLA’s familiar sugary smell comes to mind, and ABS apparently releases styrene gas in measureable quantities.

[@pizzablender]

@@MaplePanda04 Acrylic vapours are not any good either - the monomer is very toxic. The terephtalic acid used for PET is not that toxic, hence the use of PET for food storage.

[@iKingLee]

I've used it for the 1st layer to improve bed adhesion when printing directly on glass of heated bed. No glue, hair spray or any other material needed to get items to stick to the glass.

[@vincentmogy143]

U da goat in tha hood

[@JATMN]

Great analysis. Keep up the great work.

[@Rouverius]

Yeah, I've modified it when printing vase mode to beef up the print.

[@YourAverageJoe1000]

Did this for vase mode lightsaber parts, resulted in a much stronger "blade".

[@kuravax]

Just a tip if you use it. Be very careful cuz it really makes the nozzle suffer and jam (experienced during a 20h print while using overextrusion to make it quicker)

[@filipvranesevic8722]

I wonder why he doesn't print the control buttons case for his DIY Universal Test Machine :)

[@CNCKitchen]

Pure lazyness and pragmatism.

[@JefeInquisidorGOW]

Superb video, I smashed subscribe button

[@RegularOldDan]

Yep. I've played with it. On some items that had finer details, I found the defaults in Prusa Slicer would cause some of the details (such as writing) to disappear or get muddied. I've reduced the width to a straight 0.4mm on the bottom layers where the text was and I some additional detail would be preserved.

[@jdally9872]

thanks for the tip, I've noticed prusaslicer (I used superslicer but same thing) sets 0.45mm by default but never bothered changing it. I'll have to try that

[@andreasthaler7068]

Ich fasse die Extrusionsweite IMMER an! Ich habe jetzt das Video noch lange nicht fertig gesehen, aber Du stellst Di Frage ja ganz am Anfang. Mit engerer EW bekommt man geschlossene Oberflächen hin. Den "Fahrplan" kennst Du? www.3d-druck-community.de/showthread.php?tid=25055 Jetzt schau ich aber weiter...

[@andreasthaler7068]

OK. Für mich ist eben die Druckqualität im Vordergrund. Aber interessant. Und nein, es ist nicht der mangelnde Overlap, der die Löcher in der Oberfläche macht, sondern eben ganz genau die große EW. (Ein Blick in die Duckvorschau macht schlau ;) )

[@andreasthaler7068]

Hmmm, Deine Theorien über die Auswirkungen der EW auf die Extrusionsmenge ist m.E. diskussionswürdig. Da ist ein Blick auf den GCODE erhellend. Und m.E ist die EW nicht wirklich so ausschlaggebend für die Menge. Aber ich gehe, wie gesagt, in eine ganz andere Richtung: Ich fahre bei PLA meist um die 0,35 EW bei 0,88-0,85 EM. Da sieht man dan irgendwann auch fast keine Z-Naht mehr. Für schöne Drucke halt was anders als für starke Drucke. Aber ab und an brauche ich auch eine stabiles Funktionsteil. Da bin ich froh ob Deiner Ausführungen. Und das Benchy mit 1 mm EW ha schon was Künstlerisches ;)

[@yoann_m491]

The infill could be set to thicker lines to increase strength without adding time.

[@powderslinger5968]

Hello Stephan! I am always very impressed by your videos. My plan is to review all of your videos and make up a chart regarding parameters and strength and another all about print quality so I can use your data whatever my requirements are! I love the empirical approach that you use. If I had the time and could spare the materials I would do the same experiments but I don't have to since you are so thorough and all I have to do is fine tune your results to match my system. 3D printing is SO easy but getting very good quality and or very good strength is NOT so easy! Thank you for making the learning curve much flatter!

[@CharmPeddler]

If this is something you're going to share out please keep me in mind!

[@abuxxx3607]

this video deserves a sub. you’ve got mine.

[@Spajno]

Peremeters = "Wall line count" AND Extrusion width = "Line Width" in cura? Do I understand correctly?

[@techtastisch7569]

I usually change the extrusion with along with the layer height. I mostly run a layer height of 0,3mm and an extrusion with of 0,5mm on a 0,4mm nozzle so basically 66% of layer height. I adopted these values from the default setting in cura, but scaled a bit up

[@sneaky_tiki]

Yep, this parameter is definitely one of the top 3 parameters I adjust in slicers. Why? Engineer. Sometimes it's because of wall feature width, and how I want the slicer to path through the section. But it's also important to maintain the aspect ratio of your extrusion path. I don't have it readily available, but some research has been done, and you don't want the aspect ratio of your extrusion line to fall below 2:1 if you care about layer bonding. Also, the internal stresses and cooling properties depend on the aspect ratio. And you can maximize volumetric flowrate this way also. Basically, it's a super important setting :)

[@victortitov1740]

I always set extrusion width manually for all my prints. 1st layer - keeping it as wide as practical for best bed adhesion. other layers - it depends on what I need. I always chose wide extrusion for parts that need to be strong, and you seem to have confirmed that it's a good strategy, thank you very much for specific results. And narrow for best accuracy. I actually used wide extrusion for overhangs, but your tests suggest I should pick narrow extrusion for that, so I'm gonna change that habit! Every single video of yours is like a good science paper, I love it. I also take care to keep extrusion rate constant as much as possible, by manually computing print speed for the chosen width, layer height, and material and purpose.

[@Kevin-xi6pj]

As an analytical chemist I appreciate your error bars, but you have to be careful and even just do a students t-test as I suspect there is no difference in the perimeter experiment. Been watching lots of your videos, very good stuff. And great video quality! Thanks

[@truantray]

Oh dear....I hope you mean ANOVA, there are >2 data groups. At least he does three n, most US-based channels try something ONCE. But, the point of n=3 is to be able to do statistical significance, even though this is the bare minimum number of data replicates. Then, there are experimental replicates (N), which should also >3, i.e. not three parts printed at the same time, 3 part printed three separate times. The bare minimum should be N=3, and n=3, for 9 data points each condition group. Then, ANOVA can be done.

[@Kevin-xi6pj]

@okleydokley a parametric ANOVA would be proper looking at his first experiment in the video, and it would even be more proper to gauge whether or not more replicates would be needed through a power analysis, knowing the current variance in his experimental design. Personally would also prefer box and whisker plots instead of bar graphs also. I mentioned the very least of what I would expect from KZbin narrowing focus and drawing a hypothesis from from two data points with multiple variables in the last experiement. If ANOVA were to be completed, it would have to be non-parametric and I can't remember if excel has that option, been a while since I've worked outside of a suite for software.

[@kylerandall9141]

There are some print profiles out there based on this to make printing polycarbonate easier and more reliable. I *think* this is the authoritative source for the info that goes into those, but if someone knows of a prior writeup, please correct me. This is not my article, just a source I've used with good results: forum.prusaprinters.org/forum/original-prusa-i3-mk3s-mk3-print-tips-archive/polycarbonate-here-s-how-to-print-it-without-warping-delamination-or-an-enclosure-with-better-than-abs-results./paged/

[@CNCKitchen]

Very interesting!

[@ricknhelen]

I’ve had great success with this profile and eSun PC.

[@senorjp21]

Great channel. With high extrusion width, more heat is delivered to the spot under the nozzle orifice. Hot plastic warms the layer below, and then additional hot plastic replaces the cooler plastic.

[@LT72884]

Guten Tag:) Great video as usual. I have always used 0.5mm EW for mechanical parts where quality is not an issue, such as my MPCNC. I use 0.3 EW for minis. One downfall is supports with a higher EW due to to thick of supports. However, some slicers have settings to change EW on supports. I dont think mine does. I use ideamaker. Minis look amazing at 0.3mm with a 0.10 layer height. I have tried 0.3mm EW with 0.20mm LH and it did well. one test to try is 0.5mmEW with a 0.28mm LH to see if that yields better strength, but also add 3 shells and 15% infill like you say in the video, but im interested in thicker layer hieghts mixed with thicker EW. Did not know you were in aerospace as well. Me too:) in about a year a should be done with my masters in engineering! YAY!!!

[@chloemcholoe3280]

one cool idea i had that actually has been tried is making holes in your parts. maybe you should also check stuff that can affect the print strength that aren't just in the settings some people mentioned adding small hollow cylinders inside the part to force shells inside the part. but my idea was different. I wanted to make holes in the infill then after it's about to close I'd tell the nozzle to go into the hole, lower a little depending on the nozzle shape and maybe with a high temp extruder a tone of material until it fills the hole. this is perfectly fused material without layers inside so it'll greatly increase the z strength in theory! with a tool changer you could even inject resin into that hole :)

[@selvius101]

laughs in 1mm nozzle

[@court2379]

I would find it interesting to test various dips, solvent bathes and injection. For ABS I suspect you could dissolve some ABS in a solvent until it gains some viscosity, then drop your part in, vacuum the air out, and quickly take your part out. It might require a quick swish in just the solvent to clean the outside off. Another thing that would be interesting is if there were a fill pattern that changed with the distance from the wall. You could have the density high close to the wall, and low in the middle of the parts. Likely achieving the same strength as a higher fill density, but with faster print times and lower material usage. I have also thought that the nozzle tip diameter (not the oriface) has an effect on strength. That it might preheat the previous layer slightly, and definitely an adjacent line on the current layer, by rubbing across it. So perhaps a larger diameter would improve the pre melting. Even better would be a laser that tracks in front of the current line, preheating just the small area the width of the printing line to nearly its melting point. Obviously this would be a significant design challange.

[@lapidations]

I print with wet nylon so I do tweak the extrusion width a lot. I have to tweak pretty much everything to get perfect prints with it

[@tyakoss]

Stefan, I really like these series where you look at the mechanical properties. Many of these probably also apply to other materials. But really, can you stop using PLA for your tests since you would not use it for any actual mechanical parts anyways.

[@yoann_m491]

Stop living in the past, PLA of today can actually be used for almost all mechanical parts.

[@tyakoss]

@@yoann_m491 has chemistry changed since "the past"? Can PLA now do anything >45 degrees without needing to anneal it (which also deforms it, making it unusable for actual parts)? Or is it UV-resistant now, making it suitable for outdoor use? For many actual real-word applications, you usually want to have ~50 deg.C maximum operating temperature, because that is what actually occurs in the real world, at least in some countries. Hence if you want to test that your 3D-printed prototype actually works in your specified temperature range, you cannot use PLA.

[@yoann_m491]

@_ David _ I agree but, if you need something that need to be so strong that pla won't resist, maybe you should use a CNC router or a different method than 3d printing. If you have to bother with a material that warps, delaminate etc just to win 20% of strength that's not practical.

[@3dprinting_and_aviation]

The properties of materials from different manufacturers differ significantly. There are very strong PLAs. Of course, they will not replace the extremely durable or extremely lightweight materials, but will be suitable for 90% of applications.

[@MastaSquidge]

@@tyakoss I'm here from the future to tell you about high temp pla and tough pla.

[@teitgenengineering]

I use 0.8 for bottom layer and 0.4 for other layers except for strong parts thet need better layer adhesion for those I use 0.5 and up

[@karlosss1868]

Some great info here Stephan..Thx The only thing I'd like to mention is that thinner extrusions also have more surface area per volume. This means the thinner extrusion are cooled more effectively by the fan whch makes the bonds a lot weaker. I would have liked to have seen the test done with no fan. Thx again... Thumbs up!

[@whittaker007]

I have changed my extrusion width to 0.5mm when using a 0.4mm nozzle a few times to speed up printing of thin walls. A lot of designs are made with solid wall thicknesses of 1-2mm which is an obvious consequence of designing in terms of 1mm units to keep the design math simple. However if you print a solid wall of 1mm thickness with a 0.4mm nozzle you will get a straight outer wall on each side and a 0.2mm gap in the middle which gets printed with the fill pattern. This slows down the overall print a lot. If you adjust the extrusion width to 0.5mm then you get a straight line on both sides with no infill required. However I have also found that doing this reduces layer adhesion and makes such walls more fragile. But depending on your end use goals the tradeoff in strength may be worth the significant reduction in print time.

[@StephenSchlie]

I adjust it quite frequently when trying to make large, strong parts, e.g. dust collection adapters for my woodworking tools. I'll put in a larger nozzle, typically between a 0.8 and 1.2 and adjust the extrusion width to 1.2x the nozzle diameter.

[@darkthoughts1688]

I believe wall overlap percentage is related. (The overlap between the wall lines.

[@supermerill]

overlap percentage is for the infill, to go a bit inside the internal perimeter, to fuse with the shell.

[@darkthoughts1688]

@@supermerill there is a separate option for just walls on my slicer. It may be called something else on cura.

[@ninj4geek]

@@darkthoughts1688 I use Ideamaker and it has this feature, separate from infill overlap. I usually set it at 15-20% to promote good perimeter fusion

[@darkthoughts1688]

@@ninj4geek yep same here. I've used a few other slicers and Ideamaker has the cleanest UI, despite not having some advanced features.

[@margul]

With all the testing you do why not build an enclosure instead of placing the printer in the basement?

[@TheDementation]

I print everything with 3 perimeters. Early on in 3d printing I noticed I could see through some materials, and as I mostly print models without painting them, I wanted to make sure I couldn't see through them.

[@observed00]

A arrived at at an extrusion width of 120% this way: I had over-extrusion; stringing, and bumpy walls. Testing showed walls became smooth at a multiplier .87. But now there were gaps on the top and bottom infill. Adjusting the extrusion width to 120% made the tops of my parts perfectly smooth. My settings are very far from normal, but my parts a practically flawless, and strong with this material

[@deltafire5058]

120% (.48) is the stock setting in Simplified3D

[@colinehret4840]

Yep, used it to avoid the elephant foot on the first layers 😂

[@technodrone313]

You are definitely my favorite 3d printing related channel. if you're ever in Michigan and want to sample some of the best of the best of american sticky icky i got you sir. You should get a twitch channel so i can give you my prime subscription(5 bucks). I don't care if you ever stream.

[@CNCKitchen]

Used to work near Ann Arbor 10 years ago ;-)

[@technodrone313]

@@CNCKitchen Haha so much better now :D And dude i just fucked with my extrusion width in simplify on some prints i was having trouble getting thin walls to adhere and my first few prints are so much better.

[@IBICO74]

I use 210% on infill to get a better layers when printing fast and also 0.2 more on first layer

[@petersolomon3505]

Great video! Exactly the opposite of what I have been doing, I had been using smaller extrusion width settings to get better top infill on small detailed top facing parts. That is probably a good trick for those cases, but in all others I will go with larger extrusion widths as you have shown

[@chainmaillekid]

I've messed around with really low extrusion widths for the top layer, like as low as 50%, in order to get a kind of ironing effect and an improved surface finish with my larger nozzle prints. (.8mm) Mixed success dependent on the filament, but a very good surface improvement on some.

[@supermerill]

What's your slicer? You can't use the built-in ironing?

[@chainmaillekid]

@@supermerill It was Slic3r PE, I'm not sure the current version has ironing or not, but at the time it didn't.

[@ann_teator]

I use .35 on parameters and .45 for infill and solid layer with .3 nozzle To my eyes it looks better and in theory it should be faster?

[@chloemcholoe3280]

so I just set my defualt to 0.44 perimeters to 0.56 external perimeters to 0.44 infill to 0.6 solid infill to 0.56 top solid infill to 0.56 gotta love all that customization in prusaslicer.

[@marian20012]

your strength test machine sounds like "Michael Gray - The Weekend" song.

[@tokin420nchokin]

I find even just the direction of the layers can play a huge role in strength. I often have to move models around to get usable prints

[@bytesandbikes]

I actually use a 50% extrusion width, so every line overlaps two others in the same layer. Running a 150% extrusion rate.

[@MrHeHim]

Cooling fan settings? My impression is that the thicker extrusions are also more thermally insolated against cooling and/or have more thermal mass thus fuse better on top of also being pressed harder. I always have my cooling fan off if I don't absolutely need it when printing functional prototypes. Also an enclosure makes the part stay warmer and adding in better layer fusing/adhesion. I would love for there to be an option in Simplify3D so that the cooling fan only ever comes on for just bridging and steep angels where the part could curl up. For now it's just by layer which I have to do manually and compromises layer adhesion for the entire layer. Love your work, and haven't played around with my extrusion settings in a long time. This makes me rethink how I've been printing and if there's a setting so that only the outer shell could be 100% and one or two 150% inner shells for faster/stronger prints but still have the same sharp models.

[@danko6582]

Everyone always looks at strength as a factor of mass of filament used. Newtons per gram, etc. But for many of us, the real denominator isn't the amount of filament used, but the machine time. Thus I just consider how I can get the required strength with less print time.

[@nprsem]

Damn, YES! We need profiles not "draft, coarse, normal, fine", but "harder, better, faster, stronger"!

[@BrockPlaysFortnite]

very very helpful thank you

[@FlippinFunFlips]

Your awesome! If I just watch you channel, so many of my questions will get answered.

[@CNCKitchen]

Glad to hear it!

[@kurtlindner]

12:07 Next week's episode maybe? I thought I remembered you doing a nozzle diameter:strength test, resulting in larger diameter nozzles yielding weaker results, I feel like that was within the last year.

[@ddegn]

I'd love to see a comparison between different nozzle sizes with identical layer widths. As you suggest, I'd think the smaller nozzle would have the advantage.

[@erick2will]

One of the Best 3D printing channels out there!

[@TushhsuT]

I have just found this magic parameter in Cura and printing time from 3 hours droped to 1.5! For big prints where you dont need details definitely a good choice!

[@Mobile_Dom]

i mainly change the extrusion width when printing vases, makes the single wall a bit thicker and stronger withouit me needing to go through the hassle of swappign nozzles.

[@martinpirringer8055]

Great video. Greater perimeters have more problems with overhangs as the nozzle got to catch the previous layer otherwise you are just dropping filament next to it. To deal with that do the external perimeter at 110% nozzle and all other at how big you want to go and print outside perimeters first. Now you got the best of both worlds. It also pays to establish safe max volumetric extrusion rate and enter it in prusa slicer I wrote a tutorial here grabcad.com/tutorials/dialing-in-a-filament-and-specifying-the-max-volumetric-e-xtrusion-value cause even with a .4 you can get close or even exceed it and with a bigger nozzle you most likely will

[@ddegn]

In case you missed it Sefan left the following reply under a comment by @maximiliankjarnold : *"I limited flow rates to 2mm³/s to be very far away from what the hotend can handle."*

[@martinpirringer8055]

@@ddegn I missed that and yet I doubt it as that would make for extremely slow prints at the 250% 1mm at .16 layer height would have meant he printed at about 12mm/sec - where does it say that?

[@ddegn]

@@martinpirringer8055 I copied the reply from @CNC Kitchen directly. I included the original poster's name. I have no idea if the relpy as an error or not. Assuming a 1.75mm filament, I figure a material volume rate of 2mm³/s would be a linear rate of 0.83mm/s. I don't have a 3D printer yet myself. I don't know if the rate quoted is reasonable or not.

[@ddegn]

@@martinpirringer8055 After looking at your tutorial (thanks for the link) I see extrusion is usually given as mm/minute. Using Sefan's figure, I get about 50mm/minute. According to your tutorial, this is a very low value.

[@martinpirringer8055]

@@ddegn Hence I think its an error somewhere. I never tested with a .4 and only printed with one when I started and when I bought a new hotend and it got shipped with a .4 even though I ordered it with a .8... That is how I discovered you can print fast with a .4. Now for a .8 with an E3D V6 the max flow rates (add/subtract 10% for different temp) areh HIPS-17, ABS 14, PLA 13 PETG 12-14 (depending on brand) Nylons 6-8 Now there is also the max speed of your printer to consider (most are about 60mm/sec) and the max speed a filament can be printed so for example most Nylons are suggested to be printed

[@PoweredUP_]

I recently tuned my printer good enough and started to look to all cura's settings when I decided that increasing extrusion width might be a good idea for vase mode. Nozzle -0.4, extrusion width -0.6 (it actually ended up to be 0.65 when I measured the model)

[@sealco1972]

I hardly comment on stuff so when I do I feel it really good or really bad. This one is really good. Very good information, show visual results, shows textual results, very well thought out and laid out. A lot of good information in a relative short time, no "jibber jabber". What would be nice to see is a comparison on infill compression strength (and other related tests). Take for example you need to created a special part for something that is not longer available, too long to get, or expensive. What would one have as options with 3D prints. For example, maybe a busing / spacer or heavy equipment floor foot or a dampener made of ABS vs. PLA. vs. PETG vs. TPU. Any what happens at different fill patterns and fill percentage. What about lower fill percentage but different pattern and/or different infill line multipliers. What about general line width and height. And document (print) time and cost (material + power). As a hobbyist, I'm usually more concerned with lower time and cost. However, sometimes quality is more important. This probably could be split up in two (or more) parts. A) Infill compression strength B) General compression strength (ie. same infill parameters but different wall counts, layer height, etc. I did see the Gradient infill video also, more good information.

[@YungassPadawan]

This is good stuff Stefan. Keep it up! Thanks for all your hard work. I rly want this type of info I just don’t wanna research it myself :P Can’t wait til you combine all the data you’ve gathered on print strength

[@first-thoughtgiver-of-will2456]

I suspect much of the strength increase is due to heat transfer on the previous layer (until the layers slightly curl during extrusion due to being much large than nozzle and under pressure). A heated build chamber around the glassing point with minimal nozzle pressure and allowed to cool slowly and evenly may produce very interesting results.

[@dr_frost_dk]

i adjust it when i make the rings for the 3D printed Egg speakers, if i don't adjust it there will be an asymmetry in the front of the ring

[@first-thoughtgiver-of-will2456]

I'd call this academic grade research but its better than the average from academia. Thank you this is excellent.

[@CNCKitchen]

Glad you like it!

[@Gengh13]

5:06 is usually a limiting factor in wall line thickness, I would love to have in a future slicers smart enough to adapt the line width to the shape, so you only have to set the maximum and minimum width and it uses the maximum width possible to print each feature.

[@alexdashwood4976]

Simplify3d do this since version 4.0

[@FransWorkshop]

Wasn't aware of this one, great test. Thank you.

[@waylonjeffery5331]

I have always ran my Line width .04mm smaller than what my nozzle actually is. I use a .6mm nozzle, and running a .56mm line width always gives me amazing results, i've actually figured this was one of the most important settings for the quality of prints I produce.

[@pinkponyofprey1965]

This is invaluable info !!! I've just made a few test prints with a brand new first printer and I know just about nothing haha! Room for improvement taken to the next level ... I've been thinking about the print direction? Intuitively it seems like a standing object is a long spiral that's easy to snap off att the middle but a laying down object has the "fibers" in the other direction so it ought to be a whole lot stronger. Have you made any tests showing the differences if there are any?

[@davidwillmore]

I use Slic3r and I have always set the extrusion widths manually. First layer is tuned to provide the best fit for the geometry of that layer--wider being preferred as the squashing bonds to the bed better and because it makes a better looking surface finish. Other layers are set for similar reasons. This video surprised me as I thought everyone did this.

[@tobiashangler]

Great to see some more systematic and scientific approaches! Thanks for your work and sharing!

[@xl000]

I thought the S in Stefan in German was a hard S, like schtick.

[@nobocks]

On my printer it's 0.6 mm for a single wall pass.

[@fischermann5279]

I completely agree with your werdict

[@alexsie4895]

U r a legend my guy

[@OldCurmudgeon3DP]

And I had just started lowering my line width trying to eliminate gaps between perimeters. Guess I'll go back to 0.48 or maybe 0.5.

[@maximilian.arnold]

Did you change the printing speed so that the volumetric flowrate stays the same?

[@CNCKitchen]

I limited flow rates to 2mm³/s to be very far away from what the hotend can handle.

[@maximilian.arnold]

@@CNCKitchen nice! well done.

[@martinpirringer8055]

@@CNCKitchen That translates at the 1mm layer width and a layer height to a print speed of about 12.5mm/sec - are you sure that is not a typo? And did you put it in prusa slicer in max volumetric E field? Tested with a .8 PLA can handle on a V6 about 11-13 depending on temp and brand here is a link as to how to nail it down grabcad.com/tutorials/dialing-in-a-filament-and-specifying-the-max-volumetric-e-xtrusion-value

[@Gengh13]

@@CNCKitchen using the same volumetric speed means lower speed for higher line widths which means more time in contact with the nozzle per line, couldn't this be also a factor?

[@rajkumarukkuturi5756]

Hi, I noticed that support structures are hard to remove when I print parts with .5mm layer width in Simplify3D compared to when printing with .4 layer width. You havent tested the "easy of support removal test".

[@lukedavid8099]

My air gap is 0.3 vertical and 0.5 horizontal. The bottom quality is pretty good and it snaps off neatly.

[@Gengh13]

I print the support structures with a line width equal or a little less than the nozzle diameter and they are really easy to remove.

[@MrHeHim]

I have a can of compressed air next to my printer, whenever I get a part that's stuck I spray the bottom of the part with the can upside down (can be dangerous!! You could easily get a cold burn) and you can hear the part crack off the built platform.

[@j.g.3d-druckunddesign]

Where I can set the extrusion in cura?

[@phaniavireddy]

Hey Stefan, I would be interesting to see how much you need to beef up an ABS/PETG part in terms of perimeter and infil, to the same strength as an PLA part with same dimensions to that we get both strength and temperature stability.

[@exgenica]

I'm not sure we would see an equivalent strength, but anything that approaches that would be quite useful. Of course, it's seem likely this would also depend significantly on the part's design geometery.

[@arrowstheorem1881]

You done the experiments so that the rest of the world need not. Kudos!

[@billtrondsen]

Great video Stefan! Wondering if an increase in Cura Extrusion Multiplier could also help strength by forcing material overlap between lines... Keep up the great work!

[@A350Avgeek]

How much does a good (not a crazy good one just one that works) 3D printer cost?

[@jupiter2142]

250, check out the A1 mini

[@A350Avgeek]

Ok thanks

[@peekpt]

It will improve overhangs too

[@kleingarrett55]

It would be interesting to see a video of infill overlap percentage, I would expect this to increase the strength in tension in plane and have little to no effect on layer adhesion...but would wonder how much