I live in a crazy humid environment, air drying filaments does not work, it just gets steamed. I was wondering if sublimating with dry ice would work. I thought it might be an interesting experiment at least.
I couldn't see the extent of the bubbling you mentioned, and one would presume the horizontal print would reveal that decrease in integrity, so I have two other guesses for you (which you've probably considered already): 1. Assuming the stringing is the result of additional fluidity, there may be a reduced pressure of the printed paths as they are "driven" into the underlying layer. I believe you touched on a related topic in an earlier video -- wider-extrusions must deform more, under pressure, to form a thinner layer height. 2(a). The greater size-change during cooling and solidification might introducing additional tensions in the final (unbroken) part. 2(b). This may even be during cooling of an individual path, where not only is there inter-layer tension, but increased tension even within the printed extrusions individually, as their skin cools before the inside. Both the fan and the greater temperature difference between the new trace and prior layer could contribute to this -- the potential internal stresses maybe resulting in increased rigidity and brittleness.
The Cooling fan has a significant impact on layer bonding and even layer tension as it can/will cool the print, another factor is room temperature and of course enclosure. Meaning you can print at lower temperature with equivalent results of higher nozzle temps with reduced stringing or curling but keeping the better physical properties.
Good point about the cooling fan having a large effect. Higher nozzle temperatures reduce the initial viscosity of the filament, which fills in porosity more quickly. With the fan on, higher nozzle temperatures only slightly increase the time it takes the molten filament to freeze and solidify. A reduction in fan speed would have a much greater effect on the molten period. A long molten period will start to anneal the filament, by reducing internal stresses, and randomizing the orientation of the extruded polymer chains. Unfortunately, low viscosity and a long molten period also result in loss of detail, due to slumpage from gravity. With fan on, I think most of the loss of detail is due to the lower initial viscosity, because the molten period is not significantly longer. I'm not sure if air pressure from the fan is strong enough to cause significant loss of detail.
I just want to add. From my 2 year experience printing with PETG. Is that the more I use fan for the print the more stringing I get . And also print gets worse strength every time I try to increase fan speed. Also if the Filament is getting humidity it starts stringing like insane. I print PETG at the best middle temperature for my filament, and this is the best option for me, after I increase temp I get more stringing on the print. But If I decrease my direct extruder can stop skipping steps on full infill prints, or the print model starts being weak. I think temperature should be adjusted depending on the print model, sometimes you can increase it or decrease it.
Indeed. Simply upping the temperature w/o changing other settings will lead to prints sagging and melting, at least on smaller prints - and his, very small in crossection, tensile tests samples are particularly vulnerable to this effect. W/o knowing what OTHER settings did he changed to make those samples print at all given difference in temperature, the results are frankly useless, and he must know that. What matters is 'delta T' between the previous layer and layer being deposited, and you can alter this drastically with cooling settings. This video is way below the quality of his other videos when it comes to 'design of experiment', and seems like more a filler and 'discussion bait' to be fair.
whats funny Is i was looking for this video on your channel before it even existed. i was like no he HAD to have done this already. That was like 3 days ago. If you ever read this i just wanted to say thanks, you are really bringing a lot to the community by making these videos, and you are making history here
Thank you for the video Stefan. After a quick search on google, it seems the strength of PLA decreases at high temps due to thermal degradation - other compounds are introduced which may be impeding the interlayer adhesion.
The bigger factor that is unaccounted for by virtually every single 3D Printer... Is the Air temperature of the printer area. The printer should be enclosed, heated, maintained, and with steps taken to reduce moisture content.
Interesting results, makes me happy to see I'm not totally wrong with my 205°C default. I've personally always erred on the side of colder temps but had a lot of issues with underextrusion (though that was also in part due to the extruder idler slipping, which has since also been resolved) so I've inched my way up towards where I'm at right now. It wasn't quite as rigorous a process as you've shown here, more a lot of trial and error. Hope this vid helps smoothen the learning curve for beginners a bit 👍
Wow came back to this video because I have a diamondback nozzle that made me lower the temp for printing. I picked 215 just based on my temp tower for PLA and an glad that is backed up with your test!
Good information, thank you. Good resullt for you on the Polymaker Tournament. A proof of your "know how" and the quality of the information you are giving to your audiece.
Probably worth a note about the accuracy of the temperature sensors found in many printers not being anywhere near calibrated. Required set points will vary by quite a bit (+-15K) depending on the printer
It would be interesting to see what effect a heated chamber has on petg and other filaments. I would guess that the resulting lower temperature delta between layer would help with bonding.
Interesting (and reassuring) that your "sweet spots" for DF PLA and PETG match the temps I've been using for them xD Cheap Chinese "PLA+" from Sunlu worked best at 200°C for me, but the "proper" PLAs from DasFilament (and Procatec) worked best at 215 on my printer.
I think there's more to this.... travel speed for one thing. And as others have mentioned when I scroll through comments- part cooling and chamber temp.
Can you make a video combining all of your findings such as layer height, printing direction, temperatures etc etc? And, perhaps, it can be compared to the standard settings most novice makers use. I'd be interesting too see it compared to the prints that combined the worst characteristics Great video, really useful, thank you very much
You are nailing it with these videos! Yeah, I have no idea why higher temp prints act that way. Weird, huh? I usually look at the specs and pick a temp about in the middle. Seems to work well most of the time. Thanks so much for asking another important question
Hi, Stephan! I think that standing samples (PETG) printed on different temperatures gave you different results because you're over-hotting them. Every layer square is too small. So you're basically "drawing" over the same place with the hot nozzle. It constantly degenerates (on high temperatures) lower layers while printing next layers. But it makes them stronger if the temperature isn't high (a kind of a natural hardening happens). And of course the laying samples become colder pretty quick in different places because of the higher square of printing. So these results can be pretty actual for some prints.
Just a guess here, but could the delta between the extruding material and the previous layer weaken adhesion? The hotter tests, I would think, would show a greater delta between the previous layer and the extruding material. I know when mixing, materials that are closer in temperature tend to blend faster than ones that are highly variant.
Answering your question from just before 2:00 I don't have a 3D printer, I just watch as an enthusiastic fan of 3D printing and someone who's always eyeballing a printer to get it and start
my guess as to why with hotter temps there is less bonding is probably more to do with less pressure between te molten and already set plastic as the materiaal gets less viscous as temperature increases would be interesting to test if this is the case by printing very hot but decreasing layerheight and vise versa
Time between layers is also a factor. If the previous layer is still hot the temperature the resin is fusing at will be higher than if the previous layer has already cooled. Could be a reason the laying samples didn't show the same difference as the standing samples.
About the highest temperature being weaker, my thought is that the fan of the extruder is cooling the already printed part, so there is a temperature delta between the "cold" previous layer and the "very hot" temperature of the filament extruded. Maybe even some retractation (mechanical size reducing because of the cooling). Maybe you can try without the fan :O
At my workplace, getting dimensional accuracy was more paramount than overall appearance (which we can address through post-processing) or part strength. I spent weeks tuning the machine we had to minimize ABS warping, where I settled on temps between 235-238C and, ironically, with minimal part cooling fan (enough to "freeze" the plastic in place, but not enough to exacerbate warping). I'm able to get away with a little fan because the print volume is fully enclosed, and I tend to preheat the machine for at least 30 minutes before printing to make sure the air temperature is stable. There is still warping, of course, such as when the printer transitions from a low-density layer (sparse infill) to a higher-density layer (surface infill); this is where mass becomes a factor in warping. There are ways I can mitigate that, such as with orientation.
Your strength test videos have been great! Very informative and rigourous enough for hobbyists to make informed material choices. Takes me back to my engineering classes! One thing I haven't seen analysed is which filament material has the worst "creep". PLA is supposedly pretty bad, and I've heard PETG is better but there's no actual data. I request a creep test! First suggestion: cut a 100mm strand of filament, hang a weight off for some fixed amount of time and check the elongation. I'm particularly interested in PLA, PETG, polycarbonate, and carbon fiber reinforced variants.
I'm pretty new to this and I'm currently preparing for my first print with PETG. Part of it is printing my first temperature tower and therefore write my own Cura post processing plug-in so that I can set my temperature points comfortably. You've said that you set the nozzle temperature using the M104 command. So far I've gathered that that's not enough, because if you do it this way your nozzle cools/heats up while printing, leaving you with a fuzzy temperature transition. You need to send M109 as well, which tells the printer to pause until the given temperature has been reached.
In other hand if you pause it, the last layer will be a lot colder when you get to print again, which would definitely lower layer adhesion wouldn't it?
@@Kalvinjj Yes, but that's not the point of temperature towers, right? It doesn't apply to this test as Stefan has printed each batch separately to test layer adhesion at a certain temperature. But if you want to observe print quality not waiting skews the result slightly
@@saschaschneider6355 hmm I see your point now, makes sense on this case. I wonder how much it would affect tho, since the nozzle temperature change is in like 10ºc (or is it 5? 15?) intervals, it might be able to change temperature fast enough, tho I guess it depends on where it would start the next layer.
@@Kalvinjj sure, but if it heats up/cools down fast enough then M109 isn't going to slow down the process by a lot either. Hence I think it should be used in addition to M104 either way. edit: as M109 also takes temperature as an argument you could also set it slightly below the target temperature, but that's probably something I'll have to play around with - and probably will be included in my plug-in as an option
I always use 5c above the minimum manufacturer stated range. I've never printed a temperature tower nor do I tune temperature unless it's for first later adhesion.
In this test each dog-bone temp group was printed on it's own so it was printing hot filament on hot dog-bone, but some of the old test videos show many groups being printed at once. I always wondered if the hot filament on cold dog-bone, because it takes so long to get back to the first object when printing many at once, affects the test.
I haven't printed PLA in a long time but most of my ABS, ASA, PETG and PETG CF prints have been around 240-245, less and the layers (ABS + ASA) don't bond as well and a higher temp curls the edges a little effecting appearance. The size of the part will also effect the temperature I choose, a small part that will not have a chance to cool fast enough will get a lower temp while I can run a higher temp on a larger part without effecting it's appearance.
It depends entirely on the application (like with part cooling, nozzle diameter etc.). But why not also compare with ABS and ASA when talking functional prints? They practically cost the same (unlike PA/Nylon or carbon-reinforced filaments), but you have to be able to print it first, of course
You kind of answered it yourself. Unless you're already looking for something very specific that you really only get from those materials (rigidity. It's rigidity) you aren't going to be using them. PLA and PETG are general use materials. Neither ABS or ASA really is, what with the significantly increased headache that comes with them with the benefit of less strength than other things that cost the same amount. They really don't belong to the same category as PLA/PETG any more than Nylon does. (And while we're talking here, "nylon or carbon-reinforced filaments" are pretty far away from each other, given that it's already very well established that carbon-'reinforced' filaments are actually much weaker than their cheaper equivalents and the only benefit out there is that you can print them at a higher rate)
I would assume expansion coefficient plays a large factor in the higher temperature tests, causing excess layer tension compared to lower temperatures. I'd be curious to see if annealing makes any difference with the higher temperatures.
I personally use a small retraction test that prints two towers, one at a time at different temps. that way I can judge both stringing and layer adhesion without using as much filament as a temp tower though that doesn't test overhangs but I have found that if minimize stringing that overhangs usually end up looking fine.
Hi Stefan. I wonder if you can get your hands on a Xube 3D printer. It's based on a process called "xolography" in which object are printed entirely at once rather than layer by layer. They say it's meant for researchers. Maybe they can let you borrow and test one of their machines when it's released.
Since the viscosity is required to be adjusted for increased speed , it doesent really matter as you only got a fixed temp about 40mms plus to about 50, after that if you dont increase temp you will get feed issues. The temp will however not be increased per cm so an increase will only work up to a certain level ams it is linear to speed more or less
Some form of microscopy in the cross section of layer adhesion possibly provide some insight. probably some research in thermal/bonding properties of thermoplastic polymers
I always go 5C above the recommended temperature, and if that's good, I'm done. If I get weird results, then I print a temperature tower. But that's only for printing with that nozzle at that speed with that layer height. A single filament can print very differently between slow printing with a fine nozzle and thin layers, compared to fast printing with a wide nozzle and thick layers. I can't count the times I've gone from a sub-scale proof print (nozzle, layers and speed all scaled down) on my Sidewinder X1 v4 to a full-scale final print without taking this into account. This is when doing both temperature AND speed towers becomes important, especially when doing large prints after doing crude drafts and/or an accurate sub-scale print. BTW, doing perfect sub-scale proof prints lets you do photography of the proofs that you know will very closely match the final full-scale print. Yes, the print time is about the same (intentionally), so these aren't draft prints. But they do let you make changes that you know will be accurately reflected in the final part, and do so for a MASSIVE savings in filament compared to noticing a mistake on the first full-size print. But it is a lot of nozzle switching...
I normally print PLA a little warmer, 210 C, because it gives me the best results (I haven't tried higher) for the way I store my PLA (in sealed bags, dried after being out of the bag for a few days.
I once printed 12 little Lego men which worked fine, but trying to print just two with same settings ended up a globby mess. The reason was that with 12, each layer had a bit of time to cool but with 2, the layer being printed on was still mushy and so got pushed around by the extruder. I have found this effect for complex prints with some parts large and others small islands...one temp setting isn't always optimal for the whole print! I think a layer area : part volume ratio ratio may be able to be used to tweak fan speed in slicer software.
On one of my printers, 210 is the best temp. On another, with the same PLA it's 190.. It think the difference is in the quality/accuracy of the thermistor (maybe even a FW setting to set for diff style thermistor).. Another factor that may come into play here is layer time and how long the previous layer has to cool before next layer is applied.
how about printing cold with an extruder like the Volcano, which has a longer hot section? While the temperature is lower, the material has more time to heat up inside the nozzle.
Good point and something I wanted to take a look at for ages. I have some high-grade Nylon here that is basically unusable when not printed with a volcano nozzle because it needs time to heat up.
i use PETG at 245 nozzle and bed at 70 degrees and the prints i get are pretty strong and i must add that i use 1.5mm for wall and bottom and 20% infill and no part cooling fan,grüss aus holland :)
I wish you would talk and experiment more about the fumes! Maybe collaborate with experts that have fancy Test equipment? What do Pneumologists (lung-doctors) say? What do we know from the industry that works around this stuff every day?
Up to now, I've always selected nozzle temp to give the best aesthetics. But now, when I'm doing functional prints that don't need to look great, I'll consider strength over beauty.
Nice Video. I like your content. I think you should test this again with various cooling fan settings as well. Maybe this would be the solution for losing strength at higher temps.
Anecdotally I've always found hotter is better for amorphous polymers (eg abs/pc). It also makes a huge difference at a specific temp if you have a cooling fan enabled or not. For ABS, it's always stronger without a fan - hence why dual extrusion machines have a huge advantage because you can print very hot without a fan without sacrificing accuracy from support gaps. This is why stratasys machines don't have cooling fans.
@CNC_Kitchen : Stefan, have you ever tested the best layer direction for side-loaded 'column' type printed parts? I'm in a situation where I can print a square tower-like part in different alignments. It will be loaded from the side. And I'm not sure which route I should take to get the best strength. And I lack your setup to test it, of course. I love your videos especially the thorough tests about strength! :)
One possible reason the strength drops of at higer temps, might be because the already printed layer cools off too much, so the diff between the 2 layers are bigger? Would be interesting to see if there is any difference printing 5 and 1 item at a time.... Or maybe add g-code between layers to have the nozzle go park in the corner for X number of seconds.
I think that it is not really possible to conclude on the stress, because it depends on the elongation at break which is known to be very variable. I think it would have taken at least 10 tests per temperature to get an idea, and probably trying to identify a Weibull distribution would be relevant.
True, though this amount of tests is just not practical for me. I try to investigate more thoroughly than 95% of the others but just can't go into the details like in a scientific paper. Otherwise, I would have deserved a Ph.D. a long time ago 😉
@@CNCKitchen I totally understand! It's already great what you're doing, it's clear that you're the one who spends the most time looking at the impact of printing parameters on mechanical performance, it's even often impossible to find the experiments you're doing in the scientific community, so it's a great help for everyone, don't take this as a challenge. It's just that when (like me) you don't do the tests, your conclusions seem logical, but unfortunately, it's not impossible that it's the result of chance. That said, you're right, the work to be conducted is clearly a material engineer's job (not necessarily that of a PhD), and I fully understand that the time needed to conduct it is not compatible with your priorities! Maybe a little nuance in the conclusions ... Once again, thank you for this work which is useful for all of us researchers ! P.S: if I could give your Patreon to the lab, it would have been done a long time ago !
Because your Stefan... ...considering features in the firmware and slicer used - 'coming up with new features: How about a "spot welder" like feature? Where the nozzle pauses in the print at corners every few layers and semi-reheats those points through and past previously laid down layerings? At those points, the nozzle would heat up 10C or whatever for a duration, and then return to its assigned temp, and continue printing, until it got to the next "spot welding point" where it would then stop, and increase its temp' again - and repeat. Royally add print time, and start challenging image quality, but considering coming up with new features and being creative......... ;) Peace love 'n' happiness from the Pacific NW of the US. :D
I was printing PLA from DasFilament at higher speeds with 220C and 0.3 layer height. I had rare extrusion issues. I found, raising my temperature to 235C for the first layer, and 230C for all other layers solved the issues completely, and the surface looks smoother as well. I am using MK3 and MK3S printers from Prusa. The given 215C the guys from DasFilament claim, did not work for me. Me, not printing figurines and stuff. I am using my printers for creating electronics enclosures, and RC stuff.
When i want something simple with high strength, i usually use for PLA 270° 1.6mm extrusion width at 0.8mm nozzle diameter no fan speed down to 50% and 0.3 layer height in a chamber at arround 50°C At this settings layerlines disappear,the part almost looks like vapour smoothed, but with a textured surface. It looks really nice, and i suppose its pretty much perfect layer adhesion, but i never tested strength.
I have until now used my Smartfill, AzureFilm and Fiberlogy at around +5c above lowest temp range on the filament. So if rated at 190-220c, I go 195c and usually with just the right results. However I am still trying to dial in PETG and Fibersilk as these tend to either string or foam rather quickly if temp is too high. My Fiberlogy Impact PLA is spot on at 225c and have a lot of cool prints using this on my GooFoo Mido printer.
I have noticed that PETG become brittle when cooled below 40 dg to fast, I have had luck putting a fence of isolation around the build pate, so that the temperature is kept high by the build plate heating, then te strength become more like when printed flat.
Actually i start without a temp-tower and use the arithmetic middle of the temperature the filament producer writes on it :P. Usually for PLA it is around 205°C. When i see strining i drop by 5°C and in most cases land on 200°C ... that's it.
Very good video as usual. Regarding the strength of the PLA, I found that some filaments are stronger than the others from the same brand. For example, the copper color is the weakest I have (by far), all others (wood, black,..) offer higher strength. (from eryone, cyclone filament,..)
I was instantly thinking the elevated temperatures would increase bubbles and decrease surface area between layers, therefore weakening the bond between layers.
I suspect degradation and bubbling/blistering could both be playing roles. I would think that since "over-cooked" thermoplastic loses molecular weight and damages polymer chains/branches, that would to less entanglement between layer surfaces. Add to that more voids from bubbles/blisters either naturally from internal moisture or absorbing atmospheric moisture? via gassing-off? Additives vaporizing? not sure, BUT the bubbling could be tested with a microscope I suppose, and then the cause narrowed down with... more elaborate means (printing in inert gas to vacuum environment and very low humidity, etc?). On a somewhat related note, you should take a look at PLA layer adhesion with much higher bed temps and ambient heat. Anecdotally at least, it seems to have more effect than raising the nozzle temp. I have not tested directly and could be off base, but stumbled upon it when I raised the bed temp to try annealing a cooling shroud while it was printing. could be worth a peek to confirm or deny
Hey Stephan, we all know that higher temperatures increase layer adhesion, we also know that cooling decreases it. But sometimes you want stronger prints but the higher temperature might be too much (benchy smoke stack thing) there is no video regarding the combination of the too. So would it be better to print at 200 degrees but no cooling or 230 with 10% or 20% cooling. I watched both of your videos and there can be some interpolation made between them but we cannot be sure that the scaling is linear and this would really help me in lots of situations. Thank you
Did I miss something or did you not ever mention if you were using a part cooling fan or not? The temperature difference between the previous layer and the current layer being laid down is very important for comparing layer adhesion.
Interesting stuff! I print at 180c with creality pla lol. I don't have any significant problems and get low stringing, I only print models atm so I can't talk about strength.
I was having a ton of issues with the polylite PETG. I finally get it to stick at the max temp it says on the datasheet. Was anyone else having issues with PETG blobbing on the print head and ripping itself from the print bed? (I used prusa mini btw)
PETG needs high initial layer temps. Im doing fine with 240C and bed temp 85C. You may also wanna check nozzle to bed clearance. If u have trouble with bed adhesion try decreasing the clearance in .05mm increments over the z-offset in ur slicer...
I print PLA preferly at 225 because of your PLA video an I use very often the presets in prusa Slicer also for expensive material like PP (verbatim) PC (Polymaker) and Prusa PLA. But I am often to lazy to print a temperature tower so I decided a lower temperature when it should look nice somewhere in between 190-203 and when it need to be stable 225 for PLA. And after some prints I figured out, that at my machine many PETGs print well at 238, except the prusa PETG. Also for 2 Customers I made some temp Tests because of their need for a smooth surface. What I'm interested in is, how good custom nozzle materials are, my nickel plated copper nozzles are quite good.
I like to print hot so the plastic flows easier and sticks better to layers and the bed I print my petg at 240-245 and I get much better results than with the general petg preset in prusaslicer
I'd be interested in knowing what the results would be like with a volcano hotend. I think strength also highly depends on the flow rate and the layer time (small parts should cool down less before the next layer is laid down). So basically, nozzle temperature is not always equal to the filament temperature when layered.
Good point though the tensile samples are printed quite slow due to the small cross-section which should allow enough time for the material to heat up.
PLA melting temp is around 180C and PETG is around 230C, so it seems like moisture has more to do with the layer adhesion than temperature does. Makes me wonder what would happen if printing were done inside a vacuum chamber, or a chamber filled with CO2. Enclosed printers tend to produce better results since the environment is more controlled. No air blowing across the parts causing uneven cooling etc. If I had the money, I would get a mini 3d printer and make an acrylic case that can be pulled to vacuum with a gas inlet to print in an oxygen and moisture free environment. It would be quiet operation too since the entire machine would be inside a vacuum.
It could be that the layer adhesion itself becomes brittle at the higher temperatures. As you suggested, more bubbles may be present due to moisture. It makes sense that at a higher temp, more moisture would be boiled out of the material. This would create more failure points. I also wonder if the previous layer becomes brittle due to cooling from a higher temp. At higher temps, the difference in temp between the material and the surroundings makes it harder for that material to retain its thermal energy. So essentially, the material cools too fast. This would probably be present in you XY orientation models however. I was wondering, why did you use Kelvin for your delta T values? I know Kelvin and Celsius are on the same scale and a delta T of 20K is equal to a delta T of 20C but I was just curious if that was some convention. It makes perfect sense if you’re doing thermodynamic calculations with it because those are mostly based on absolute temperatures. Just wondering because I don’t have any experience with engineering outside of college. Thank you.
3D print pieces to make an anechoic chamber section of wall... I really want to know if it works with hard plastic and how much of the wall you need to cover to make a difference to a room.
Show your friends how to properly chose a nozzle temperature and share this video on Facebook, Reddit, Twitter, and other platforms! ❤
The CR 30 is now a Vidiography tool to you...
Nice to have a Printer for that.
Hey! Thanks for the guide. Helped a lot. Keep up the great work :)
I live in a crazy humid environment, air drying filaments does not work, it just gets steamed.
I was wondering if sublimating with dry ice would work. I thought it might be an interesting experiment at least.
Would love to see what filament material has the greatest layer bond adhesion comparison?
I couldn't see the extent of the bubbling you mentioned, and one would presume the horizontal print would reveal that decrease in integrity, so I have two other guesses for you (which you've probably considered already): 1. Assuming the stringing is the result of additional fluidity, there may be a reduced pressure of the printed paths as they are "driven" into the underlying layer. I believe you touched on a related topic in an earlier video -- wider-extrusions must deform more, under pressure, to form a thinner layer height. 2(a). The greater size-change during cooling and solidification might introducing additional tensions in the final (unbroken) part. 2(b). This may even be during cooling of an individual path, where not only is there inter-layer tension, but increased tension even within the printed extrusions individually, as their skin cools before the inside. Both the fan and the greater temperature difference between the new trace and prior layer could contribute to this -- the potential internal stresses maybe resulting in increased rigidity and brittleness.
The Cooling fan has a significant impact on layer bonding and even layer tension as it can/will cool the print, another factor is room temperature and of course enclosure. Meaning you can print at lower temperature with equivalent results of higher nozzle temps with reduced stringing or curling but keeping the better physical properties.
+1. This is a factor that needs to be accounted for. It's well known that cooling impacts PETG layer bonding.
Good point about the cooling fan having a large effect.
Higher nozzle temperatures reduce the initial viscosity of the filament, which fills in porosity more quickly. With the fan on, higher nozzle temperatures only slightly increase the time it takes the molten filament to freeze and solidify. A reduction in fan speed would have a much greater effect on the molten period. A long molten period will start to anneal the filament, by reducing internal stresses, and randomizing the orientation of the extruded polymer chains. Unfortunately, low viscosity and a long molten period also result in loss of detail, due to slumpage from gravity. With fan on, I think most of the loss of detail is due to the lower initial viscosity, because the molten period is not significantly longer. I'm not sure if air pressure from the fan is strong enough to cause significant loss of detail.
I just want to add. From my 2 year experience printing with PETG. Is that the more I use fan for the print the more stringing I get . And also print gets worse strength every time I try to increase fan speed. Also if the Filament is getting humidity it starts stringing like insane. I print PETG at the best middle temperature for my filament, and this is the best option for me, after I increase temp I get more stringing on the print. But If I decrease my direct extruder can stop skipping steps on full infill prints, or the print model starts being weak. I think temperature should be adjusted depending on the print model, sometimes you can increase it or decrease it.
Very true. Plus print speed and the size of your part. Reprint temperature is key but very hard to measure on a desktop printer.
Indeed. Simply upping the temperature w/o changing other settings will lead to prints sagging and melting, at least on smaller prints - and his, very small in crossection, tensile tests samples are particularly vulnerable to this effect. W/o knowing what OTHER settings did he changed to make those samples print at all given difference in temperature, the results are frankly useless, and he must know that. What matters is 'delta T' between the previous layer and layer being deposited, and you can alter this drastically with cooling settings. This video is way below the quality of his other videos when it comes to 'design of experiment', and seems like more a filler and 'discussion bait' to be fair.
At 4:25 this is the best use of a CR30 I've seen so far! :D
Who needs a camera slider if he has a belt printer? 😅
Didn't even notice this 😂
Awesome information as always! The strain graph is a great addition to help visualize the loading keep up the great work!
whats funny Is i was looking for this video on your channel before it even existed. i was like no he HAD to have done this already. That was like 3 days ago. If you ever read this i just wanted to say thanks, you are really bringing a lot to the community by making these videos, and you are making history here
Great video. I have suspected that the higher temps I have been using with PETG are not helping, now I know where to adjust them to!
Thank you for the video Stefan. After a quick search on google, it seems the strength of PLA decreases at high temps due to thermal degradation - other compounds are introduced which may be impeding the interlayer adhesion.
The bigger factor that is unaccounted for by virtually every single 3D Printer... Is the Air temperature of the printer area. The printer should be enclosed, heated, maintained, and with steps taken to reduce moisture content.
that "sub scr ibed?" sign was like the "dont dead, open inside" meme man
Interesting results, makes me happy to see I'm not totally wrong with my 205°C default. I've personally always erred on the side of colder temps but had a lot of issues with underextrusion (though that was also in part due to the extruder idler slipping, which has since also been resolved) so I've inched my way up towards where I'm at right now. It wasn't quite as rigorous a process as you've shown here, more a lot of trial and error. Hope this vid helps smoothen the learning curve for beginners a bit 👍
Wow came back to this video because I have a diamondback nozzle that made me lower the temp for printing. I picked 215 just based on my temp tower for PLA and an glad that is backed up with your test!
Thank you Stefan. I love seeing your test results.
Good information, thank you. Good resullt for you on the Polymaker Tournament. A proof of your "know how" and the quality of the information you are giving to your audiece.
4:22 great use of the CR6 lol
Hey Stephan, always great to see you uploaded a new video, please do more tests with ABS.
That's the best.尼龙也经过测试。
Your info is so useful to the community! Thank you for your dedication!
Probably worth a note about the accuracy of the temperature sensors found in many printers not being anywhere near calibrated. Required set points will vary by quite a bit (+-15K) depending on the printer
It would be interesting to see what effect a heated chamber has on petg and other filaments. I would guess that the resulting lower temperature delta between layer would help with bonding.
Interesting (and reassuring) that your "sweet spots" for DF PLA and PETG match the temps I've been using for them xD
Cheap Chinese "PLA+" from Sunlu worked best at 200°C for me, but the "proper" PLAs from DasFilament (and Procatec) worked best at 215 on my printer.
I think there's more to this.... travel speed for one thing. And as others have mentioned when I scroll through comments- part cooling and chamber temp.
Can you make a video combining all of your findings such as layer height, printing direction, temperatures etc etc? And, perhaps, it can be compared to the standard settings most novice makers use. I'd be interesting too see it compared to the prints that combined the worst characteristics
Great video, really useful, thank you very much
You are nailing it with these videos!
Yeah, I have no idea why higher temp prints act that way. Weird, huh?
I usually look at the specs and pick a temp about in the middle. Seems to work well most of the time.
Thanks so much for asking another important question
I run 210 for "PLA+" and 240 for PETG. Seems I'm close enough to the sweet spot vs stringing for my needs. 👍
I like your scientific approach 👍🏼
When are you going to talk about the pellet extruder?
Hi, Stephan! I think that standing samples (PETG) printed on different temperatures gave you different results because you're over-hotting them. Every layer square is too small. So you're basically "drawing" over the same place with the hot nozzle. It constantly degenerates (on high temperatures) lower layers while printing next layers. But it makes them stronger if the temperature isn't high (a kind of a natural hardening happens).
And of course the laying samples become colder pretty quick in different places because of the higher square of printing.
So these results can be pretty actual for some prints.
My teacher usually printed at around 190 so that's what I've always printed my PLA at but now i know that higher temperatures equals higher strength
Just a guess here, but could the delta between the extruding material and the previous layer weaken adhesion? The hotter tests, I would think, would show a greater delta between the previous layer and the extruding material. I know when mixing, materials that are closer in temperature tend to blend faster than ones that are highly variant.
I like this theory. There could be a difference in the internal stresses as the new layer cools and shrinks.
Fascinating and super helpful! Thank you very, very much!
Wie immer perfekt. Dir und Deiner Familie schöne Weihnachten!
Answering your question from just before 2:00 I don't have a 3D printer, I just watch as an enthusiastic fan of 3D printing and someone who's always eyeballing a printer to get it and start
my guess as to why with hotter temps there is less bonding is probably more to do with less pressure between te molten and already set plastic as the materiaal gets less viscous as temperature increases
would be interesting to test if this is the case by printing very hot but decreasing layerheight and vise versa
Good point.
Time between layers is also a factor. If the previous layer is still hot the temperature the resin is fusing at will be higher than if the previous layer has already cooled. Could be a reason the laying samples didn't show the same difference as the standing samples.
Great video as always
I love that book, It is twice as scary as the film!
Your studio looking great!
Thanks for sharing :-)
About the highest temperature being weaker, my thought is that the fan of the extruder is cooling the already printed part, so there is a temperature delta between the "cold" previous layer and the "very hot" temperature of the filament extruded. Maybe even some retractation (mechanical size reducing because of the cooling). Maybe you can try without the fan :O
At my workplace, getting dimensional accuracy was more paramount than overall appearance (which we can address through post-processing) or part strength. I spent weeks tuning the machine we had to minimize ABS warping, where I settled on temps between 235-238C and, ironically, with minimal part cooling fan (enough to "freeze" the plastic in place, but not enough to exacerbate warping). I'm able to get away with a little fan because the print volume is fully enclosed, and I tend to preheat the machine for at least 30 minutes before printing to make sure the air temperature is stable.
There is still warping, of course, such as when the printer transitions from a low-density layer (sparse infill) to a higher-density layer (surface infill); this is where mass becomes a factor in warping. There are ways I can mitigate that, such as with orientation.
Your strength test videos have been great! Very informative and rigourous enough for hobbyists to make informed material choices. Takes me back to my engineering classes!
One thing I haven't seen analysed is which filament material has the worst "creep". PLA is supposedly pretty bad, and I've heard PETG is better but there's no actual data.
I request a creep test! First suggestion: cut a 100mm strand of filament, hang a weight off for some fixed amount of time and check the elongation. I'm particularly interested in PLA, PETG, polycarbonate, and carbon fiber reinforced variants.
Have you tried to test print speed vs strength? Does printing super slow give you higher strength? How about "ironing" each layer.
I would like to see a speed test too. Everyone wants to print faster, but it seems fairly clear that slower is better.
I'm pretty new to this and I'm currently preparing for my first print with PETG. Part of it is printing my first temperature tower and therefore write my own Cura post processing plug-in so that I can set my temperature points comfortably.
You've said that you set the nozzle temperature using the M104 command. So far I've gathered that that's not enough, because if you do it this way your nozzle cools/heats up while printing, leaving you with a fuzzy temperature transition. You need to send M109 as well, which tells the printer to pause until the given temperature has been reached.
In other hand if you pause it, the last layer will be a lot colder when you get to print again, which would definitely lower layer adhesion wouldn't it?
@@Kalvinjj Yes, but that's not the point of temperature towers, right? It doesn't apply to this test as Stefan has printed each batch separately to test layer adhesion at a certain temperature. But if you want to observe print quality not waiting skews the result slightly
@@saschaschneider6355 hmm I see your point now, makes sense on this case. I wonder how much it would affect tho, since the nozzle temperature change is in like 10ºc (or is it 5? 15?) intervals, it might be able to change temperature fast enough, tho I guess it depends on where it would start the next layer.
@@Kalvinjj sure, but if it heats up/cools down fast enough then M109 isn't going to slow down the process by a lot either. Hence I think it should be used in addition to M104 either way.
edit: as M109 also takes temperature as an argument you could also set it slightly below the target temperature, but that's probably something I'll have to play around with - and probably will be included in my plug-in as an option
I always use 5c above the minimum manufacturer stated range. I've never printed a temperature tower nor do I tune temperature unless it's for first later adhesion.
In this test each dog-bone temp group was printed on it's own so it was printing hot filament on hot dog-bone, but some of the old test videos show many groups being printed at once. I always wondered if the hot filament on cold dog-bone, because it takes so long to get back to the first object when printing many at once, affects the test.
I haven't printed PLA in a long time but most of my ABS, ASA, PETG and PETG CF prints have been around 240-245, less and the layers (ABS + ASA) don't bond as well and a higher temp curls the edges a little effecting appearance. The size of the part will also effect the temperature I choose, a small part that will not have a chance to cool fast enough will get a lower temp while I can run a higher temp on a larger part without effecting it's appearance.
It depends entirely on the application (like with part cooling, nozzle diameter etc.). But why not also compare with ABS and ASA when talking functional prints? They practically cost the same (unlike PA/Nylon or carbon-reinforced filaments), but you have to be able to print it first, of course
You kind of answered it yourself. Unless you're already looking for something very specific that you really only get from those materials (rigidity. It's rigidity) you aren't going to be using them. PLA and PETG are general use materials. Neither ABS or ASA really is, what with the significantly increased headache that comes with them with the benefit of less strength than other things that cost the same amount. They really don't belong to the same category as PLA/PETG any more than Nylon does. (And while we're talking here, "nylon or carbon-reinforced filaments" are pretty far away from each other, given that it's already very well established that carbon-'reinforced' filaments are actually much weaker than their cheaper equivalents and the only benefit out there is that you can print them at a higher rate)
This guys voice haunts my dreams but goddamnit he’s got good content...
Don't know if I should find that creepy or charming...
Nice ! you should do a book with all this info, maybe aiming to make tooling and fixtures with 3d printing.
The petg printed very hot expands more than the recommended, which creates residual stresses at the adhesion layers.
I would assume expansion coefficient plays a large factor in the higher temperature tests, causing excess layer tension compared to lower temperatures. I'd be curious to see if annealing makes any difference with the higher temperatures.
I personally use a small retraction test that prints two towers, one at a time at different temps. that way I can judge both stringing and layer adhesion without using as much filament as a temp tower though that doesn't test overhangs but I have found that if minimize stringing that overhangs usually end up looking fine.
Hi Stefan. I wonder if you can get your hands on a Xube 3D printer. It's based on a process called "xolography" in which object are printed entirely at once rather than layer by layer.
They say it's meant for researchers. Maybe they can let you borrow and test one of their machines when it's released.
Since the viscosity is required to be adjusted for increased speed , it doesent really matter as you only got a fixed temp about 40mms plus to about 50, after that if you dont increase temp you will get feed issues. The temp will however not be increased per cm so an increase will only work up to a certain level ams it is linear to speed more or less
Some form of microscopy in the cross section of layer adhesion possibly provide some insight. probably some research in thermal/bonding properties of thermoplastic polymers
I always go 5C above the recommended temperature, and if that's good, I'm done. If I get weird results, then I print a temperature tower. But that's only for printing with that nozzle at that speed with that layer height.
A single filament can print very differently between slow printing with a fine nozzle and thin layers, compared to fast printing with a wide nozzle and thick layers. I can't count the times I've gone from a sub-scale proof print (nozzle, layers and speed all scaled down) on my Sidewinder X1 v4 to a full-scale final print without taking this into account.
This is when doing both temperature AND speed towers becomes important, especially when doing large prints after doing crude drafts and/or an accurate sub-scale print.
BTW, doing perfect sub-scale proof prints lets you do photography of the proofs that you know will very closely match the final full-scale print. Yes, the print time is about the same (intentionally), so these aren't draft prints. But they do let you make changes that you know will be accurately reflected in the final part, and do so for a MASSIVE savings in filament compared to noticing a mistake on the first full-size print.
But it is a lot of nozzle switching...
I normally print PLA a little warmer, 210 C, because it gives me the best results (I haven't tried higher) for the way I store my PLA (in sealed bags, dried after being out of the bag for a few days.
I use a temp tower and pick the highest temp that gives me results with acceptable amounts of stringing. I always aim for hotter temps generally
I once printed 12 little Lego men which worked fine, but trying to print just two with same settings ended up a globby mess. The reason was that with 12, each layer had a bit of time to cool but with 2, the layer being printed on was still mushy and so got pushed around by the extruder. I have found this effect for complex prints with some parts large and others small islands...one temp setting isn't always optimal for the whole print! I think a layer area : part volume ratio ratio may be able to be used to tweak fan speed in slicer software.
On one of my printers, 210 is the best temp. On another, with the same PLA it's 190..
It think the difference is in the quality/accuracy of the thermistor (maybe even a FW setting to set for diff style thermistor)..
Another factor that may come into play here is layer time and how long the previous layer has to cool before next layer is applied.
how about printing cold with an extruder like the Volcano, which has a longer hot section? While the temperature is lower, the material has more time to heat up inside the nozzle.
Good point and something I wanted to take a look at for ages. I have some high-grade Nylon here that is basically unusable when not printed with a volcano nozzle because it needs time to heat up.
Great work!
Pretty nasty teaser there in the intro. Belt printer working on a Starship?
Edit: jup, starship in the audible section.
Had to find a way that you watch the whole promo 😎 More shots of the CR-30 at 9:26.
@@CNCKitchen Definitely worked :D . Great video!
i use PETG at 245 nozzle and bed at 70 degrees and the prints i get are pretty strong and i must add that i use 1.5mm for wall and bottom and 20% infill and no part cooling fan,grüss aus holland :)
I wish you would talk and experiment more about the fumes! Maybe collaborate with experts that have fancy Test equipment? What do Pneumologists (lung-doctors) say? What do we know from the industry that works around this stuff every day?
Up to now, I've always selected nozzle temp to give the best aesthetics. But now, when I'm doing functional prints that don't need to look great, I'll consider strength over beauty.
Nice Video. I like your content. I think you should test this again with various cooling fan settings as well. Maybe this would be the solution for losing strength at higher temps.
Anecdotally I've always found hotter is better for amorphous polymers (eg abs/pc). It also makes a huge difference at a specific temp if you have a cooling fan enabled or not. For ABS, it's always stronger without a fan - hence why dual extrusion machines have a huge advantage because you can print very hot without a fan without sacrificing accuracy from support gaps. This is why stratasys machines don't have cooling fans.
@CNC_Kitchen : Stefan, have you ever tested the best layer direction for side-loaded 'column' type printed parts?
I'm in a situation where I can print a square tower-like part in different alignments. It will be loaded from the side. And I'm not sure which route I should take to get the best strength. And I lack your setup to test it, of course.
I love your videos especially the thorough tests about strength! :)
One possible reason the strength drops of at higer temps, might be because the already printed layer cools off too much, so the diff between the 2 layers are bigger?
Would be interesting to see if there is any difference printing 5 and 1 item at a time.... Or maybe add g-code between layers to have the nozzle go park in the corner for X number of seconds.
I think that it is not really possible to conclude on the stress, because it depends on the elongation at break which is known to be very variable. I think it would have taken at least 10 tests per temperature to get an idea, and probably trying to identify a Weibull distribution would be relevant.
True, though this amount of tests is just not practical for me. I try to investigate more thoroughly than 95% of the others but just can't go into the details like in a scientific paper. Otherwise, I would have deserved a Ph.D. a long time ago 😉
@@CNCKitchen I totally understand!
It's already great what you're doing, it's clear that you're the one who spends the most time looking at the impact of printing parameters on mechanical performance, it's even often impossible to find the experiments you're doing in the scientific community, so it's a great help for everyone, don't take this as a challenge. It's just that when (like me) you don't do the tests, your conclusions seem logical, but unfortunately, it's not impossible that it's the result of chance. That said, you're right, the work to be conducted is clearly a material engineer's job (not necessarily that of a PhD), and I fully understand that the time needed to conduct it is not compatible with your priorities!
Maybe a little nuance in the conclusions ...
Once again, thank you for this work which is useful for all of us researchers !
P.S: if I could give your Patreon to the lab, it would have been done a long time ago !
Because your Stefan... ...considering features in the firmware and slicer used - 'coming up with new features: How about a "spot welder" like feature? Where the nozzle pauses in the print at corners every few layers and semi-reheats those points through and past previously laid down layerings? At those points, the nozzle would heat up 10C or whatever for a duration, and then return to its assigned temp, and continue printing, until it got to the next "spot welding point" where it would then stop, and increase its temp' again - and repeat. Royally add print time, and start challenging image quality, but considering coming up with new features and being creative......... ;) Peace love 'n' happiness from the Pacific NW of the US. :D
Nice work and presentation. Will you do the same for ASA and Nylon? Thanks!
I was printing PLA from DasFilament at higher speeds with 220C and 0.3 layer height. I had rare extrusion issues. I found, raising my temperature to 235C for the first layer, and 230C for all other layers solved the issues completely, and the surface looks smoother as well. I am using MK3 and MK3S printers from Prusa. The given 215C the guys from DasFilament claim, did not work for me. Me, not printing figurines and stuff. I am using my printers for creating electronics enclosures, and RC stuff.
When i want something simple with high strength, i usually use for PLA 270° 1.6mm extrusion width at 0.8mm nozzle diameter no fan speed down to 50% and 0.3 layer height in a chamber at arround 50°C
At this settings layerlines disappear,the part almost looks like vapour smoothed, but with a textured surface. It looks really nice, and i suppose its pretty much perfect layer adhesion, but i never tested strength.
I have until now used my Smartfill, AzureFilm and Fiberlogy at around +5c above lowest temp range on the filament. So if rated at 190-220c, I go 195c and usually with just the right results. However I am still trying to dial in PETG and Fibersilk as these tend to either string or foam rather quickly if temp is too high. My Fiberlogy Impact PLA is spot on at 225c and have a lot of cool prints using this on my GooFoo Mido printer.
I have noticed that PETG become brittle when cooled below 40 dg to fast, I have had luck putting a fence of isolation around the build pate, so that the temperature is kept high by the build plate heating, then te strength become more like when printed flat.
I love your work, videos AND style! Just saying 😊🥰
Actually i start without a temp-tower and use the arithmetic middle of the temperature the filament producer writes on it :P. Usually for PLA it is around 205°C. When i see strining i drop by 5°C and in most cases land on 200°C ... that's it.
Very good video as usual. Regarding the strength of the PLA, I found that some filaments are stronger than the others from the same brand. For example, the copper color is the weakest I have (by far), all others (wood, black,..) offer higher strength. (from eryone, cyclone filament,..)
Did anyone else notice that the supports of the rocket shifted on the belt during the print?
I was instantly thinking the elevated temperatures would increase bubbles and decrease surface area between layers, therefore weakening the bond between layers.
Your t-shirt is brilliant 😂😂😂 👍
Thanks 😆
I wonder if there's any atmospheric contamination happening as well. 3d printing in a purge chamber next? :D
I suspect degradation and bubbling/blistering could both be playing roles. I would think that since "over-cooked" thermoplastic loses molecular weight and damages polymer chains/branches, that would to less entanglement between layer surfaces. Add to that more voids from bubbles/blisters either naturally from internal moisture or absorbing atmospheric moisture? via gassing-off? Additives vaporizing? not sure, BUT the bubbling could be tested with a microscope I suppose, and then the cause narrowed down with... more elaborate means (printing in inert gas to vacuum environment and very low humidity, etc?).
On a somewhat related note, you should take a look at PLA layer adhesion with much higher bed temps and ambient heat. Anecdotally at least, it seems to have more effect than raising the nozzle temp. I have not tested directly and could be off base, but stumbled upon it when I raised the bed temp to try annealing a cooling shroud while it was printing. could be worth a peek to confirm or deny
I see you're working with your print mill now. Very excited to see some of the things you make with it.
This contridicts your test results with PETG vs tempature in 2017
Thanks for sharing Stefan! Could you do a similar test with ASA? There doesn't seem to be allot of data with this filament.
Thanks!
Hey Stephan, we all know that higher temperatures increase layer adhesion, we also know that cooling decreases it. But sometimes you want stronger prints but the higher temperature might be too much (benchy smoke stack thing) there is no video regarding the combination of the too. So would it be better to print at 200 degrees but no cooling or 230 with 10% or 20% cooling. I watched both of your videos and there can be some interpolation made between them but we cannot be sure that the scaling is linear and this would really help me in lots of situations. Thank you
Maybe off-gassing prevented layer adhesion at the very high temperatures? Or the leidenfrost effect with the moisture your mentioned?
Very interesting would be to know at which speeds you print - DANKE
Wish you tested abs too! And enclosure/room temperature might be a fun thing to try
Did I miss something or did you not ever mention if you were using a part cooling fan or not? The temperature difference between the previous layer and the current layer being laid down is very important for comparing layer adhesion.
Part cooling for PLA and PETG is always on for me. But that's a valid point and I investigated that in a previous video.
Interesting stuff! I print at 180c with creality pla lol. I don't have any significant problems and get low stringing, I only print models atm so I can't talk about strength.
I was having a ton of issues with the polylite PETG. I finally get it to stick at the max temp it says on the datasheet. Was anyone else having issues with PETG blobbing on the print head and ripping itself from the print bed? (I used prusa mini btw)
PETG needs high initial layer temps. Im doing fine with 240C and bed temp 85C. You may also wanna check nozzle to bed clearance. If u have trouble with bed adhesion try decreasing the clearance in .05mm increments over the z-offset in ur slicer...
I print PLA preferly at 225 because of your PLA video an I use very often the presets in prusa Slicer also for expensive material like PP (verbatim) PC (Polymaker) and Prusa PLA.
But I am often to lazy to print a temperature tower so I decided a lower temperature when it should look nice somewhere in between 190-203 and when it need to be stable 225 for PLA.
And after some prints I figured out, that at my machine many PETGs print well at 238, except the prusa PETG. Also for 2 Customers I made some temp Tests because of their need for a smooth surface.
What I'm interested in is, how good custom nozzle materials are, my nickel plated copper nozzles are quite good.
I like to print hot so the plastic flows easier and sticks better to layers and the bed I print my petg at 240-245 and I get much better results than with the general petg preset in prusaslicer
Perhaps the parts printed too hot are delaminating like a Damascus steel bar when it is forged too hot?
I'd be interested in knowing what the results would be like with a volcano hotend. I think strength also highly depends on the flow rate and the layer time (small parts should cool down less before the next layer is laid down).
So basically, nozzle temperature is not always equal to the filament temperature when layered.
Good point though the tensile samples are printed quite slow due to the small cross-section which should allow enough time for the material to heat up.
PLA melting temp is around 180C and PETG is around 230C, so it seems like moisture has more to do with the layer adhesion than temperature does. Makes me wonder what would happen if printing were done inside a vacuum chamber, or a chamber filled with CO2. Enclosed printers tend to produce better results since the environment is more controlled. No air blowing across the parts causing uneven cooling etc. If I had the money, I would get a mini 3d printer and make an acrylic case that can be pulled to vacuum with a gas inlet to print in an oxygen and moisture free environment. It would be quiet operation too since the entire machine would be inside a vacuum.
It could be that the layer adhesion itself becomes brittle at the higher temperatures. As you suggested, more bubbles may be present due to moisture. It makes sense that at a higher temp, more moisture would be boiled out of the material. This would create more failure points. I also wonder if the previous layer becomes brittle due to cooling from a higher temp. At higher temps, the difference in temp between the material and the surroundings makes it harder for that material to retain its thermal energy. So essentially, the material cools too fast. This would probably be present in you XY orientation models however.
I was wondering, why did you use Kelvin for your delta T values? I know Kelvin and Celsius are on the same scale and a delta T of 20K is equal to a delta T of 20C but I was just curious if that was some convention. It makes perfect sense if you’re doing thermodynamic calculations with it because those are mostly based on absolute temperatures. Just wondering because I don’t have any experience with engineering outside of college. Thank you.
You should do a video about PLA deforming over time.
3D print pieces to make an anechoic chamber section of wall... I really want to know if it works with hard plastic and how much of the wall you need to cover to make a difference to a room.
Perhaps the surface irregularities at high temp printing causes localized stress concentrations cause the coupon to fail.