Did you ever notice a difference in your 3D prints, depending on the material color? Also: check out our *CNC Kitchen products* at cnckitchen.store/ or at resellers www.cnckitchen.com/reseller and on AMAZON (EU) geni.us/s8rYtQ
Hi CNC Kitchen, may I make a suggestion? Your thoughts on residual stresses are a lot more relevant than you think it is. Up for having a conversation? You're starting to delve into principles around weld buildups. My day job is on the metal side of buildups, but it has a significant amount of overlap with what you are exploring.
I print with ABS and i had noticed some diferences, specially with red and Brown, but i recently changed my supplier to a better and did not test different colors yet. I discarded the color theory thinking it was my previous supliers that was very bad. Thanks for letting us know that this can be a problem even with good suppliers.
Yes, BUT I think the amount of difference between colors is almost similar to the difference between brands. My takeaway here is to maybe stick with blue or translucent colors for strength from now on. I totally agree on staying away from silk or matte filaments for functional parts, though.
I work in a masterbatch laboratory and can confirm, pigments can make a colour behave very differently. Usually it's way less pronounced in a finished product, with dispersion values ranging usually between 1 and 4%, but as seen in the video - it still makes a difference. A thing I noticed for example, which wasn't mentioned here but I can tell from experience, is that silvery/glitter colorants make the extruded material way more prone to absorb moisture, it's worth keeping that in mind
This works for other products as well. My wife and I have a commercial embroidery business and one of the first things we learned was that black embroidery thread and white embroidery thread are more prone to breaking than other colors. Black thread is usually another color that didn't come out well, so it was bleached and redyed black. This is also true of white thread. It was usually another color then bleached heavily to make it white. The bleaching and redying process weakens the thread.
Good to know. Yeah i'm sure some black plastics from less trustworthy manufacturers have black be all-regrind with extra masterbatch; which is a problem for ABS in particular since it degrades heavily in the process, but would be noticeable for other polymers as well. I think some other colours can be full of regrind too occasionally, grey for example.
That makes a lot of sense, so dark colours are dyed black because cost wise it is easier and more cost effective to dye it black than it is to try (and fail) to bleach out all of the existing colour.
Very interesting results, with ABS on Vorons we`ve always recommend avoiding the use of white filament as its shown in the past to have poor layer adhesion and increased chance of part failure, but with PLA it looks like thats not the case.
Ah, so that's the reason for the red/black theme /s. This is absolutely fascinating and I never would have guessed that the difference between pigments was this large. I wonder if it's the pigment itself or the different sourcing of the masterbatches used by dasFilament?
I suspected this after getting some white hatchbox abs and it was crap to print and for strength. Normally I use esun so I didn't know if it was just the brand but, now I know it's white in general, thanks!
@@ska042 yeah, realised this after buying four 4kg spools of the white PETG. Destroyed our RatRig on an overnight print, had to order new hotend, new sensor and print new head assembly using other printers. Now have a load of this filament condemned.
I'm a paint department manager in a small town hardware store and I found it interesting when I would pick up the quarts of colorant to refill the dispenser, and they would all weigh significantly different. For instance, a quart of white weighs 3.6lbs and a quart of black weighs 2.4lbs. That's when I learned about pigment properties lol
I work in printing and used to order ink. The we were charged by the pound from our supplier. A can of black would be 5.0 pounds and reflex blue was 5.7 pounds for the same size can.
Every single silk silver I have bought has been the biggest pain to print. I’ve been finding color makes a huge difference and no body talks about it. I really appreciate All of your videos for the conversations they start online but especially thank you for this one, I feel validated lol
Typical i hate the "DrY yOuR fiLiMeNt" response but, I've noticed that the silk PLAs are actually hygroscopic and definitely benefit from drying. But also they need to be printed at much higher temps than typical PLAs. I had a gold silk pla that liked to be printed at 250c and only printed properly if it was dry.
The most interesting thing to me was that natural wasn't always the winner, sometimes the pigment was making it stronger! I wonder how the difference would be in premium PLA blends that already have additives.
More interesting to me in the test results was the wild variation in the error bars per color. That tells me that the color not only influences the macro level performance of the plastic, but also introduces microvariation that can have a notable impact.
Didn't even notice that at first, maybe the weaker prints had slightly more pigment than the stronger ones? Many of these pigments are so concentrated that even a large difference in concentration wouldn't be very noticeable. Stephan has a filament extruder, this would make for an excellent follow up experiment. 50% recommended masterbatch ratio, 100% ratio, 150% ratio, 200% ratio, etc.
That's the problem is that each brand has different mixes for different pigments and when we are testing all similar materials with simple tests we are going to start seeing the variation in the results itself where each individual tested filament roll might behave differently, what if the weight is different from pigment as well as moisture. So in turn we eventually do need to start testing further into the results to get a definitive conclusion
That or another variable, it's not a super scientific test. I would say almost none of the tests are really significant, with the exception of the hook tests.
or... or this guy didn't isolate a true independent variable and we can draw basically no conclusion from these results other than 3d printing parts have highly unpredictable properties that are still poorly understood.
My company works with PLA+. If a client needs a strong piece (within the possibilities of what PLA can achieve), white is everyone's colour of choice and there are no complains. Other than white, blue. Orange, grey, red and green were other colours I've worked with, but white is by far THE one requested.
Actual cool fact from a chemist working a lot with dyes/pigments: White pigment/master batch includes titanium dioxide which gives the PLA/material better UV resistance
Please create a series of videos comparing the properties of filaments types across manufacturers. - The manufacturers probably won’t release their raw materials manufacture and mix recipes. As well as added modifiers. I would really like to know which manufacturers ASA, ABS, PC, and nylons have the properties I’d expect when choosing a material. I’d like to know if a manufacturer adds modifiers that improve general user performance but negatively affect properties.
Interesting video! I noticed a big difference with the semi-transparent green PLA that I used for printing a guitar neck. This material felt harder and more brittle, yet tougher than the PLA I was used to. By the way, a good reason to print white is that you can draw on it with a pencil. Ideal for prototypes, like a 3D notepad!
It's interesting but its because the pigment pellets have tons of extra additives and plasticizers. which are going to make a big difference because most basic or ""natural"" polymers because like amber or like old plastic beads.
@@infernaldaedra Yeah, that's sort of what I was thinking? I can't imagine the actual pigment causing much influence, more the material its suspended in. 🤔
@@BRUXXUS that's one assumption, but it doesn't make sense to me. I know pigments are not just a drop of food coloring. GITD and white tear through nozzles because they are not pigments, they're colorful minerals, and much harder than any plastic. You can manipulate colors with particle size too. When there's poor adhesive, one cause is the amount of non sticky additives.
I've noticed difference in printing between colored vs. translucent PETG with translucent being the easiest to get to adhere to the bed. In woven fabrics the dye used to color them will also to an extent change how they behave with blue color making the fabric more firm and unyielding.
PETG has no problem sticking to smooth PEI sheets, it actually sticks too well the solution I found is plastic razor blades. After 2 kg of filament the PEI sheet is not damaged and still going.
They must be adding other additives specifically to compensate for the property changes brought on by the pigments, so they have to tune each color to be roughly the same "strength," hence the relatively small spread on the hook test, which is the most practical. So then maybe a material tuned specifically to be strong might see less of a spread between different colors, since the manufacturing goals are different. Would be interesting to see some tests on PLA+.
I'd be very interested in seeing if there was a difference between different nozzles and annealing warpage. Particularly, I wonder if the CHT nozzle (or the diy versions) would make much difference.
This is literally why I create a profile for different colors of filament using a temperature tower and extrusion calibration cube. Once they add a different color component to the plastic, every color takes on their own unique properties. I have found that by starting with the existing profile for the manufacturer and material, when one exists, and generic profile when one doesn't, that different colors from the same manufacturer typically require slightly different extrusion multipliers and temperature to achieve best results. Of course I'm only going for appearance and extrusion accuracy, not strength testing.
Hey Stefan, are the print layer inconsistencies you show at 6:34 fixable on a Voron 2.4? I have a 2.4 that I have tuned a lot and my layers look similar to yours in this video. I had watched your video "(HOW) I fixed the ugly layers on my 3d printer" and the layers you showed post fix seemed absolutely perfect. So I'm wondering if these inconsistencies on the V2.4 could be related to the Clockwork extruder? I'm currently running CW2 in a Stealthburner. Thanks!
Just yesterday I was looking at taulmans website and they say for their “black” filaments, they use a dark blue color because of carbon’s negative effects on performance
This is very interesting, but I have to admit when printing PLA, the colors matter for the design. If strength is needed other filaments should be used. This would be interesting to repeat for PETG and ABS or ASA.
Yes and no. PLA is not very strong but it is relatively stiff, which along with its low price and ease of printing means it does have a role in structural parts.
@@cooperised I use it to prototype, and then decide if I need to use another material. Since I received my Bambu X1 Carbon, that has allowed me to print almost any material, with great success.
Good topic- thanks! For what it’s worth I had to dry out several spools of PLA and kept track of progress via weighing. I found the silk spool had absorbed dramatically less moisture than other colors.
Hi Steafan! Are you shure that the warping diffrence druring your "Baking Experiment" is representative? Have you tried to bake multible specimen at the same time? I would guess it comes from unevenness in your oven.
i have to agree with Steam´n stuff, if i place mini pizzas in my oven for the recommended time the outer ones have brown crispy cheese while the middle ones have perfectly melted cheese. 😓
Thanks, Stefan, for another good video. I no longer use PLA because of its distortion over time when under load, such as shelf brackets, and always found white to be a poor performer with its layer adhesion. In those days, printer issues were more common, and filament dimensions, nozzle temperature control, and cooling were major issues. Today, I use PETG and Amphora based carbon fibre filaments. I find that black PETG is more brittle and that transparent is more ductile, while layer adhesion is similar and much better than PLA. I do print at a fairly high temperature, usually 240°C with a 0.4mm hardened nozzle on a Mk3S for all filaments. Sunlu PETG is the brand I use at the moment. I am now trying the Revo nozzle, so there are no substantative answers yet on that one, but the first models appear fine. I'm still waiting for the Obxidian nozzles for the CF filaments that you might consider investigating some time.
I'd heard about certain colors resulting in different plastics being stronger/weaker back when quadcopters were using ABS propellers. Supposedly the (matte) black ones were weaker. Great to see a good analysis of this!
I got a few pla colors when I went shoping at microcenter. One was an extremely nice Glass Purple and I printed a little drawer attachment to my ender 3 pro. However after a few days the filament was breaking in places and when I picked it up, it was like I left a cheap toy in the sun for a year because of the way it broke apart like an autumn leaf. Maybe it could be something wrong with my printer setting but I'm sure some of my other prints would have acted similarly. You should give it a try, see if other "transparent" fillaments have the same issue.
If you print 25-50 mm/s (more relax time), 125-150% extrusion rate (counters stretching), gyroid infill (single chamber), and alternating wall directions (stresses cancel out), does it reduce warping when annealing?
I went through three different manufacturers of silk PLA and two of them had major problems with delamination no matter what temperature I printed at, I have sacked silk PLA now but I am using a lot of matte PLA and carbon fibre strand PLA and PETG-CF and PA-CF Nylon, the first matte PLA I used was problematic to start with but I persevered with it and now can print perfect parts with all these new filaments.
In my experiences, silk and matte finish pla has always given me problems, mostly with layer adhesion and bed adhesion. For colors, I’ve found to like grey and black, I feel they warp less and offered measurable difference in just about all parameters. Blue has always seemed to be very easy to print and offered very smooth, clean prints. And red the worst, I feel like it absorbs moisture much more then others, anytime I need to print in red I dry it for a minimum of a day and I store it in desiccants too, I also feel red, in my experiences, are the most fragile, but I feel that’s because the extra moisture absorption I’ve experienced. I do a lot of prints for orders, like custom Starratt tool boxes for clients, friends and family, so I do have a ton of experience in red pla and pla + printing. And I feel the differences between colors in pla+ are less then plain pla. But these are my simple experiences and thanks for the video, glad to see some I’m not just going crazy with noticing differences between colors of the same brand and type filament.
no no i think you are on to something here. i use Eyrone filament mainly. Black seems almost seamless and perfect. Blue seems sturdy as heck, The red one i used for a MK50 infinite articulating pistol imediatly snapped of after pulling the slider a few times. Printed one in Eyrone Blue and its sturdy AF printed another one with the red one for a budy to pain and again snapped off at the seams. So i do think Red is a poopy PLA to print if you need something sturdy and infact the blue stuff i print rarely breaks even when dropped wherass my Silk and Red just are so brittle as well as my OD green matt filament which breaks easily at the seams. Printing at different temperatures doesnt help at all when testing. Oh yeah PLA+ Deffinatly worth it if you need stronger material without switching to a different plastic type. Ofcourse not a huge difference compared to normal but a bit more sturdy is often enough for Airsoft purposes
Another spectacular testing round! Well done! The variance in properties with color would reinforce why folks have developed different parameter sets for them. Absolutely fascinating. Great work as always.
Great video. I have experienced silks having poor layer adhesion but not mattes. Also, good to know that I don’t have to worry about my recent printer build with white filament. (Unless the base resin and base filament would yield different results) affected by the colored pigments used.
Great video! It would be interresting to see the see results of each color printed at their optimal temperature, and if the colours affects strength when printed as optimal as best
thanks for the video. The difference colorants , especially titanium Oside (white), can make are one of the reasons why I sell my Resins in "natural" color. To achieve peak performance the unaltered material cannot be beaten , and users can color them as they need. same Effects are there with Filament.
Beautiful work, one test i would have loved to see would be thermal conductivity of said colors. That would give a great insight on print temperatures and layer bonding
I've noticed differences in colours of the same materials before. Most recently, when printing a lot of Overture pla for accent colours, it's been very noticeable how much more difficult to remove the supports are of orange, whereas white almost falls off on its own. So very interesting to see the results in the video.
I wish you have checked the flow rate. From what Ive observerd, I needed the least temperature for blue, then black, green and white to achieve acceptable flowrate.
Indeed, I experienced a difference between different colors of PETG. I use PETG for parts which get hit by metal parts. My prototype was made with transparent green PETG, and had no problems. Then I printed some final parts with black PETG by M4P. The behaved really well. Last I tried a military green PETG, and it failed where the others already had more mileage that this. The green one broke not along the layer lines, but in a diagonal direction. I improved the part at that spot, now it holds, but I was really confused at first.
When i print parts that i have to paint afterwards i always print with uncolored material aka natural. Parts come out stronger, have less stringing, curling, warping... your get the point. White filament gives me the most trouble, while "galaxy" colors even clogged the nozzle a few times when changing to other filaments after printing.
speaking of different color properties, I want to share, from experience having selling custom order, using White Black and Grey p with many times of experience in printing with it ( maybe around 10 spools or more of each color) They behave really different. I'm using e-sun PLA +, specifically Cold White, Black and Gray. The Biggest difference is the Gray, it tends to warp!!! I even use PEI both smooth and textured. Even using brim doesn't help much. What really helps a lot is lower the temp. Weird but it works. usually on 60C but for gray it has to be 55C or lower, then it wont warp. I've make sure the BED is PID tuned, cleaned & I double check with infrared Thermometer for the temp. TLDR : E-sun PLA+ Gray ten to warping even with brim if you put bed 60C so i put it 55C its fine, white & black fine with 60C bed. (this is exclude the condition wetness of my spool, sometimes i print with wet & fresh Oven dry spool with or without spool enclosure)
I enjoyed this video and appreciate it very much (which is why I'm coming back to it). This time reading many of the comments. I feel like the series of tests you did is only the beginning (not that I expect you to do all the work), but I'm not set up to test things as easily so perhaps you/others can dive more into this and related factors? Thanks Stephan! :)
I definitely ran into this in my school's 3D printing lab. Hatchbox red PLA gave me much more consistent and high quality parts with better layer adhesion than any other color of hatchbox PLA.
Yep, I've worked with thermoplastics for decades, and colors always make a difference. Black PVC takes a hit on strength and dimensional stability because it's packed so dense with colorants. Machining colored cast acrylic, yellow cuts smooth, beautiful shavings of plastic come off the part. With red, small chips and a rougher finish, so I had to compensate for it.
From my experience Silk PLA is the worst in regards to strength, it just loves to split between layers and you can bump up the temparature however you like, it still would not be as strong as plain PLA
I always had layer adhesion issues with different brown colored filaments. I always wondered if my print settings are not right or if this color is really more difficult to print. Fantastic video!
Brilliant. I thought I was going mad. I printed the exact same model in esun blue and esun red and the red snapped in the same place on both prints, yet did not with the blue 👍 Colour does play a part depending in what you're using it for.
Yes, noticed a lot of difference in layer adhesion and strength while using Gembird filaments, over 70 KG of it. Lately switched to Plasty Mladec and C-Tech, first is very very good, second also good and i found out they have Orange HIPS boxes with Orange PETG inside, so i keep buying it as it is cheaper here! Also got some other HIPS colors from C-Tech that i will have to test. Gembird Red and White HIPS is very good, much easier to print than ABS or ASA by a long stretch.
Answering Question at 4:20 ; I recently did a project that used around 4kg of White Filament (Sunlu) and it destroyed my nozzles way faster than any other filament I've used so far. I did some research and I believe it's because of what they use for the master batch.
As usual, great video Stefan! I think a good baseline would be to do a tensile strength on the filament itself right off the roll. Measuring the diameter with a micrometer would give you great precision since a small variant on 1.75mm makes a big difference.
If you revisit this, you should control for batch variation within the same colour. This variation could be partly explained by each colour necessarily being from a different mix. Is the relationship consistent between different batches of the same colour?
Hej Stefan, I noticed at 8:53 that the purple hook of your tesile strength mashine is bending a little bit. Maybe you should replace this simple part with a metal one because this could give you little inconsistancy (especially over time). At least u are testing allways on the same mashine, so it should not be that bad. But maybe if you are looking for some upgrades in the future that could something to concider. Thanks for the good work, love your videos :)
Used to work for LEGO back in the day, supervising various design teams. With LEGO, it's mostly about "clutch power" (strength required to pull two elements apart). Around 2005, we ran into a lot of trouble with our new light blue: apparently the pigment had an impact on shrinking during the cooling phase, leading to our models literally falling apart as all elements made in that color were oversized by a few 1/1000 of a millimeter...
I’ve never had a unusually big issue with silks / rainbows but when I want a consistently great print without a headache I always reach for my black Hatchbox pla cause it just works.
For your extension metering, I know there is software that can use video footage to calculate the displacement field in a specific sample, also resulting in a strain field. The software I know of is called DICengine. I haven't used it before, but seeing as you already film your tests this might be an easy addition.
Out of the PLA varieties I use, Clear PLA feels like it has the best resistance to flexing to me. It also seems less vulnerable to becoming brittle due to moisture, which is invaluable, as I'm based out of Japan where summer gets unpleasantly hot and humid.
You should do a test on annealing/warping based on print speed. It would make sense that a model printed slower would have less internal stresses, so should warp less when annealing, in theory.
Interesting ! My intuition has always been white is the best overall, seem to be held out by these results. Black I have always had issues with. Silver is an eye opener though - always assumed it would be super weak, and have not used it, tended to print white and then spray paint, but looks like if I want to make a mount that I need to anneal in particular its worth a try - cool to know ! Contains real metal particles I guess ?
I really like your comment about white amplifying any inconsistencies in your prints -- I've been trying to dial in my settings lately, and been getting really frustrated that I can't seem get a perfect result like I usually can. Now I'm starting to wonder if it's because of my white filament, just making issues more obvious than normal. I'm gunna swap colors tomorrow (:
Very interesting test! Thank you for the great videos and the invested time. I have also observed that filaments behave differently across different colors. Basically, each color variant requires its own calibration. This means that the extruder steps, flow, and PA/LA must be adjusted and determined for each color variant, taking into account the different shrinkage of the materials and compensating for it. Have the appropriate calibrations been performed for each color variant in the test trials? I'm looking forward to many more videos.
One property that was not mentioned in the video is that certain colors of filament can be more abrasive to the Nozzle than others. One example of this is the use of titanium dioxide in white filament, wich is a very hard ceramic and i´ve heard it´s bad for the durability of especially brass nozzles. Interestingly it seems to perform quite well mechanicly, so it would be interesting to know if Das Filament uses titanium dioxide. Thanks for the informative video!
I am basically only using DasFilament filament in matte black and white. From time to time I have issues with bed adhesion, but that's more on my leveling and my extruder. I really like their filament
I haven't done actual measurements, but my prints with "natural" (pale translucent white) have seemed strongest and have had a little more flexibility, making the "minies" for TTRPG break less often if dropped.
What polymer blend? Or filament type? Pale translucent white doesn't sound the same as natural but neither is regulated or standardized but usually natural means free from additives. Translucent white might mean they actually tried to brighten up and clear the filament
That actually sounds like it's the weakest of the materials with low rigidity and higher flex and ductility, which is a good thing for when you accidentally drop them.
Just got an Ankermake and haven't even printed anything yet, although the affect of pigment was one of the first things I researched when shopping, and found that same study you referenced! 😆 So of course this video is of interest!
The testing and results you show are amazing. Thank you. In the future, can all graphs start at 0? Some of the results for strength and such were trunkated and make the difference seem more significant than they are.
Very interesting, I've wondered about bed adhesion as a function of colour, seems it also affects other properties. As a point for potential improvement of such tests, there are some missing graph numbers (ductility) and some graphs with supressed zeros which make small differences appear larger than they are.
Glitter filaments are usually significantly more hygroscopic, which makes sense, since many contain mica. I have to regularly dry my galaxy-black Prusament ASA, but never have to dry non-glittery colors. The same seems to hold true for PLA, and glitter PLA also becomes more brittle as it sits on the shelf vs. non-glitter. For PETG I find that the more heavily pigmented, opaque colors have better print quality with less stringing and blobbing. The biggest problem with PETG is that it's very sticky and goopy, and pigments seem to reduce this property. But it probably lowers strength.
I have natural PETG here behaving like a dream, and sunlu opaque blue and it's giving me trouble, like i sure can get decent looking prints out of it, but i do not get any sort of semi decent layer adhesion! Infuriating. I have to drive the temperature way up to get it to stick to itself, and then it blobs and strings more than the natural.
I actually disagree with using the exact same gcode for the tests. Different colours from the same supplier need very different extrusion multipliers to hit the same part geometry. This can be confirmed with a weight measurement, or by measuring a single wall vase mode cube with proper micrometers (read: not calipers). Some of the PLA we use can range from 0.92 all the way to 0.98 EM, across the same product range where the only difference is the colour. I suspect the results could have been affected by the printed parts all being slightly different in the end.
Sky blue seems like the most well rounded in this test. I was just going to leave a comment saying I wonder how the different colors anneal and then got to the part of the video where you tested that! Fun stuff.
White is better. Look at the scatter on the sky blue plots, it's all over the place. White also did better in the axial tests on both xy and z, again with less scatter than sky blue. Finally, sky blue deforms 3 times as much when annealed. The only test sky blue clearly won is ductility, but not by much. So white is far more predictable while being just as strong as sky blue IMO, so I'd say it's best as a general choice. Too bad it gets a bad rap. The only potential downside is that I've heard it's more abrasive than other colours.
I did watch your High Flow adapter video. In one part you are talking about internal stresses in the filament when printing above 25 mm3/s flow rate. I was thinking if lower flow rate (hence lower stress) could yield better results in annealing process? To me, it looks like the infill structure is "dragging" walls inwards. Or maybe different infill patterns. Just thinking it could be interesting video.
I have some orange PLA hatchbox filament that is running on 5+ years old now, it sat around in it's vacuseal bag for about 3 years, but so did 2 other colors of that brand, it prints like -nothing- else I've managed to find yet! I can print at 200c easy and the layers turn out glassy smooth and filled in but not melty at the edges etc, layer adhesion also. For all my other filaments 205-210c at minimum to run the same 100mm/s. A red spool from same company/year prints totally different and needs 210c, yet still looks a rather dry surface-finish wise. Nothing like the glass sheen the orange creates. It totally spoiled me and I didn't even know it until I swapped over to a new color finally. Most of my printer upgrades were made with that orange and all pieces have held up fantastically. fully printed belt tensioners and 'fang' style part cooling near the hotend included.
I was printing a stiff part out of TPU. Printing with grey everything worked perfectly. printing with black it was much more flexible and even after dramatically increasing the infill and walls it was still more flexible than the grey (same manufacturer)
Before seeing results: I always treat all filaments of the same type and manufacturer the same, regardless of color. I've not been able to correlate any issues/quality/printability results to any color. But I can certainly tell there are differences between filaments because I use a Palette + RepBox. Manufacturer A sitting in the RepBox after half a day (or less) snap at the exit hole... while Manufacturer B has been there for a month and exhibits no difference in filament property. And for the Palette to splice things together, you (should) do splice tuning... and it's kinda wacky how 2 manufacturers, with the same color, both who proudly proclaim they use pure resin + master batch end up having two very different splice characteristics. To my expectation: I expect everything has some variance and also some differences... but I do think scale of difference will be important. Lining everything up and saying "red can handle 10x as load as blue" is a great headline, but if the difference is 0.5g then it can probably be ignored. If that difference is actually 100g, then that's significant and I'll certainly be planing some print usage by color. Lastly, regardless of the result, I'm glad to see my systematic/deterministic results for 3D printing which is very anecdotal, touchy/feely, and "works for me" right now. The more "here are numbers to make decisions off of", the further we get from "the print won't stick to my Ender 3 v2" getting 50 different responses... and the more we can turn 3D printing into something the tinkerers and hobbiests do for fun and work while the average person can just go "I put in blue and it printed" for every printer. Post see results: I feel my only desires are to see where silk compares but also that one comment you made... "does printing slower change the properties?" as not only is that an interesting one to think about, I feel it could be a hot topic given the, um, arguments going over right now around "why buy a slow printer when you can buy a fast printer?". As for the results, I think it was interesting... the difference in matte filament makes sense. I generally don't use special materials (like silks) for mechanical parts. But it seems like, while there were properties that did have some drastic differences, the results were loosely correlated. Maybe that was just for Das Filament, but the range of it seemed to make it a little less useful. I'm going to be literally printing parts of a floor for a workshop in dark grey PLA to match the injection molded flooring that will take up a majority of it. I'm less concerned on pulling layers apart and more on impact and general strength. From these results, being dark grey may work in my favor as silver had the most impact strength... but I also know multiple perims/top layers/some amount and type of infill may have more significant impact then the color. Still, interesting stuff and maybe going further and testing more options can flesh out results to try and build a stronger relationship between colors and properties.
This is a very interesting investigation. I hope you will take a look at dimensional accuracy on fast prints from klipper driven machines versus slower commercial printers. I wonder if the steppers skipping steps on the faster ones will be an issue or if the faster printing actually yields better results because of less warping!
Very interesting! I suggest you built a comparison test for the bare filament. Is it any link between the filament properties and the printed part in terms of mechanical properties? Last but not least, what about the humidity? Was all samples the same humidity content to start with?
Have had same-filament-color-to-color differences several times, but never had enough batches or precise knowledge of filament condition to distinguish between color-to-color or spool-to-spool dependency. Though always thinking about it because you know - additives.
I use Zyltech PLA a lot for prototyping and making brackets and enclosures for my various vehicle projects because it's really cheap but still reasonably good quality. I have found that silver, clear, and clear blue are the most durable and are the least likely to have threaded inserts spin or get torn out. But if I want something to be super precise; I use milky white as it seems to be the most consistent with almost zero blobbing or stringing.
I work with industrial plastic extrusion machines used to put plastic/PVC/silicone insulation on wires. We used all sorts of colors, even custom ones. From day one during training we drill in the fact that color pigments are actually a contaminate for the plastic, reducing their effectiveness, thus the percentages we use are important and not to be deviated from since it will change the characteristics of the insulator. While we generally don't care about the mechanical strength of the insulation, we are aware that the natural/clear version is generally stronger than the colored added version. It is especially noticable with purple. For whatever reason, all the purple pigments we have do not show as well as the others and we end up using a much higher percentage than other colors. It creates a minor but noticeable difference in the quality of the finished plastic.
With some of the graphs the standard deviation (thanks for putting those in!) makes it a bit tricky to see if it's very significant (looking at the layer adhesion test). Super interesting though and quality content as usual!
Expansion during annealing might also stem from expanding air trapped inside the part. It might be worthwhile to repeat the test with open-pored infill (e.g. gyroid) and a small hole drilled into the part to let air escape.
I've never had to use any of my prints for strength comparison purposes but I have found different colours from the same manufacturer will have different properties in the quality (eg. stringing), so I've been ordering different options as I continue to explore.
![Lp. Сердце Вселенной #60 РОЖДЕНИЕ ЛОЛОЛОШКИ [Финал] • Майнкрафт](http://i.ytimg.com/vi/YoR0pAV9FVQ/mqdefault.jpg)
Did you ever notice a difference in your 3D prints, depending on the material color?
Also: check out our *CNC Kitchen products* at cnckitchen.store/ or at resellers www.cnckitchen.com/reseller and on AMAZON (EU) geni.us/s8rYtQ
for the best print temperature to combat stringing, I encountered up to 15°C difference from the same manufacturer.
Hi CNC Kitchen, may I make a suggestion? Your thoughts on residual stresses are a lot more relevant than you think it is. Up for having a conversation? You're starting to delve into principles around weld buildups. My day job is on the metal side of buildups, but it has a significant amount of overlap with what you are exploring.
I print with ABS and i had noticed some diferences, specially with red and Brown, but i recently changed my supplier to a better and did not test different colors yet.
I discarded the color theory thinking it was my previous supliers that was very bad. Thanks for letting us know that this can be a problem even with good suppliers.
YES and its fkn annooyinggggggggggg
Yes, BUT I think the amount of difference between colors is almost similar to the difference between brands. My takeaway here is to maybe stick with blue or translucent colors for strength from now on. I totally agree on staying away from silk or matte filaments for functional parts, though.
I work in a masterbatch laboratory and can confirm, pigments can make a colour behave very differently. Usually it's way less pronounced in a finished product, with dispersion values ranging usually between 1 and 4%, but as seen in the video - it still makes a difference. A thing I noticed for example, which wasn't mentioned here but I can tell from experience, is that silvery/glitter colorants make the extruded material way more prone to absorb moisture, it's worth keeping that in mind
I believe it is because it makes it more porous, which is why carbon fiber filament (especially stuff like Onyx) absorbs water very quickly
We've found our bronze petg absorbed way more moisture (or just faster) than our gold and silver petg. Same manufacturer
Which Masterbatch Company do you work for?
I am asking out of interest.
@@schizophrenicgaming365 That's very probable, or at least a contributing factor
@@longnamedude3947 It's Avient
This works for other products as well. My wife and I have a commercial embroidery business and one of the first things we learned was that black embroidery thread and white embroidery thread are more prone to breaking than other colors. Black thread is usually another color that didn't come out well, so it was bleached and redyed black. This is also true of white thread. It was usually another color then bleached heavily to make it white. The bleaching and redying process weakens the thread.
Good to know. Yeah i'm sure some black plastics from less trustworthy manufacturers have black be all-regrind with extra masterbatch; which is a problem for ABS in particular since it degrades heavily in the process, but would be noticeable for other polymers as well. I think some other colours can be full of regrind too occasionally, grey for example.
That makes a lot of sense, so dark colours are dyed black because cost wise it is easier and more cost effective to dye it black than it is to try (and fail) to bleach out all of the existing colour.
Very interesting results, with ABS on Vorons we`ve always recommend avoiding the use of white filament as its shown in the past to have poor layer adhesion and increased chance of part failure, but with PLA it looks like thats not the case.
For white specifically the titanium dioxide is also an abrasive, so it may eat up nozzles just like carbon fiber or glow in the dark filaments.
Ah, so that's the reason for the red/black theme /s.
This is absolutely fascinating and I never would have guessed that the difference between pigments was this large. I wonder if it's the pigment itself or the different sourcing of the masterbatches used by dasFilament?
I suspected this after getting some white hatchbox abs and it was crap to print and for strength. Normally I use esun so I didn't know if it was just the brand but, now I know it's white in general, thanks!
@@ska042 yeah, realised this after buying four 4kg spools of the white PETG. Destroyed our RatRig on an overnight print, had to order new hotend, new sensor and print new head assembly using other printers. Now have a load of this filament condemned.
@@jbradleyk If you have that much you could probably buy a hardened nozzle and get good use out of it lol
I'm a paint department manager in a small town hardware store and I found it interesting when I would pick up the quarts of colorant to refill the dispenser, and they would all weigh significantly different. For instance, a quart of white weighs 3.6lbs and a quart of black weighs 2.4lbs. That's when I learned about pigment properties lol
This is also the reason for the so called "pink tax" Pink plastic is not identical to blue, green, or black plastic.
I concur as an ex-paint store employee! The white colorant was a massive pain to lug!
I work in printing and used to order ink. The we were charged by the pound from our supplier. A can of black would be 5.0 pounds and reflex blue was 5.7 pounds for the same size can.
Every single silk silver I have bought has been the biggest pain to print. I’ve been finding color makes a huge difference and no body talks about it. I really appreciate All of your videos for the conversations they start online but especially thank you for this one, I feel validated lol
Same, cant get decent wall adhesion
Typical i hate the "DrY yOuR fiLiMeNt" response but,
I've noticed that the silk PLAs are actually hygroscopic and definitely benefit from drying. But also they need to be printed at much higher temps than typical PLAs.
I had a gold silk pla that liked to be printed at 250c and only printed properly if it was dry.
Well, silk PLA isn't just PLA. They're an alloy with other plastics. So it's really a different material entirely.
Same but in my experience had clear/yellow/gray pla always seemed more pliable to me
Slant 3D made a video called "White Filament Will Destroy Your 3D Printer"
I don’t think I’ll ever get tired of these analysis videos. Great work!
That Safety Third shirt's got you looking extra handsome today Stefan
this is the comment I was looking for
The most interesting thing to me was that natural wasn't always the winner, sometimes the pigment was making it stronger! I wonder how the difference would be in premium PLA blends that already have additives.
More interesting to me in the test results was the wild variation in the error bars per color. That tells me that the color not only influences the macro level performance of the plastic, but also introduces microvariation that can have a notable impact.
Didn't even notice that at first, maybe the weaker prints had slightly more pigment than the stronger ones? Many of these pigments are so concentrated that even a large difference in concentration wouldn't be very noticeable.
Stephan has a filament extruder, this would make for an excellent follow up experiment. 50% recommended masterbatch ratio, 100% ratio, 150% ratio, 200% ratio, etc.
That's the problem is that each brand has different mixes for different pigments and when we are testing all similar materials with simple tests we are going to start seeing the variation in the results itself where each individual tested filament roll might behave differently, what if the weight is different from pigment as well as moisture. So in turn we eventually do need to start testing further into the results to get a definitive conclusion
That or another variable, it's not a super scientific test. I would say almost none of the tests are really significant, with the exception of the hook tests.
or... or this guy didn't isolate a true independent variable and we can draw basically no conclusion from these results other than 3d printing parts have highly unpredictable properties that are still poorly understood.
@@googleyoutubechannel8554 exactly my thoughts
My company works with PLA+. If a client needs a strong piece (within the possibilities of what PLA can achieve), white is everyone's colour of choice and there are no complains. Other than white, blue.
Orange, grey, red and green were other colours I've worked with, but white is by far THE one requested.
Actual cool fact from a chemist working a lot with dyes/pigments: White pigment/master batch includes titanium dioxide which gives the PLA/material better UV resistance
Please create a series of videos comparing the properties of filaments
types across manufacturers.
- The manufacturers probably won’t release their raw materials manufacture and mix recipes. As well as added modifiers.
I would really like to know which manufacturers ASA, ABS, PC, and nylons have the properties I’d expect when choosing a material. I’d like to know if a manufacturer adds modifiers that improve general user performance but negatively affect properties.
It’d be interesting to test the tensile strength of the filament before extrusion, would serve as a useful baseline.
Nice video and much more acurate testing than I did back 4 years ago in my video on the subject "Layer adhesion strength test based on filament color"
Interesting video! I noticed a big difference with the semi-transparent green PLA that I used for printing a guitar neck. This material felt harder and more brittle, yet tougher than the PLA I was used to. By the way, a good reason to print white is that you can draw on it with a pencil. Ideal for prototypes, like a 3D notepad!
Quite shocking results! I’m most surprised by the natural PLA performing so poorly compared to some colors. Super interesting.
It's interesting but its because the pigment pellets have tons of extra additives and plasticizers. which are going to make a big difference because most basic or ""natural"" polymers because like amber or like old plastic beads.
@@infernaldaedra Yeah, that's sort of what I was thinking? I can't imagine the actual pigment causing much influence, more the material its suspended in. 🤔
@@BRUXXUS it does but in most cases it's the blend thats needed to get a good color that does it in my opinion
@@BRUXXUS that's one assumption, but it doesn't make sense to me. I know pigments are not just a drop of food coloring. GITD and white tear through nozzles because they are not pigments, they're colorful minerals, and much harder than any plastic. You can manipulate colors with particle size too. When there's poor adhesive, one cause is the amount of non sticky additives.
@@BRUXXUS The masterbatch usually comes loaded in EVA from what i heard.
I've noticed difference in printing between colored vs. translucent PETG with translucent being the easiest to get to adhere to the bed. In woven fabrics the dye used to color them will also to an extent change how they behave with blue color making the fabric more firm and unyielding.
I use a lot of the eSun Blue PETG which is translucent and prints beautifully. It was cheap too and seems very strong.
PETG has no problem sticking to smooth PEI sheets, it actually sticks too well the solution I found is plastic razor blades. After 2 kg of filament the PEI sheet is not damaged and still going.
That was very interesting Stefan. Very nicely made too. Thanks for sharing. Good stuff
Glad you enjoyed it!
@@CNCKitchen I enjoy all of your vid and when I see the notification, I go all excited and watch :)
I ALWAYS look forward to a new CNC Kitchen video... Excited for this one and I jsut started
They must be adding other additives specifically to compensate for the property changes brought on by the pigments, so they have to tune each color to be roughly the same "strength," hence the relatively small spread on the hook test, which is the most practical.
So then maybe a material tuned specifically to be strong might see less of a spread between different colors, since the manufacturing goals are different.
Would be interesting to see some tests on PLA+.
I'd be very interested in seeing if there was a difference between different nozzles and annealing warpage. Particularly, I wonder if the CHT nozzle (or the diy versions) would make much difference.
This is literally why I create a profile for different colors of filament using a temperature tower and extrusion calibration cube. Once they add a different color component to the plastic, every color takes on their own unique properties. I have found that by starting with the existing profile for the manufacturer and material, when one exists, and generic profile when one doesn't, that different colors from the same manufacturer typically require slightly different extrusion multipliers and temperature to achieve best results. Of course I'm only going for appearance and extrusion accuracy, not strength testing.
I would love to see something similar with PETG. Thank you for such comprehensive tests!
Hey Stefan, are the print layer inconsistencies you show at 6:34 fixable on a Voron 2.4? I have a 2.4 that I have tuned a lot and my layers look similar to yours in this video. I had watched your video "(HOW) I fixed the ugly layers on my 3d printer" and the layers you showed post fix seemed absolutely perfect. So I'm wondering if these inconsistencies on the V2.4 could be related to the Clockwork extruder? I'm currently running CW2 in a Stealthburner. Thanks!
Just yesterday I was looking at taulmans website and they say for their “black” filaments, they use a dark blue color because of carbon’s negative effects on performance
This is very interesting, but I have to admit when printing PLA, the colors matter for the design. If strength is needed other filaments should be used. This would be interesting to repeat for PETG and ABS or ASA.
Yes and no. PLA is not very strong but it is relatively stiff, which along with its low price and ease of printing means it does have a role in structural parts.
@@cooperised I use it to prototype, and then decide if I need to use another material. Since I received my Bambu X1 Carbon, that has allowed me to print almost any material, with great success.
I use pla for structural parts too, sometimes the stiffness of pla is more appropriate than the toughness of petg for example
Good topic- thanks! For what it’s worth I had to dry out several spools of PLA and kept track of progress via weighing. I found the silk spool had absorbed dramatically less moisture than other colors.
Hi Steafan! Are you shure that the warping diffrence druring your "Baking Experiment" is representative? Have you tried to bake multible specimen at the same time? I would guess it comes from unevenness in your oven.
i have to agree with Steam´n stuff, if i place mini pizzas in my oven for the recommended time the outer ones have brown crispy cheese while the middle ones have perfectly melted cheese. 😓
Thanks, Stefan, for another good video. I no longer use PLA because of its distortion over time when under load, such as shelf brackets, and always found white to be a poor performer with its layer adhesion. In those days, printer issues were more common, and filament dimensions, nozzle temperature control, and cooling were major issues. Today, I use PETG and Amphora based carbon fibre filaments. I find that black PETG is more brittle and that transparent is more ductile, while layer adhesion is similar and much better than PLA. I do print at a fairly high temperature, usually 240°C with a 0.4mm hardened nozzle on a Mk3S for all filaments. Sunlu PETG is the brand I use at the moment. I am now trying the Revo nozzle, so there are no substantative answers yet on that one, but the first models appear fine. I'm still waiting for the Obxidian nozzles for the CF filaments that you might consider investigating some time.
I'd heard about certain colors resulting in different plastics being stronger/weaker back when quadcopters were using ABS propellers. Supposedly the (matte) black ones were weaker. Great to see a good analysis of this!
I recently purchased your heat set insert soldering tips. I love them, thank you so much for a great product and wonderful videos like this.
I got a few pla colors when I went shoping at microcenter. One was an extremely nice Glass Purple and I printed a little drawer attachment to my ender 3 pro. However after a few days the filament was breaking in places and when I picked it up, it was like I left a cheap toy in the sun for a year because of the way it broke apart like an autumn leaf. Maybe it could be something wrong with my printer setting but I'm sure some of my other prints would have acted similarly. You should give it a try, see if other "transparent" fillaments have the same issue.
If you print 25-50 mm/s (more relax time), 125-150% extrusion rate (counters stretching), gyroid infill (single chamber), and alternating wall directions (stresses cancel out), does it reduce warping when annealing?
I went through three different manufacturers of silk PLA and two of them had major problems with delamination no matter what temperature I printed at, I have sacked silk PLA now but I am using a lot of matte PLA and carbon fibre strand PLA and PETG-CF and PA-CF Nylon, the first matte PLA I used was problematic to start with but I persevered with it and now can print perfect parts with all these new filaments.
In my experiences, silk and matte finish pla has always given me problems, mostly with layer adhesion and bed adhesion. For colors, I’ve found to like grey and black, I feel they warp less and offered measurable difference in just about all parameters. Blue has always seemed to be very easy to print and offered very smooth, clean prints. And red the worst, I feel like it absorbs moisture much more then others, anytime I need to print in red I dry it for a minimum of a day and I store it in desiccants too, I also feel red, in my experiences, are the most fragile, but I feel that’s because the extra moisture absorption I’ve experienced. I do a lot of prints for orders, like custom Starratt tool boxes for clients, friends and family, so I do have a ton of experience in red pla and pla + printing. And I feel the differences between colors in pla+ are less then plain pla. But these are my simple experiences and thanks for the video, glad to see some I’m not just going crazy with noticing differences between colors of the same brand and type filament.
no no i think you are on to something here. i use Eyrone filament mainly. Black seems almost seamless and perfect. Blue seems sturdy as heck, The red one i used for a MK50 infinite articulating pistol imediatly snapped of after pulling the slider a few times. Printed one in Eyrone Blue and its sturdy AF printed another one with the red one for a budy to pain and again snapped off at the seams. So i do think Red is a poopy PLA to print if you need something sturdy and infact the blue stuff i print rarely breaks even when dropped wherass my Silk and Red just are so brittle as well as my OD green matt filament which breaks easily at the seams. Printing at different temperatures doesnt help at all when testing. Oh yeah PLA+ Deffinatly worth it if you need stronger material without switching to a different plastic type. Ofcourse not a huge difference compared to normal but a bit more sturdy is often enough for Airsoft purposes
Another spectacular testing round! Well done! The variance in properties with color would reinforce why folks have developed different parameter sets for them. Absolutely fascinating. Great work as always.
Have you tried any layer adhesion tests with/ without an enclosure?
Seems like it would also create less stresses in the part aswell.
Great video. I have experienced silks having poor layer adhesion but not mattes. Also, good to know that I don’t have to worry about my recent printer build with white filament. (Unless the base resin and base filament would yield different results) affected by the colored pigments used.
Great video! It would be interresting to see the see results of each color printed at their optimal temperature, and if the colours affects strength when printed as optimal as best
thanks for the video. The difference colorants , especially titanium Oside (white), can make are one of the reasons why I sell my Resins in "natural" color. To achieve peak performance the unaltered material cannot be beaten , and users can color them as they need. same Effects are there with Filament.
Beautiful work, one test i would have loved to see would be thermal conductivity of said colors. That would give a great insight on print temperatures and layer bonding
I've noticed differences in colours of the same materials before. Most recently, when printing a lot of Overture pla for accent colours, it's been very noticeable how much more difficult to remove the supports are of orange, whereas white almost falls off on its own.
So very interesting to see the results in the video.
I wish you have checked the flow rate. From what Ive observerd, I needed the least temperature for blue, then black, green and white to achieve acceptable flowrate.
Indeed, I experienced a difference between different colors of PETG. I use PETG for parts which get hit by metal parts. My prototype was made with transparent green PETG, and had no problems. Then I printed some final parts with black PETG by M4P. The behaved really well. Last I tried a military green PETG, and it failed where the others already had more mileage that this. The green one broke not along the layer lines, but in a diagonal direction. I improved the part at that spot, now it holds, but I was really confused at first.
When i print parts that i have to paint afterwards i always print with uncolored material aka natural. Parts come out stronger, have less stringing, curling, warping... your get the point.
White filament gives me the most trouble, while "galaxy" colors even clogged the nozzle a few times when changing to other filaments after printing.
speaking of different color properties, I want to share, from experience having selling custom order, using White Black and Grey p with many times of experience in printing with it ( maybe around 10 spools or more of each color) They behave really different. I'm using e-sun PLA +, specifically Cold White, Black and Gray. The Biggest difference is the Gray, it tends to warp!!! I even use PEI both smooth and textured. Even using brim doesn't help much. What really helps a lot is lower the temp. Weird but it works. usually on 60C but for gray it has to be 55C or lower, then it wont warp. I've make sure the BED is PID tuned, cleaned & I double check with infrared Thermometer for the temp.
TLDR : E-sun PLA+ Gray ten to warping even with brim if you put bed 60C so i put it 55C its fine, white & black fine with 60C bed. (this is exclude the condition wetness of my spool, sometimes i print with wet & fresh Oven dry spool with or without spool enclosure)
I enjoyed this video and appreciate it very much (which is why I'm coming back to it). This time reading many of the comments.
I feel like the series of tests you did is only the beginning (not that I expect you to do all the work), but I'm not set up to test things as easily so perhaps you/others can dive more into this and related factors? Thanks Stephan! :)
I definitely ran into this in my school's 3D printing lab. Hatchbox red PLA gave me much more consistent and high quality parts with better layer adhesion than any other color of hatchbox PLA.
I noticed this in CNC machining too with different colors for the same plastic (Propylux) Some were harder to cut and had worse surface finish
Yep, I've worked with thermoplastics for decades, and colors always make a difference. Black PVC takes a hit on strength and dimensional stability because it's packed so dense with colorants.
Machining colored cast acrylic, yellow cuts smooth, beautiful shavings of plastic come off the part. With red, small chips and a rougher finish, so I had to compensate for it.
From my experience Silk PLA is the worst in regards to strength, it just loves to split between layers and you can bump up the temparature however you like, it still would not be as strong as plain PLA
I always had layer adhesion issues with different brown colored filaments. I always wondered if my print settings are not right or if this color is really more difficult to print. Fantastic video!
Brilliant. I thought I was going mad. I printed the exact same model in esun blue and esun red and the red snapped in the same place on both prints, yet did not with the blue 👍 Colour does play a part depending in what you're using it for.
Yes, noticed a lot of difference in layer adhesion and strength while using Gembird filaments, over 70 KG of it. Lately switched to Plasty Mladec and C-Tech, first is very very good, second also good and i found out they have Orange HIPS boxes with Orange PETG inside, so i keep buying it as it is cheaper here! Also got some other HIPS colors from C-Tech that i will have to test. Gembird Red and White HIPS is very good, much easier to print than ABS or ASA by a long stretch.
Answering Question at 4:20 ; I recently did a project that used around 4kg of White Filament (Sunlu) and it destroyed my nozzles way faster than any other filament I've used so far. I did some research and I believe it's because of what they use for the master batch.
As usual, great video Stefan!
I think a good baseline would be to do a tensile strength on the filament itself right off the roll. Measuring the diameter with a micrometer would give you great precision since a small variant on 1.75mm makes a big difference.
If you revisit this, you should control for batch variation within the same colour. This variation could be partly explained by each colour necessarily being from a different mix. Is the relationship consistent between different batches of the same colour?
Hej Stefan,
I noticed at 8:53 that the purple hook of your tesile strength mashine is bending a little bit. Maybe you should replace this simple part with a metal one because this could give you little inconsistancy (especially over time). At least u are testing allways on the same mashine, so it should not be that bad. But maybe if you are looking for some upgrades in the future that could something to concider.
Thanks for the good work, love your videos :)
Used to work for LEGO back in the day, supervising various design teams. With LEGO, it's mostly about "clutch power" (strength required to pull two elements apart). Around 2005, we ran into a lot of trouble with our new light blue: apparently the pigment had an impact on shrinking during the cooling phase, leading to our models literally falling apart as all elements made in that color were oversized by a few 1/1000 of a millimeter...
I’ve never had a unusually big issue with silks / rainbows but when I want a consistently great print without a headache I always reach for my black Hatchbox pla cause it just works.
For your extension metering, I know there is software that can use video footage to calculate the displacement field in a specific sample, also resulting in a strain field. The software I know of is called DICengine. I haven't used it before, but seeing as you already film your tests this might be an easy addition.
Out of the PLA varieties I use, Clear PLA feels like it has the best resistance to flexing to me. It also seems less vulnerable to becoming brittle due to moisture, which is invaluable, as I'm based out of Japan where summer gets unpleasantly hot and humid.
Thanks for this test. Just in time- I watched this video before ordering PLA black matt from das filament. Now I will order a usual black one.
You should do a test on annealing/warping based on print speed. It would make sense that a model printed slower would have less internal stresses, so should warp less when annealing, in theory.
Was thinking exactly that. 🤔
Interesting ! My intuition has always been white is the best overall, seem to be held out by these results. Black I have always had issues with. Silver is an eye opener though - always assumed it would be super weak, and have not used it, tended to print white and then spray paint, but looks like if I want to make a mount that I need to anneal in particular its worth a try - cool to know ! Contains real metal particles I guess ?
I really like your comment about white amplifying any inconsistencies in your prints -- I've been trying to dial in my settings lately, and been getting really frustrated that I can't seem get a perfect result like I usually can. Now I'm starting to wonder if it's because of my white filament, just making issues more obvious than normal. I'm gunna swap colors tomorrow (:
White filament is notorious for being harder to print.
Very interesting test! Thank you for the great videos and the invested time. I have also observed that filaments behave differently across different colors. Basically, each color variant requires its own calibration. This means that the extruder steps, flow, and PA/LA must be adjusted and determined for each color variant, taking into account the different shrinkage of the materials and compensating for it. Have the appropriate calibrations been performed for each color variant in the test trials? I'm looking forward to many more videos.
One property that was not mentioned in the video is that certain colors of filament can be more abrasive to the Nozzle than others. One example of this is the use of titanium dioxide in white filament, wich is a very hard ceramic and i´ve heard it´s bad for the durability of especially brass nozzles. Interestingly it seems to perform quite well mechanicly, so it would be interesting to know if Das Filament uses titanium dioxide. Thanks for the informative video!
I am basically only using DasFilament filament in matte black and white. From time to time I have issues with bed adhesion, but that's more on my leveling and my extruder.
I really like their filament
I haven't done actual measurements, but my prints with "natural" (pale translucent white) have seemed strongest and have had a little more flexibility, making the "minies" for TTRPG break less often if dropped.
What polymer blend? Or filament type? Pale translucent white doesn't sound the same as natural but neither is regulated or standardized but usually natural means free from additives. Translucent white might mean they actually tried to brighten up and clear the filament
That actually sounds like it's the weakest of the materials with low rigidity and higher flex and ductility, which is a good thing for when you accidentally drop them.
Just got an Ankermake and haven't even printed anything yet, although the affect of pigment was one of the first things I researched when shopping, and found that same study you referenced! 😆 So of course this video is of interest!
Amazing information... thank you so much for taking the time to do these important tests. Great Video.
The testing and results you show are amazing. Thank you. In the future, can all graphs start at 0? Some of the results for strength and such were trunkated and make the difference seem more significant than they are.
Very interesting, I've wondered about bed adhesion as a function of colour, seems it also affects other properties. As a point for potential improvement of such tests, there are some missing graph numbers (ductility) and some graphs with supressed zeros which make small differences appear larger than they are.
Second the point about misleading graphs. Quite annoying having to pause and try to figure out what the actual difference is
Glitter filaments are usually significantly more hygroscopic, which makes sense, since many contain mica. I have to regularly dry my galaxy-black Prusament ASA, but never have to dry non-glittery colors. The same seems to hold true for PLA, and glitter PLA also becomes more brittle as it sits on the shelf vs. non-glitter.
For PETG I find that the more heavily pigmented, opaque colors have better print quality with less stringing and blobbing. The biggest problem with PETG is that it's very sticky and goopy, and pigments seem to reduce this property. But it probably lowers strength.
I have natural PETG here behaving like a dream, and sunlu opaque blue and it's giving me trouble, like i sure can get decent looking prints out of it, but i do not get any sort of semi decent layer adhesion! Infuriating. I have to drive the temperature way up to get it to stick to itself, and then it blobs and strings more than the natural.
Very well, I have been waiting for this test for a long time.
I actually disagree with using the exact same gcode for the tests. Different colours from the same supplier need very different extrusion multipliers to hit the same part geometry. This can be confirmed with a weight measurement, or by measuring a single wall vase mode cube with proper micrometers (read: not calipers). Some of the PLA we use can range from 0.92 all the way to 0.98 EM, across the same product range where the only difference is the colour. I suspect the results could have been affected by the printed parts all being slightly different in the end.
Sky blue seems like the most well rounded in this test. I was just going to leave a comment saying I wonder how the different colors anneal and then got to the part of the video where you tested that! Fun stuff.
White is better. Look at the scatter on the sky blue plots, it's all over the place. White also did better in the axial tests on both xy and z, again with less scatter than sky blue. Finally, sky blue deforms 3 times as much when annealed. The only test sky blue clearly won is ductility, but not by much.
So white is far more predictable while being just as strong as sky blue IMO, so I'd say it's best as a general choice. Too bad it gets a bad rap. The only potential downside is that I've heard it's more abrasive than other colours.
Love your blue and black 2.4. I went "stock" red and black and wish I had seen before I did.
Well, the great thing about the Voron is, that if you get tired of the colors, then just re-print the parts and re-build it 😅
I did watch your High Flow adapter video. In one part you are talking about internal stresses in the filament when printing above 25 mm3/s flow rate. I was thinking if lower flow rate (hence lower stress) could yield better results in annealing process? To me, it looks like the infill structure is "dragging" walls inwards. Or maybe different infill patterns. Just thinking it could be interesting video.
Impressive video as always
Great walk-through of the problems I didn't know I had!
Thanks again for sharing your expirence with all of us 👍 😀
I have some orange PLA hatchbox filament that is running on 5+ years old now, it sat around in it's vacuseal bag for about 3 years, but so did 2 other colors of that brand, it prints like -nothing- else I've managed to find yet! I can print at 200c easy and the layers turn out glassy smooth and filled in but not melty at the edges etc, layer adhesion also. For all my other filaments 205-210c at minimum to run the same 100mm/s. A red spool from same company/year prints totally different and needs 210c, yet still looks a rather dry surface-finish wise. Nothing like the glass sheen the orange creates. It totally spoiled me and I didn't even know it until I swapped over to a new color finally. Most of my printer upgrades were made with that orange and all pieces have held up fantastically. fully printed belt tensioners and 'fang' style part cooling near the hotend included.
I was printing a stiff part out of TPU. Printing with grey everything worked perfectly.
printing with black it was much more flexible and even after dramatically increasing the infill and walls it was still more flexible than the grey (same manufacturer)
Before seeing results: I always treat all filaments of the same type and manufacturer the same, regardless of color. I've not been able to correlate any issues/quality/printability results to any color. But I can certainly tell there are differences between filaments because I use a Palette + RepBox. Manufacturer A sitting in the RepBox after half a day (or less) snap at the exit hole... while Manufacturer B has been there for a month and exhibits no difference in filament property. And for the Palette to splice things together, you (should) do splice tuning... and it's kinda wacky how 2 manufacturers, with the same color, both who proudly proclaim they use pure resin + master batch end up having two very different splice characteristics. To my expectation: I expect everything has some variance and also some differences... but I do think scale of difference will be important. Lining everything up and saying "red can handle 10x as load as blue" is a great headline, but if the difference is 0.5g then it can probably be ignored. If that difference is actually 100g, then that's significant and I'll certainly be planing some print usage by color. Lastly, regardless of the result, I'm glad to see my systematic/deterministic results for 3D printing which is very anecdotal, touchy/feely, and "works for me" right now. The more "here are numbers to make decisions off of", the further we get from "the print won't stick to my Ender 3 v2" getting 50 different responses... and the more we can turn 3D printing into something the tinkerers and hobbiests do for fun and work while the average person can just go "I put in blue and it printed" for every printer.
Post see results: I feel my only desires are to see where silk compares but also that one comment you made... "does printing slower change the properties?" as not only is that an interesting one to think about, I feel it could be a hot topic given the, um, arguments going over right now around "why buy a slow printer when you can buy a fast printer?". As for the results, I think it was interesting... the difference in matte filament makes sense. I generally don't use special materials (like silks) for mechanical parts. But it seems like, while there were properties that did have some drastic differences, the results were loosely correlated. Maybe that was just for Das Filament, but the range of it seemed to make it a little less useful. I'm going to be literally printing parts of a floor for a workshop in dark grey PLA to match the injection molded flooring that will take up a majority of it. I'm less concerned on pulling layers apart and more on impact and general strength. From these results, being dark grey may work in my favor as silver had the most impact strength... but I also know multiple perims/top layers/some amount and type of infill may have more significant impact then the color. Still, interesting stuff and maybe going further and testing more options can flesh out results to try and build a stronger relationship between colors and properties.
This is a very interesting investigation. I hope you will take a look at dimensional accuracy on fast prints from klipper driven machines versus slower commercial printers. I wonder if the steppers skipping steps on the faster ones will be an issue or if the faster printing actually yields better results because of less warping!
Very interesting! I suggest you built a comparison test for the bare filament. Is it any link between the filament properties and the printed part in terms of mechanical properties? Last but not least, what about the humidity? Was all samples the same humidity content to start with?
Thank you for conducting these experiments. I did notice differences between blue red white and clear PLA. Good to know it wasn’t just me ;)
Have had same-filament-color-to-color differences several times, but never had enough batches or precise knowledge of filament condition to distinguish between color-to-color or spool-to-spool dependency. Though always thinking about it because you know - additives.
This is the reason why voron design PIF wont provide white color since titanium oxide or what ever is used in white color weakens ABS :)
I use Zyltech PLA a lot for prototyping and making brackets and enclosures for my various vehicle projects because it's really cheap but still reasonably good quality.
I have found that silver, clear, and clear blue are the most durable and are the least likely to have threaded inserts spin or get torn out. But if I want something to be super precise; I use milky white as it seems to be the most consistent with almost zero blobbing or stringing.
Very interesting video, I never thought colors would have an impact on the strength of a print and now I know :)
Good basic research. Thanks, Squarespace!
It would be interesting mixing the top ones of each category to see if they average out or get better.
I work with industrial plastic extrusion machines used to put plastic/PVC/silicone insulation on wires. We used all sorts of colors, even custom ones. From day one during training we drill in the fact that color pigments are actually a contaminate for the plastic, reducing their effectiveness, thus the percentages we use are important and not to be deviated from since it will change the characteristics of the insulator. While we generally don't care about the mechanical strength of the insulation, we are aware that the natural/clear version is generally stronger than the colored added version. It is especially noticable with purple. For whatever reason, all the purple pigments we have do not show as well as the others and we end up using a much higher percentage than other colors. It creates a minor but noticeable difference in the quality of the finished plastic.
With some of the graphs the standard deviation (thanks for putting those in!) makes it a bit tricky to see if it's very significant (looking at the layer adhesion test). Super interesting though and quality content as usual!
Great insight presented in a clear way. Thanks for sharing 👍🏼
This was incredibly interesting! Would you be testing the same in PETG?
Nice findings! Generally I was only aware of matte and silk colors to be weaker. Never thought about the "standard" colors
Expansion during annealing might also stem from expanding air trapped inside the part. It might be worthwhile to repeat the test with open-pored infill (e.g. gyroid) and a small hole drilled into the part to let air escape.
I've never had to use any of my prints for strength comparison purposes but I have found different colours from the same manufacturer will have different properties in the quality (eg. stringing), so I've been ordering different options as I continue to explore.