Prusa Unveils Their Own Line Of PLA Filament

There’s little debate that the Original Prusa i3 MK3 by Prusa Research is just about the best desktop 3D printer you can buy, at least in its price bracket. It consistently rates among the highest machines in terms of print quality and consistency, and offers cutting edge features thanks to its open source iterative development. Unless you’re trying to come in under a specific budget, you really can’t go wrong with a Prusa machine.

But while the machine itself can be counted on to deliver consistent results, the same can’t always be said for the filament you feed into it. In a recent blog post, [Josef Prusa] explains that his team was surprised to see just how poor the physical consistency was on even premium brands of 3D printer filament. As a company that prides itself with keeping as much of the 3D printing experience under their control as possible, they felt they had an obligation to do better for their customers. That’s why they’ve started making their own filament which they can hold to the same standards as the rest of their printer.

Their new filament, which is aptly called “Prusament”, is held to higher physical standards of not only diameter but ovality. Many manufacturers simply perform spot checks on the filament’s diameter, but this can miss bulges or changes in its cross-sectional shape. On your average 3D printer this might cause some slightly uneven extrusion and a dip in print quality, but likely not a failure. But the Prusa i3 MK3, specifically with the Multi Material upgrade installed, isn’t most printers. During testing even these slight variations were enough to cause jams.

But you won’t have to take their word for it. Every spool of Prusament will have a QR code that points to a page which tells you the exact production date, length, percent ovality, and standard diameter deviation of that particular roll. An interactive graph will even allow you to find the filament’s diameter for a specific position in the spool, as well as determine how much filament is remaining for a given spool weight. It should be very interesting to see what the community will do with this information, and we predict some very interesting OctoPrint plugins coming down the line.

Prusament is currently only available in PLA, but PETG and ASA variants are coming soon. You can order it now directly from Prusa Research in Prague for $24.99 per kilogram, but it will also be available on Amazon within the month for help keep the shipping costs down.

33 thoughts on “Prusa Unveils Their Own Line Of PLA Filament

    1. There’s a pretty strong case to be made than PETG/Nylon can fulfill the niche roles in which ABS would have been used over PLA in the past. So there’s less and less interest in dealing with the peculiarities of ABS now that we have a wider selection of materials available.

      1. “niche roles” ?!?!

        I hardly consider being able to survive if left in a hot car to be a “niche” role. Maybe if you live in a big city with great public transportation and rarely leave said city… for the rest of us I would think surviving in a car is a necessity for all but shelf-bound nick nacks. I’ve had PLA parts soften and deform just left outside on a very hot sunny day.

        PETG might be ok. I intend to play with that as soon as I upgrade my hotend. Still, PETG’s glass transition is about 20 degrees lower than ABS so I don’t know if it is “good enough” yet.

        As for Nylon, if you need something extra tough or low-friction then that’s a great “niche” application for Nylon. I thought the toxicity of ABS fumes was the main reason to get away from ABS though and that Nylon if overheated a little is potentially much worse.

      2. Not sure why my lenghty comment wasn’t good enough for the moderators. So I will try to use very simple language now.
        I like ABS, it’s a material with all the right properties, except for warping, which is annoying but manageable.
        PLA is too brittle, so you can’t make anything that encounters even small amounts of flexing.
        PET-G fails catastrophically without any hint beforehand, so it’s not reliable.
        Nylon is very expensive, needs design modification (too flexible) and is really annoying to print (water, print speed)
        ABS can flex, stress whitening tells you of impending failure. It’s also inexpensive. For me, an ideal go-to material when I don’t need any special properties. PLA might be nice for decoration, but I don’t really care about this.

        1. Because why bother using a heated bed if you don’t need to? Most of my printers don’t have heated beds, and while I occasionally do appreciate the flexibility of having it, it’s certainly not an ability I need on a day to day basis.

        2. I my case, it’s because the amount of time it takes to heat the bed to temperature is significantly longer than the time it takes to heat the extruder. Not a long time in the grand scheme of things but I don’t keep my printer running continuously and consider waiting that time before I can print to be a hassle.

        3. well.. the heater itself is not a problem, but I’m pretty sure that most 3D printer can’t supply the current for the bed. Simply because of a cheap under powered powersupply. Or by the absence of thermal control making adding a heated bed a problem. We must be aware of the fact that there are still people out there who aren’t technically skilled or equipped to modify their printer to add a heated bed.

          Personally when I’ve build my printer I didn’t have a heated bed, but when I noticed things were not sticking the way the could I added the bed and a lot of my problems were solved. From the moment the bed was added I tried ABS and from there I never looked back. But things really went great when I also added the buildtak sticker, but that’s a different story. Buildtak is great but the cheap ali/ebay knock off works just as fine.

          So Although I understand that people are “scared” of heated beds, I can only say that it is worth the time and effort.
          The statement that it takes more time to heat up the printer is correct, but it that really an issue to wait ten minutes longer on a 3 hour print? Ohhh and another 10 minute to wait for the bed to cool to remove the print without damaging it OR damaging the buildtak, because both are easily damaged and more sticky when hot.

          1. Actually one of the things I miss about a heated bed is how easy it was to remove the prints. Just let the bed cool and they slipped right off.

            With PLA I use blue printer tape and rubbing alcohol. The prints stick so well I sometimes have trouble removing them.

        4. @Joseph – “because the amount of time it takes to heat the bed to temperature is significantly longer than the time it takes to heat the extruder.”

          Ok, I can see that.

          I usually start my bed heating first while I’m doing the slicing. Of course if I’m printing something I’ve printed before it’s already sliced but that’s not my most common situation. In those cases I go get a drink or something.

          I haven’t really even tried printing even PLA on my current printer without my bed heated up though. I know when I first built it and I was experimenting with the temperature lower settings lead to prints that sometimes popped off early, even with PLA.

          In the past when I did use a printer without a heated bed I always had to use blue painters tape. I hated that stuff as it worked too well making prints overly hard to remove. Also sometimes it would leave blue die and/or paper fibers embedded in the bottom of my prints. Having a heated bed allowed me to stop using that. I do still use glue stick for prints that are more prone to bed adhesion problems. I like when I can avoid even that though as the naked glass makes a very smooth and shiny surface on the print!

          I hope to one day upgrade to a silicone mat heater running on line voltage for faster heating times. I do like being able to use PWM though. I am considering trying to rectify the line voltage to DC then running that through a big SCR to allow me to keep using PWM.

          Another crazy idea I sometimes ponder but would probably never actually attempt… butane powered bed pre-heater! I love my butane powered soldering iron as it heats immediately. Why souldn’t my printer?

          @Jan – “Simply because of a cheap under powered powersupply. Or by the absence of thermal control ”

          I opted to build my printer but had I not done so I likely would have bought one of those “cheap” printers you are describing. I came close to going that route. I knew that I wanted a heated bed so I put some thought into that. I don’t see any reason that the bed HAS to be ran off of the printer’s own power supply nor does it even HAVE to be controlled by the printer’s electronics or even G-code.

          I’m surprised I haven’t seen this here yet. A cheap Fleabay heated bed + an SCR which could also be on a ready-made module from Fleabay or dx.com or something like that + an arduino and whatever power supply one can find laying around ought to do the trick. For a deluxe version one could add a character display and a rotary encoder to set the temperature.

          That sounds to me like exactly the kind of beginner friendly Arduino project that this site loves! The bed could be clipped on with bulldog clips and the electronics all sit next to the printer keeping the printer completely unmodified. Well.. except for leveling the bed anyway.

          “We must be aware of the fact that there are still people out there who aren’t technically skilled or equipped”

          Ugh! Don’t they have enough websites, magazines, tv channels even?!?! Can’t this be HACK a day?

          “I understand that people are “scared” of heated beds”

          Then it’s even worse than I thought! I didn’t suspect that there was actual FEAR involved. :-(

          “Ohhh and another 10 minute to wait for the bed to cool to remove the print without damaging it OR damaging the buildtak”

          Just use glass clipped to the bed by bulldog clip, no buildtak. Unclip the glass and it will cool quicker. If you are really in a hurry you can pry with a razor blade and be pretty abusive to the glass without damaging the surface.

          1. I agree with you about the blue painter tape. That’s what I use on my bed with PLA and it can sometimes be difficult getting the print to release. It’s one of the things I do miss about the heated bed.

    2. I still use ABS sometimes, it doesn’t print as well as PLA but it has better mechanical properties (more flexible, higher melting point). I have been having better and better luck with PETG as an ABS replacement, it doesn’t warp and has similar mechanical properties. The problems with it are it is more stringy so prints with lots of small islands will be connected with strings, also I find you have to print slower with it to get good results.

    3. I havent used ABS in a long time. PETG just prints easier and gives about the same material qualities. I only print in ABS if I need that extra flexibility or if i plan to acetone smooth a model afterwards.

    4. Chemists at my work pretty much stated “ABS isn’t really good for anything, and it’s only used because everyone uses it and it’s average at everything”.

      For printing, PLA gives much better aesthetics, PETG gives the same or better strength and temperature resistance as ABS. Nylon is much tougher. For every goal, there is generally a better alternative then ABS. And seeing the issues with printing ABS, and the huge difference in printing properties within grades of ABS, it’s not strange that people are moving away from it.

      1. That’s also why ABS became popular.
        It’s a allrounder for when you don’t know exactly what it’ll be exposed to, just that it’s something that takes its toll on plastics.

        Master of none is usually accompanied by Jack of all trades.

  1. I’m more of a 3D printer (l)user than a developer so maybe this is out there, but has anyone looked at a rolling micrometer setup to compensate the feed rate for the diameter of the filament during the print? I always take a measurement of the diameter in the next few loops and plug that into the slicer before I send the file, and yes it does vary a little bit. That would probably be a total PITA for a production setting since it can trigger a re-slice before every job.

      1. I’ve seen this a couple of years ago, good to see it’s been developing. As in the article, the measurement is often only the diameter, not ovality. Maybe two sensors could improve it, measuring the filament with in two directions, getting a better idea of deviation from a perfect circle. Noise in the system could introduce artifacts as well. It would be interesting to see a comparison. This could also double as an out of filament. I’ll see if I can add this to my diy 3D printer :)

  2. Has anyone here ever printed with filament that had lousy dimensional tolerances?

    Some years ago I remember reading a blog about someone making a recycler. He said the resulting filament had a lot more variance in it’s diameter than what one usually buys commercially. He claimed that this did not matter, when he printed with it everything came out fine!

    Unfortunately I don’t remember a lot of specifics including the URL.

    Maybe this was long enough ago that his extruder used a DC motor instead of a stepper. That kind of makes sense as it would probably take less pressure to push a thinner section of filament thus the motor would turn quicker and vice versa. That might help to regulate the volume extruded.

    1. In a process like this motor speed would be very carefully controlled. The diameter is measured with 3 digit precision and a closed loop control system will be regulating the speed of the filament output to get _exactly_ the right diameter. I’ve printed with cheap inconsistent filament and some of the good stuff and the difference is huge.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.