Announcement

Collapse
No announcement yet.

Cetus 3d printer & ABS

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Cetus 3d printer & ABS

    I've had really good results with my Cetus printer with pla filament, so I decided to try some abs. I installed the heated bed and placed my coated bed on top with some longer screws. On my first couple of prints, the raft was warping up off the bed before the print was finished.


    Do I need to put some hair spray on the old bed , or will putting the printer inside and enclosure solve this problem.
    Gary Davison
    Tarkio, Mo.

  • #2
    Just went through this with my Maker Select. I didn't care for the idea of using tape, hair spray or glue and was determined to get it working with either BuildTak or PEI. General feelings are to use PEI on glass for the surface but I was able to get good adhesion on the original BuildTak surface by cleaning it with acetone. Problem was I ended up destroying the BuildTak removing the finished prints. Think heating the bed to 90 and the acetone also contributed to the destruction of the BuildTak.

    I ended up using glass with PEI, printed a spacer that clips above the Z axis limit switch that raises the home position just enough to add the glass to the bed. This also allows me to quickly switch back and forth between the original BuildTak surface and the glass plate with PEI.

    The single most important item I discovered with any surface is bed leveling and nozzle height. The first layer needs to be smashed onto the build surface, if it's not being laid down oval it will never stick. Try cleaning the surface with denatured alcohol and if that doesn't do it try acetone. ABS on BuildTak was sticking almost too well, it was a bear to remove the print and the reason I ended up destroying it. So far the PEI seems to be working and once it has cooled down the parts are fairly easy to remove.

    Almost forgot, my printer is not in an enclosure but is in a warm office with little air movement.

    Comment


    • #3
      I also have a Cetus and, so far, I have only tried printing with PLA. I do have a reel of ABS and plan to use it soon.

      I also had a failure of my first somewhat larger object to adhere to the build plate. It's base was about 3" x 5" and one corner simply lifted off the plate.

      Cetus advertises that it's printer does not have to be leveled. POPPYCOCK! Otherwise known ad BS!. I suspect that ALL 3D printers DO need to be leveled. Leveling refers to making the build plate as perpendicular to the vertical axis of the printer as possible. It certainly needs to be within a few thousandths over the print area. And that would be the area of the mat if you use one.

      So, Cetus, which claims it does not need to be leveled, has a screen/function just for that. I used the readings I got at the nine points across the build plate to shim the build plate to make it physically level. I think the Cetus allows you to enter values for these nine points so the "leveling" is accomplished in the software, but I prefer it to be physically as close as possible first. I haven't had any more failures since I added that shim.

      Cetus does not explain this very well. Their instructions and other information leaves a LOT to be desired.

      PS: I used aluminum foil for the shims between the plate and the mounting surface that it screws down to. I did the math and wound up with two thicknesses at one of the screw locations, one thickness at two others and none at the fourth. I haven't checked the flatness of the several plates that I have, but I suspect that it may vary so different shims may be needed for different plates.
      Last edited by Paul Alciatore; 03-28-2018, 01:29 PM.
      Paul A.
      SE Texas

      And if you look REAL close at an analog signal,
      You will find that it has discrete steps.

      Comment


      • #4
        I found ABS sticks too well to my PEI coated spring steel sheet (Prusa MK3). The glue stick or hairspray trick is to act as a release agent, the way I understand it. Same with PETg and nylon, it sticks too well.
        I found for long or thick parts I have to crank up the bed temp to 115C and slow the print down to about 60% speed, this has worked well for me. I also find the parts pop off easier when the bed is at 115C, not cold. I even tried putting the bottom plate in the freezer to get some ABS prints off with no joy, only heat worked.
        Cheers,
        Jon

        Comment


        • #5
          Originally posted by BobinOK View Post
          Just went through this with my Maker Select. I didn't care for the idea of using tape, hair spray or glue and was determined to get it working with either BuildTak or PEI. General feelings are to use PEI on glass for the surface but I was able to get good adhesion on the original BuildTak surface by cleaning it with acetone. Problem was I ended up destroying the BuildTak removing the finished prints. Think heating the bed to 90 and the acetone also contributed to the destruction of the BuildTak.

          I ended up using glass with PEI, printed a spacer that clips above the Z axis limit switch that raises the home position just enough to add the glass to the bed. This also allows me to quickly switch back and forth between the original BuildTak surface and the glass plate with PEI.

          The single most important item I discovered with any surface is bed leveling and nozzle height. The first layer needs to be smashed onto the build surface, if it's not being laid down oval it will never stick. Try cleaning the surface with denatured alcohol and if that doesn't do it try acetone. ABS on BuildTak was sticking almost too well, it was a bear to remove the print and the reason I ended up destroying it. So far the PEI seems to be working and once it has cooled down the parts are fairly easy to remove.

          Almost forgot, my printer is not in an enclosure but is in a warm office with little air movement.
          I've not heard about BuildTak or PEI. Did some research and looks like PEI would be the way to go. Just peel and stick. A 220x 220mm on Amazon for 13 dollars.

          Just how difficult is it to replace??
          Gary Davison
          Tarkio, Mo.

          Comment


          • #6
            Your build plate needs to be 100 - 110c for ABS to stick. Make sure there are no drafts, turn off the print cooling fan. Only use the nozzle fan.
            Or use PETG, all the orange parts on a Prusa are made with PETG...

            Comment


            • #7
              Changing to a PEI bed will not help with the warping, the warping is from the ABS cooling too fast. An enclosure would definitely help.
              Slowing down the hot end and a high bed temp will help slow down the cooling.
              Cheers,
              Jon

              Comment


              • #8
                Originally posted by RB211 View Post
                Your build plate needs to be 100 - 110c for ABS to stick. Make sure there are no drafts, turn off the print cooling fan. Only use the nozzle fan.
                Or use PETG, all the orange parts on a Prusa are made with PETG...
                110c wow. I'll need an external power supply and a good regulator to get that high. It barely gets to 50c now. I see an enclosure build in the near future.

                And maybe some PETG.
                Gary Davison
                Tarkio, Mo.

                Comment


                • #9
                  Originally posted by gld View Post
                  110c wow. I'll need an external power supply and a good regulator to get that high. It barely gets to 50c now. I see an enclosure build in the near future.

                  And maybe some PETG.
                  For my own purposes, I won’t bother with ABS anymore, and I say this as some one who doesn’t really have many issues printing with it. PETG offers the strength and heat resistance of ABS with the ease of PLA. It is in the middle.


                  Sent from my iPhone using Tapatalk Pro

                  Comment


                  • #10
                    I print ABS on a glass bed, heated to 100C. I have a thin layer of Elmer's Purple Glue Stick on the glass.

                    Prints stick very well during printing, and then after the bed cools, they come loose on their own.

                    Comment


                    • #11
                      ABS is going to shrink different than PLA no matter what you do. All of the tricks and tips listed in here are "correct" they will assist the process.

                      The REAL trick is to design differently to work with the shrink rather than against it. Long flat sections with sharp corners will be troublesome. Large thick sections as well. Print some test samples and watch how they distort and curl, and orient your features accordingly.

                      Once you learn how to do that through iteration and practice ABS will become your goto material. ASA and Trans-ABS as well.

                      Comment


                      • #12
                        Originally posted by Paul Alciatore View Post
                        Cetus advertises that it's printer does not have to be leveled. POPPYCOCK! Otherwise known ad BS!.
                        My Cetus was dead level straight out of the box to within a couple of thou (most of which was probably variation in the adhesive layer), no adjustment necessary. It's far from ad BS with the Cetus, and that's one of the reasons why it's probably the best value consumer printer available.


                        Originally posted by Paul Alciatore View Post
                        I suspect that ALL 3D printers DO need to be leveled. Leveling refers to making the build plate as perpendicular to the vertical axis of the printer as possible. It certainly needs to be within a few thousandths over the print area. And that would be the area of the mat if you use one.
                        They really don't, at least, any more than the first time, and IMHO that should be done at the factory. The whole fixation with bed leveling in the 3D printing community is laughable coming from a machining perspective, because the entire necessity for bed leveling is due to the unnecessary design compromises made by the designers of 3D printers, and I think that's a consequence of where the impetus of consumer 3D printing came from - it was primarily driven initially from the electronics and software end of the design world, where the mechanical realities didn't match up to the expectations of what electronics and software could compensate for.

                        Printers initially were homebuilt jobs mostly assembled by people with no experience or tooling for precision assembly. Once they hit the big time, they were asian import consumer devices built to the lowest price-point possible, trying to squeeze every penny, while still being sold effectively as 'kits' for otherwise inexperienced, untrained, and ill-equipped consumers to assemble. Removable beds came along, but they needed to be affordable, so there wasn't any attempt to mechanically key them to the printer to ensure repeatability. It was impossible to ensure that the bed was level from the factory, because the factory had no control over how the machine was assembled, nor whether removable build surfaces were installed level.

                        As soon as you begin to design a 3D printer with the same mechanical principles you would any other standard cartesian mechanical device, almost all of the things that make 3D printing problematic for most people go away. The problem is it isn't done because of the price point the machines are being built to - and a part of that is the need to ship them affordably direct to consumer from China.

                        The Cetus is revolutionary in my mind as an engineer, specifically because it was designed from the outset to be affordable, but not the cheapest, with solid mechanical and motion control fundamentals as you would use for designing any other cartesian machine, and only then had the print-specific parts tacked on. That's specifically what sold me on it almost overnight, when I had been dismissing 3D printers against all evidence of practicality for 5+ years on grounds that I now realize weren't problems with the concept of 3D printing, but were issues with the specific tropes with which it was being implemented.


                        Originally posted by Paul Alciatore View Post
                        So, Cetus, which claims it does not need to be leveled, has a screen/function just for that.
                        The Cetus software has a bed leveling function only because it's a reskinned version of the control software for the UP Plus series of printers made by the same company as Cetus (Tiertime). The Up Plus and the Cetus share, AFAIK, mostly the same control electronics. The UP Plus printers need bed leveling because they have removable print beds. It's really not necessary, and I would agree that mechanical leveling of the build surface, if necessary, is infinitely superior to software compensation.

                        Comment


                        • #13
                          Try to design your parts as if they were injection molded. What I mean is, if you ever really studied an injection molded part, you would notice a very important design feature, uniform thickness. Sidewall thickness is the same across the part. Where it is not, you get curling / warpage. You can print a raft or use five or six skirt passes to help hold down the part, but it is the differeing thicknesses causing the most grief.

                          Comment


                          • #14
                            Perhaps you have more experience with 3D printing in general and with Cetus in particular, but I can not automatically agree with your comments.

                            My comments are below, in red.



                            Originally posted by Sun God View Post
                            My Cetus was dead level straight out of the box to within a couple of thou (most of which was probably variation in the adhesive layer), no adjustment necessary. It's far from ad BS with the Cetus, and that's one of the reasons why it's probably the best value consumer printer available.

                            Well, YOURS may have been that well adjusted at the factory, but mine definitely wasn't. And there was absolutely no sign of any damage in shipping. Even the corners of the outer box were not beat up.

                            The Cetus plate measures about 10 inches corner-to-corner, diagonally. Two of the screws in that same direction, also measured diagonally, measure about 1 inch. That is a 10:1 ratio. I had to shim those screws about 0.0014" to get it level. That means that the plate was off by about 0.014". I do not consider that to be good enough and my failed part seems to verify that. If your printer arrived from China with better alignment than that, I would guess that you were lucky and conversely, I was not that lucky. I don't think that Cetus is doing anything prior to shipping to guarantee the needed accuracy. I stand by my comment that they or at least some of them DO need leveling.

                            As for factory adjustment, WHERE is that? I do not see any places for adjustment of the Cetus printer. They seem to depend of precision parts for proper leveling. Is this adjustment just in the slop in the assembly screws? That's not a very good design.


                            They really don't, at least, any more than the first time, and IMHO that should be done at the factory. The whole fixation with bed leveling in the 3D printing community is laughable coming from a machining perspective, because the entire necessity for bed leveling is due to the unnecessary design compromises made by the designers of 3D printers, and I think that's a consequence of where the impetus of consumer 3D printing came from - it was primarily driven initially from the electronics and software end of the design world, where the mechanical realities didn't match up to the expectations of what electronics and software could compensate for.

                            I couldn't agree with you more about the code geeks not understanding accuracy in assembly. And YES, they do think that everything should be done in the software. But then, that is all they know. Most schemes that I have seen for "leveling" the plate seem to be for installing a second plate on top of the original one and using set screws in the corners to level it; talk about a Mickey Mouse solution.

                            Printers initially were homebuilt jobs mostly assembled by people with no experience or tooling for precision assembly. Once they hit the big time, they were asian import consumer devices built to the lowest price-point possible, trying to squeeze every penny, while still being sold effectively as 'kits' for otherwise inexperienced, untrained, and ill-equipped consumers to assemble. Removable beds came along, but they needed to be affordable, so there wasn't any attempt to mechanically key them to the printer to ensure repeatability. It was impossible to ensure that the bed was level from the factory, because the factory had no control over how the machine was assembled, nor whether removable build surfaces were installed level.

                            As soon as you begin to design a 3D printer with the same mechanical principles you would any other standard cartesian mechanical device, almost all of the things that make 3D printing problematic for most people go away. The problem is it isn't done because of the price point the machines are being built to - and a part of that is the need to ship them affordably direct to consumer from China.

                            I agree that standard mechanical principles should be used. Cetus did seem to take a lot of trouble in packaging my printer to survive the trip. I would give them an A++ on that one.

                            The Cetus is revolutionary in my mind as an engineer, specifically because it was designed from the outset to be affordable, but not the cheapest, with solid mechanical and motion control fundamentals as you would use for designing any other cartesian machine, and only then had the print-specific parts tacked on. That's specifically what sold me on it almost overnight, when I had been dismissing 3D printers against all evidence of practicality for 5+ years on grounds that I now realize weren't problems with the concept of 3D printing, but were issues with the specific tropes with which it was being implemented.

                            The Cetus software has a bed leveling function only because it's a reskinned version of the control software for the UP Plus series of printers made by the same company as Cetus (Tiertime). The Up Plus and the Cetus share, AFAIK, mostly the same control electronics. The UP Plus printers need bed leveling because they have removable print beds. It's really not necessary, and I would agree that mechanical leveling of the build surface, if necessary, is infinitely superior to software compensation.
                            I am glad that you agree that mechanical "leveling" (alignment, actually) is superior. I can not talk about the software as this is my first exposure to 3D printing and to this company.
                            Paul A.
                            SE Texas

                            And if you look REAL close at an analog signal,
                            You will find that it has discrete steps.

                            Comment


                            • #15
                              Oh, and while I am thinking about it, there is one serious problem with software leveling. It does not correct the basic problem that the Z axis needs to be perpendicular to the X and Y axis. If it is not and you just compensate in the software, then you are still building your part at an angle to perpendicular. You could call this the "Leaning Tower of Pisa" effect. If this is severe, it can also produce failed prints. You really do need proper mechanical alignment FIRST. And hopefully, that is all you really need.
                              Paul A.
                              SE Texas

                              And if you look REAL close at an analog signal,
                              You will find that it has discrete steps.

                              Comment

                              Working...
                              X