Announcement

Collapse
No announcement yet.

Make A Switch Assembly For 3D Printer

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

  • Make A Switch Assembly For 3D Printer

    I am getting acquainted with my 3D printer and one thing that bugs me is the processes of leveling the print plate and setting the height of the print head above it. In my current office set-up I can not locate the printer next to my computer so I have to run back and forth sever or many times when I do these set-ups. They tell you to test the print head height above the plate with a piece of paper. Now, while that works OK, it is a real PITA to run back and forth, setting the print head one step further down, and testing with the piece of paper. Over and over, until you hit the right point. PITA, BIG TIME.

    It occurs to me that if I could just have a plate that fits the build plate and which has a series of switches that make contact with the appropriate distance between the print head and the build plate, then I could just put that switch assembly on the build plate and operate the vertical feed in the desired position until a switch closes and a light comes on. No back and forth.

    After finding the position of the print head with the piece of paper (0.003", +/-) the instructions say to back the print head up for a half mm (0.020") before setting it to zero. So, about 0.023" above the build plate is the zero position for the first layer of a 3D print.

    My thoughts are as follows:

    1. Start with a square of heavy, flexible, insulating material the size of the build plate (about 150mm square in my case). This should sit down nicely on the build plate with no gaps or at least just small ones.

    2. On top of that place a square of SS shim material. This has no particular features except for a tab for attaching an electrical wire. It should be thin enough so that it's shape is determined by the insulating material of #1 above. Perhaps 0.003" thick.

    3. On top of that a square of insulating material with some cutouts at the switch locations. These cutouts would be round or square, in the range of 1/4" to 1/2" in size. This would form a space between the Conductive layer above and the one below. First guess would make this layer at about 0.005" thick. I need a gap, but not too much of one.

    4. Finally another square of SS shim material with Holes that match the ones in #2 above, but that have a bridge that goes across the center of those holes and is around 0.100" to 0.200" wide. The print head would hit these bridges and press them down and into contact with the lower SS layer for the switching action. It would also have a tab for attaching another wire.

    If I want the light to come on when the print head is about 0.023" above the build plate, then the first, insulating layer would need to be about 0.023" - 0.005" - 0.003" = 0.015" thick. Of course, adjustments may be needed on a second, third, etc. prototype. But the total would remain at about 0.023".

    My questions:

    1. Does this seem like a good way to make this?

    2. What can I use for the first, insulating layer? It needs to be both flexible and somewhat heavy so it sits flat and holds the layers above it down. And it would need to have good dimensional stability in terms of it's thickness. The pressure of the print head on it should not make more than a thousandth or two of deformation/indentation.

    3. How can I cut the holes in the insulating and SS layers? The tolerances do not really need to be very tight: +/- 0.005" would be great and +/- 0.010" is probably OK. But they need to stay as flat as possible after being cut and have NO burrs or turned edges. Nothing should change the spacing determined by the insulating layer. I can't even begin to think about how to deburr SS shim stock. Water jet sounds like a good way, but I don't have one or know where to find one. But I could work on that. Laser? I don't know what kind of edges to expect and even a thousandth or two would/could be problematic. Perhaps etching?

    Any other thought on this would be appreciated.

    Sorry, this is in the spitball stage so there is no drawing yet.
    Paul A.

    Make it fit.
    You can't win and there is a penalty for trying!

  • #2
    Can you get an app or vpn to your desk top for your phone?

    Comment


    • #3
      You could actually make a little mechanical gauge using the leverage effect to amplify small motions. Having only about 20 thou to work with makes it harder- but you would be able to see a needle deflect from several feet away. A custom miniature balance no more than 18 thou or so high at the sensor end.
      I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

      Comment


      • #4
        Originally posted by Paul Alciatore View Post
        ...Any other thought on this would be appreciated. ...
        Solve the original problem? If your printer doesn't have a jog wheel then get a wireless keyboard for your computer. I have a little "remote style" one with a touch pad. Direct from China for $12 or so. Works across the room just fine.

        But, assuming your printer does have limit switches...

        I have one of those that require manual levelling and height adjustment, same 4 corner screws. I too had thoughts about automating it in some fashion. But, you know, it's so easy now I couldn't bother. Paper once in a while, but mostly just a 1/2 turn tight to make way for using double-sided carpet tape or glass, back for just masking tape. If a print looks wrong going down on a skirt, stop the job and run each or the appropriate corner up or down to compensate. It's pretty obvious just by looking at that first few lines if it's the right height or not. Getting a feel for how much to adjust the height doesn't take long.

        Or, if the above is all wrong for your particular printer, your "running across the room" issue is easily solved by a simple continuity light. Either clip onto the print head or make an insulated probe tip that clips on, ground plate on the build plate. Advance until the light comes on, then adjust a known amount. The light doesn't have to come on exactly at zero. So long as it's consistent that light-on point is always going to be the same amount of adjustment for a given plate material. Software or hardware, up or down, say .22mm or 1/4 turn on the screws. If you have something different on the build plate, it's a different adjustment constant. You'll figure out what those are quick enough.

        David...
        http://fixerdave.blogspot.com/

        Comment


        • #5
          Why are you setting the nozzle height often enough for it to be irritating? You have a Cetus, the software retains the nozzle height offset. Unless you're changing out nozzles or the bed on a daily basis (again, why?), the nozzle height is set and forget. I haven't changed any of my bed leveling settings, or the nozzle height offset setting, since the last time I took the bed off, over 6 months ago, and I'm knocking out 5-10 reliably successful prints a week. I can't remember the last time a print failed that wasn't a design problem or a slicer fault. My parts are coming straight off the printer and easily holding ±0.1mm dimensional tolerances without any preparation or interaction from me beyond pre-warming the bed for good first layer adhesion.

          You're making this way more complicated than it needs to be.

          Even then, the technique you propose is absurdly overwrought. 3D printing is a 'near enough is good enough' process. If you're having to set nozzle height or change the bed settings that frequently that it's irritating you walking back and forth (again, the first question you should be asking is why you're having to change those settings?), engage your machinist brain and solve the problem. Fabricobble (3D print, even!) a bracket to mount a dial indicator to the print head. Power off the machine, rest the nozzle and the indicator probe on the bed, zero the indicator. Turn on the machine, home, rapid the head down to wherever it's zeroed, see whats on the dial indicator, and change the nozzle height offset by the difference between what you see and what you want it to be. You should be more than close enough on the very first attempt to successfully print.

          But again, I'll ask - why the heck are you worrying about this stuff? You level the bed once, you set the nozzle height once, then start printing, and forget about it. If the settings are changing often enough that you're having to recalibrate, then there's something mechanically wrong with your Cetus and that should be investigated in detail first.

          Comment


          • #6
            Probably not, I have a $25 phone.



            Originally posted by no704 View Post
            Can you get an app or vpn to your desk top for your phone?
            Paul A.

            Make it fit.
            You can't win and there is a penalty for trying!

            Comment


            • #7
              Why indeed! Have you followed the Cetus complaints on Facebook? Over and over, a rat's nest because the print did not stick to the build plate. I have had only one failure and, contrary to Cetus' claim that you do not need to level the bed, I found that my BRAND NEW printer, with only a couple of previous prints to it's credit, DID have a plate that was NOT LEVEL. I didn't get a rat's nest, but one corner of the object I was printing was not adhered to the build plate. Once I figured out how to level it and did so, I have not had any more print failures. QED!

              Now I have purchased an extra plain plates and a heated one so I will be changing them back and forth. How do I know that my Cetus, that does not need to be leveled, is actually level. I think:

              YES, it does need to be leveled! At least if you change the build plate. I will agree that leveling seems to be OK after doing it once with a particular build plate.

              But YES, you probably need to level it if you change build plates. I haven't tried reinstalling a build plate with the shims that I created for it when leveling, so I don't know if you can do that.

              And when I have checked the head height before printing a new object the next day, after turning the Cetus off and back on with the same build plate, I almost always get a different reading for the correct head height. Am I doing it wrong? Is it my technique? Or what? Until I can get consistent readings the next day or week, I think I need to continue to check this each time.

              Anyway, overthinking it? I don't think so. I do like things to be under control. I do like your DI idea. I am going to think about that.

              I am in the process of making a cabinet and moving it closer to my computer.



              Originally posted by Sun God View Post
              Why are you setting the nozzle height often enough for it to be irritating? You have a Cetus, the software retains the nozzle height offset. Unless you're changing out nozzles or the bed on a daily basis (again, why?), the nozzle height is set and forget. I haven't changed any of my bed leveling settings, or the nozzle height offset setting, since the last time I took the bed off, over 6 months ago, and I'm knocking out 5-10 reliably successful prints a week. I can't remember the last time a print failed that wasn't a design problem or a slicer fault. My parts are coming straight off the printer and easily holding ±0.1mm dimensional tolerances without any preparation or interaction from me beyond pre-warming the bed for good first layer adhesion.

              You're making this way more complicated than it needs to be.

              Even then, the technique you propose is absurdly overwrought. 3D printing is a 'near enough is good enough' process. If you're having to set nozzle height or change the bed settings that frequently that it's irritating you walking back and forth (again, the first question you should be asking is why you're having to change those settings?), engage your machinist brain and solve the problem. Fabricobble (3D print, even!) a bracket to mount a dial indicator to the print head. Power off the machine, rest the nozzle and the indicator probe on the bed, zero the indicator. Turn on the machine, home, rapid the head down to wherever it's zeroed, see whats on the dial indicator, and change the nozzle height offset by the difference between what you see and what you want it to be. You should be more than close enough on the very first attempt to successfully print.

              But again, I'll ask - why the heck are you worrying about this stuff? You level the bed once, you set the nozzle height once, then start printing, and forget about it. If the settings are changing often enough that you're having to recalibrate, then there's something mechanically wrong with your Cetus and that should be investigated in detail first.
              Paul A.

              Make it fit.
              You can't win and there is a penalty for trying!

              Comment


              • #8
                How many of the current auto-calibration techniques have you investigated: piezo-electric, Fsr, z-probe, etc.? Your proposed method sounds overly complex for such a simple task.

                Before I started using Printbite for a build surface, I used plate glass mirrors. Once calibrated, I changed mirrors for every print without having to re-calibrate.

                .023" (.58mm) seems rather high for the first layer setting. That's nearly three times higher than the .2mm I use for ABS. Different filaments may require different first layer heights to adhere to the bed. How will your method handle that? The normal way is to set machine 0 at .003" above the bed and the slicer sets the first layer height. Slic3r has the .003" compensation built in.

                p.s. Kapton is a good insulator.

                Comment


                • #9
                  Why are you changing the beds out? There's something fundamentally, conceptually wrong here. You only need one bed, whether it's heated or not.

                  I mean, no wonder you're having problems. Of course if you swap out the build plate you need to re-level the bed. The same as you'd need to tram a bridgeport if you took the head off. That's not a design flaw of the printer, that's misuse by the user.

                  You shouldn't need to change the bed out, until the adhesion goop wears off or gets too scarred, and even that's a stretch.

                  I don't take much stock in complaints about misbehaving 3D printers anymore, without proof of the problem. Failed prints and birdsnests are not evidence, they're symptoms. Too many misdiagnosed ID 10T errors.
                  Last edited by Sun God; 05-03-2018, 02:52 AM.

                  Comment


                  • #10
                    Cetus recommends raising the head 0.5mm after establishing the position with a piece of ordinary paper. I mike the paper at about 0.003" and 0.5mm is about 0.020" so that adds up to 0.023". This is for the 0.4mm nozzle. I haven't tried other nozzles yet but I would think that would vary with different nozzle sizes. And I have been using PLA so far, but hope to try ABS soon. Apparently the Cetus slicer does not add any offset: they specifically tell the user to manually add that half mm. I am a beginner at this and have not experimented with that recommendation yet. I don't know about other printers or slicers. I do want to find out how to use a different slicer but one thjing at a time.

                    You talk about piezoelectric and other ways of doing this. And say a simple switch is overly complex. Really? But I do not want to reinvent the wheel so, where do I read about any of them?

                    As for handling different heights, I would just manually move the print head after establishing whatever distance it is made for. That is what I do now after using the scrap of paper and before I hit the Zero button. In short, no problem there.

                    As for changing build plates, I am not sure that their aluminum plates are made to the same tolerances as your glass mirrors.

                    Why did you talk about Kapton being a good insulator? Is that something I could use for the first layer of my switch assembly? How flexible is it? How heavy? I will look it up.



                    Originally posted by elf View Post
                    How many of the current auto-calibration techniques have you investigated: piezo-electric, Fsr, z-probe, etc.? Your proposed method sounds overly complex for such a simple task.

                    Before I started using Printbite for a build surface, I used plate glass mirrors. Once calibrated, I changed mirrors for every print without having to re-calibrate.

                    .023" (.58mm) seems rather high for the first layer setting. That's nearly three times higher than the .2mm I use for ABS. Different filaments may require different first layer heights to adhere to the bed. How will your method handle that? The normal way is to set machine 0 at .003" above the bed and the slicer sets the first layer height. Slic3r has the .003" compensation built in.

                    p.s. Kapton is a good insulator.
                    Paul A.

                    Make it fit.
                    You can't win and there is a penalty for trying!

                    Comment


                    • #11
                      Paul, its sort of a solved solution in some quarters. My printrbot arrived some years ago with a inductive Z probe and inductive probing is in pretty much most of the firmware variants out there, depending on how many points you want to take the average level off.
                      I don't take my bed off or alter the z setting at all, for months on end. I let it warm up to temps so everything expands first then let it run a auto level every print cycle generated as part of the header in the slicer config. I use kapton across my whole bed exclusively now, but i used to mess round with masking tape and hairspray etc and had the same experience.

                      Reprap wiki has some stuff on z probes.
                      http://www.reprap.org/wiki/Z_probe

                      Comment


                      • #12
                        Paul

                        I purchased but did not install a BLTouch probe.
                        see:
                        https://www.indiegogo.com/projects/b...-3d-printers#/

                        I have a Tevo Tornado. The touch probe integrates with the firmware of the printers controller. The adjustments described below have worked out so far. Instead of the probe or whatever else, you could print up a holder for a dial indicator fastened to the extruder. Don't know if it works for the Cletus...

                        I have my extruder set much closer to the bed plate, about 4-5 thou. I had a lifting problem as well, but raising the bed temperature slightly to 20C has solved it using my PLA filament. My extruder is at 220C. If your control program comes with a few slicers, try a different one. I use the Cura instead of the default Slic3r package.
                        I am too much of a rookie to mess with density and such as yet. Still using the factory bed plate material, but I will prolly move to a glass bed soon.

                        If you are using PLA check these folks out.
                        https://coexllc.com/
                        shipping was cheap and the service was great.

                        I have no fiduciary interest in any of the aforementioned products, just happy to use them.
                        paul
                        ARS W9PCS

                        Esto Vigilans

                        Remember, just because you can doesn't mean you should...
                        but you may have to

                        Comment


                        • #13
                          I have a MakerGear, level the plate one time (2 years ago) and then use the height adjustment I did away with the bare glass plate, hair spray and the like and just use PEI sheet on the glass. Never issues, its a heated plate, turn off the heat if not needed. The MakerGear and I have an older one just works.

                          I just checked out the Cetus 3D printer. It seems to have a design flaw that would make accurate printing difficult. Instead of the single X arm support most all others have a boxed in design that supports the X axis on both sides. The flex in the printer is always going to be an issue. Its not the plate.
                          Last edited by wmgeorge; 05-03-2018, 09:57 AM.
                          Retired - Journeyman Refrigeration Pipefitter - Master Electrician - Fine Line Automation CNC 4x4 Router

                          Comment


                          • #14
                            Loving my Prusa MK3, nothing to level, ever.
                            Cheers,
                            Jon

                            Comment


                            • #15
                              I have a Prusa I3 clone and because I used a square when assembling the kit and have the printer on a level base it's easy to set the bed level and it then stays that way.
                              If you benefit from the Dunning-Kruger Effect you may not even know it ;-)

                              Comment

                              Working...
                              X