Modified shutter

AD001AD001 Member, Backers
In an effort to maximally control light bleed on long prints, I have my shutter setup to be closed whenever I am not actively projecting (layer changes etc). This means my shutter actuates.. a lot. So much, in fact, that I wrecked the original stepper. My fixing of that stepper is documented elsewhere on the forums, but unfortunately the problem returned. My fix worked- the stepper wore out elsewhere in the mechanism. Oops! Analyzing the failure mode(s) of the stepper tells me that this much actuation will wear down the stock stepper sooner than later.

Thus, I have done two things- replaced the stepper with a superior model, and replaced the shutter. The stepper is an MG996R, and the shutter is a smaller, thinner one. I'm quite pleased with the results! My new 'design' weighs in at almost exactly half of the stock model, significantly reducing strain on the stepper and speeding up actuation a bit.

See the imgur link for some pics: http://imgur.com/a/3QqTW

I have enough plexi to make a few more if people are interested. I can also order some more and make some to send out. I'm guessing they'd run maybe $15 incl. shipping. Maybe less (definitely less if more than a few are interested)..

Comments

  • AD001AD001 Member, Backers
    edited April 14
    As a side note, I have my projector on the 8th slot and set to take up almost the entire build plate. The light profile still does not reach near the edges of the new, reduced shutter; additional light bleed isn't a concern. The only disadvantage I could see is slightly reduced protection from resin drips. As you can see, I've derped and done those once or twice.. usually during cleaning when my projector isn't in the machine anyway.
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    Nice mod Andrew! 

    It seems hard to believe that your brass overhauled servo died again! So moving to the MG996R is a better choice? Is that the same as the vat tilt servo?
    Ron
  • GravelleGravelle Member, Backers
    Does anyone know if there is a source for the Serial Port commands to change brightness, contrast and gamma on the any of the projectors we use.  It occurred to me yesterday that instead of physically blocking UV light you might be able to digitally keep the projector from emitting it in the first place.  This could be a better long term answer over actuating a shutter between layers.

    -Gravelle
    Check out my Youtube channel Garage Science for videos and tutorials about printing with the Draken.
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    You can find the serial codes for those commands on the projector, but you would have to write custom firmware modifications to convert them into gcode that you can add to your scripts.

    I am not sure however how well that will work unless you can actually turn off the lamp. Since it is a constant source of light emission and the light path of the projector is not a perfect light sink, those commands operate more on the digital signal I think.

    What they need is an iris in front of the lens that can be adjusted. High-end projectors have this capability, but I don't think it makes it into the projectors we have now.
    Ron
  • AD001AD001 Member, Backers
    edited April 15
    rkundla said:

    [...]


    I am not sure however how well that will work unless you can actually turn off the lamp. Since it is a constant source of light emission and the light path of the projector is not a perfect light sink, those commands operate more on the digital signal I think.
    [...]
    Yup, unless you physically block the lens, the 'black screen' still has plenty of light. When you work with 4+ hour prints (especially in light-colored resin), this adds up.
    rkundla said:

     So moving to the MG996R is a better choice? Is that the same as the vat tilt servo?
    I think so. I'd have to un-dynamat my tilt servo to see what 3DFacture put in there so I can't currently confirm if it is the same. I CAN confirm that I believe it to be a better choice. I changed two variables at once (servo and shutter) so I can't be 100% sure how each contributes independently. The brass-reinforced portion of the servo still works great, by the way. The gears on the inside had already begun to grind down from before that mod, however, and had a bit of play in them. I believe the GFRP wore out elsewhere in the motor, on the bottom. An all-metal housing would solve this, likely. If this somehow happens to my new servo I'll stop being a lazy bum and just mill one out of 6061. Until then... the MG996R is what I'm using. I've tossed the old servo in my 'motors' drawer since it still works aside from having a lot of play in the gearing.

    As a final note, the new servo/shutter combo DID noticeably reduce printer noise, if anyone is concerned about noise.
  • AD001AD001 Member, Backers
    edited April 15
    Update:

    I noticed a bit of vibration on the new setup! I have concluded that this is due to the shutter being entirely undamped. The shutter size is small enough that it is well within the load capacity of the stepper, but there's still opportunity for harmonics to make it vibrate every so often. I believe I have solved the problem by further reducing the size of the shutter slightly- it still has a healthy amount of extension over the projected light. For reference..

    Stock shutter weight: ~140g
    Modded shutter weight: ~65g
    Modded Rev. 2 shutter weight: ~50g

    We are now sitting at ~1/3 of the stock shutter weight. Noise and speed were not affected between 65->50g. ~Engineering intuition~ tells me that the 65g shutter was fine, but that the particular shape presented an opportunity for some dickery re: vibration. I'm not going to bother running equations on it because it isn't important enough. The 50g model works just fine, and the only real difference between the two is that I can fit more 50g shutters on a sheet of plexi than I could of the 65.

    Upon reflection, I think the optimal solution to preventing vibrational feedback in the servo would be to include weak dampers on the ends of the swing. Some quick experimentation shows that a pen spring is more than sufficient. If I end up designing a mount for dampers on the end, I'll toss it on the forums.
Sign In or Register to comment.