Sandbox Constructability Advice

Home Forums AR Sandbox Forum Sandbox Constructability Advice

Viewing 14 posts - 1 through 14 (of 14 total)
  • Author
    Posts
  • #125169
    rysheki
    Participant

    I’ve basically never built anything before in my life, have no tools, first project, etc. My background is on the tech side of things and I’ve pretty much got everything else working except for the sandbox construction. I’ve been trying to research what might work and mocking up some potential designs, but I need a reality check on if I’m anywhere remotely in the ballpark of constructable, and what issues I might run into. I’m using bolts in certain areas to disassemble and move the structure if necessary.

    Here is a link to the model in PDF, DXF, and SketchUp formats: https://1drv.ms/u/s!AgHX_kP_GBGygqZfQASgx0AvEpy0rw?e=kv05CW

    There are a lot of these out there according to the registered map – is anyone aware of any other build schematics, etc. that might be out there for other’s sandboxes?

    Any feedback / advice is greatly appreciated.

    ISO View:
    ISO View

    Top View:
    Top View

    Front View:
    Front View

    Right View:
    Right View

    #125179
    Sean Robinson
    Participant

    I’ll give feedback, but don’t take me as a construction expert. If you have never built something like this before, I recommend trying to partner with someone with carpentry experience. This is a big piece of work and you may need another pair of hands to hold things while working/assembling.

    I like your plan. Yes, keep the camera centered on the sandbox and offset the projector. And it looks like you’ve accounted for the projector lens being off center from the body of the projector. Don’t worry about the projected image perfectly fitting the box; overshoot the box edges and let the calibration crop the image for you.

    It looks like the sand is about 20″ off the floor. Who is your target user? This seems low for anyone other than small children.

    Why mount the CPU in the air? It works, but another option is to use long USB and video cables with the CPU nearer the ground.

    Make sure your projector can handle the short throw needed. How are you attaching the projector and camera to the frame? Your projector mounting plans may go through big changes after you have the projector in hand.

    I have as-built drawings for a different style of cart/sandbox at https://showcase.scottsdalecc.edu/geology/arties-arsandbox/ It’s served us well in the limited use we had before our COVID shutdown.

    #125184
    rysheki
    Participant

    Sean –
    I responded earlier, but it’s it showing up yet. I think it may be some manual post moderation requirement when posting embedded photos or links. I’ll try and reply with them stripped out to see if it makes a difference. Apologies if this ends up being a double post.

    —–

    Sean – Thanks for the feedback.

    Yeah, I had planned on working with friends / family that might have the necessary tools, etc.

    I had opted for keeping the projector closer to the Kinect sensor and just dealing with the keystone over-projection and edge focus issues. As you stated, the over-projection should be handled by the calibration, etc.

    Yes – the target users are actually my children – ages 2 & 5. 22″ was waist high for one and armpit for the other, so seemed like a good compromise between the two. I fastened the legs using bolts so I could replace them as the kids outgrow the height, but I fully expect them to play with it for all of 5 minutes now that I’ve put this much time and money into it haha.

    The CPU is in the air because I didn’t really care for the other mounting positions I mocked up. It’s pretty low at this height if you try to mount it under the sandbox, and I wasn’t sure about ventilation / cable management. Up top it’s well ventilated, I know all the cables will reach, and I only have to deal with one cable (power) coming off the structure. It has the added benefit of the power button being out of reach of my 2 year old, and I think the plywood will help prevent any parallelogram bending all those end joins might have – a happy accident another person pointed out.

    Projector is being attached via a
    Fino FUP-150 low profile universal projector mount.
    http: //www.promounts.com/oneproducts/fup-150

    Kinect sensor is attached via a 3D printed mount based off of a model on Thingiverse by jirisliz.
    https ://www.thingiverse.com/thing:2741802

    Projector is an InFocus IN1501
    https ://www.projectorcentral.com/InFocus-IN1501.htm
    and my design takes into account the throw characteristics of the projector.

    I have been testing with different projector positions which is why the upper rigging extends so far back. The projector I’m using has a -4″ offset throw characteristic at that distance, so my initial builds had a non keystone projection requiring two independent supports for the upper rigging to avoid interfering with the projected image. A lot of the builds I’ve seen just use a single support for the upper rigging in the middle of the back, but I’m keeping the two supports in case I feel like I need to move the projector back to that non-keystone position.

    I actually have everything setup horizontally in my basement with a cardboard sandbox placeholder against the wall, so I’m pretty confident the positioning will work. Only thing left is the sandbox construction.

    Thank you for sharing your design! It gives me some additional ideas. I have been looking into extruded aluminum use for some of the upper rigging to allow for flexibility in positioning if I don’t like the way things are working out, but I haven’t mocked anything up yet. I thought some extruded aluminum rails on one of the supports holding a small monitor & mouse keyboard for the independent display capability as well, to allow for height adjustment over time.

    #125187
    Sean Robinson
    Participant

    My questions/feedback were mainly to ensure you had thought about those things and it seems you have. I like that you’ve mocked up your configuration; I think you’ll have fewer surprises.

    I see that you have a native 1024×876 XGA projector. I found it hard to get a new short-throw, XGA projector and opted for higher resolution,

    And I hope when you have your sandbox constructed that you’ll post your finished design and announce it here for other to learn from. I published my design because I also had trouble finding others plans when planning our sandbox.

    #125206
    rysheki
    Participant

    I appreciate you asking your questions. Answering them makes me feel more confident about starting the process.

    I spent many a night filtering through eBay projectors and cross-referencing projector central for throw ratio, etc. to find one that fit the preferred criteria outlined in some of the original guidance. Finally snagged one for $160 with only 32 hours on the bulb (if that can be trusted) due to a few black / white pixels which won’t matter for this application. Didn’t realize until later that the -4″ offset from the projected image might be an issue.

    Yes – I do plan on posting the finished product and model when completed. Experimenting with extruded aluminum modifications now. Could end up being pretty adjustable for use with different projectors potentially.

    • This reply was modified 3 months, 2 weeks ago by rysheki.
    #125183
    rysheki
    Participant

    Sean – Thanks for the feedback.

    Yeah, I had planned on working with friends / family that might have the necessary tools, etc.
    I had opted for keeping the projector closer to the Kinect sensor and just dealing with the keystone over-projection and edge focus issues. As you stated, the over-projection should be handled by the calibration, etc.
    Yes – the target users are actually my children – ages 2 & 5. 22″ was waist high for one and armpit for the other, so seemed like a good compromise between the two. I fastened the legs using bolts so I could replace them as the kids outgrow the height, but I fully expect them to play with it for all of 5 minutes now that I’ve put this much time and money into it haha.
    The CPU is in the air because I didn’t really care for the other mounting positions I mocked up. It’s pretty low at this height if you try to mount it under the sandbox, and I wasn’t sure about ventilation / cable management. Up top it’s well ventilated, I know all the cables will reach, and I only have to deal with one cable (power) coming off the structure. It has the added benefit of the power button being out of reach of my 2 year old, and I think the plywood will help prevent any parallelogram bending all those end joins might have – a happy accident another person pointed out.
    Projector is being attached via a Fino FUP-150 low profile universal projector mount. Kinect sensor is attached via a 3D printed mount based off of a model on Thingiverse by jirisliz.
    Kinect Sensor Mount
    Projector is an InFocus IN1501 and my design takes into account the throw characteristics of the projector. I have been testing with different projector positions which is why the upper rigging extends so far back. The projector I’m using has a -4″ offset throw characteristic at that distance, so my initial builds had a non keystone projection requiring two independent supports for the upper rigging to avoid interfering with the projected image. A lot of the builds I’ve seen just use a single support for the upper rigging in the middle of the back, but I’m keeping the two supports in case I feel like I need to move the projector back to that non-keystone position.
    I actually have everything setup horizontally in my basement with a cardboard sandbox placeholder against the wall, so I’m pretty confident the positioning will work. Only thing left is the sandbox construction.
    Thank you for sharing your design! It gives me some additional ideas. I have been looking into extruded aluminum use for some of the upper rigging to allow for flexibility in positioning if I don’t like the way things are working out, but I haven’t mocked anything up yet. I thought some extruded aluminum rails on one of the supports holding a small monitor & mouse keyboard for the independent display capability as well, to allow for height adjustment over time.

    #125260
    Sean Robinson
    Participant

    How secure is the Kinect on the mount you posted? I looked at this style of mount and decided it looked too likely to drop the camera while rolling our sandbox cart across campus.

    One of our students designed a cradle we currently use. The design needs revision for better rigidity, but the COVID lockdown stopped our ability to access the 3D printer and trial new versions. But it’s better than the aluminum plate and nylon zip cord I used at first.

    #125264
    rysheki
    Participant

    Wow, that first response finally posted! Haha
    Apologies for the double. I would edit one, but I don’t appear to have the option at this point on either.

    I’ll have to provide an update on the 3D printed mount once the sandbox is closer to being finished. I’m refraining from attaching the mount to the Kinect prior to affixing the mount to the upper rigging which hasn’t been built yet. I’m not sure that it will be easily separated from the Kinect once attached and I won’t have the clearance to screw the mount to anything once the Kinect has been attached.

    As I recall the comments on the thingiverse listing had mentioned having to file some parts of the mount down a little to get the Kinect to snap on, which lead me to believe it would be a snug fit, but I could be remembering one of the other models I looked at. Holding it up next to the Kinect I almost feel like I may need to scale it down a percent or so for it to fit. Would you mind sharing your cradle mount STL?

    Unfortunately work has really picked up leaving me less time to finalize my model,and it shows little sign of letting up in the next month or two. I’ll be sure to remark on the Kinect mount fit in my final build post.

    • This reply was modified 3 months, 1 week ago by rysheki.
    #125266
    Sean Robinson
    Participant

    My impression at the time I was looking for a 3D printable Kinect mount was that the style you have printed was for a tripod mount. I could not find anything about using it to mount the camera on a vertical surface. It might work for that, but I could not convinve myself that the posters and commenters I read had tried it how I planned to use it.

    I will post our current STL file, but it will be a few weeks. Our WordPress site would not let me attach an STL file and I’ll need to find a work-around. And I’ll be away for almost two weeks, so I won’t have time to work on it until I get back.

    P.S. I thought I had read that posting before… I suspect posts with external links require a moderator to approve. As long as I’m linking only to another forum posting, it seems to work right away.

    #125321
    Sean Robinson
    Participant

    With the help of our campus IT people, I have been able to post the STL 3D model for what I call our KinectCage. A preview image (and link) is available about halfway down our Artie’s ARSandbox page.

    Here is a direct link to the current KinectCage model: https://showcase.scottsdalecc.edu/geology/files/2021/04/KinectCage-20191105.stl

    #125556
    rysheki
    Participant

    Thanks Sean! I imported your STL with the intent to modify it for my project, but found it easier to just create my own based on the idea instead. Here’s what I came up with in case it’s useful to anyone else here.

    Cheese Grater Kinect Cage Mount v2:
    https://1drv.ms/u/s!AgHX_kP_GBGygqg23fGRG2WIV5lXpA

    Test fit is great on my 8020 extruded aluminum rail. Washers can be ignored for most, I just needed the extra width to use the bolts / tnuts that came with my 8020 kit. Top chamber access holes were designed for “L” shaped Allen key access. Seems to fit my Kinect (360) perfectly. Still have to construct and assemble the rest of the sandbox, so haven’t had the chance to actually try it out yet though.

    #125566
    Sean Robinson
    Participant

    I like your design. It looks like you have plenty of air vents, which was a potential problem with our first draft version. I assume the two halves slide together around the Kinect and are held by the 80/20 Al system. The “arrows” all pointing toward the joint/center is a nice touch.

    Thank you for mentioning 80/20 Aluminum, I’d never heard of it before. I’ll keep it in mind for future projects.

    Another benefit of you posting your STL file is that I finally got an STL viewer. Which showed me that the model file I published was the wrong version. The box I posted before was an early version we never used. I’ve uploaded the model for the mount we currently use.

    #125568
    rysheki
    Participant

    Ha! That’s funny you should mention that! One of my first modifications of your design looks very similar, just geared to slip over that 1″ rail: https://1drv.ms/u/s!AgHX_kP_GBGygqdZHHjBO1HiQPuDkA

    The 80/20 was great. I’ve got a rig that can adjust to whatever changes may be necessary due to a multiple rail configuration. I’ll post a completed model once I make a few final adjustments and finish my build.

    The triangles were something my father in law added (who was printing the STL for me) to avoid the print needing support structure that would just need removed once the print was complete. The printer we used couldn’t fit the entire cage at once so we split it in two to facilitate printing but that had some impacts on the design that needed adjustment afterward. Appears to be pretty solid now though *knock on wood*.

    #125591
    Sean Robinson
    Participant

    Our KinectCage v1 would not fit in our local printer and we asked a sister college to print and mail it to us. Of course, we had the wrong dimensions and the Kinect would not fit in this first printing. One reason for the minimalist (and smaller) redesign was to speed up the print-revise iterative process by allowing us to print it locally.

    Although KinectCage v2 is used on our AR Sandbox, I think it needs more stiffness on the back plane. When faculty and students are back on campus in Fall and Spring (assuming the latest timeline holds), I hope to revise and print v3.

Viewing 14 posts - 1 through 14 (of 14 total)
  • You must be logged in to reply to this topic.

Comments are closed.