Friday, February 6, 2015

Ciclop turntable build

I've finished the Ciclop turntable assembly (using the files from GitHub) and it is solid and functional, and looks quite nice in my opinion.

The base is chunky but with pleasantly rounded lines; the turntable support and bearing clips go together well without any cleanup or fitting needed; the motor-to-turntable connection is well thought out and makes assembly very easy. The only problem is that some of the nice features also make it hard to print.











The pretty arch on the bottom of the base is a classic no-no in FFF printing, starting off easy but becoming an impossible overhang. And not visible in my photo but just as important, the flat area where the motor attaches is also unprintable, even as a bridge, because it has to contain the holes for the motor shaft and the four attachment screws.










There are a couple of ways this could be dealt with. One would be to slice the entire piece in half horizontally, to allow the curved base to be printed upside-down and the flat top upright, with screws holding them back together. A small redesign might even allow for the four motor screws to to the job. I don't know for sure whether the flipped lower part would print cleanly, but I think the fillets along the curved edges would help.

All that said, the motor base does print; even though the finished product is a little bit sloppy the problem areas aren't readily visible, and it is still entirely functional. I did some basic cleanup of strings and drooping lines, and had to drill the holes for the M3 screws to mount the motor since they were completely closed over; I was able to use the top infill pattern to locate them pretty well. It is printable because the designers at BQ included hand-drawn support material, visible at the edge of the arch. Slicer-generated support material is a mixed bag in my experience, typically effective but often very difficult to remove. In this case it might be a real improvement though, since the supports as supplied were barely functional. The very large and solid puck at the bottom printed fine, but the tiny, single-walled tubes that were intended to do the actual support work were too weak and didn't print cleanly, with only about half of them making it all the way to the point where they were supposed to hold up the overhangs.






The finished product is the first functional part of the scanner, so naturally I had to jury-rig a circuit to allow the Raspberry Pi to drive it. I have more appropriate prototyping parts coming in the mail: an Adafruit T-Cobbler to connect the GPIO pins instead of using spare motor and endstop cables, and a USB serial console cable (also from Adafruit) that connects directly to the right pins. But this bodged version was able to make the turntable spin, and by good luck I actually connected it to turn in the correct counterclockwise direction.





Power is coming from a tiny switching supply that I salvaged from an old external disk enclosure; it produces 12 and 5 volts at 1.5 amps each, which ought to be enough for this project. While testing I'm using an old Model B Pi, but I have a shiny new Pi 2 waiting in the mailbox to replace it.

Next up will be printing the parts for the other end of the scanner; the camera and laser holders. I've had to redesign both of them in order to match my parts and the physical layout that the FreeLSS software prefers, so while those print I'll document the Bill of Materials, so far.

16 comments:

  1. Cool stuff Bill.

    I'm going to start this print this evening. Will be so kind to share the STL's of the modified parts after you tested them?

    Regards

    ReplyDelete
    Replies
    1. Absolutely - I've printed everything now, and just need some time to wire up the electronics and test. I don't want to post the parts until I'm sure they work ;)

      Delete
    2. I was able to get the Piclop working last night, and have made a couple of scans but there are clearly issues with the calibration, and possibly with the way I have the camera mounted. Right now it's held on with scotch tape, but that's strictly temporary! I need to understand how it should be positioned, and that may change the design of the camera holder. But if you want to start with what I have, I plan to post the parts tonight in their current state.

      Delete
    3. That is great. I'm still busy printing all the turn table stuff and I'm still waiting for my motor and other spare. Thus no rush.

      Do you have any thoughts on the Methacrylate disc? From what I've read Methacrylate is nothing but acrylic. I've found couple of guys that can laser cut it for me and they asure me that perspex would do the same job.

      Thanks for the help

      Delete
    4. I haven't decided how to handle the turntable plate yet. I did my test scans last night with the object just sitting on the turntable base (the part with the three bolt holes). I think that FreeLSS expects the turntable to be higher though, and that may be part of my problem. Plastic is difficult for me to cut neatly and I don't have a laser cutter, so I might just make one out of plywood and paint it black. I don't think it is in any way critical as long as it is not obviously warped or twisted.

      Delete
    5. I found your project while researching both the BQ and Atlas 3d scanner. I like what you are doing by combining the best of both concepts. I'm trying to follow your idea, but I guess I need more of a 1-2-3 build plan. I've been to github and any other page you mentioned and bookmarked all of them. I also ordered the things (well most of them) on your BOM list and hope to have them in the next week or so. I already tried the Kinect version of scanner, but it doesn't do small objects.(made a neat bust of my grandson though) I have a DaVinci 3d scanner and hope to start making parts Easter weekend. Did you post the modified STL's you referred to earlier? Thank you for any help or direction to do this project. It's been a few years since I had to do hands on and I'm looking forward to getting back into the game.

      Delete
    6. I haven't done any further writeups yet, but if you want to get started printing you can get the turntable files for the Ciclop and print those without any changes, and then print the modified camera and laser files from Thingiverse: http://www.thingiverse.com/owens/designs Other than that, it's a matter of deciding how large you want the scanner to be and cutting the 5/16" or 8 mm threaded rod to length; I'll measure mine to give you something to start with. Assembling the mechanical parts is pretty simple, but I'll also add some pictures of the finished scanner so you can see how everything fits together; the laser arms in particular are slightly different than on the Ciclop (simpler, though).

      Delete
    7. Ok, now comes the inexperience on my part. Went to GitHub/ciclop and did a fork to my repository pawizard/ciclop. So I have all the files, but can't figure out how to download them or manipulate them to a destination on my desktop. I see the 'download zip' but can't make a zip file from them. I'm guessing that all I really need would be the STL file contents to build the turntable? Or do I need to use the 'freecad' files?

      Delete
    8. All you need is the package of STL files; the Freecad would be useful if you wanted to modify the parts but for printing they're not needed. I just tried the Download Zip button and it worked for me, but the easier thing to do is go to Thingiverse and search for 'ciclop' and it should be the first result; the folks from bq uploaded the files a couple of days ago.

      Delete
    9. This comment has been removed by the author.

      Delete
    10. The basic diagram is on the FreeLSS website (freelss.org) and that's what I started from. I made a modification to let the lasers run from a separate 3.3 volt power supply since the ones I bought were intended for 3 volts input. Uriah Liggett has indicated that the Kickstarter kits will come with 5 volt laser modules, but I didn't find those for sale when I looked. I have a couple of other modifications to the circuit but I'm still working on those; I'm not an electronics expert, I'm just taking what other people have done and bodging it together so it takes me a while to make sure it is all right ;)

      Delete
  2. For the disc you could try these. I have them and all going well will get the last of the turn table bits done tonight / tomorrow.

    I was going to try to post the link directly but it is pretty long. But search on ebay for this text.

    Acrylic Plexiglass Green Edge Round 3/8" x 8" Diameter Circle Clear NEW

    Also the 12" ones would work if you wanted to scan a bigger object possibly.

    ReplyDelete
  3. Hi Bill,
    It's me again. I've printed everything that I can find and I have a question about the laser mounts and the clamps. The clamps don't fit into the mounts and I'm not sure where they would go. Do you have a pictogram of the setup you're using. I know it's still in progress as you mentioned. I also have the 5/16 threaded rod and I would like to cut them to the approximate sizes. I'm guessing that the sizes mentioned in the original 'ciclops' BOM is right from the turntable to the camera mount. But I can see that you (maybe) have the rods/struts going from one laser to the other laser straight through the camera mount piece.
    Thank you,
    Michael

    ReplyDelete
  4. hi does anyone have a technical drawing or sketch of the pattern holder? i have not got the facility to print yet but i am building a metal ciclop and will need the holder for calibration.
    thanks everyone

    ReplyDelete