Welcome to Project Dalek

You need to be a member in order to access our content.

 

Click Here To Join!

 

For instant access to the plans and all Forum features, choose Active Membership.

trickster

Members
  • Content count

    60
  • Joined

  • Last visited

  • Days Won

    3

trickster last won the day on April 2

trickster had the most liked content!

Community Reputation

41 Good

About trickster

  • Rank
    At Maxtible's mansion
  • Birthday

Profile Information

  • Gender
    Male
  • Location
    Eugene, Oregon, USA

Recent Profile Visitors

473 profile views
  1. Have tryed to download the file for the bottom right part of the cowl for a NSD but it seems to be corrupted . please would you  attach a good file to my e- mail

    robin.wakefield@baesystems.com

    All the other files work o.k 

     

    Regards

          Robin

    1. Ferrain

      Ferrain

      I just downloaded & gave it a trial on my Makerbot print software, looks good on the bed & in the print preview export.

       

      Where is the problem when it's being sliced or in the printing?

       

      Whats the make\model of your printer & software?

       

      Also using status as a PM method is not effective - You have no guarantee that the recipient will see the message before it sinks to the bottom, then disappear as more & more status updates come in above this one. 

       

    2. trickster

      trickster

      I know the file is flagged as corrupted when it's opened with 3d Builder but it prints. That's why the screen shots have the red boxes. I've tried running it thru a repair but it still gets flagged. But Replicator didn't complain. Much. May have thrown a few warnings about weird triangles, but it printed ok.

      Ferrain is right too. Status posts might not only get lost, but they may not be seen by others with input. Of course that goes with PMs too. Probably the best place for something like this would be the feedback section in the 3d download area. Please let me know how it works out tho.

    3. robin.wakefield

      robin.wakefield

      Ok

        but on the software it's looks corrupt

       

      Kind regards

         Robin

  2. Well I've got the cowl done! Thanks to Jbourd who gave me a 3d scan of a complete dome. I was able to use 3d Builder (the brain dead buggy tool that comes with Windows 10) to carve away the dome and leave the cowl. Well, mostly. I think there's a chunk of dome that just wouldn't delete. Still, considering the original props had cowls that didn't quite fit, I decided it was close enough for rock and roll. But of course a cowl is much too big for my printer's build space. I split it into two halves. The only way each would fit in my build space was on an edge. I printed one and the measurements along the top and bottom seemed short. I thought it might be the original model but I checked and the model was accurate. Then I realized I had forgotten to punch bolt holes. So I did that and printed again. This time it was obvious that the print was warping. The bolt hole came out as a flattened oval instead of round. I tried flipping it over. Bolt holes were OK but the join edge was warped. No matter what I did the halves warped beyond use. I reasoned that if the part could be built flat on the build plate any warpage would be in thickness which I didn't think would matter too much. But half a cowl won't fit. So I decided to split the cowl into quarters. Now each piece would fit on my build plate without issue. I punched some peg and bolt holes and added some pegs. I was able to save three out of the four pieces, but no matter what I tried, 3d Builder refused to save the last piece. Even if I took the opposite piece which did save and mirrored it, the bottom left quarter would still not save. I realized that my build program Replicator also had a mirror function. I was able to take the right bottom, mirror it there, and print it successfully. When the parts were built flat on the build plate, something else happened. Because the curves were close to the horizontal plane, each threshold of .2mm left a curvy perimeter line. The pattern looks just like wood grain! On the two bottom parts which are mirrors of each other, the pattern is perfectly symmetrical. But the top parts which I had made separately were at slightly different orientations and the patterns were different. So I ditched one of the top quarters and reprinted it as a mirror of the other. Now the patterns were all symmetrical and looked great! I took a look at the warped pieces I had made earlier and none had similar patterns. Since the curve was around the vertical plane it just looked like straight lines for each layer along the vertical surfaces. Next I inserted the pegs and glued the pieces together. Trouble. There was warpage. As I expected, since it was in the part thickness, it was not visually apparent. But since the squared off part of the cowl is tall in this position, that warped enough that the eyestalk pivot would not fit through it. The front of the opening was barely wide enough, but the back was too small by about half a centimeter. So I made a shim to fit between the two top quarters and even up the opening to the proper width. This shim lines up with the groove and looks like it belongs there. A tiny little mohawk. I glued the ID tag in place. Having learned a little from the plunger incident, I test fit the eyestalk and cowl to the dome rather than just attaching the cowl first. Good thing. The pivot has to be trapped behind the cowl. I got some of those great looking socket head screws like I used on the gun box trim and attached the cowl with the pivot in place. The cowl does not fit flush to the dome. It's solid against the dome in the middle but there's a gap of < 1cm from the bottom to the join point. I don't care. It's done. The eyestalk pivot is not attached yet. I haven't quite decided how I'm going to do that. I'll get to that when I put the iris assembly in the eye. Here's an exploded view of the cowl (less shim) Here's the assembled cowl without the shim. Notice how the opening at the top narrows towards the back Here's the cowl with shim in place. The shim itself is wider towards the back of the cowl to even out the gap This picture illustrates the wood grain effect. A very happy surprise! Here's the painted cowl from several angles And here it is with Kai's ID tag in place Getting the cowl positioned before drilling Cowl bolted on Here you can see the gap between the dome and cowl Tag in place. Yes, Daleks can look straight up! Decided I better turn the shoulder section before something bad happens to the plunger The model files for the cowl can be found here. They include a full cowl in case you'd like to slice and dice it yourself.
  3. I've uploaded the cowl models. There's a complete cowl if you want to slice and dice it yourself. There's also a set of models for printing the cowl as four main parts plus pegs (and optional shim). here's the files.
  4. The inner arm tubes are each 30cm long, outer is half cm over standard at 40cm. As for the ring inner and outer diameters, it's a big list of numbers that probably won't work unless you have the exact same tubes and paint thicknesses. I've also found that different filaments can produce different results. I was about to say yeah sure about indenting the text but then got to thinking. Yes you can indent text into an object. But if you were to use a separate piece for the ident, you would have to also form fit the back of it to the cowl because it will be thicker and won't bend. My tag is only 1mm thick and I can bend it around the curve of the cowl. You would have to make the tag thick enough to cut letters into. You might be able to pull off a thin tag with really shallow letters but it would be difficult to paint. You could also carve the text directly into the cowl if you have that modeled. Working on that. Oh wait! Yeah you can! Simply make a rectangle 1mm thick and cut the letters right thru it! paint the cowl behind the tag so it shows thru the holes!
  5. Short entry today. The connector I was waiting for on the iris arrived. It is the wrong size. So I'm going out today trying to find a match. Might have to tear apart an old cd-rom drive (that's the original app for my motor). In the meantime I made an ID tag for Kai. I downloaded and installed the Dalek logo font and used it to extrude his name. I sprayed the tag black and then sprayed a board silver and pressed the tag against it. Now if I can just get a decent build on the cowl...
  6. Thanks. The rings are all 3d printed test pieces. I made lots of them in various inner/outer sizes to test fits.
  7. Today I finished the plunger arm. As I mentioned in my previous entry, I decided to use plastic tubes and found 3 out of 4 sizes I needed at an aquarium supply shop. The largest tube was purchased from Amazon. I gave each tube several spray coats of metallic silver, and then a clear coat. I couldn't find any official lengths for the two extension tubes so I decided to make them as long as made sense to me. I figured I would use spacer rings between the tubes and allow for 10cm of support for each tube. I'd put a limit spacer inside the end of the two bigger tubes and place a pair of spacers on each of the smaller tubes to slide inside the parent. I also decided to use the same 5/8" (19.5mm) inner support tube as I used for the eyestalk. I ran this tube thru the gazing globe and made a pair of spacers to mount the largest diameter tube on it. I made lots of test rings to get the dimensions perfect. Some rings had to fit snugly inside and slip outside, some had to slip inside and fit snug outside, some had to be snug both ways. I got a wooden dowel which fits inside the support tube and made an adapter to attach it to the plunger. Eventually I want to replace the dowel with an antenna motor so I made it easy to remove if necessary by using a very snug fit between the plunger and the smallest extension tube. I also put a stop on the end of the dowel to prevent shoving the plunger off the pipe. So I had everything figured out and assembled the arm. Perfect! Now to mount it in the shoulder section. Oh. Plunger doesn't fit thru the hole. I had to pop the stop off the end of the dowel and remove the dowel/plunger. I then installed the arm and once it was in replaced the plunger, dowel, and stopper ring.
  8. Today's build diary entry is entitled Eyestalking to You. There was a several years gap in my Dalek building. I'd pretty much done everything I could think of with plywood and I had no idea how I was going to do the dome and appendages. I found a solution for the dome which I've already completed and documented. But eyestalks, plungers, and gun parts had me stumped. Then 3D printers became available at a realistic price. So I got one and started cranking out Dalek parts and various Dalek building tools. BIG thanks and shout out to Cdngoose, Slythenperior, Elmoray, and Chrisosborne for the model files they uploaded and I used for gun, plunger, and eyestalk parts. So I printed the two piece eye and an eyestalk pivot. I decided to use plastic tubes for the eyestalk and plunger arm. The local aquarium supply store had tubes that were close enough for the middle and small plunger arm segments. They also had a 5/8" (19.5mm) tube but no 7/8" (22mm) tube. Since the part isn't visible I decided to go with the smaller size. The largest plunger tube I had to order from Amazon. I'll say more about the plunger next entry when it is done. I sprayed the 1" (25mm) tube with metallic silver, several coats, and a clear coat. Then I cut it into the various segments using a miter box and a hack saw. I made plastic rings that fit inside the tube and placed one on either side of the cut line so that the tube wouldn't compress while sawing. This really helped. I also printed a bunch of spacer rings to go between the inner tube and the outer and adapters for the pivot and eye. Then I looked at the original part description for the eyestalk disks. They looked like spherical caps to me. So I did the math based on the part dimensions to determine the size of each sphere, then used the Dalek part modified dimensions to model out the disks. I used translucent filament to make them but was not happy with the results. Even with only 10% infill they were mostly opaque. They looked more like a human iris than a Dalek eyestalk disk! So I got the idea to break each disk into two halves which could be assembled into a hollow disk. They were much better. Then I decided to try an acetone vapor bath to see how they would come out. I was very pleased with the results. I'll describe how to do the vapor bath in a little while. Here's pics of a solid disk, a hollow disk, a vapor treated hollow disk, a pair of finished disks viewed from opposite sides, and a comparison of both halves: Ah yes, the acetone vapor bath worked! It was also real easy to do. I found two metal baking pans, one steel and one aluminum. I used the aluminum for the bottom because of its heat transfer properties and the steel one as a cover. I made a wire frame that fits inside the pan. To do a treatment, I'd pour some acetone (quarter cup or so) in the aluminum pan, insert the frame, place a piece of foil over the frame, parts on top, cover with other pan. Then I set the whole thing in a pan of boiling water. Acetone vaporizes at 54C/134F. I used boiling water because this seemed by far the safest way to deal with acetone. Fumes are not good for you and are flammable. No active heat source eliminates any danger of combustian and doing it outside takes care of the fumes. I used an oven thermometer to track the water temperature. I did some experiments. Turns out that my chamber leaked lots of fumes so when I thought I was doing a 10 minute treatment it was more like 5 because all the vapor had escaped. Learned this the hard way with a large pour of acetone that lasted longer than 5 minutes. Also it is very important to keep the water temperature up there. If the temperature inside the chamber drops too far acetone will condense on the part and it will be ruined. I also tried vapor welding the two halves of the disk together. This did not go as planned. I didn't realize that any vapor inside the disk would condense and rot it from the inside out. Here's pics of the vapor chamber and what happens when you vape a hollow piece: Now I had this pile of parts: After gluing spacer rings in the little tube segments I had this pile of parts: A little more gluing and I had this eyestalk: Here's pics of a spool rack I had to cobble together because the filament reel I bought didn't fit my machine and a pic of the eyestalk before the disks were vapor treated: Oh, and were you wondering about that mystery piece: It's an iris insert. I'm waiting on a connector for the motor zif cable so I can figure out where that goes. I'll tell you all about it when it's done. Then I'll have this thing make it go: Here's the model file for the disks
  9. ...and then followed a bunch of private chatter between Jbourd and myself about modeling the cowl. He gave me the dome model and I did my best to carve off the dome and leave the cowl. The model was not exactly to spec as per the plans here but very close. I'm not sure the original scan is completely accurate or that it's a perfect scan of a real world "imprefect" Dalek. Also I am using the incredibly primitive 3D Builder which comes with Windows 10. It has issues with subtracting and splitting, which of course are the exact functions needed. I managed to isolate the cowl but it is way too big for my build plate. So I sliced it in half with holes on the join faces and pegs to fit. It still only fit in my build space at odd angles. I tried repeatedly on each orientation but always had too much warp. I think the halves might print flat but they don't fit like that. I also tried to slice into three parts: bottom and two top halves. 3D Builder wouldn't have any of that. So I have cowl models available but I can't print them. I don't want to upload them to the downloads section until there's a clean print I can verify. If there's someone who'd like to give it a go I'll package up the files for you. If you get a clean print (or two nudge-nudge-wink-wink) let me know and I'll put them on the site.
  10. Is anyone currently working on this sort of thing? I've been researching Raspberry Pi and want to give it a go. The Pi would run headless and operate all the Dalek automation. The wifi control interface would be on a touch screen and include camera feed. Here's what I'm thinking (without vid window): The Pi software will be called Dalek Audio Visual Robot Operating System (DAVROS) and the control program is Simulated Killer Automaton Remote Operation (SKARO). On power up the Dalek would initialize all the onboard devices. Each range of motion will have endpoint toggles and would be calibrated as part of the boot. This particular design is based on a shoulders section that can rotate 90 degrees left or right from the skirt and a dome that can rotate 90 degrees left/right from the shoulders. This effectively gives the Dalek 360 degrees of vision without twisting wires too much. The wheelchair base is controlled from both the joystick emulation or automatically with the rotation widget. If the operator chooses to ride inside the Dalek, the tablet would still be the only control necessary. This might make it a bit less cramped in there too! My eventual goal is to ride a Davros build while controlling the Dalek remotely with certain manual controls hidden, maybe work up some comedy routines. Any comments, tips, caveats, experiences, etc more than welcome.
  11. The fun never stops! Yesterday was dome light day. I decided to use incandescent bulbs over leds even with the increased power draw because they can dim. So I picked up a couple 1156 bulbs and sockets to match. I 3d printed bases to fit the sockets. Then I discovered I couldn't solder to the socket. So I took some paper clips and wrapped them around a socket (the bolt turning kind) into small springs that fit inside the printed base. I was able to solder to the coils and they provided a good contact for the socket. I glued the sockets into the bases making sure the electrical contact was ok. Mounted the bases to the dome and then mounted the Moflash assemblies over them. I connected the bulbs in parallel and ran about 2 meters of wire for eventual connection. Testing with a 9v battery proved that the lights work, but with the heavy draw from two bulbs and puny power source, all I could get was an "eggs..."
  12. Seriously impressed with your build, the wooden dome looks fantastic! 

     

    You ever come to visit TX stop by and teach me some stuff!

     

    :) 

    1. trickster

      trickster

      Thanks! I had my fill of Texas tho.

    2. jbourd

      jbourd

      Too bad lol!

      Keep up the good building

  13. Lots of progress on the dome. I 3d printed a set of tools to augment my Dremel router attachment to do the dome detailing. I had varying degrees of success with them as I'll describe below. The complete kit is available here. Here's a picture: OK so I belt sanded the dome down, filled gaps in the wood, sanded some more. Then I marked the dome at 45 degree angles using the dome protractor. I them used the dome ruler to extend the lines down to the chamfer line. I used the cone ruler to square off from the bottom of the cone and finish out the lines. I measured and marked the eyestalk port and indents using 3d printed templates. I measured out where the lights go and marked a center point and then used the drill guide to drill a hole for each light and another in the front at the same position. I'll get to that in a minute. . I attached the Dremel to the router base, and the router base to the my adapter. Then I used the arm attachments to cut the grooves. The latitude arm has a single hole that fits a peg at the dome center hole. It made a perfect cut. Unfortunately I forgot to take pictures of the steps involving the base adapter. I then used a peg in the front hole to guide the longitude arm and cut the front top groove. I used pegs in the light holes to guide the arm over the dome top from each side, then used a peg in the center hole to finish out the bottom of each groove. I did not like the results. The Dremel router bit is too narrow and the tool/my jigs did not produce perfect cuts. In fact I had several screw ups which I filled, sanded, and recut. So I took what we call a doob tube (container for prerolled joint at weed dispensary), wrapped it with sand paper, and sanded the grooves wider, straighter, and smoother. I used a flat bit to do the dome indents. The tabs broke off the base adapter a couple times but I was able to glue it back together. If I did it again I would add reinforcements to the tabs and probably more fill(I did everything at only 10%) Next I removed the base adapter and made a rail system for the router base using some 3d printed parts, a pair of 3/16" x 16" steel rods, and the frame from my belt sander rig. This worked well. Next I used a cutter bit and the base adapter to cut out the eyestalk port. This also gave me a look at the dome profile with the part cut out. Then primer, flat black, and metallic black
  14. Do you think that you could take the cowl off of a complete 3d dome and make it so the only thing that prints is the cowl?

     

    I have a full dome but obviously no where near a printer big enough!

     

    Cheers

    1. Show previous comments  26 more
    2. trickster

      trickster

      I think I've got the cowl nailed. Broke it into 4 pieces. Top two quarters successfully printed. Once I've got all 4 confirmed I'll upload.

    3. trickster

      trickster

      I did it. I have a 4 (7 including pegs) piece 3d printed cowl that looks great! All parts are small enough to print in any reasonable build space. I will package and upload in the next day or so.

    4. redman99

      redman99

      Awesome. I'll give it a try as soon as I get home. :)

  15. It was worth waiting 3 weeks for delivery from China. Twice. Eventually it will sound like gun noises instead of Lou Reed. Thanks to whoever edited the post to embed the video!