News

Jigsolve: Rotate + vacuum assembly

Here are the 3d printed blocks from the previous post and the rotation/nozzle assembly together on a car that will ride on the X axis of the corexy frame. Currently I’m building and testing the lifting mechanism. The lifter moves Z0-2cm.

Now anyone who’s ever done a jigsaw puzzle knows they’re not well cut pieces. Sometimes I lift a piece and the neighbor pieces stick to the one I want. The robot has to deal with this undesirable behavior. my lifter has a spring-loaded ring around and slightly below the nozzle. When a piece is being lifted the ring should be holding the neighbor pieces down. When the nozzle is placing a piece the ring should not interfere. When the head is moving across the table, with or without a jigsaw piece, the ring should not touch the pieces on the table.

Today I’m 3D printing and laser cutting the wood parts to assemble this version and do a live test.

News

Jigsolve: nozzle mounting

We’re waiting for parts to be delivered from our suppliers, and since the website is moving smoothly and tutorials are getting written I’ve taken a few days to prototype mounting systems for the pick and place nozzle. Above are some of the test prints.

I’ll probably leave it open-faced and zap-strap around the parts to hold them in place while still being accessible. The next step is to mount them on a system that can be raised and lowered. Then that system has to mount on the gantry and I can finally get back to lifting a jigsaw piece.

The biggest question right now is which way to point the web camera. I think straight down, but still able to see the nozzle tip is the best. I suspect the camera should NOT move up/down, to make it easier to aim the nozzle.

News

Jigsolve: New pnp nozzle arrived

The new nozzle arrived from RobotDigg. It’s two months late and never included the tracking number I asked for. Also I told them I wanted 150cm wire from the motor and I got 25. Also I told them my hose is 1/4″ ID. Did they include a male/male 1/4″ connector? No.

On the plus side the manufacturing of the part seems OK. @Aidan Leitch‘s soft nozzle seems to fit like a dream.

Every design problem trickles down from the nozzle. Tomorrow I hit the local hose & fitting supplier to get the missing adapter. Once nozzle tests pass, then i mount the camera above the nozzle and build a new z actuator. Then redesign the X carriage to hold the actuator. Then the Y axis. Then the cable management. With a little luck this thing will be working for the holidays.

These days… I find my patience is short and life moves WAY too slowly. I will pay for talented help. Marginally Clever is doing well and could be going so much faster.

News

Jigsolve: Nozzle work

The point of the robot is to let the internet finish a jigsaw puzzle. All problems trickle down from there. Starting at the frame was great fun, but maybe not the best place to begin. The nozzle head has gone through three redesigns and I suspect it will take three to five more before I get it right. It will affect the X carriage and gantry, which may affect the Y carriages, which may affect the corner mounts.

Anyways! Here’s the hollow shaft NEMA17 stepper motor I ordered.

Here’s the 3D printed adapter that will hold the vacuum hose on the back of the stepper.

I’ve sprayed the outside with white primer paint in the hope that it will seal microscopic holes in the print and maintain a good air-tight seal. I believe I need a thin layer of rubber between the plastic and the metal body of the stepper for a good seal. I tried a thin bead of hot glue, but testing shows the suction isn’t enough to lift a bottle cap.

Also the tonal pitch of the pump rises significantly when I put my finger over the pump inlet. It changes much less when I do the same at the silicone nozzle on the stepper, or on the stepper directly. I conclude there is an air gap in the 3D printed adapter and that this gap is on my hit list.

News

JigSolve: on loan for art project!

Steven, a member of the local hackspace, is building an art machine. Imagine a table with a thin bed of sand. a metal ball rolls through the sand, making spirograph patterns all day. The metal ball is moved by a magnet from beneath the table, giving it an almost magical appearance. If you’ve been to the San Mateo fair, you know what I’m talking about.

So with two weeks before the local maker fair, his mechanism to move the magnet crapped out and he called me for the design files to JigSolve. “Why not just use the whole machine?” I asked. So he did.

Apparently he’s gone the extra mile and redesigned most of the parts and then laser cut them from acrylic. I’m not sure if he’s building a complete second system or if he’s modding the original. Either way, I should soon have some video of JigSolve being used for something other than jigsaw puzzle solving, which makes me feel better.

It’s odd – My booth has gone sideways and my hot new thing has fallen apart, but I’ll have four or five other booths showing stuff in which I played a significant role. Does it help my maker business? Only time will tell.

Check back ~June 12, I should have video from the Vancouver Mini Maker Fair of the system in action.