Project Logs

Collapse

• Many more photos

  AKA • 10/31/2014 at 18:04 • 0 comments

  I just posted many photos of the production and prototyping process, link is here: https://plus.google.com/photos/107206617969726519022/albums/6070219415626136945?authkey=CMTj1q7kupZN

  I also delivered the final product (the non-silicone potted, 16-pixel one) to the new parents, who report it to be working well. (Phew!)

• Fail / curing update

  AKA • 10/27/2014 at 19:57 • 0 comments

  So it's been about 24 hours since I poured the Smooth-On "Dragon Skin" platinum cure silicone to pot the electronics for one of the milklichts...and I have a lesson to report:

  *Don't* use gaffer tape as a mold enclosure! The silicone at the very edges of the piece appears to be fouled and will not, I fear, ever cure. This has happened to me a couple of times in the past, but I've never had the wisdom to reflect on exactly which part of the process was at fault (I'm also guilty of habitually trying to use years-old silicones, because I begrudge throwing out unused raw materials).

  I was able to rescue the piece in a pretty ugly, but usable way: I got a very sharp hobby blade and carefully cut away the outer ~2mm of each side, exposing a fully-cured area underneath.

  The downside to the silicone-potting approach (even if the edges _had_ cured properly) is that it more or less guarantees that the user will pick at it. If I had milled a mold into which I cast the silicone, I'd avoid rough/imperfect interface between silicone and wood, but that would have required a lot more time and planning and a lot less beer...

• version 3 now curing

  AKA • 10/27/2014 at 02:02 • 0 comments

  Today I milled a third enclosure and mounted electronics inside it and made the second enclosure more ready for handoff. I am dictating this entry on mobile because my internet has been out but I just wanted to upload some more build photos (after encountering a 2mb upload limit, it seems I will have to post photos tomorrow from work)

  I potted one light in silicone, though as of the next morning the edges are still quite sticky. The silicone I used was open for several months, so there's a chance I've permanently ruined this one...

  The other light is somewhat inelegantly JB Welded in place so it sits flush with the wooden surface. I like this arrangement because, unlike the version potted in silicone, the master on/off switch is still accessible here (as are the 6 programming pins)

  I used another rare earth magnet in the light unit this time, rather than just using something ferrous (there's another magnet in the charging base). This way, you can affix the light to anything ferrous to hold it in place.

  Don't mind the weird tool marks on the edges! I was hurrying to get some prototypes to the new family, who had their baby a week ahead of the build schedule ;-)

  Here are both lights and the charger. I'll be milling another charging base sometime this week.

• State as of mid-October

  AKA • 10/15/2014 at 00:06 • 0 comments

  Right now, I have a v2 prototype kicking around for testing with the parents-to-be.

  The v2 version is the first to have a CNC milled enclosure, so I'll be revising the CNC designs in the next round. The software is more or less done, at least until I can hear some feedback from the actual users. If you look at the repo, you can see comments (and earlier commits) that show an idea about allowing the brightness of the light to be controlled - I tested that a bit in v1 but took it out for simplicity while I test it.

  The v1 version was a gross affair molded out of shapelock. More like a Louise Bourgeois nub than a prototype. Its main purpose was to let me measure how much coaxial alignment the coils need (the answer: a lot).

View all 4 project logs