Fixing a toaster handle the hard way: a photo essay

December 27, 2009

Jeremy Elson, jelson at google mail

A few days ago, the handle of our toaster broke. I've been meaning to learn how to create 3D objects ever since a store called Metrix Create:Space opened right up the street from us; they have (among other cool toys) a 3D printer called a MakerBot. The broken handle was just the right opportunity for a first attempt at a new skill.

Below I present a brief photo-essay of the experience.

1) 7:30pm. The sad, forlorn handle of our toaster after the plastic handle broke off. Step 1 was to measure its dimensions with a cheap pair of Harbor Freight calipers.

2) 9:32pm. A 3D model of a replacement handle I drew in SolidWorks. It's pretty simple; just basically a rectangle with a slot cut in the back for the lever. I also added two channels in the slot that would match the protrusions of the toaster's handle. And, since I'm making my own custom 3D object anyway, I may as well customize it -- I put "L&J" on the front (for Liesl and Jeremy).

3) 10:58pm. I arrive at Metrix Create:Space, which is open until midnight 7 days a week, and wait about 30 minutes while the woman working the swing shift fusses with the MakerBot. The store has only been open for a few weeks and they don't quite have their procedures down yet. Plus, these machines are fussy, as their focus was more on keeping cost low rather than quality high.

4) 11:48pm. Printing in progress. The object was printed in 24 minutes. Metrix charges $0.50/minute so the handle cost $12 to print.

5) 12:23am. Back at home with the completed object!

Unfortunately it didn't fit on the toaster because I didn't leave enough clearance in the slot. It was so tight that the protrusions didn't make it into the slot - i.e., there was no way for them to reach their channels. The next evening, I printed one with a slightly larger slot. I also rounded the corners out and made the body a little smaller.

6) Success! Printing cost of the final version: $5.60. The only problem is that the "L&J" didn't come out very clearly. The very thin bottom strokes of the art deco font I used didn't render well on the bot whose minimum feature size is quite a bit larger than the laser cutters I'm used to.

Overall, a fun experience!