【IC】Stemshaver - a tool to fix BOX switches

类型: IC
作者: lutchbu
发布时间: 2018-10-31 18:48:53
更新时间: 2019-03-21 11:07:32
原链接: https://geekhack.org/index.php?topic=97974.0

---

Hi geekhack!
Two weeks ago I posted on reddit about a little piece of brass I machined to fix the old stemmed BOX switches I have. I also made a v2 out of steel.

Reddit post v1: https://www.reddit.com/r/MechanicalKeyboards/comments/9p2ay5/i_present_to_you_the_stemshaver/
Reddit post v2: https://www.reddit.com/r/MechanicalKeyboards/comments/9p97bd/stemshaver_v2_adventures_in_metal/
Imgur album: https://imgur.com/a/zgmJw6z
This got some attention and lots of people told me they’d be interested in buying one, so here we are…
IC Form: https://docs.google.com/forms/d/e/1FAIpQLSdqCEiw2psQEP_6m0DztjzyczzvGgwjO_JBK3t7tdOsohNagw/viewform
If there are any questions, feel free to get in touch but it’s the first time I’m doing something like this, so please be gentle.
frequently asked questions
Price
Material costs are low. I don’t know for sure but probably under 0.5 USD. The expensive part is my time and the machining. It took me about an hour to make the first one.
To explain my current manufacturing steps a bit:

• cutting stock to length

• milling the slots

• grinding the bevels and deburring

• fine tuning the distance between the cutting edges to 1.25mm by hand sanding the inside faces of the tool The new version does not require this anymore.
  Current estimate: 10-18 USD depending on order quantity

• Shipping*
  I’m in the EU and I plan to ship worldwide if we get that far. The tool is small, this should keep the shipping costs fairly low. Current estimate is 8-10 USD tracked and insured up to ~40 USD. No promises though.

• Durability*
  It turns out there are lots of you with several hundred switches they’d like to fix. This rises the reasonable question if the tool will hold up/stay sharp for that many switches, especially the brass version.
  I don’t think the sharpness will be a problem. Even if the shaver blunts it could easily be resharpened with some fine grit sandpaper and a flat surface. However, I am a bit concerned about the cutting edges bending apart. While the tool wouldn’t become unusable, it would become less effective, requiring multiple applications per stem or application at an angle, which in turn would require more care by the user. The new version should not have this problem.
  BUT, to be honest I don’t know and this is all speculation on my part.
  Testers report minimal to no wear after testing the stemshaver on up to 200 switches.

• *How to use it

• *Ideally, you push it on the stem in the right orientation, pull it off and you’re done. I would recommend testing the fit every few switches and adjusting the tool or your technique accordingly to produce the desired tightness.
  Short video of me using the tool:

• UPDATE 2*
  New version:
  
  In this version the sides of the x-axis are closed. This should prevent the prongs from deforming but is a bit more difficult to machine.
  I also made prototypesof the regular and deluxe version.
  Regular:
  
  Deluxe:
  
  more pictures here: https://imgur.com/a/wnLvcYB
  The next step is to send a few stemshavers out for testing.

• UPDATE 3*

• OVERVIEW*
  testresults, i’ve been takling to a cnc shop and decided to drop the artisan idea

• TESTRESULTS*

• design: Material will be steel and the stemshaver does not need a bevel. No bevel would cut costs by ~25% but means that there might be a slightly higher force required to operate

• function: Prototypes produced the desired results but had to be applied a second time rotated 180°. This is due to my milling setup and should not be an issue with the ones produced by the cnc shop.

• durability: Prototypes were tested on up to 200 switches with minimal to no wear observed

• PRICE*
  Current estimate is 10-18 USD for the regular version, depending on quantity.

• and now?*
  I’m not quite sure how to proceed from here to be honest. Ideally, I would want to send out a batch of prototypes made by the cnc shop to test them out. However I do not have the means to finance a small prototype run.
  So… any advice?

• Changelog:*
  2018-11-01: added FAQ
  2018-11-11: added Update 2, updated FAQ
  2019-03-21: added Update 3, updated FAQ