geekhack
geekhack Projects => Making Stuff Together! => Topic started by: Wingklip on Sat, 07 January 2017, 06:30:34
-
Before the dawn of the IBM buckling spring I read somewhere a patent that involved a single spring. That's it. It buckles and hits the wall of the barrel. No need for directing. Good for a very silent buckling spring, amirite?
Here's the link https://www.google.com/patents/US3699296
We all know the problem of using one spring for the bucking spring, and the complexity and adulteration that is when you use a follower hammer. But what if you thought outside the box, and used 2 springs of smaller diameter and height to bring the profile down to cherry compatible height? If I were to make the device capacitive as well, it would be easy, as cherry already has the capability of that - all we need is a contact pad in a coat of polymer and redesign the keyboard :)
-
I'm planning to integrate this into a cherry module btw through the tried and tested manufacture of made in China 3d printing.
I also have an idea, IBM had trouble making their springs buckle when they had less and less travel distance but
WHAT IF!
We had multiple smaller springs? :))))))))) Then the force on them swill equate to the large one.... Or WHAT IF!
OVULAR SPRINGS!!!11!!1!1111
Then one side will experience less force required to buckle than the other :0000
-
Wouldn't this require a metal spring pinging against a metal wall to connect the switch? Without much room for dampening materials that doesn't seem like it would be very quiet...
-
Alps made a buckling spring design (https://deskthority.net/wiki/Alps_buckling_spring) which uses a hammer and follower, which allows for a shorter spring.
In my experience, switches with longer coiled springs tend to have higher starting force than switches with shorter springs. That is something to consider. Personally, I prefer Cherry MX over Alps SKCL/SKCM primarily because Cherry MX has a higher starting force.
Metal-on-metal contacts need to be gold, or there will be corrosion over time. Gold-plated springs are obtainable, but I would be worried about wear and tear. Gold is actually a relatively soft metal.
IBM's buckling spring is only a spring and a hammer or tilting plate. That is incredibly simple already compared to other mechanical switches.
The click from a buckling spring, surprisingly, comes from the spring itself when it reaches the "catastrophic buckling" state - not from anything hitting anything. You can test this for yourself by pressing a spring from a IBM keyboard between your thumb and forefinger.
The foot at the end of a Model M's buckling spring does not clack against the membrane - there is a piece more akin to a wheel at the bottom that presses the membrane together. The flipper only stops it from rotating further.
You can reduce the ringing of a buckling spring switch by doing the floss mod (https://deskthority.net/wiki/Dental_floss_mod) - if the floss is just the right length, the ringing goes away completely.
Using a thick lubricant could also remove ringing. Too much thick lubricant will reduce both click sound and click-feel, however - you can't reduce the sound without losing the feel.
-
The spring itself? I would only assume that it would ring, and not necessarily click to any loud degree as the model F series do, which I have. That thing is designed to click using the slap of the flippers against naked PCB.
Using a single spring also means the highest reliability of any bucklespring switch because it only has 1 moving part. Also because it hits a metal wall, it would scrape the corrosion present on it without issue.
For sound though, maybe a single straight metal leaf would work better than a spring?
-
We all know the problem of using one spring for the bucking spring, and the complexity and adulteration that is when you use a follower hammer. But what if you thought outside the box, and used 2 springs of smaller diameter and height to bring the profile down to cherry compatible height? If I were to make the device capacitive as well, it would be easy, as cherry already has the capability of that - all we need is a contact pad in a coat of polymer and redesign the keyboard :)