A unusual keyboard key switch | Hacker News Reader

A unusual keyboard key switch

twitter.com

132 pointsredshirt27 comments

This just reminded me of the most unusual switch I've ever heard of: Smith-Corona's acoustic switch from the 80s.

There's no per-switch sensor at all!

Instead, each key-press just results in one wide metal bar being hit. And to figure out which key was pressed, there's a sensor on the right and the left side which are used to calculate what key was pressed based on how long it took for the signal (i.e. vibration in the bar) to reach the left and right sensor.

It's so clever and so ridiculous at the same time!

[1] Video presentation (Talk about acoustic switches starts at 1:46:00): https://youtu.be/AvszDsr1js8?t=6361

[2] Slides: https://docs.google.com/presentation/d/1jfWpf8cABnH54yVjPIU3...

Once dealt with a 2d digitising pad that worked off the audio click of a spark in a similar way. Wild.

A flip-flop would work to convert the analog to 1-bit digital, but you could also tweak the magnetics and with a better ADC get velocity measurements! Add some software and you can type ANGRY MESSAGES by just pounding on your keyboard, no shift/cruise control needed!

Cute idea but I’m not sure of that. The sensing method is by multiplexed polling.

Realforce did this a few years ago with Topre switches (which are capacitive). Here's a video demonstration of someone using a keyboard like that as a piano [1].

[1] https://www.youtube.com/watch?v=mhCSIy-Qt0M

that's a cool idea!

Slightly more readable version: https://threadreaderapp.com/thread/1260688848104771586.html

Shocked Twitter has chosen to make their reading experience even worse. Now I get linked to a tweet thread, Twitter knows it's a tweet thread of one user but what does it choose to show?

Just the single tweet, a small "Show this thread" link and then 5 random other tweets from other people not connected or related to this tweet at all.

It's almost like they don't want you to spend time reading but "engaging"...

That's not what I get. I get the 13 tweets in the thread and then replies to those tweets. Are you logged in?

This glorious hunk of junk

https://oldcomputers.net/compucolor-8001.html

had a keyboard that was optical, it ran beams of light through masks attached to the keys, and looked at the beam interruptions as keys were pressed.

The keyboarding experience was absolutely miserable; you had to COMPLETELY release a key before pressing another. Ghosting and mangled keystrokes were almost impossible to avoid. I'd rather type on a card punch than one of those . . . well, no. But it's darned close.

The claim that it was patents that prevented it from seeing widespread use doesn’t seem legitimate, especially as the list of cons doesn’t mention the high production and component costs of such a mechanism. The reason why crappy dome switches won out despite their really long list of cons is extreme price superiority. The linked design is simply far too expensive and complicated by comparison.

dome switches are definitely cheaper, but i was comparing this design to mechanical switches. the magnetic switch has far fewer moving parts and the tolerances don't have to be very tight.

It looks a like a binary LVDT [1] sensor.

These are linear sensors that work by measuring the output of a transformer with a movable core. They have a reputation for being extremely reliable, even under harsh conditions. Therefore they are commonly used in aircraft, turbines, nuclear reactors, etc...

[1] https://en.wikipedia.org/wiki/Linear_variable_differential_t...

Very cool! Old keyboards and key-switches are an endless source of fun trivia.

I think Topre is the only one that manufactures keyboards with contactless key switches today. Topre key-switches measure the change in capacitance instead, and their keyboards (of the HHKB fame) sell for greater than $200.

https://www.keychatter.com/topre-switches/

You can also get Topre switches in Leopold keyboards, models ending with letter C (for capacitive).

Another unusual key switch is the opto-mechanical like the one used in https://wooting.io/wootingone (not affiliated).

They use infrared and they are also analog. You can bind different event based on the actuation force you put on the switch.

Interesting! I'm curious how this would sound on a built board compared to the other unique mechanical switches out there.

Sound is determined by the springs and mechanical action. This design could potentially be made silent. More interestingly, it could give you pressure-sensitive switches ...

You might want to take a look at this : https://wooting.io/wootingone (not affiliated)

it doesn't have any snap action. the force is set by a spring so it's just linear. the sound is fairly quiet.

So, how's the tactile feel on this switch?

Probably something like a U curve, with no explicit tactile bump, since the magnets would be attracted to the ferrite core, and when that force falls off, the spring force would pick back up.

I wonder if this will make it into a modern mechanical keyboard.

Hall effect switches are coming. Some manufacturers have already started. The cool part is that you can set the actuation point in software on them.

Hall effect switches have existed in keyboards for a long time [1]. I don't think there are many of them being produced nowadays though, but there's at least one new Chinese Hall effect keyboard that came out a few years ago [2].

Wooting also has a Hall effect switch [3]. You can't buy any keyboards with it yet but you can pre-order one at least [4].

[1] https://deskthority.net/wiki/Keystroke_sensing#Hall_effect

[2] https://geekhack.org/index.php?topic=84525.0

[3] https://blog.wooting.nl/introducing-the-lekker-switch/

[4] https://wooting.io/wooting_two_lekker_pre_order

Oh yes, I should have written "coming back". I have one from the 70s in my posession somewhere. I am more a clicky kind of guy (or at least proper tactile, not mx brown mush), but for linear switches I would love to be able to set the actuation point!

Reply on news.ycombinator.com