The Clickiest Keyboard Ever

No matter how clicky your keyboard is, nothing compares to the sensory experience of using a typewriter. The sounds that a typewriter makes, from the deep clunk of hitting the spacebar to the staccato of keys striking paper to the ratchety kerchunk of returning the carriage, are a delight compared to the sterile, soulless clicks of even the noisiest computer keyboard. Oh, and the bell — who doesn’t love the bell?

Unwilling to miss out on the feel of real typing, [Jatin Patel] whipped up this solenoid-powered typewriter simulator. The first version had the core functionality, with a line of six solenoids mounted to a strip of wood. The coils are connected to an Arduino through a relay board; a Python program running on his PC reads every keypress and tells the Arduino which solenoid to fire. Each one sounds different somehow, perhaps due to its position on the board, or maybe due to differences in mounting methods. Whatever the cause, the effect is a realistic variability in the sounds, just like a real typewriter.

Version two, shown in the video below, ups the simulation with a motor that moves the solenoid rack one step with each keypress, to simulate the moving carriage of a typewriter. The last solenoid rings a bell when it’s time to return the carriage, which is done with a combination wrench as a handle. Weird hex, but OK.

Can’t get enough typewriter action? We understand; check out this typewriter-cum-USB keyboard, the tweeting typewriter, or this manual typewriter that pulls some strings.

61 thoughts on “The Clickiest Keyboard Ever

  1. IBM used to have a block-mode terminal that had a built-in solenoid to whack the keyboard case with each keystroke. When the power was off, the keyboard was terrible, but with the power on it was the best typing experience I ever had (just like typing on a Selectric). 35 years later and I still miss that keyboard.

    1. That’s a curious effect, since the tactile feel of the key being pressed travels faster from your fingertip through your nerves than the sound goes through the air. The clicky sound is just superfluous to the point, and with the delay between the key being pressed and the solenoid acting, you would have felt the key pressed before you would have felt OR heard the solenoid click.

      On the other hand, people adapt to input lag to the point that removing the lag makes you feel like the computer is predicting what you’re about to press. Therefore, its highly likely that the thing was actually slowing your writing to wait until you hear/feel the solenoid click, for the satisfaction of the confirmation that something happened.

      1. Even if the haptic feedback from banging the keyboard case comes later by a few tens of milliseconds, considering the way the brain processes the real world, thats virtually simultaneous.

        1. It can be up to 200-300 ms. The brain locks on to the stronger stimulus.

          It’s common for music software to provide a virtual keyboard that maps to the letters on the PC USB keyboard. The latency for the PC keyboard is ideally about 8 ms, whereas a standard hardware MIDI keyboard goes down to about 0.3 ms per byte, and a MIDI USB interface can be set to a refresh time of 1 ms at the fastest. With the audio stack latency at somewhere around 10 ms for optimized drivers, you’ve got about 10-20 ms lag between you pressing a key and hearing the sound – assuming there’s no additional DSPs and buffers along the way.

          Adding 10 ms doesn’t sound much, but when you’re playing the virtual instrument, the PC keyboard makes you play slower than the MIDI keyboard because you’re instinctively waiting for the notes to come out. It’s harder to keep a melody going the longer the delay is, because you are playing a fraction of a note ahead of the tune.

          It’s the same effect as the “speech jamming gun” that projects your own voice back at you with a delay and makes it nearly impossible to talk.

          1. Of course sound travels at around a foot per millisecond, so how close you are to the monitors also affects your ability to play. It’s frankly a miracle that a band can play in sync with everyone hearing everyone else’s playing at different delays.

          2. i’ve made some softsynths and played some piano and so on and so long as the delay is less than about 20ms it doesn’t bother me. if it’s more than that though, hoo boy, hold onto your hat! it’s suddenly a nightmare.

            but i mean the comment at the top of this thread compared it to a selectric. the delay just needs to be comparable to the selectric. i’m not sure how it sequences hammer strokes but its ability to keep up with 150wpm typing suggests loosely the delay is probably less than 100ms. but either way, it’s part of the selectric experience and we all know and love it if we’ve enjoyed typing on that beast.

        2. What I meant, the brain can integrate “now” up to about a quarter of a second before it starts to consider things as separate events – but you have to get used to it. If you suddenly introduce a 250 ms delay, you’ll notice the gap.

      2. I read that as the solenoid becoming part of the tactile experience rather than acoustic, so rather than relying on the sound it would add a click to what your fingers feel.

        1. Sound travels a lot faster than nerve impulses. It is why a sniper bullet (Mach 2 to Mach 4) to the medulla oblongata can shut down a brain before it can even twitch a trigger finger.

          1. The touch signals from your fingertips travel at around 1/3 the speed of sound, so they need a 2 millisecond head start. Normally you’ll feel the key buckling down under your finger only just before the click of the switch reaches your ear, because the buckling of the spring happens first.

            A common solenoid takes about 30-50 ms to move, so it’s really late to the party.

    2. IBM used to make a typewriter called the Selectric. It was a really nice piece of equipment. They then added set of interposers to it and made it into an output printer. The kit for it was available back in the day and a friend of mine at the time got the kit and converted one. It was a marvel to watch and worked really well. It was also very useful that in his day job he worked for IBM and repaired these keyboards. As time went on the fell out of favour as most things do but it was really something to watch it print with different fonts just by changing the ball on it.

  2. I wonder, did Carpal Tunnel Syndrome exist back in the 1950’s-1960’s?
    Did people who developed CTS back then just “drop out” of the workforce, or did the longer travel of the keystroke with a manual typewriter have less of an impact (pun intended?) on the wrists?

    1. Back then, you learned to type from an instructor who would flunk you out if you didn’t have correct posture, and then you wouldn’t get a job. Self solving problem.

      1. I decided as a kid that I wanted to be a programmer. Computers weren’t a thing then but we knew they were coming.

        So I did a typing course in preparation. We learnt on mechanical unpowered typewriters. We did have some electric typewriters but they weren’t for learners.

        It was odd for me (as a boy) because all the other students were girls. I was questioned about my choice to take the course often as it was considered a female task.

        Anyway TLDR it was the best choice I could have made.

        I can code very rapidly. By removing the cognitive load of typing, I can far better focus on the code itself. Also having learnt on a typewriter that has no edit, delete or cursor keys (except tab, line feed, carriage return … That’s where these names came from) I get things write the first time and don’t loose my train of thought by having to edit a typo.

        1. Lol @ write the first time.

          Before anyone chimes in … I’m now “typing” on a smart phone complete with auto correct, a whole different story. Perhaps this exemplifies my point.

          1. It’s a whole different story for people who haven’t explicitly learned to type, but have learned to type by things like chat rooms. It becomes second speech to the point that you make the same type of errors, such as stuttering, when you’re not paying attention.

          2. I was actually wondering whether you were joking when I read ‘write’ and ‘loose’ instead of ‘right’ and ‘lose’ when you were talking about not making typos.

        2. ROB

          Same here…took a term of typing in Jr. High. Mechanical typewriters, no letters printed on the key caps (to force memorization), a braille-like dot on “F” and “J” on the home row so you could find your default finger placement by touch…that’s it.

          First week…asdf asdf asdf….then they added another letter, then another and another until I knew the whole keyboard by heart and could type with my eyes closed. Like you, I was one of a few males in that class. Nobody outside of class understood why I was wasting an elective on that.

          But I too was interested in computers and engineering, I saw what was coming. For the record, I believe I was the first person in my high school to turn in a term paper edited on a computer (my TRS-80 mod I) and printed on a printer (modified Decwriter LA36).

          Deciding to learn how to type was one of the better decisions I’ve made in my life.

      2. I don’t believe in “correct posture”. What is important is keyboard height and variation. “Correct posture” usually involves fixed-height desks that are at the wrong height for most users, and a complete lack of variation.

        I truly do not understand how RSI was not a bigger deal back then. Maybe maladies that afflict women weren’t considered to be problematic? Maybe people tended to do a lesser volume of typing? Probably a personal secretary only had to do so many letters per day to justify her worth?

        1. RSI is all about women’s finger nails and bad keyboard shapes. It wasn’t a problem in the pre-computer age when people learned to type and people who had to type as part of their living trimmed nails short. To understand RSi frequency, watch someone with long nails type on flat keyboards with fingers splayed out and up as far as possible. Plus most computer keyboards have abandoned sculpting like a typewriter or Apple II.

    2. Yeah, but it was mostly associated with telegraph operators. ‘Straight’ morse keys with the traditional up/down movement, can really mess with your wrist if you use it multiple hours per day.

      CTS or RSI are more things associated with mouse use, i suspect (i haven’t checked the literature about it). So back in the early days of the computer, you wouldn’t have much trouble with it.

    1. Exactly!

      “No matter how clicky your keyboard is, nothing compares to the sensory experience of using a typewriter. ”

      I really miss having my pinky fingertip bending in unnatural ways and getting stuck between keys and levers just because I made the sin of trying to type a capital T. How much fun we had meticulously manipulating the strike levers with one hand just to free a finger from the tenderly loving innards of the old Olivetti!

      The 10 degree bent in my right pinky and 5 on the left one are good mementos of those amazingly ergonomic devices called typewriters!

  3. I like that the carriage return is a wrench, now that’s hacky! I kind of wish that it actually did something use but I realize this is HAD and usefulness is surplus to requirements when it comes to cool hacks.

  4. Nice one, next step should be a custom printer driver which instantly prints each character, so you can get rid of that shiny monitor thing :-D (and the option of deleting)

    1. Yes and no. In a typewriter with hammers, the platen (paper) travels in the direction seen in the video. So the carriage would move in the correct direction.
      In teletypes and the ibm selectric typewriter, the platen stays in place and the hammers or ‘golf ball’ go in the opposite direction.

      This project is a mish mash of both technologies. It has a carriage return and next-line lever, so that means the paper/platen is moving, and it would be correct. It, however, also has the soleniods on the moving part, and in that case it would be incorrect.

      Long story short, this is a fun project and not attempting to be reflective of the reality of type writers. There is no wrong or right direction.

  5. Like the others above, I learned in high school on manual typewriters (late 1970’s). Met my first girlfriend in that class so double win! Touch typing has been so useful to me that I made sure all four of my kids learned to touch type. They started much younger than I did, though. I think the carriage is traveling in the correct direction. He could extend the hack by trying to simulate the line feed mechanism that was part of the carriage return. Just for completeness. ;-)

  6. It wasn’t the sound of the solenoid that made the keyboard so good, it was the tactile feedback. It actually allowed me to type faster, since there was no doubt about the keystroke being registered. Modern “mechanical” keyboards attempt to replicate that with a “click”, which does help.

  7. Touchtyping was the most useful class I took in high school. I have friends who claim to be able to “hunt and peck” as quickly (which is not true), but aside from the speed, touchtyping allows me to type without looking at the keyboard at all.

    I always find it humorous when computer reviews make a big deal about keyboard backlighting. If you’re looking at the keyboard, you’re doing it wrong.

      1. Also, just because you’re not looking doesn’t mean you don’t see the keyboard in your peripheral vision. Put a screen between yourself and the keyboard and watch your error rate go up.

          1. You wouldn’t make out the keycaps in your peripheral vision anyhow – what you see is where your fingers are relative to the board, which helps keeps your hands and fingers positioned correctly.

            The human proprioception isn’t absolute. There’s a common party trick to demonstrate the effect where you stretch your middle finger up and then let it sink down – it feels like your finger is sinking down through the table. The same effect happens with the keyboard – your finger positions start to drift with changes in hand pose, and seeing your hands over the keyboard helps to keep that in check.

          2. Luke, a touch typist doesn’t rely only on proprioception, that’s for sure, but feeling the keys is more than enough for recentring the hand position. Returning to home keys, feeling the bumps on f and j keys, and your thumbs on the space bar is more than enough to centre without seeing your hands at all.

  8. Luke,

    Nope. I don’t need to see the keyboard to type quickly and accurately. I have used keyboards with blank key tops, and sometimes type with my eyes closed while typing if I’m thinking/composing on the fly.

  9. When I was working at Marconi Instruments in the 80’s there was a big guy who used to PUNCH his code in on a DEC VT52. You could hear that thing right over the other side of the office – CLACK CLACK CLACK … CLACK CLACK CLACK CLACK CLACK …

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.