A proposal for a new SSK

[M'er Forever]

[Account and posts deleted on request]

[Muirium]

Ping is the Model F's hallmark, it's true. Buckling spring was at its loudest right at the start; and yet the ancestor it was trying to emulate, beam spring, was nice and quiet in comparison.

I don't have an M2 of my own, but several people do around here so hopefully you'll get the answers you need.

Your plan to work with Unicomp is laudable, but they don't seem to be quite the most cooperative of keyboard manufacturers when it comes to the community. It might be wise to develop ideas that don't depend on them joining in the fun.

[Daemon Raccoon]

Lexmark took out a patent on an even lower profile buckling spring variant, but to my knowledge no prototypes were ever produced. Probably ditched the effort due to the complexity.

[Muirium]

A lateral spring?

They were up to something, all right.

[Findecanor]

Are the barrels shorter? If so, then the springs must be shorter (and perhaps smaller diameter) than the M's springs.

The barrels are shorter, as are the one-piece keys, but the springs and hammers are the same.
The big difference is that the keyboard has a flat backplane made of plastic, as opposed to a curved backplane made of steel.
I have been thinking of actually bolt-modding a M2.. with a metal backplate, but then the case wouldn't fit together.

BTW, the floss mod (done correctly) will remove the twang on a Model M or M2, leaving the click without any detriment to tactile feel.

[M'er Forever]

[Account and posts deleted on request]

[kyb]

Hmm, I ask myself why do people still add a Capslock key - even in brand new designs?

[M'er Forever]

[Account and posts deleted on request]

[Muirium]

The same can be said for the entire numpad!

Caps Lock makes a great source to remap for your layer key. It lives in a prime spot. My favourite place to put actual caps lock is in a macro: press both Shift keys to toggle. Stays right out of the way, then, and it's still there when you need to get all shouty about serial numbers etc.

[M'er Forever]

[Account and posts deleted on request]

[Halvar]

Very interesting, thanks Mu! Do you have a photo of the topside of that barrel frame? Are the barrels shorter? If so, then the springs must be shorter (and perhaps smaller diameter) than the M's springs. That might explain the difference in key feel and/or sound.

Quick & dirty pictures of the spring and a cap from Model M (left) vs. Model M2 (right).

While the key is higher on the original Model M, the functional part of it (stem) has pretty much the same size, as has the spring.

[M'er Forever]

[Account and posts deleted on request]

[Broadmonkey]

So.. can you use an M2 key cap on a Model M keyboard? or does the stem not fit?

[Halvar]

I'll try it again when I'm at home. I know that you can use a Model M key on an M2 without problems, but IIRC it doesn't work the other way around because the stem of the M2 cap is too short (or the Model M barrel too long).

[rodtang]

It is possible to have diodes in the membrane, you'd just have to license it from Microsoft. They use it in their sidewinder keyboards.

[M'er Forever]

[Account and posts deleted on request]

[rodtang]

It is a printed resistor, printed directly on the membrane.

download.microsoft.com/download/D/8/5/D85E2240-8688-4D49-BFD3-763E61A7D90E/AntiGhostingBrochure.pdf

I don't know the patent number, but I wouldn't be surprised if someone else here knew it.

EDIT: I think I found them:
https://www.google.com/patents/US20100066567
https://www.google.com/patents/US20100066572

[Daemon Raccoon]

A resistor is not a diode.

[rodtang]

A resistor is not a diode.

Yes, and I should pay more attention. Turns out I have no idea how the sidewinders work.

[Daniel Beardsmore]

A resistor is not a diode.

That's the whole point. You can't print diodes onto a membrane, but you can print resistors. I think essentially they're using bitwise maths with resistances to detect which combinations of keys are pressed. Pretty clever really, but that falls into my Senseless Over-engineering Over Nothing category, rather like the Oriental Tech OK-100M. A lot of research consumed for the sake of converting a lousy rubber dome into a lousy rubber dome.

[M'er Forever]

[Account and posts deleted on request]

[Kurk]

... You can't print diodes onto a membrane, but you can print resistors...

I think you can, at least goole gives a lot of hits. Mostly on printed OLEDs but also on ordinary ones. Of course printing diodes is more involved than depositing a film of pourly conducting material.

[JBert]

A lot of research consumed for the sake of converting a lousy rubber dome into a lousy rubber dome.

Ehr no, it's converted to a lousy rubber dome with NKRO. And you could actually use the technique to make a NKRO model M, if you can print your own membrane.

[TacticalCoder]

Hmm, I ask myself why do people still add a Capslock key - even in brand new designs?

Indeed. But not everyone does: for example on the amazing HHKB CTRL is where it was in the beginning of time (or was it?) and there's no physical key in the lower left corner of the keyboard (because it doesn't make much sense to go reach for a key there)...

Of course the HHKB doesn't have any numpad either. I'm not saying the HHKB is perfect but at least it got these two things (ctrl where it should be and no numpad) very right.

But anyway it's no big deal: you can simply remaps caps-lock to act as the modifier you fancy the most there and simply not use modifiers which are poorly located (for example the ones preventing you from touch-typing correctly).

I like very much that new SSK. I'd love it with a detachable cable and in plastic / industrial color

[M'er Forever]

[Account and posts deleted on request]

[Compgeke]

Black casing with white on black keycaps could be nice if you wanted to render that.

[Parak]

From my experience the 3d printed barrels are not of sufficient smoothness or precision. I've aimed to recreate the F switch design (as well as improve on it) for a while now, but the biggest barrier is getting the new parts injection molded, which requires quite precise measurements of existing barrels and flippies. In addition, the flippies need to be made out of hard and mildly conductive plastic in order for capsense to work properly. My ideal goal in this scenario is creating discrete F type barrels with captive flippies, similar to the Brother switch design, so that the assembly can be taken apart without the need to remove keycaps to avoid springs flying every whichway..

[wcass]

new model M flippies will not work for cap sense; they do not add capacitance the way that Model F flippies do. i believe that the F flippies are made of plastic with something mixed in with it to add capacitance (graphite maybe). We are also waiting on a capacitive controller (see first link below), but there are several very bright people working on that now that the Kishsaver has been found. Also a capacitive matrix (as IBM implemented it) is quirky in design (see third link). I have done some experimentation and design with both cap sensing and embedded diodes and can tell you that embedded diodes does work and is relatively inexpensive (see second link).

i would be happy to share whatever knowledge i have with you or anyone working to bring more buckling spring goodness to the world. i could use some 3D help making a case for my M-60 (second link).

design info on the capacitive switch PCB
http://deskthority.net/workshop-f7/brin ... t3047.html

M style switch but bottom membrane is a PCB with diodes. in this example, i am using 5x15 matrix, but i have also found 8 element diode arrays that would allow the more traditional 8xX matrix.
http://geekhack.org/index.php?topic=40111.0

capacitive switch matrix design info
http://geekhack.org/index.php?topic=51767.0

[M'er Forever]

[Account and posts deleted on request]

[M'er Forever]

[Account and posts deleted on request]