keyboard shape experiment - Geek to Layman
-
- Location: Germany
- DT Pro Member: -
Hey everyone! I would appreciate infos or ideas to summarise some of the comprehensive know-how from this forum so that a layman such as myself could experiment with different keyboard switches and shapes and share feedback about them on this forum. Here's what I'm looking for:
1) A plug and play matrix for 60 to 80 keys with a USB on one end and plugs on the other.
Where can one purchase this?
Which type of plugs would be smart to use?
2) Key switches wired (say 20 cm wires) to plugs that attach to the matrix.
Where can these be purchased? Or which keys are good for wiring to plugs? maybe these? https://www.conrad.de/de/mikroschalter- ... 94406.html
3) Individual thin casings for each switch so that they can be pressed into a casing that holds them together. The bottom of the casing has an opening for the wires that lead to the plug.
4) A supercasing to hold the individual switch casings together and give your keyboard its shape. The design would fit to the design of the individual casings. A standard design would be thin enough that a the individual casing fits under a regular keycap. A list of casing design files for use with a 3D printer would be great!
5) Keycaps that fit to each design.
Purchase infos for standard keycaps.
CAD files or 3D print files (is that the same thing?) for individual keycaps.
If anyone is interested in CAD desiging some of the unusual shapes that I would like to try, please contact me. With all these components one could easily try different keyboard shapes by pressing the individual casings into a newly shaped casing and pluging the switches into the plugs on the matrix.
1) A plug and play matrix for 60 to 80 keys with a USB on one end and plugs on the other.
Where can one purchase this?
Which type of plugs would be smart to use?
2) Key switches wired (say 20 cm wires) to plugs that attach to the matrix.
Where can these be purchased? Or which keys are good for wiring to plugs? maybe these? https://www.conrad.de/de/mikroschalter- ... 94406.html
3) Individual thin casings for each switch so that they can be pressed into a casing that holds them together. The bottom of the casing has an opening for the wires that lead to the plug.
4) A supercasing to hold the individual switch casings together and give your keyboard its shape. The design would fit to the design of the individual casings. A standard design would be thin enough that a the individual casing fits under a regular keycap. A list of casing design files for use with a 3D printer would be great!
5) Keycaps that fit to each design.
Purchase infos for standard keycaps.
CAD files or 3D print files (is that the same thing?) for individual keycaps.
If anyone is interested in CAD desiging some of the unusual shapes that I would like to try, please contact me. With all these components one could easily try different keyboard shapes by pressing the individual casings into a newly shaped casing and pluging the switches into the plugs on the matrix.
Last edited by Ladybug on 20 Apr 2016, 11:30, edited 1 time in total.
- Scarpia
- Location: Sweden
- Main keyboard: F77 / Alps SKCM Brown TKL
- Main mouse: Logitech MX Anywhere 2
- Favorite switch: Capacitive BS, Alps SKCM Brown
- DT Pro Member: 0223
Sure, I'll indulge your idea. What you are proposing isn't impossible, but it would take a massive effort to achieve, since none of the individual components exist off-the-shelf, meaning you'll need to fabricate essentially everything you need. However, if it were me, I would:
A) Definitely get a 3D printer because you're going to go through a lot of iterations
B) Keyboard controller: To connect the keyboard to a computer, you need a controller board, and most people use the inexpensive Teensy for that. However, if you want something that is truly plug and play, that is, something you won't need to re-program each time you move keys around, you'll probably want to hook up each switch individually, forming a single row with N columns, where N is the number of keys you want.
This is going to require at least N+1 pins on the controller, and the Teensy simply doesn't have enough pins for a full keyboard if you wire it like that. So again, you're in for a bit of research and development, because the regular way (a rows*cols matrix with a fixed layout) doesn't really work for your purpose.
As for the 'plugs' you're talking about, any plug with two leads that comes in a male and a female should work just fine. Simply solder male connectors to the wires from the controller (positive side goes straight to VCC and the negative side goes to a key-specific pin on the controller), and maybe wrap a little label around each with the assigned key written on it.
C) Key switches: If what you are looking for is flexibility to experiment with ergonomics etc., I assume you will want to support as many different types of switches as possible, which basically means support for Cherry MX compatible as well as Alps compatible switches. The good news is in your case it's not a big problem, as you're 3D printing each keyboard and customization isn't an issue. Stabilised keys could be a big problem though.
From a cost/effort point of view, this is the easy part. Simply buy a bunch of switches from Cherry/Matias and you're good to go.
D) The matrix: Good news - since your 'matrix' only has 1 row, there is no need for diodes in your circuit. However, you will still need a way to connect your keyswitches to the connectors from your controller, and we've already settled on a male-female connector type, so you'll need to solder the female connectors on to the pins on the underside of every keyswitch. This is tedious, but easily doable.
Alternatively, you could design a single-switch PCB like the Amoeba and PCB-mount the female connectors to the PCB itself.
E) Mounting the switches: You could try to fashion a friction bracket that snaps firmly around those single-switch PCBs, but it could be pretty wobbly. Instead, you could 3D print a switch mounting plate for each keyboard shape and snap the switches in like a regular plate mounted keyboard. Having a 3D printer (and not worrying too much about longevity or achieving an ideal 'feel' of the keyboard) helps a lot here.
Finally, the assembly. In order to run an experiment on a new shape, you'll need to print a new case with all the mounting holes matching the layout you want. Then, you'll snap the switches onto the top plate, flip it around and attach the corresponding wires from your controller (with 60-80 keys this is going to be a MESS of wires, mind you). Flip it back around, and if you've connected the correct wires to the correct switches, you can plug a USB cable into your controller board and you're off.
Well, in theory. You'll also have spent at least a few thousand dollars (3D printer, PCB design+production runs, custom controller board, and the switches) before you ever get to this point, so there's that.
A) Definitely get a 3D printer because you're going to go through a lot of iterations
B) Keyboard controller: To connect the keyboard to a computer, you need a controller board, and most people use the inexpensive Teensy for that. However, if you want something that is truly plug and play, that is, something you won't need to re-program each time you move keys around, you'll probably want to hook up each switch individually, forming a single row with N columns, where N is the number of keys you want.
This is going to require at least N+1 pins on the controller, and the Teensy simply doesn't have enough pins for a full keyboard if you wire it like that. So again, you're in for a bit of research and development, because the regular way (a rows*cols matrix with a fixed layout) doesn't really work for your purpose.
As for the 'plugs' you're talking about, any plug with two leads that comes in a male and a female should work just fine. Simply solder male connectors to the wires from the controller (positive side goes straight to VCC and the negative side goes to a key-specific pin on the controller), and maybe wrap a little label around each with the assigned key written on it.
C) Key switches: If what you are looking for is flexibility to experiment with ergonomics etc., I assume you will want to support as many different types of switches as possible, which basically means support for Cherry MX compatible as well as Alps compatible switches. The good news is in your case it's not a big problem, as you're 3D printing each keyboard and customization isn't an issue. Stabilised keys could be a big problem though.
From a cost/effort point of view, this is the easy part. Simply buy a bunch of switches from Cherry/Matias and you're good to go.
D) The matrix: Good news - since your 'matrix' only has 1 row, there is no need for diodes in your circuit. However, you will still need a way to connect your keyswitches to the connectors from your controller, and we've already settled on a male-female connector type, so you'll need to solder the female connectors on to the pins on the underside of every keyswitch. This is tedious, but easily doable.
Alternatively, you could design a single-switch PCB like the Amoeba and PCB-mount the female connectors to the PCB itself.
E) Mounting the switches: You could try to fashion a friction bracket that snaps firmly around those single-switch PCBs, but it could be pretty wobbly. Instead, you could 3D print a switch mounting plate for each keyboard shape and snap the switches in like a regular plate mounted keyboard. Having a 3D printer (and not worrying too much about longevity or achieving an ideal 'feel' of the keyboard) helps a lot here.
Finally, the assembly. In order to run an experiment on a new shape, you'll need to print a new case with all the mounting holes matching the layout you want. Then, you'll snap the switches onto the top plate, flip it around and attach the corresponding wires from your controller (with 60-80 keys this is going to be a MESS of wires, mind you). Flip it back around, and if you've connected the correct wires to the correct switches, you can plug a USB cable into your controller board and you're off.
Well, in theory. You'll also have spent at least a few thousand dollars (3D printer, PCB design+production runs, custom controller board, and the switches) before you ever get to this point, so there's that.
- vvp
- Main keyboard: Katy/K84CS
- Main mouse: symetric 5-buttons + wheel
- Favorite switch: Cherry MX
- DT Pro Member: -
Or you he can add e.g. 14 8-bit parallel-in serial-out shift registers with bushed resistor network (0.3+0.15 cents for one pair) to support up to 112 switches. And use maybe even teensy 2.0 as a controller.Scarpia wrote: ↑ B) Keyboard controller: To connect the keyboard to a computer, you need a controller board, and most people use the inexpensive Teensy for that. However, if you want something that is truly plug and play, that is, something you won't need to re-program each time you move keys around, you'll probably want to hook up each switch individually, forming a single row with N columns, where N is the number of keys you want.
But there is no reason not to do a regular e.g. 11x10 matrix on the PCB with diodes (on the pcb too) but instead of each switch the PCB will contain a connector to which the switch cable can plug in.
- Plasmodium
- Location: UK
- Main keyboard: QPAD MK80
- Main mouse: Logitech LX3
- Favorite switch: Cherry brown
- DT Pro Member: -
PCB is printed circuit board.
So, I take it what you are trying to do is make a keyboard with an easily swappable layout? Do you want it to look like a grid? If so, you could start by looking at Tipros - they were originally 'point of sale' keyboards, but the layout can be customised to whatever you want: you can 'blank' off keys, swap the caps, change the function of each key, use larger keycaps and so on. What you can't do is to change the type of switch (clicky, tactile, linear) without de-soldering the old one and re-soldering in a new one.
Notice how some keys are 2 units long, and others are blank. The keyboard comes with software that allows you to remap each to to any function.
[Edit after jacobulus linked to Ladybug's original comment] I see now that you are interested in more obscure/ergonomic physical layouts, so the Tipro isn't what you're looking for, I guess.
If you're looking for greater flexibility than a grid (ie, you might want keys to be 'staggered', or at different angles), that could be more difficult.
RE: keycap purchasing info, you can check out http://pimpmykeyboard.com/blank-key-packs/ for blank keycaps. They have some other printed sets as well. If you are buying in bulk or need custom printing, you might be better off contacting them for a quote.
So, I take it what you are trying to do is make a keyboard with an easily swappable layout? Do you want it to look like a grid? If so, you could start by looking at Tipros - they were originally 'point of sale' keyboards, but the layout can be customised to whatever you want: you can 'blank' off keys, swap the caps, change the function of each key, use larger keycaps and so on. What you can't do is to change the type of switch (clicky, tactile, linear) without de-soldering the old one and re-soldering in a new one.
Notice how some keys are 2 units long, and others are blank. The keyboard comes with software that allows you to remap each to to any function.
[Edit after jacobulus linked to Ladybug's original comment] I see now that you are interested in more obscure/ergonomic physical layouts, so the Tipro isn't what you're looking for, I guess.
If you're looking for greater flexibility than a grid (ie, you might want keys to be 'staggered', or at different angles), that could be more difficult.
RE: keycap purchasing info, you can check out http://pimpmykeyboard.com/blank-key-packs/ for blank keycaps. They have some other printed sets as well. If you are buying in bulk or need custom printing, you might be better off contacting them for a quote.
Last edited by Plasmodium on 20 Apr 2016, 13:02, edited 1 time in total.
-
- Location: geekhack ergonomics subforum
- Favorite switch: Alps plate spring; clicky SMK
- DT Pro Member: -
Folks reading this should also see Ladybug’s original comment, post299513.html#p299513
Ladybug wrote: ↑Hey everyone. I'm a doctor interested in using my anatomical knowledge towards reducing impingement and RSI due to keyboard and mouse usage. I'm new to this subject so I will express myself simply. Does someone know of a ready made kit for experimenting with keyboard shapes and layouts?
Here's what I am looking for or propose as an idea for someone to build at least semi en mass as I believe it would be interesting to enough people for it to be produced:
A device with around 60 keyswitches which are each connected to the matrix (or whatever you call that complicated board with all the crossovers that is connected to a controller ready to use, meaning the user could choose the function of each key out of a template a program that comes with the device. Each key would have a number which s shown in the program.
The key switches would each individually have 20-30 cm wire connection to the board so they can be placed at will. Maybe offer some tubing that is open on one side to hold them together with once in place.
The key switches would have individual encasing so that they can be (hot) glued onto a given surface or pressed into a provided casing mold (I'll call them molds) as above in the deskthority picture. Several useful shapes of mold like flat or bowlshaped could be offered to press the individual encasings into e.g. in in 5 rows and 6 columns (or even just three).
I believe keycaps with individual curves and height can be purchased already, but if someone is able to do this, why not make a website you can go through and choose your individual keyboard including the shape. All the user would have to do is press the keys into place and choose a layout. If the shape is uncomfortable a new one can be tested using the same device as the individual key cases can be pressed back out and put into a new mold. There could be key cap sets to go with the standard curved molds.
So you would go through the site and choose the number of keys (say 20 to 80) that are attached to the controller, the type of key switches, the mold shape (say curved, flat, with or without thumb buttons, that can be easily glued or stuck together like lego), the amount of columns and rows for each mold, and keycaps to complement the chosen mold shape (casings). Stick everything together at home and choose the layout, even leave out some keys if you don't end up needing them. Try out your custom keyboard and If you don't like it purchase a different mold and fitting keycaps if necessary and recycle the device into a new shape.
In short what is new would be that the keyswitches can be places at will, around 20-30 cm or more from the matrix board. -- PLus The individual keycap casing with fitting holes in prefabricated casings of different shapes.
And a user friendly program that allows someone without programming skills to assign keys or functions to the keyswitches.
If such a set, or parts of it exist please write here or at offers@wolke7.net
-
- Location: geekhack ergonomics subforum
- Favorite switch: Alps plate spring; clicky SMK
- DT Pro Member: -
Ladybug: how much have you tried existing “ergonomic” keyboards? You might be able to learn some about what is comfortable to type on by trying various prior ideas.
I would recommend trying to borrow or buy a Maltron, and possibly an Esrille (since old Japanese TRON project and NEC M-Type keyboards are effectively unobtainable in the West). You might also be interested in the keyboard.io project, a recent Kickstarter which is still taking pre-orders and should hopefully ship sometime in the next year.
There were also a number of otherwise-standard-layout-but-split keyboards made in the 90s, such as the Apple Adjustable, Cherry G80-5000, IBM model M15, Omnikey Evolution, Chicony KB-7000, etc. Today, the best comparable keyboard is the Matias ErgoPro.
There were a number of interesting designs made at various points which are now defunct. The one I find most inspiring is the DataStealth, but see this Geekhack topic for discussion of several old keyboards https://geekhack.org/index.php?topic=55099.0
I would recommend trying to borrow or buy a Maltron, and possibly an Esrille (since old Japanese TRON project and NEC M-Type keyboards are effectively unobtainable in the West). You might also be interested in the keyboard.io project, a recent Kickstarter which is still taking pre-orders and should hopefully ship sometime in the next year.
There were also a number of otherwise-standard-layout-but-split keyboards made in the 90s, such as the Apple Adjustable, Cherry G80-5000, IBM model M15, Omnikey Evolution, Chicony KB-7000, etc. Today, the best comparable keyboard is the Matias ErgoPro.
There were a number of interesting designs made at various points which are now defunct. The one I find most inspiring is the DataStealth, but see this Geekhack topic for discussion of several old keyboards https://geekhack.org/index.php?topic=55099.0
Last edited by jacobolus on 20 Apr 2016, 13:19, edited 1 time in total.
-
- Location: geekhack ergonomics subforum
- Favorite switch: Alps plate spring; clicky SMK
- DT Pro Member: -
If you decide you want to make a custom shape, I recommend not trying to build something that supports on-the-fly reconfiguration. That would be much more complicated and expensive to build than the alternative.
Instead, find a local place (e.g. some kind of maker space) where you can use a laser cutter, and prototype your keyboard switchplates out of laser-cut acrylic (this takes a lot less skill than CAD modeling and operating your own 3D printer, though that would also be an option). For your initial prototypes, I recommend not even bothering to wire them together, but just iterating on the physical design a few times until you have an idea you like. I recommend buying 100–200 keyboard keyswitches. The easiest to find are Cherry MX and Matias switches. I recommend using standard keycaps for the moment. Cheap 3D printing technology is not sufficient for producing keycaps, so if you want custom shapes, you’ll need to pay quite a bit to have them produced on a high-end machine.
Once you have a rough prototype you like, you can hand wire the switches into a matrix on the back. You’ll need a soldering iron, a pile of diodes, some wire, and some little $5–15 microcontroller such as an Arduino Nano or Teensy. You’ll need to put some kind of custom firmware on the microcontroller. Fortunately, there are community-developed open source firmware options, so it’s something you can learn to do without too much trouble, even as a non-programmer.
To learn how to do all of these things, refer back to the other thread you posted in, post141635.html#p141635 and also check out a similar thread at geekhack, https://geekhack.org/index.php?topic=69582
Instead, find a local place (e.g. some kind of maker space) where you can use a laser cutter, and prototype your keyboard switchplates out of laser-cut acrylic (this takes a lot less skill than CAD modeling and operating your own 3D printer, though that would also be an option). For your initial prototypes, I recommend not even bothering to wire them together, but just iterating on the physical design a few times until you have an idea you like. I recommend buying 100–200 keyboard keyswitches. The easiest to find are Cherry MX and Matias switches. I recommend using standard keycaps for the moment. Cheap 3D printing technology is not sufficient for producing keycaps, so if you want custom shapes, you’ll need to pay quite a bit to have them produced on a high-end machine.
Once you have a rough prototype you like, you can hand wire the switches into a matrix on the back. You’ll need a soldering iron, a pile of diodes, some wire, and some little $5–15 microcontroller such as an Arduino Nano or Teensy. You’ll need to put some kind of custom firmware on the microcontroller. Fortunately, there are community-developed open source firmware options, so it’s something you can learn to do without too much trouble, even as a non-programmer.
To learn how to do all of these things, refer back to the other thread you posted in, post141635.html#p141635 and also check out a similar thread at geekhack, https://geekhack.org/index.php?topic=69582
- vvp
- Main keyboard: Katy/K84CS
- Main mouse: symetric 5-buttons + wheel
- Favorite switch: Cherry MX
- DT Pro Member: -
Man, I think you should include also Kinesis Advantage which is a lot cheaper than Maltron. But probably not better. But it is also cheaper than Estrille and even cheaper than the keyboard.io kickstarter. And probably better than those. At least you get a contoured keyboard which saves a lot of finger travelling:jacobolus wrote: ↑ I would recommend trying to borrow or buy a Maltron, and possibly an Esrille (since old Japanese TRON project and NEC M-Type keyboards are effectively unobtainable in the West). You might also be interested in the keyboard.io project, a recent Kickstarter which is still taking pre-orders and should hopefully ship sometime in the next year.
http://www.kinesis-ergo.com/shop/advantage-for-pc-mac/
-
- Location: Germany
- DT Pro Member: -
I want to design something that puts together the advantages of all of those plus solving the problem of strain due to retroflection in the wrist, and reducing impingement in the shoulder. I would like to try an extreme variation: a supination keyboard which allows you to type palms up. Try meditating with your palms upwards and see what an effect that has on the mind and posture. Or even just laying your hands on your desk palm up when you're not typing. It has an effect on over all body postion and therefore also the psyche.
I meant breathing meditation, where you focus your attention or what your breathing is doing without your purposeful effort, whether you are breathing in or out without your intention or if your breathing is momentarily paused without you holding your breath. In this sensitive state it is easier to notice the effect your body's position has on your level of happiness.
A supination keyboard sounds like a freak idea, but I think it would help at least as a temporary measure against shoulder impingement.
I meant breathing meditation, where you focus your attention or what your breathing is doing without your purposeful effort, whether you are breathing in or out without your intention or if your breathing is momentarily paused without you holding your breath. In this sensitive state it is easier to notice the effect your body's position has on your level of happiness.
A supination keyboard sounds like a freak idea, but I think it would help at least as a temporary measure against shoulder impingement.
-
- Location: Stockholm, Sweden
- DT Pro Member: 0011
That's interesting. Very few keyboards slant more than 45°. I know of two that are vertical (90°) but both have staggered QWERTY layout.
I think that when you can't see the keyboard at all then being guided by feel to the right keys is more important - so I think that something with keys in contoured columns like the Maltron and Kinesis might be the way to go.
The "Safetype" (as seen on TV on "Parks and recreation") has side mirrors like a car so that the user could see the keys
while typing.
The "Ergomaster" was mounted to the edge of the desk with the mounts adjustable from 0° to 90° but with no wrist rests.
You could of course 3D-print frames that hold key-switches. A couple of people here has done that before.
Whether you go for 3D-printing or laser-cut acrylic do make sure that you have enough thickness in the plastic so that it is sturdy enough. You could glue two layers of thin acrylic together - one layer for snapping switches and one layer for extra strength.
I think that when you can't see the keyboard at all then being guided by feel to the right keys is more important - so I think that something with keys in contoured columns like the Maltron and Kinesis might be the way to go.
The "Safetype" (as seen on TV on "Parks and recreation") has side mirrors like a car so that the user could see the keys
while typing.
The "Ergomaster" was mounted to the edge of the desk with the mounts adjustable from 0° to 90° but with no wrist rests.
You could of course 3D-print frames that hold key-switches. A couple of people here has done that before.
Whether you go for 3D-printing or laser-cut acrylic do make sure that you have enough thickness in the plastic so that it is sturdy enough. You could glue two layers of thin acrylic together - one layer for snapping switches and one layer for extra strength.
-
- Location: Germany
- DT Pro Member: -
mirrors? way too confusing. What I would love to see but have no idea how to create would be a shell type curvature that is transparent except that the letter you have configured for a certain area of the shell is shown there. It would essentially be a transparent touchpad that is also able to show which letter is currently configured on that spot
The convex side of the shell points towards you and is suspended like a lamp. I'm not smoking anything
But until that is available, where can I order keyswitches online? I live in Germany but the Cherry site is confusing for this.
The convex side of the shell points towards you and is suspended like a lamp. I'm not smoking anything
But until that is available, where can I order keyswitches online? I live in Germany but the Cherry site is confusing for this.
-
- Location: geekhack ergonomics subforum
- Favorite switch: Alps plate spring; clicky SMK
- DT Pro Member: -
By “retroflection” (retroflexion?) do you mean the wrists bent upward? (If so, the typical term in keyboard ergonomics literature is “extension”.) Wrist extension during typing can be avoided on any keyboard by repositioning it relative to the body, and is mostly caused by poor technique among untrained typists.Ladybug wrote: ↑I want to design something that puts together the advantages of all of those plus solving the problem of strain due to retroflection in the wrist, and reducing impingement in the shoulder.
The biggest issue with standard one-piece keyboards is the severe wrist pronation they force on the typist. Sometimes to reduce wrist pronation, typists cope by pushing the keyboard away from the body and rotating the elbows outward and forward. Because this puts a lot of static load on the shoulders and upper arms, they then compensate by resting their elbows, forearms, or palms on chair armrests, the table, or a palmrest. Sometimes this involves uncomfortable wrist extension.
I think poor shoulder posture when typing has as much to do with the chair and desk as the keyboard per se. Even using a standard keyboard a typist can keep the shoulders in a neutral position without any static strain. It takes sitting up straight, pulling the keyboard close enough to the torso, and tilting the top surface of the keyboard to be parallel to the forearms.
Having a split keyboard with sufficient separation between halves and any amount of tenting helps a lot with this, because it reduces the need to substitute elbow movement for twisting in the forearm/wrist.
Well, give it a shot!Ladybug wrote: ↑I would like to try an extreme variation: a supination keyboard which allows you to type palms up.
My own preference is about a 30–50° "tent" on each half: steep enough that the wrist isn’t excessively pronated, but still getting a bit of gravity assistance, and using the keyboard itself as a surface on which to rest the fingers and letting the big muscles in my upper arms absorb any shock from typing. A keyboard with the keys on the bottom for typing with palms up seems like it would either require holding the hands suspended above nothing, or else statically resting on the table, making it hard to reach more than a few keys and increasing strain on the fingers.
Can you elaborate? I don’t find the direction of my palms while sitting to have any particular effect on my mind or on my body’s posture.Try meditating with your palms upwards and see what an effect that has on the mind and posture.
* * *
As a general matter, if you could draw a picture, or better yet build a little 3-dimensional model (using plasticine, taped together pieces of cardboard, bent coat hangers, whatever ...) it might be a bit easier to communicate your idea.
Last edited by jacobolus on 22 Apr 2016, 00:17, edited 1 time in total.
- vvp
- Main keyboard: Katy/K84CS
- Main mouse: symetric 5-buttons + wheel
- Favorite switch: Cherry MX
- DT Pro Member: -
Cherry will ship only huge orders with delivery time of about a year.Ladybug wrote: ↑ But until that is available, where can I order keyswitches online? I live in Germany but the Cherry site is confusing for this.
You can get Cherry and Alps switches here on deskthority from 7bit. You can buy also some keycaps from 7bit. Check the market place forum. Order only the stuff he has in stock. Otherwise you will wait way too long.
Gaterons can be bought almost always from amazon.com or probably also ebay. Massdropp also does group buys for them.
Mouser has Cherry switches too but way too expensive (about 1 EUR per switch). Cherry switches can be bought for about 0.65 EUR a piece. Gaterons and Alps can be bought for about 0.3 EUR a piece; maybe even less if you search/wait a lot. That is for small quantities of about 100 pieces.
-
- Location: Stockholm, Sweden
- DT Pro Member: 0011
Now you made me remember that there are already several "keyboards" made from somewhat similar concepts.Ladybug wrote: ↑What I would love to see but have no idea how to create would be a shell type curvature that is transparent except that the letter you have configured for a certain area of the shell is shown there. It would essentially be a transparent touchpad that is also able to show which letter is currently configured on that spot
There is a category of handheld keyboards called "back-typing" keyboards, mostly intended for use with mobile devices.
Those are held with both hands like a gamepad and typed with the fingers on the bottom (back) of the device.
Several of these have been on Kickstarter but failed ...
(I didn't remember their names, so I had to google them:)
* The Grippity - is a transparent touch-keyboard like your idea, but flat.
* The TREWGrip has real keys so it is opaque but it has legends on the top mirroring the keys on the back. Each of these legends also light up when you press the corresponding key. (TREWgrip on Youtube)
* Research project on Microsoft: RearType - a prototype made simply from a tablet with movable keys held upside-down.
* AlphaUI - at the back of a smartphone.
Best way to find out how they work is probably to search for images on tech-reporting sites and for videos on Youtube.
Edit: BTW. After compiling this list, I also added the info to the Wiki.
-
- Location: Germany
- DT Pro Member: -
Spello, txs. I didn't mean reflecting on all the good times in the past. But that's nice tooBy “retroflection” (retroflexion?) do you mean the wrists bent upward? (If so, the typical term in keyboard ergonomics literature is “extension”.) Wrist extension during typing can be avoided on any keyboard by repositioning it relative to the body, and is mostly caused by poor technique among untrained typists.
I think ideally the keyboard would be tilted slightly away from the typist and and slightly inwards to avoid wrist extension. (As if you had your hands on your girlfriend's hips from behind ) Basically this idea incorporated into a split DataStealth. The photo above shows that the typist has wrist extension and his shoulders will be rotated slightly inwards which works towards shoulder impingement.I think poor shoulder posture when typing has as much to do with the chair and desk as the keyboard per se. Even using a standard keyboard a typist can keep the shoulders in a neutral position without any static strain. It takes sitting up straight, pulling the keyboard close enough to the torso, and tilting the top surface of the keyboard to be parallel to the forearms.
The difference is slight, but pronation of the hand rotates the radiohumeral joint inwards which turns rotates the humerus inwards which causes pressure on nerves in the area between the humerus and the rib cage on the front side of the shoulder. You won't notice much if you don't have shoulder impingment already. While walking in a natural postition the forearm is in a position that would have the thumb side up on a flat table. If you stand comfortably in a natural position your thumb points forward, your hands seem slightly flexed in the wrist (I would say slightly limp if you are relaxed) and your shoulders aren't turned inward. If you keep everything as is and only bend the elbow as much as necessary ( say 80° for a keyboard slightly lower than the elbow) you have the position that the keyboard should be made to fit to. If the keys are placed so that the arms do the work of repositioning the fingers when moving away from home keys you have some movement in the shoulder as well, to avoid muscle cramps. Also you reduce strain to the muscles in the forearms that extend the hand and fingers. I think that it would generally be best give preference to wrist flexion over extension when placing keys as our hand flexors are much stronger than their extensors. Working like this mimics a digging motion or swimming, or even swinging from trees, keeping us in touch with how we evolved ( or consider comparative anatomy and a man's best friend).Can you elaborate? I don’t find the direction of my palms while sitting to have any particular effect on my mind or on my body’s posture.Try meditating with your palms upwards and see what an effect that has on the mind and posture.
Looking forward to my keyboard however long it takes to make...
-
- Location: Germany
- DT Pro Member: -
https://www.youtube.com/watch?v=2hSbmM8GdR0* The TREWGrip has real keys so it is opaque but it has legends on the top mirroring the keys on the back. Each of these legends also light up when you press the corresponding key. (TREWgrip on Youtube)
http://www.trewgrip.com/
nice! I want one of those split, and with braces for the arms on my desk chair and one for the top of my desk. I just wrote them
I know it is a different discussion, but since you knew about this, do you also know of a mouse that can be used with the palm facing upwards?
-
- Location: geekhack ergonomics subforum
- Favorite switch: Alps plate spring; clicky SMK
- DT Pro Member: -
When I stand or walk in a completely relaxed, neutral position, my palms don’t directly face towards each-other, but are rotated about 30–40° back. If I start sprinting they rotate forward a bit. No noticeable wrist flexion.Ladybug wrote: ↑While walking in a natural postition the forearm is in a position that would have the thumb side up on a flat table. If you stand comfortably in a natural position your thumb points forward, your hands seem slightly flexed in the wrist (I would say slightly limp if you are relaxed) and your shoulders aren't turned inward. If you keep everything as is and only bend the elbow as much as necessary ( say 80° for a keyboard slightly lower than the elbow) you have the position that the keyboard should be made to fit to.
Bending the elbow up 80° or similar sounds fine. (Whatever amount you bend the elbow, you should tilt the keyboard so that its top is parallel to your forearms.) That is going to require a taller chair and/or lower desk than the vast majority of existing office furniture. If sitting on a regular chair with a flat lap, it would probably be best to just set the keyboard on the lap, if you want elbows with an interior angle of ~100°.
In practice, there will be a mix of subtle arm motions and finger extension, mostly the latter. Your finger extensors don’t need to be too strong to reposition the fingers on different keys.If the keys are placed so that the arms do the work of repositioning the fingers when moving away from home keys you have some movement in the shoulder as well, to avoid muscle cramps.
There are some folks who recommend leaving fingers in a completely limp position and only using arm movements to reposition the fingers, but if you watch them demonstrate such typing style, it’s incredibly slow and clunky. I suspect when they actually are typing for themselves, they go back to using finger movement like everyone else.
For example, this lady, whose experience was with piano technique for RSI rehabilitation, goes so far as to recommend using arm movements even to press the keys – http://www.golandskyinstitute.org http://www.healthytyping.com/about-moveright/overview/ – Hilariously, if you watch her actually playing the piano, her arm and finger movements are entirely different than what she claims.
I really want to see a video of someone typing with a digging motion.Working like this mimics a digging motion or swimming, or even swinging from trees,
-
- Location: Germany
- DT Pro Member: -
So the 30-50° "tent" which is comfortable for you keeps them in the same angle of pronation or slightly more pronation. If you don't have any symptoms that must be fine for you.When I stand or walk in a completely relaxed, neutral position, my palms don’t directly face towards each-other, but are rotated about 30–40°back. No noticeable wrist flexion.
Have you heard of the MBT ( Massai Basic Technology) shoe which was made after interviewing peoples with very little or no back and knee joint pain. It was found that they mainly walked barefooted on soft ground and this was simulated in the shoe. The company doesn't exist any more but they were great. I believe some Sketchers use the technology now.
Anyway, it would be interesting to know if peoples that have never used a keyboard have the same amount of hand pronation in neutral position that we do. To avoid shoulder impingement It can be helpful to stretch the pectoral muscles until the palms face each other when in neutral position. As an experiment I stretched mainly one side over the course of a few months and the palm on the side I stretched now faces my side, while the other hand is still rotated inwards to around 40°. Apparently this can be caused by strong or tight pectoral muscles (e.g. from a lot of bench presses or push ups) without stretching them enough or building enough outer rotary shoulder muscles to balance them. Among other functions the pectoral muscles rotate the arm inwards.
But unfortunately many "ergonomic" keyboards are tilted towards the typist so that the fingers have to strain to extend to higher keys .Your finger extensors don’t need to be too strong to reposition the fingers on different keys.
With a two letter keyboard of the right design my dog could type the word "me" 300 times a minuteI really want to see a video of someone typing with a digging motion.Working like this mimics a digging motion or swimming, or even swinging from trees,
But really,some if not all of the keys on my keyboard experiment really will face away from me so that I can press them towards me or towards my palm. I'll post some pics once I have something worth showing. Hopefully there'll be no pain in my flexors, but I'll try it and comment here. Thanks for all the tips and pics. [/quote]
Last edited by Ladybug on 22 Apr 2016, 22:58, edited 1 time in total.
-
- Location: geekhack ergonomics subforum
- Favorite switch: Alps plate spring; clicky SMK
- DT Pro Member: -
In general, if a keyboard is tilted up at the back, it was designed to be used with standard office furniture: low chair, high desk (height designed for writing with pen and paper or eating, or sometimes even higher than comfortable for those). For such furniture, if you make the keyboard completely flat or tilt it down at the far side, it will be less comfortable/healthy to type on.Ladybug wrote: ↑But unfortunately many "ergonomic" keyboards are tilted towards the typist so that the fingers have to strain to extend to higher keys .
Personally, I wish keyboards were as thin and flat as possible, because I’m happy to pile stuff under one end or the other, depending on the setup.
When I use a laptop in a café, I almost always either set the laptop on my lap, or put it on the (too tall) table but pile a book or two under the far end.
Other people I see leave the laptop flat on the table, and either flex their wrists or lift their arms up and rest their palms on the laptop "chin". Both are substantially worse IMO than tilting the keyboard, lowering the table, or raising the chair.