Bringing the IBM PC XT into the 21st Century
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
The idea with the XTant was to see if it would work and to reuse as much of the donor board as possible. The XT case is very sturdy and case design is difficult (for me anyway). I am sure that someone will design a case and have it machined. Maybe M'er Forever will share one of his SSK models if Unicomp is not using it.
-
- DT Pro Member: -
Would it be possible to fit one of your assemblies in a regular Model M case (101, SSK, EnduraPro, etc)? Also, I really like the idea of an XTant with a full numpad. Could that be produced as an option, or would it increase your production costs too much?wcass wrote: The idea with the XTant was to see if it would work and to reuse as much of the donor board as possible. The XT case is very sturdy and case design is difficult (for me anyway). I am sure that someone will design a case and have it machined. Maybe M'er Forever will share one of his SSK models if Unicomp is not using it.
EDIT: Actually, if you ever want to consider doing a kit using the 83 XT hammers and barrels but with a completely custom case, I would love to see this layout:
http://www.keyboard-layout-editor.com/# ... 8891fef08e
![Cool 8-)](./images/smilies/icon_cool.gif)
By the way, a late congrats on reaching a working prototype, this has been a long time coming.
![Very Happy :D](./images/smilies/icon_e_biggrin.gif)
- Muirium
- µ
- Location: Edinburgh, Scotland
- Main keyboard: HHKB Type-S with Bluetooth by Hasu
- Main mouse: Apple Magic Mouse
- Favorite switch: Gotta Try 'Em All
- DT Pro Member: µ
As Wcass says: cases are hard. Even seemingly simple mods require a lot of hacking and finesse. Here's 002's well illustrated SSK mod on a G80 for instance:
http://deskthority.net/workshop-f7/g80- ... t8486.html
Model F insides are shorter (5 rows) than Model Ms. I'm not sure this is something practical to alter either! But the XT's shell is a nice blank canvass and this project has a world of potential there.
http://deskthority.net/workshop-f7/g80- ... t8486.html
Model F insides are shorter (5 rows) than Model Ms. I'm not sure this is something practical to alter either! But the XT's shell is a nice blank canvass and this project has a world of potential there.
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
As it so happens, I was planning a similar layout for my next project. We are limited to the key cap shapes that Unicomp has - that 1.75 right shift will be stepped like a caps lock - but that is a good thing in that it creates a tactile buffer between the shift and up arrow. I liked that so much that i put a 1.75 Ctrl to the left of left arrow - this really sets the arrow cluster apart and easy to find without looking. Backspace can be split. ANSI or ISO Enter and left shift are supported and user selectable. The controllers are fully programmable, so the users can move caps around (like swap Caps Lock with right Ctrl).REVENGE wrote: ... Actually, if you ever want to consider doing a kit using the 83 XT hammers and barrels but with a completely custom case, I would love to see this layout:
http://www.keyboard-layout-editor.com/# ... 8891fef08e 8-)
By the way, a late congrats on reaching a working prototype, this has been a long time coming. :D
http://www.keyboard-layout-editor.com/# ... f45fd75154
Just need someone to design the case. Any volunteers?
Last edited by wcass on 28 Sep 2014, 20:41, edited 1 time in total.
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
The plate/PCB/barrel/plate sandwich is mostly 8mm thick. Hardware thickens that to about 13mm in a few spots 10mm from the top and bottom edge and at the horizontal center-line. The number of rows should not be much of an issue. The curvature of the plates is about 260mm radius.Muirium wrote: As Wcass says: cases are hard. Even seemingly simple mods require a lot of hacking and finesse. Here's 002's well illustrated SSK mod on a G80 for instance:
http://deskthority.net/workshop-f7/g80- ... t8486.html
Model F insides are shorter (5 rows) than Model Ms. I'm not sure this is something practical to alter either! But the XT's shell is a nice blank canvass and this project has a world of potential there.
- Muirium
- µ
- Location: Edinburgh, Scotland
- Main keyboard: HHKB Type-S with Bluetooth by Hasu
- Main mouse: Apple Magic Mouse
- Favorite switch: Gotta Try 'Em All
- DT Pro Member: µ
Ah, that's good then. I'll just go crazy and ponder the practicality of a Model F based SSK! Any obvious gotchas designing an "SSKant" to fit in (a chopped) Model M case?
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
- Muirium
- µ
- Location: Edinburgh, Scotland
- Main keyboard: HHKB Type-S with Bluetooth by Hasu
- Main mouse: Apple Magic Mouse
- Favorite switch: Gotta Try 'Em All
- DT Pro Member: µ
Intriguing. The ultimate custom Model F, for me, would be either a 60% in a chopped Model M shell — quite like a Kishsaver, but with HHKB inspiration too — or a nice Model F SSK.
- Hypersphere
- Location: USA
- Main keyboard: Silenced & Lubed HHKB (Black)
- Main mouse: Logitech G403
- Favorite switch: Topre 45/55g Silenced; Various Alps; IBM Model F
- DT Pro Member: 0038
My preferences for a custom Model F:
+ Form factor: 60% or TKL (SSK).
+ Layout: ANSI with OS X and HHKB configuration.
+ Switches: IBM capacitive buckling spring (Model F).
+ Case: Derived from Model F Kishsaver, 77, or 107, or XT (I like as much metal as possible).
I much prefer the feel of the 107 or XT cases to AT, M, or M-SSK cases; much more solid and free of creakiness.
-
- DT Pro Member: -
Nice, layout looks good! I wouldn't even mind buying a case-less kit and improvising with the case build (Legos, wood, etc).wcass wrote: As it so happens, I was planning a similar layout for my next project. We are limited to the key cap shapes that Unicomp has - that 1.75 right shift will be stepped like a caps lock - but that is a good thing in that it creates a tactile buffer between the shift and up arrow. I liked that so much that i put a 1.75 Ctrl to the left of left arrow - this really sets the arrow cluster apart and easy to find without looking. Backspace can be split. ANSI or ISO Enter and left shift are supported and user selectable. The controllers are fully programmable, so the users can move caps around (like swap Caps Lock with right Ctrl).
http://www.keyboard-layout-editor.com/# ... f45fd75154
Just need someone to design the case. Any volunteers?
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
Another option might be to go without the curve. Flat metal bottom plate, .25" acrylic spacer, flat top plate, .25" acrylic trim. It might look like this:
-
- Location: UK
- Main keyboard: Filco ZERO green alps, Model F 122 Terminal
- Main mouse: Ducky Secret / Roller Mouse Pro 1
- Favorite switch: MX Mount Topre / Model F Buckling
- DT Pro Member: 0167
only problem i see with that is the metal plate and the metal back-plate are part of the quality feel of an XT
i think a mod of the case to cover the gap left with the nav cluster is the way to go at least for me
i think a mod of the case to cover the gap left with the nav cluster is the way to go at least for me
- Hypersphere
- Location: USA
- Main keyboard: Silenced & Lubed HHKB (Black)
- Main mouse: Logitech G403
- Favorite switch: Topre 45/55g Silenced; Various Alps; IBM Model F
- DT Pro Member: 0038
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
I think so. Not 100% sure. Could you post a few pictures with the case open viewing the bottom of the back plate?
The XT back plate has a hole that we use to mount the PCB and controller. The hole is on a tab extending out the "top" edge, just right of center. If the back plate for the 4176191 has a hole in the same spot, I think it should work just fine.
- Muirium
- µ
- Location: Edinburgh, Scotland
- Main keyboard: HHKB Type-S with Bluetooth by Hasu
- Main mouse: Apple Magic Mouse
- Favorite switch: Gotta Try 'Em All
- DT Pro Member: µ
If I had one of these Bigfoots, I'd be prying it open to spam pictures of its innards. But I don't, so here's some inside pics I found in a search. Can you see what you need here, Wcass?
XTant would be a great upgrade for these guys with their notorious attached controller:
![Image](http://geekhack.org/index.php?action=dlattach;topic=56747.0;attach=59814;image)
XTant would be a great upgrade for these guys with their notorious attached controller:
-
- Location: UK
- Main keyboard: Filco ZERO green alps, Model F 122 Terminal
- Main mouse: Ducky Secret / Roller Mouse Pro 1
- Favorite switch: MX Mount Topre / Model F Buckling
- DT Pro Member: 0167
That is a totally different controller than an XT the component count is so low and the connector is different
same backplate as far as i can see
same backplate as far as i can see
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
That does look a lot like the XT back plate.
The plate that the PCB/controller is attached to, is it 17.5" wide? If you measured from the center of that bolt to the sides is it 9.9" and 7.6"? If so, then the plate is the same and the XTant kit will work.
The plate that the PCB/controller is attached to, is it 17.5" wide? If you measured from the center of that bolt to the sides is it 9.9" and 7.6"? If so, then the plate is the same and the XTant kit will work.
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
A working XTant will be at the monthly keyboard meetup in San Francisco - October 25.
http://deskthority.net/off-topic-f10/ki ... t8819.html
Attendees will get to see/handle most of the parts that will go into the kits.
http://deskthority.net/off-topic-f10/ki ... t8819.html
Attendees will get to see/handle most of the parts that will go into the kits.
- Muirium
- µ
- Location: Edinburgh, Scotland
- Main keyboard: HHKB Type-S with Bluetooth by Hasu
- Main mouse: Apple Magic Mouse
- Favorite switch: Gotta Try 'Em All
- DT Pro Member: µ
Are you there too or is it out the home, making its own way in the world already? They grow up so soon…
Oops, I see you already said:
http://deskthority.net/post187086.html#p187086
Oops, I see you already said:
http://deskthority.net/post187086.html#p187086
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
If the reviews go well, look for kits in a month or so.
I’ve been asked a few times (and wondered myself) about how I will do spacebar stabilization. I’ll use Unicomp space bars of course, but how to hold the bar against the barrel frame is the question.
I did a mod a few years back that replaced an AT’s spacebar for two Alt keys and a model M spacebar. For that, I bent a piece of metal and glued this to the barrel frame. It worked, but the metal was not as thick as I would have liked (too bendy) and the connection to the barrel frame was a weak point on a keyboard famous for being almost indestructible. I wanted to try out some other options.
My first idea was to 3D print a copy of the retention clips used for the spacebar on the 122 key Model F, but there are two reasons why this is not ideal. First, mounting the clips to the barrel frame would require glue again. Second, the barrels on an XT have a alignment stud different from the alignment notch on the barrels of an AT or 122. The stud sticks up through the barrel frame slightly and would block the stabilizer from sliding easily into a clip as the spacebar is mounted. This second issue would also affect stamped or welded on clips too.
So my next idea was to use the geometry of the keyboard to hold the clips on - cantilevered off of a barrel. I would need a piece of metal with a hole in it; the hole must be close to the diameter of the barrels and the outer edge must extend over the stabilizer bar. I picked up two 7/16” SAE washers from my neighborhood hardware store for less than a dollar, dropped them over the C and < keys and they work perfectly with nothing but the washer's weight and rigidity keeping the washer edge from tilting up – and this keeps the spacebar stabilizer tight against the barrel frame. The spacebar registers properly and reliably even when I press at the extreme left and right of the key.
I’ve been asked a few times (and wondered myself) about how I will do spacebar stabilization. I’ll use Unicomp space bars of course, but how to hold the bar against the barrel frame is the question.
I did a mod a few years back that replaced an AT’s spacebar for two Alt keys and a model M spacebar. For that, I bent a piece of metal and glued this to the barrel frame. It worked, but the metal was not as thick as I would have liked (too bendy) and the connection to the barrel frame was a weak point on a keyboard famous for being almost indestructible. I wanted to try out some other options.
My first idea was to 3D print a copy of the retention clips used for the spacebar on the 122 key Model F, but there are two reasons why this is not ideal. First, mounting the clips to the barrel frame would require glue again. Second, the barrels on an XT have a alignment stud different from the alignment notch on the barrels of an AT or 122. The stud sticks up through the barrel frame slightly and would block the stabilizer from sliding easily into a clip as the spacebar is mounted. This second issue would also affect stamped or welded on clips too.
So my next idea was to use the geometry of the keyboard to hold the clips on - cantilevered off of a barrel. I would need a piece of metal with a hole in it; the hole must be close to the diameter of the barrels and the outer edge must extend over the stabilizer bar. I picked up two 7/16” SAE washers from my neighborhood hardware store for less than a dollar, dropped them over the C and < keys and they work perfectly with nothing but the washer's weight and rigidity keeping the washer edge from tilting up – and this keeps the spacebar stabilizer tight against the barrel frame. The spacebar registers properly and reliably even when I press at the extreme left and right of the key.
- Hypersphere
- Location: USA
- Main keyboard: Silenced & Lubed HHKB (Black)
- Main mouse: Logitech G403
- Favorite switch: Topre 45/55g Silenced; Various Alps; IBM Model F
- DT Pro Member: 0038
@wcass:
Brilliant!!!
Thanks for sharing your washer stabilizer invention for putting Model M spacebars into Model F keyboards. It is an elegant solution.
You probably should have filed for a patent.
Edit: What keeps the stabilizer wire and spacebar from shifting laterally? What about sliding too far toward the front of the keyboard?
Brilliant!!!
Thanks for sharing your washer stabilizer invention for putting Model M spacebars into Model F keyboards. It is an elegant solution.
You probably should have filed for a patent.
Edit: What keeps the stabilizer wire and spacebar from shifting laterally? What about sliding too far toward the front of the keyboard?
-
- DT Pro Member: -
That's pretty clever. It's very satisfying when there's a simple solution to a complex problem.wcass wrote: So my next idea was to use the geometry of the keyboard to hold the clips on - cantilevered off of a barrel. I would need a piece of metal with a hole in it; the hole must be close to the diameter of the barrels and the outer edge must extend over the stabilizer bar. I picked up two 7/16” SAE washers from my neighborhood hardware store for less than a dollar, dropped them over the C and < keys and they work perfectly with nothing but the washer's weight and rigidity keeping the washer edge from tilting up – and this keeps the spacebar stabilizer tight against the barrel frame. The spacebar registers properly and reliably even when I press at the extreme left and right of the key.
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
The key cap plunger keeps the cap in the right spot north/south/east/west. The stabilizer wire is fixed to the cap but free to rotate. It acts exactly like an anti-roll bar does on a car. The last direction of unwanted movement is twist; this is where the barrel alignment studs help. The length of the Model M wire is such that it reaches just past this - effectively preventing rotation. The picture with the caps off shows it.
- Muirium
- µ
- Location: Edinburgh, Scotland
- Main keyboard: HHKB Type-S with Bluetooth by Hasu
- Main mouse: Apple Magic Mouse
- Favorite switch: Gotta Try 'Em All
- DT Pro Member: µ
Reckon this little bit of genius would work on a Kishsaver? Does it matter whether the space bar is Model M or F stab-style?
-
- Location: Houston, Texas
- Main keyboard: IBM Bigfoot
- Main mouse: CST trackball
- Favorite switch: IBM Model F
- DT Pro Member: -
That's a great idea, wcass. You could add a dab of silicone adhesive if you were worried about rattling. I may do that on one of my ATs if I ever get around to transplanting a Model M space bar.
- wcass
- Location: Columbus, OH, USA
- Main keyboard: ibm model m
- Main mouse: kensington expert mouse
- Favorite switch: buckeling spring
- DT Pro Member: 0185
The F-122, and 4700 keyboards have retention clips already - though sized for thinner stabilizer wire. I would recommend using the F-M stabilizer adapter bars that I make for this. I ordered more tubing recently, so I will be making more. The washer trick would work if you had a broken stabilizer clip on one of those models, or on a model without clips (XT, Bigfoot, AT). Twist action might be an issue without the alignment studs. I'll have to check it on an AT and Kishsaver to see.