F104+SSK+122+62+77+50+Ergo orders now open! New Kishsaver+Industrial Model F Keyboards

[mrprofessor]

Days!? The OGs hold up better over decades…

Unfortunately my repro didn't. I am the 2nd owner of the board, but it started chipping from the corners.

[depletedvespene]

The factory can even do green, red, and blue for the sublimation text/graphics and multiple colors can be printed at the same time and on the same keycap. If anyone wants one-off sets in various colors it is possible, though some sublimation colors come out better than others and you'd have to make them to find out. So far green, blue, and red have come out nicely.

This could well be used to do "binational" layouts, as some companies did back in the day (although it'd definitely require using front-printing for the tertiary layer).

Also... for the navigation/numpad block on the F77, come to think of it (top left, in black, numbers; bottom right, in blue, navigation icons).

[Bjerrk]

This could well be used to do "binational" layouts, as some companies did back in the day

The keyboard on my wife's laptop is even trilingual.

[depletedvespene]

This could well be used to do "binational" layouts, as some companies did back in the day

The keyboard on my wife's laptop is even trilingual.

One of those "generic Nordic" keyboard prints?

[Bjerrk]

One of those "generic Nordic" keyboard prints?

Y(æ/ä/ø)s.

[depletedvespene]

One of those "generic Nordic" keyboard prints?

Y(æ/ä/ø)s.

I ain't Nordic and I find those things disgraceful. It could be worse, though (by forcing Estonian into the mix; Latvian and Lithuanian are... just too far out).

[Bjerrk]

One of those "generic Nordic" keyboard prints?

Y(æ/ä/ø)s.

I ain't Nordic and I find those things disgraceful. It could be worse, though (by forcing Estonian into the mix; Latvian and Lithuanian are... just too far out).

It's incredibly annoying, I'll give you that. But hey, as long as it isn't ANSI

[Ellipse]

Here are some great photos shared with permission from a user of the Russian-Hebrew-English set - nice colors and nice work by Zed on this special request layout.

20220714_195846.jpg (334.48 KiB) Viewed 12328 times

20220714_195811.jpg (275.33 KiB) Viewed 12328 times

20220714_195805.jpg (376.76 KiB) Viewed 12328 times

[mrprofessor]

What are the color codes for powdercoated industrial and beige cases?

Do you mean before or after it has been used for a few days?
powderuncoating.jpg

(It has almost attained a vintage look, like stone-washed jeans )

Wondering if this is a trade secret for this project.

[milk_stain]

I just received my F62 and so far I've noticed some weird behavior/chatter with the 'F' key and the spacebar. This is using the default QMK firmware, which the manual says should be working correctly. I've also double checked that the keys are seated correctly and click like the rest of the functioning keys. The chatter seems to only happen if I'm typing fast, and appears every several presses or so. Sometimes there will be 2 keypresses, sometimes 3 or 4.

I tried using pandrew's utility's keypress monitor, and pressing either the 'F' or spacebar multiple times results in sporadic keypress detection for various other keys. Futhermore, holding the 'F' sometimes causes the keypress monitor to flicker very fast between "pressed" and "not pressed". And other times, holding the 'F' will show a keypress, then no keypress for a second or two, and then back to a keypress.

Does anyone have any idea what the issue could be?

[Ellipse]

I am posting with permission another nice photo someone sent me: a customized Swedish-Finnish layout

milk_stain as noted in the manual, please tighten the 2 controller screws.

Swedish layout - Copy.jpg (1.16 MiB) Viewed 12205 times

[milk_stain]

milk_stain as noted in the manual, please tighten the 2 controller screws.

I checked the controller screws and they were already pretty tight. I tightened them a little bit more (as much as I could), and the issue persisted. I also unscrewed the controller screws and screwed them in again and the issue still persisted.

The issue seems to be easily reproducible by repeatedly holding the the F key for a second, which results in several keypresses during the hold (this is without repeated key presses enabled).

The issue appears both on my Windows 10 PC and my Chromebook.

Here's two videos of rapid tapping and hold-tapping showing the issue https://photos.app.goo.gl/BtWywpY1TtRLTUvj6

[Ellipse]

milk_stain please follow the manual and Quality Control secrets video linked to in the manual. If only one or two keys is acting up, is likely that the spring needs adjustment or replacement or debris has entered the inner assembly between the flipper and PCB (less likely).

[milk_stain]

or debris has entered the inner assembly between the flipper and PCB (less likely).

Debris, you say?

...I put some compressed air through the barrel and now everything works. Go figure

Thanks!

[Ellipse]

Thanks to the final remaining keys going out to folks over the past month or so, they have been able to complete their custom layouts and I am glad that folks are sending me photos of their setups. Here's another nice setup I am posting with permission - a layout that includes black keys and dark gray keys:

layout.jpg (580.04 KiB) Viewed 11967 times

[darkcruix]

I can finally post a picture of the finished HHKB inspired Compact F62. I received the front printed keys a few days ago and finally had time to take a photo.

IMG_4696.JPG (204.2 KiB) Viewed 11948 times

[Muirium]

Needs a side by side real HHKB shot. I’m guessing FHKB is thiccc.

The blue diamond is a nice touch. Bit of a cheat sheet for HHKB newbs!

Looking at the front legends, the absent Volume Down and Up remind me of a niggle about the real HHKB: I’ve long mapped those audio keys to Mute, Vol- and Vol+ in that order, to match everything else. The HHKB’s native order was forever catching me out. Consistency is king.

[daedalus]

Having recently reassembled my F62 in order to move to the compact case, I have had issues with spring verb and inconsistent key feel. Multiple attempts at assembly/reassembly have caused the affected keys to change, and swapping springs has not helped.

After doing some more testing, I noticed that some of the affected keys had loose barrels (I could wiggle them around slightly). I tested the tightness of the keyboard assembly with a clamp and noticed that the top of the keyboard assembly has quite a lot of give in it, I was able to squeeze it tighter by about 1mm. The bottom part has little to no give. If I squeeze the top part of the assembly with the clamp and attempt to tighten the top latches with a pair of pliers, I find that there's a gap between the latch and the backplate, so no matter how I adjust the top latches, once I take away the clamp the assembly will go back to being loose again.

I have been following this video (https://www.youtube.com/watch?v=xaa23N9JDBs) for reassembly. The one place where I find the process diverging is that after I snap the lower latches together, I need to bend the top latches out of the way or else I can't snap shut. I have to move them back, snap, and then readjust.

Any idea what may be going on? I can take some pictures, but it's getting dark, so it's hard to take good ones right now.

[Ellipse]

Some more nice photos! Posting with permission from a user of both Colemak and QWERTY who was wondering if the keys could take a few dozen removals/insertions if they switch frequently between layouts - any experience with the durability of keys? Mine are the same exact material as what IBM used (as relayed by an original Model F production supervisor, as noted before) so it should have similar properties.

The first one is a nice photo showing off the Deskthority home page.

daedalus there should be some ability to wobble the barrels a bit, even with most IBM Model F originals. Tightening the top inner assembly plate when it is not installed with the rest of the parts is something I have done with a large vice and has improved the sound a bit on some F107 originals formerly in my collection. Maybe adding extra scrap foam or cutting up a second F77 foam might prove more effective than making physical adjustments to the inner assembly plates, or a combination of both approaches.

regular and colemak (1) - Copy.jpg (315.03 KiB) Viewed 11673 times

regular and colemak (2) - Copy.jpg (331.32 KiB) Viewed 11673 times

regular and colemak (3) - Copy.jpg (389.08 KiB) Viewed 11673 times

regular and colemak (4) - Copy.jpg (237.84 KiB) Viewed 11673 times

[Ellipse]

And here's another great photo posted with approval - a customized UK set as well as the black unprinted set.

black unprinted and UK.jpg (962.13 KiB) Viewed 11377 times

[Arkku]

Finally got mine, and completely uniquely for my keyboards: almost all of the keys actually have legends that match what happens when pressed. Initial impressions are very positive!

IMG_5830 (1).jpeg (1020.86 KiB) Viewed 11310 times

Still experimenting with the layout a bit, and working on getting my own firmware working for this. Also waiting for the F62 to arrive.

[yac]

Both of my keyboards came, only had time to assemble the compact one for now.

compact_77.jpg (589.39 KiB) Viewed 11252 times

layout.png (63.49 KiB) Viewed 11247 times

Thanks in no small part to the amazing QC of Joe, and me looking through the videos in advance. I just had a single key with a buzz, fixed it right up, did not have to remove and re-attach nor use a spare. I use the web-hosted configurator and dfu-programmer to flash from my Mac.

Would be interesting to see if one could add holding down keys at startup and having the keyboard send them? So we can use the bootup modes without keeping a separate keyboard around.

I haven't looked at the code myself yet, but the idea i have is as soon as the keyboard starts, we assume it's exactly as last time, so we don't perform any calibrations/startup procedures beyond absolutely necessary , instead we look for keys pressed. If keys are being pressed we send them immediately as being pressed to the host. If no keys are being pressed we proceed with regular calibration/startup procedure.

The actuation force seems a little lighter than my XT Model F, i will try moving my XT springs over. Still would wish we could buy some sets of your old rejected springs, get a feel for what you tried over time, Ellipse.

[Kugelkopf]

...
Would be interesting to see if one could add holding down keys at startup and having the keyboard send them? So we can use the bootup modes without keeping a separate keyboard around.

I haven't looked at the code myself yet, but the idea i have is as soon as the keyboard starts, we assume it's exactly as last time, so we don't perform any calibrations/startup procedures beyond absolutely necessary , instead we look for keys pressed. If keys are being pressed we send them immediately as being pressed to the host. If no keys are being pressed we proceed with regular calibration/startup procedure.

The actuation force seems a little lighter than my XT Model F, i will try moving my XT springs over. Still would wish we could buy some sets of your old rejected springs, get a feel for what you tried over time, Ellipse.

Well, as noted earlier you can at least enter startup manager by manipulating a Mac's NVRAM. (Otherwise I wouldn't type on the Model F anymore, because I have to switch systems often.) The first thing I would check is, if the disability to enter startup options is a timing issue caused by the signal calibration procedure at all. My idea was, to compile a QMK test firmware with the calibration functions replaced by a routine that simply (and faster) sets a common threshold determined by the signal untility, e.g. If that works, it would be time to tackle a more elegant solution. As far as I understood earlier reports, the original firmware with fixed signal levels wasn't equally reliable, hence the calibration procedure itself seems worth retaining.

I must admit, that after the final tweeks of my QMK firmware and the NVRAM trick, I've enjoyed typing instead of exploring QMK test versions until today. I'll report anything noteworthy, whenever I find time for testing.

And I agree, the actuation force could be a little stronger. But this is just a personal guess and since I have never had the pleasure to type on an original F, this does not say anything at all about how well our new Fs achieve the primary project goal to match the originals from the eighties.

[Shinryuu]

I finally got my hands on F62 keyboard! I didn't join this project directly in the past because it was still too expensive for me to do so. I found a person who was willing to sell this keyboard which was still unopened. It came with first aid kit as a bonus.

springyboba.jpg (2.02 MiB) Viewed 11008 times

IMG_20220721_220415.jpg (2.54 MiB) Viewed 11008 times

I am posting with permission another nice photo someone sent me: a customized Swedish-Finnish layout

milk_stain as noted in the manual, please tighten the 2 controller screws.

Swedish layout - Copy.jpg

That layout looks pretty much I've been trying to figure out. I own F62 with Swedish/Finnish layout as well. Here's my setup what I've been trying to fiddle on with. I still don't quite understand how to align right corner for 1.75U shift and 1U button (might need to read more docs?). Here's a picture so it might be easier to tell what I'm after. Layout flipper placement: ISO Enter (Vertical enter and split Left Shift) if it matters.

f62test.jpg (180 KiB) Viewed 11008 times

[Meowmaritus]

I prefer my keyboards propped at a pretty high angle like the original model M with the feet flipped out and using my F77 flat on my desk is killing my wrists.
I placed an order for keyboard feet / bumpers in the shop but the list of options is kinda confusing and it's unclear if what I ordered ("2 Large rubber bumpers (left of photo) with screw and nut plus 2 dome (middle of photo)") is what I'm looking for. Additionally, it will likely take a while to ship.
In the meantime, has anybody had success with propping their boards up at an angle like I'm trying to do? If so, what did you use?

[yac]

I prefer my keyboards propped at a pretty high angle like the original model M with the feet flipped out and using my F77 flat on my desk is killing my wrists.
I placed an order for keyboard feet / bumpers in the shop but the list of options is kinda confusing and it's unclear if what I ordered ("2 Large rubber bumpers (left of photo) with screw and nut plus 2 dome (middle of photo)") is what I'm looking for. Additionally, it will likely take a while to ship.
In the meantime, has anybody had success with propping their boards up at an angle like I'm trying to do? If so, what did you use?

I haven't tried the full-size yet but typing on my compact keyboard i too would like more angle. I think the way to truly be able to dial in the ergonomics is one of those under-desk keyboard trays, and find a way to adjust the angle, maybe with a platform and a screw from underneath? With a bit of woodworking i think it will be OK. Just need to find a tray rated for the weight...

[Ellipse]

Shinryuu confirmed you cannot do a 1.75U+1U HHKB style split right shift on the non-HHKB boards as the barrels and PCB pads are in slightly different locations. You can do the 1U+1.75U JIS style split right shift on the non-HHKB boards like the one in your photo.

Thanks yac and Arkku for the feedback and photos! Confirmed you can adjust the height through various bumpers such as the screw on bumpers. Here are the bumper heights offered through the project web site, though others should work: cork – 1/16 inch (1.6 mm). 4 square – 1/8 inch (3.2 mm). cylindrical = 1/4 inch (6.4 mm). dome (both adhesive and “medium with threads and nut”) = 1/2 inch (12.7 mm). large with screw and nut = 1 inch (25.4 mm)

[Meowmaritus]

Confirmed you can adjust the height through various bumpers such as the screw on bumpers. Here are the bumper heights offered through the project web site, though others should work: cork – 1/16 inch (1.6 mm). 4 square – 1/8 inch (3.2 mm). cylindrical = 1/4 inch (6.4 mm). dome (both adhesive and “medium with threads and nut”) = 1/2 inch (12.7 mm). large with screw and nut = 1 inch (25.4 mm)

Thanks for the specifics and it might be a good idea to clean up the feet/bumpers store page to be a bit more concise, maybe with one image for each selection with heights written on it.

[Arkku]

Also received my black + dark gray F62 now. (I made a DIY ISO Enter hack, but hopefully the actual ISO Enter will be available in dark gray later.)

E0BD7067-E019-47F6-8A83-C94BF6186169.jpeg (1.76 MiB) Viewed 10527 times

It would also be nice to get the arrow keys (blue inverted T) with printed arrows, as well as the Fn key in a bigger size but offset. But this is very usable as is.

A nice thing about this case size is that it's exactly the same width as common TKL wrist rests.

[Sam]

Hence I decided to drop Vial for the time being and compile a custom QMK layout including tap dances. Thanks to the post of fellow forum member "BuccoBruce2" installation and setup of a local QMK environment suitable for Ellipse's Model F went largely without problems (Catalina) and the layout developed under Vial could soon be reconstructed under QMK. (I must admit, though, that I've used the QMK webconfigurator to design everything except for functions not available there, like tap dances, downloaded the layout's JSON representation and just added the few missing crumbs before compiling the completed firmware.) Compiling finished way faster than in the webconfigurator, although I missed the potatoe animation.

Making my own TRUE QMK layout sounds like a perfect solution to me. Doing software development and using the F1 to F12 keys extensively while working, I could never be as productive on my F77 compared to IBM layouts with dedicated F keys. Needing to hit another key to activate the F layer, along with various possible modifiers just makes too much finger gymnastics which, while absolutely possible to do, just slow down things. So what sounds perfect to me is having all the F keys on the base layer, but doing a double tap to access them, while a single tap accesses the numerals/symbols. That way, any modifiers can easily be combined without the need for pressing another key to access a different layer. Not just for the F keys, but for several other keys, putting them on a double tap I think will be so much easier.

Additionally, I think the caps WORD function will be very handy. For example, typing #defines in C where some words in all caps, separated with an underscore, and then when finished typing it (i.e., hitting space, tab, etc.), it'll revert back to non-caps automatically. Of course I can go into caps lock, then out of caps lock to do the same thing, but reducing the operation by one keystroke helps out. Those types of constants in C are about the only time I use all caps, so no need for an all caps key if using the QMK caps Word function.

Now, the big question is, is there an easy to follow guide to compiling QMK under Windows? Thank you, Kugelkopf and BuccoBruce2 for the information. It just seems though that the information necessary to do what I want is divided into at least three different sets of instructions. Would be great if there was one guide to explain this well, with some examples of how to add Tap Dance and other standard QMK features which aren't available in the default firmware installed on the keyboards.

Afterall, one of the major points of this project is getting a FULLY customizable keyboard, but which at the moment it seems to me you have to jump through hoops to do. Also, this one MASSIVE thread is really a huge detriment to getting useful information. Does anyone really have the time to wade through many thousands of posts to patch together the right information? Would be great if there was an easy way of breaking out the information into various subjects so that a person could zero in on what they're looking for without the need to read thousands of pages of material unrelated to what's needed.

One more thing, based on some of the posts I've read, it seems the MCU used is severely underpowered for this keyboard, especially with respect to RAM memory. If my reading of the posts is correct, I would suggest serious consideration be given to updating the controller board with a MCU having more RAM memory capacity. so that the full of QMK functions can easily be added to the standard firmware along with the capsense functions, etc. For anyone buying such a premium (quality and price) keyboard, and getting a cheaper version of the MCU which doesn't have enough memory capacity to do all that's needed seems to be a mismatch. Even if we had to pay a few dollars more to get a larger capacity MCU would be an option I would have jumped at when I ordered my keyboards.

Anyways, having the Tap Dance available, and specifically related to making Double Tap keys seems like a game-changer to me. I guess I'll try to wade through the information presented and try to compile QMK on my Windows machine and see if it works.