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

[RedESC]

Leyden Jar update:
Rico has just published the open source Leyden Jar diagnostic tool. Now all the project files are ready for future development work to add support for additional keyboards, maybe even non-Model F / beam spring keyboards!

https://github.com/mymakercorner/Leyden ... ostic_Tool

RedESC yes, at least one board did end up being returned which was listed on eBay. The seller was a third party that did not realize the provenance and was fully cooperative. Even if the serial is defaced that would be a giveaway that the board is suspect. I log who gets each serial so I can confirm the board is ok. Some people thankfully have been contacting me to check that a particular listing is legit.

Thanks for the update. Also, neat.

[Ellipse]

I have updated the Model F product pages based on feedback that the web site should be more organized to help people who are not as familiar with all of the various options and features. There is now an explanation of each customizable feature and option plus a FAQ section on the various product pages.

Feel free to check these pages out and let me know of any additional changes that would be helpful or things people wished were written on the site instead of having to dig down to find an answer.

[krastatt]

I could not find the PC AT Enter (big enter) on any of the ordering pages. Are these not yet available to order?

[Ellipse]

krastatt correct, a batch of the 3 new keys arrived but I was not happy with the quality of the keys so I am having the molds updated and the parts remade. The keys did not buckle the springs reliably, which is not acceptable.

[lisdude]

Does the classic style FSSK use the same powder coating as the F77? There was mention a few million pages back about possibly switching to a more modern and durable coating. I love my F77, but the powder coating didn't last a year before it started to come off. It's easily the biggest disappointment of an otherwise amazing product. Really ruins the aesthetic.

(for reference, here's what mine looks like about a year after switching to the beige case)

[Ellipse]

Glad to see the original batch keyboard has been well used! Having refurbished dozens of original IBM bank branch terminal keyboards, I saw many with similar wear patterns and chipped paint.

The new keyboards all use the more durable coating similar to the updated production compact cases (the two piece ones that are powdercoated die cast aluminum instead of anodized aluminum). I still have spare die cast zinc classic F62/F77 cases from the latest production batch and they are $50 off at the moment if you wanted to pick up any extras. The updated 2022 batch with its slightly reformulated powdercoating seems to have fewer reports of issues.

[wobbled]

Yeh those IBM boards were heading for 40 years old Ellipse. Your cases haven’t lasted a year.

[Ellipse]

Everything is about tradeoffs. If you want an original formulation textured, matte appearance you have to accept the non-automotive quality paint finish. Uncoated paint does wear down after heavy usage (think how easy it is to scuff and chip painted walls), but the improved powdercoating of the most recent production run and the new formulation used with the recent cases should slow down the wear process. The tradeoff is that the new process is more of a mix between the Model M style finish and Model F style, and not exclusively the Model F style, but the durability is higher.

So far I have not received even one report of wear for the most recent formulation of the ultra compact cases (the two piece die cast aluminum cases).

Regarding a comparison to the originals: my guess is a lot of the old 4700 systems and their keyboards were only used for several years in the 80s before they were shelved at the recycler companies after the banks upgraded to more powerful systems as technology rapidly advanced from when these came out in the early 1980s, so much of the wear was expected to be sustained in the first year or first few years that they were actually used (even though I have heard reports from my recycler contacts of some small branches that still used the 4700 system consistently for decades, as of about 10 years ago)!

Having refurbished many of these original IBM keyboards, I have seen that many of them were repainted by third parties because the paint does not withstand usage for long. Also IBM had an option to rent these keyboards - my guess is that after a couple years many of them were replaced and refurbished just for cosmetic reasons, as part of a service contract or rental agreement. Often times the inner assembly and case manufacture dates were years apart (sometimes the reverse was true and the case was older than the inner assembly as they would just replace the inner assembly due to technological failure from what I have seen - many inner assemblies of my 4704 collection were refurbished in 1996).

[wolfman]

It's time for a status update for my QMK pull request for the F62.
https://github.com/qmk/qmk_firmware/pull/21193

Six weeks ago I updated the readme.md file with the details on how to access the boot loader. I am waiting to be told what's the next step to get the pull request accepted.

The QMK pull request review process is S..L..O..W.. QMK is a volunteer open source project. One's personal life should have priority over volunteer open source.

One of the changes caused by the review process is the renaming of the key layouts for the F62. There is a community name for the F62's physical key layout: tsangan. I had to update the source to use QMK's community tsangan layouts. The non-community layouts have been renamed to reflect the tsangen layout.

The left side of the F77 is the tsangan layout. The right side of the F77 is a macro pad. I changed the F77 layouts to reflect community naming.
F77 Example: 60_ansi_tsangan_15_macro

Updated F62 Physical Key Layout Names:
all
60_ansi_tsangan
60_ansi_tsangan_split_bs
60_ansi_tsangan_split_rshift
60_tsangan_hhkb
60_iso_tsangan
60_iso_tsangan_split_bs
60_iso_tsangan_split_rshift
60_iso_tsangan_split_bs_rshift

[hentaijoe]

Hi, Can someone please provide an update on the status of the keyboard I ordered from Model F Labs? It was Order#27342, for a Ultra Compact F104 Model F Keyboard (Recommended Choice) plus extra parts. Please any and all info would be much appreciated, I just want to know what to expect. Thanks!

[Ellipse]

I just replied to your email. In stock keyboards typically go out within 1-4 weeks as noted at the top of the homepage of the project website.

[nihilist000]

Referring back to the F77 powder coating, can anyone comment on whether it would be a good idea to apply a coat of clear varnish to the front bar of my F77 case to avoid similar wear?

[Ellipse]

I don't know which batch you ordered and whether you ordered the compact or classic style case (both are powdercoated in the latest batch) but the latest batch's (Jan. 2022) updated coating was reformulated and should better resist wear. So far I believe the reports are noting keyboards from earlier batches even though the latest batch of keyboards has been out there for over two years.

I would actually leave it be if I were you; I like the sight of a well-worn bookmark, a professor's old briefcase, etc.! It shows that you actually use something and it is not only something to be displayed!

Touch up paint may be better if you prefer your keyboard to look as close to original as possible. I went to Home Depot to order sample sizes of paint - about $8 each I think. Here's a photo of the codes of each that should give you a pretty close match, but not perfect. The paint employee can custom mix the paint with these codes.

viewtopic.php?p=513650#p513650

Some other folks talked about changing up the color by stripping and repowdercoating their keyboards every now and then, or stripping and then spray painting their keyboard once things wear from usage.

Maybe something like this would offer a similar texture: https://www.amazon.com/gp/product/B074MDTYZF/

[zappazappa]

Anyone have an F15 and is on Linux? How did you make it work?

I'm stuck trying to overcome the two keyboard issue:

https://gitlab.gnome.org/GNOME/gnome-sh ... ote_818548

https://gitlab.gnome.org/GNOME/gnome-sh ... ote_645551

https://gitlab.gnome.org/GNOME/mutter/- ... 3143/diffs

important note - this is not an issue with the keyboard, any two keyboards from any manufacturerer exhibit this problem in every debian-based linux build I've tried.

[Ellipse]

As we have been discussing over email - my comments are copied below. In summary there have been no reports besides this one of the split xwhatsit keyboards not functioning well on Unix/Linux-based systems so the below comments may not be necessary. The current firmware allows sharing of layers and functionality between keyboard halves without any additional hardware or software required, outside of the need to install Karabiner on Mac (no configuration is needed for that program - only to install it). If anyone's split xwhatsit board is working fine on these OSes please do reply with the OS version that you are using.

I wanted to tell you about a project I was considering to combine 2 keyboards into one USB device. The project is called HID remapper and uses 2 raspberry Pi picos and a USB 2.0 hub (not 3.0).

https://github.com/jfedor2/hid-remapper/tree/master

I am copying and pasting my emails and posts for your reference. This may not be accurate or relevant and some posts are out of date from later posts. The stuff about layer sharing through the num lock state was already implemented in the current firmware by NathanA. If you do get this working please let me know so I can share your results and details.

The hid-remapper tool is based around using 2 Raspberry Pi Picos and a USB 2.0 hub, with output of only one device, the HID remapper device, reported to the computer. There is a single pi pico option with the hid-remapper project but it did not work properly. Both keyboards plug into this device and the computer sees it as one device.

Given the trickiness I went through in my testing of the hid-remapper solution involving the PCB, a USB hub, plugging in the second keyboard only afterwards, etc. I think the solution of the software remapping with an unused key that you specified would probably be the one I recommend as easier to set up and use, but I can link to the hardware-based solution for those who are interested in it. The hardware cost is low (2 picos, some wiring, a micro USB cable, a $4 USB hub, and a micro USB OTG cable that is micro USB male to usb female; alternatively my above-linked USB hub includes adapters for micro USB so the OTG cable is not needed). Having to do the series of unplugging and replugging every time the computer is restarted seems not as tenable.

The only oddity for all of the working hubs I tested was that you have to leave the second keyboard unplugged when you are plugging in everything else (the PCB, hub, and one keyboard). Then wait a bit and plug in the 2nd board and everything works fine.

A powered USB hub may eliminate the above trickiness.

The hid-remapper tool is only needed if you want to use function layers. Regular operations such as Ctrl, Command, Alt, and Shift do not require the hid-remapper so you don't need it if you are not using function layers (for example, you don't need it to press Ctrl with your right hand and S with your left hand).

Per NathanA: "You can't program layers that are shared between keyboards with Vial. You would have to do all of your layer mappings within the hid-remapper -- which has its own separate configuration interface (not Vial) -- for any layer actions that you wanted to have shared across keyboard halves. Also, hid-remapper layers cannot trigger proprietary QMK features that are normally activated through special QMK keycodes. For example, if you have a Fn button set up on the right-half keyboard, and you wanted to have Fn+R (with R of course being on the left-half) kick both keyboard halves into bootloader mode, that's not going to be possible. (Not that you would likely want to do that, mind you.) You would need to have separate bootloader trigger keys defined on each half of the keyboard, within QMK/Vial itself, since the hid-remapper has no clue about these QMK features. Or perhaps a better example is that if somebody was crazy enough to install two solenoids, one on each half, and wanted to toggle the solenoids on and off, you couldn't accomplish that through a single key combo...you would have to have two separate combos, one on each keyboard half, and trigger each of them independently of each other."

So you might have to define some layers within Vial, and other layers within hid-remapper, depending on exactly what it was that you were trying to accomplish.

From the project coordinator:
Not all hubs seem to work with it, so you might have to try another if the first one doesn't work.

Perhaps I wasn't clear about what I claim HID Remapper can do. It has its own mapping and layers mechanism, which you can use while having multiple keyboards connected.

It doesn't automatically pass layer state between connected QMK keyboards, it knows nothing about QMK, its layer mechanism is separate. An MO(1) mapping in QMK doesn't actually send anything over USB, HID Remapper doesn't see it as a keypress and it doesn't normally send anything to the other devices connected.

That said I think there are ways to achieve what you want. I said that HID Remapper doesn't send anything to other devices, but it's not strictly true, it can send keyboard LED state (CapsLock, NumLock, etc.). I think this can be used to pass layer state information between the two halves. You'd have to let the layer trigger key be a normal key in QMK, configure HID Remapper to map this key to, say, ScrollLock LED, and also configure QMK to react to ScrollLock LED by activating a layer. This would activate the layer on both halves.

[zappazappa]

Yes, no reports of the issue means people have solved this .... hence I'm here fishing for one of these talented people to show me the way!!

[genericusername57]

Yes, no reports of the issue means people have solved this .... hence I'm here fishing for one of these talented people to show me the way!!

Just skimmed through but from the first link it seems to be an Xorg issue and that it works fine in Wayland? So indeed the best solution would be to switch to Wayland and get a much nicer experience at the same time.

[zappazappa]

Yes Wayland works out of the box, which is great.

I'm on Xorg due to work requirements, but maybe that can be solved. If anyone here has been able to fix this on Xorg, please let me know!

To be clear, this is not an issue with the keyboard, any two keyboards from any manufacturerer exhibit this problem in every debian-based linux build I've tried.

[genericusername57]

I'm on Xorg due to work requirements, but maybe that can be solved.

Booo! Why? Xwayland is a thing, i mean. Lack of wayland support is the main hurdle for me to use *BSD as a daily driver (even though sway works on OpenBSD now, but I'm too old for tiling window managers now). I'm sorry you have to live with Xorg.

[Ellipse]

Progress update:

The factory is finishing up assembly of the F104 and FSSK this month.

The F122 has also completed production, everything except for the inner and outer/packaging foam, so I approved assembly to start. This week I ordered the F122 PCBs and the Leyden Jar controllers. The factory had some issues with their F122 sample even though mine worked fine but they just ended up fixing it by using another capacitive board (one of the sample boards was defective or damaged).

To save time in the past the factory would start assembly with the foam/barrels/flippers (the most time consuming portion of assembly) and they store the inner assembly upside down while they wait for the PCBs to arrive and for the time they spend soldering the controller/ribbon cable/capacitive PCB. Also the Leyden Jar firmware has been fine tuned with some slight updates for the F122 and now it is working well.

2019-04-28 imgpsh_fullsize_anim1.jpg (261.71 KiB) Viewed 8081 times

2019-05-16 May 16-2019 progress (2).jpg (2.33 MiB) Viewed 8081 times

2019-05-29 2019 05 29 fixture to hold unfinished inner assemblies (1).jpg (1.14 MiB) Viewed 8081 times

2019-05-29 2019 05 29 fixture to hold unfinished inner assemblies (2).jpg (846.69 KiB) Viewed 8081 times

[applehugger]

I got my Ultra Compact F104 today and WOW, what a beast! It's so heavy and well-built. Assembly (i.e., putting in the stabilizer inserts and keys) was straightforward, although I read the entire how-to guide on the Model F website several times and watched the entire setup video. I had a few stabilized keys that needed some attention and my numpad enter was spamming keystrokes until I re-seated the spring, but otherwise I didn't have any major issues.

I ended up floss-modding mine. I intended to try it out sans floss first, but from the moment I got the spacebar installed, I knew I couldn't handle it. The ping is just so intense! I don't like the pinging on my Model M, and this seems about twice as bad. However, with the floss mod, I like the sound. I don't think it's amazing sounding like some of my Alps keyboards, but the click sound is very clean and cool sounding. It definitely sounds uber-mechanical.

So far, the key feel isn't blowing me away, but I felt the same way about genuine Topre switches when I first tried them and after a few days, they became my favorite. I like the very clean tactility of the Model F keys, but the weighting is a bit light for my taste. I prefer the weight of the Model M's switches. However, I will have a more well-formed opinion after a few days typing with this keyboard.

Overall I'm kind of astonished that I was able to buy a brand-new Model F in a never before seen form factor. I think the look of the Ultra Compact F104 is awesome, and I think it's amazing that Ellipse was able to pull this all off. I always wanted to try out a Model F, and several years ago, I missed out on a good deal on an F122, but I really don't like any of the original layouts. This is a dream come true and I am seriously considering one of the Classic 104 or 122 key layouts in the future. Major, major props to Ellipse to bringing this project to fruition!

I got mine in black with grey keys. Here are some pictures:

[RedESC]

Kinda Curious,

Did this current batch of keyboards cause any major issues with the tooling used for the f77 with these boards being larger and having more keys? (E.g. tooling to press inner assembly together)

[Ellipse]

Nope, they made the stamper larger than needed. I posted a photo of the stamper a while back.

[wobbled]

Are there any plans to update the f104 case so that the slope is correct and doesn’t make the keys look recessed into the case? Literally the only thing putting me off it

[Ellipse]

Pad print update: Here is the latest factory photo showing a test print. This is not yet approved for production. They still need to fine tune the alignment to make it more consistent; a few keys on one side look to have been double printed and a number keys did not have full contact, such as tab.

Not sure if you are referring to production photos or earlier prototype photos. The keys in the production version do align with the case similar to the F and M originals: https://www.modelfkeyboards.com/wp-cont ... -40-51.jpg

The new F122's inner assembly did sit a little too high so the factory added some metal washers to the previous production sample, just as IBM did by adding a plastic spacer to some early 4704 F62 boards (the new Model F ones use metal as the plastic was prone to cracking).

2024 05 15 pad print.jpg (152.65 KiB) Viewed 6316 times

[Ir0n]

Black caps will be really nice on those red cases.
Those are looking good so far.

[Ellipse]

The next NYC Mechanical Keyboard meetup has just been announced and you can get tickets at the link below (each meetup sells out quickly so pick up your ticket soon!). Due to its popularity the venue has moved to one with a much larger capacity.

https://www.eventbrite.com/e/nyc-mechan ... 8852882437

PC World has posted an article on the NYC Mechanical Keyboard meetup last month.

The last photo in the article shows the new Model F project keyboards table (with an original 122 on it) and me testing out a nice original IBM F122 that someone else brought to the meetup:

https://www.pcworld.com/article/2303216 ... mecca.html

[engr]

Now that some manufacturers are offering reverse dye sub keycaps, could this become an option for Model F?

[Ellipse]

Unfortunately the dye sublimation tooling custom made for the new Model F project cannot sublimate the entire keycap; it can only sublimate the top. A separate jig was made to do the front sublimation at a different time from when the top is sublimated. Given the height of the keycaps and available equipment currently out there, there was no good quality way to sublimate the top and front at the same time, so it is unlikely that the entire key could be sublimated at once with the current equipment. If anyone has videos of an entire key set being sublimated with the entire surface area please do share as it may offer some inspiration, though the cost may not be feasible.

[engr]

Here you go. There is even a company, YuzuKeycaps, making custom reverse dye sub sets for only $100 a piece (Keyreative is their manufacturer). They only offer Cherry profile sets, but Keyreative has done other profiles before.