deskthority

The solenoid isn't stuck--if I keep pressing the toggle key it will click away happily. I don't think there's a key stuck down on the keyboard--I tested them all several times after setting up-- but it's an interesting theory: because of nkro it may confuse the software but the keyboard may still appear to be operating normally even if F22 is stuck down, etc. I'll check again when I get back to the office, and I'll put the 'scope on the solenoid output and see if it's trying to pull it.

If that fails, I'll hook it up to my other ModelFKeyboard and see if it works there--that would indicate a problem in the LeydenJar board or the software.

Meanwhile I finally implemented media keys support in my firmware… only to find out that Windows doesn't properly support the optimized USB HID report I used for them. (I put the media keys into the reserved byte of the boot protocol -compatible report, which Mac and Linux support just fine, as they should, but Windows simply ignores it so I ended up just assigning those keys to F19–F24 and mapping them to the media keys in Windows registry with SharpKeys.)

At the same time, I think I might have finally come up with the "final" layout for the F50 (at least until I decide to change it again). I originally had the F-keys set up quite differently (four down on each column, to keep the original Model M -style groups of four), but it never felt quite right. Now I have them arranged in the same way as the numpad next to them, with F1 to F9 in the same positions as the numpad 1 to 9 keys.
It looks quite nice next to my F62 (seen here in my more typical night owl mood lighting):
(On the F62 I gave up on trying to learn to live with a split backspace and put the 2U backspace back in that space. Meanwhile living with the split ISO enter has been no problem at all: apparently I always press that small lower part of enter anyway.)

Takashi wrote: 08 Apr 2025, 08:24Switching the keys is a good idea. Only problem is that the keyboards are 800km apart from each other (using them in two different locations).

Where in Sweden are you exactly? I have a bunch of spare keycaps you could borrow if you want to (random alphas and modifiers like numpad, F-keys etc). Depending on how many keys you need to check with I guess, I don't think there's enough to outfit an entire keyboard using these spares.

Ellipse wrote: 08 Apr 2025, 04:46 Yes, as I mentioned over email, it is expected that the uncoated original style paint will wear where it is rubbed during contact. To keep the project's goal as a reproduction of the IBM originals, the paint has no glossy automotive style protective finish.

Several years ago I changed the paint formulation for a less historically styled, tougher modern powdercoating which currently has an excellent track record of little to no wear (this formulation was first used on the die cast compact case F62/F77 cases, while the very first production run from 2019 used anodized cases for the compact cases). The original F62/F77 project's remaining keyboards were not discarded and will probably sell out over the coming months.

We discussed your issue with some keys having chipped stems at the bottom and those keys being the ones that required more careful on-center pressing - I don't think you caught all of them from the originally damaged batch but I noted that I am happy to replace the broken keys with the order you were placing. Broken keys should not be used.

Why not switch all the keys from one keyboard to the other to see if what you call the scratchy keys are due to the keycaps or due to the barrels? I suggest using them for a month or so before evaluating.

Thanks for the info! I'll make sure to order the spare keys!
Is there a way to know whether which paint formulation my keyboards are using?
I made an earlier post about it, it seems the texture of my paint coatings is vastly different between the two keyboards:

The upper is my "old" keyboard (serial number 6xx) and the lower one is the new (serial number 9xx).


Switching the keys is a good idea. Only problem is that the keyboards are 800km apart from each other (using them in two different locations).

So I have two F77 that I've had for a while. One almost three years and the other one almost 2 years.
One came in very good condition while the other one came quite brutally beaten up by the shipping. The keyboard itself was quite intact, but a big bunch of the keycaps and springs were crushed but Joe was kind and sent replacements for these.
Now that I've used the beaten up keyboard a while, it doesn't "feel" as good as the first one.
I guess all the springs and caps were slightly off. It's been almost 2 years now so I just have to accept it I guess. The keys on the second keyboard are just more "scratchy" and not as smooth as on the first one. The left shift for example I have to press in the center for it to actuate. If I press it in the corner, it doesn't want to go down.

I've noticed that the color now gets loose (discoloring) on both of them. It's on the place where my hand is resting.

Keyboard #1


Keyboard #2


Not living in the warmest or most humid of places (Sweden).

Anyone has suggestions of what can be done? Don't feel like doing a total repaint of them. Maybe someone has some good ideas as I cant' be the only one with this problem?

Irving wrote: 26 Feb 2025, 19:46 I like your caps lock macOS mapping proposal. Long term I would prefer not to give up the Caps Lock functionality.

Couple of updates on this.

First, from the reading I've done so far, it sounds like mapping Caps Lock to Fn is a feature that only appeared in MacOS as of the most recent version (MacOS 15 "Sonoma"). So you won't see that option unless you are fully up-to-date.

I have an old Mac that is stuck on MacOS "Catalina", and so it is not possible for me to re-map Caps Lock just by going into the Keyboard Settings panel. However, I did download and install Karabiner-Elements, and not only can it remap Caps Lock to Fn, but it can remap *any* arbitrary key to Fn, even keys that don't normally act as modifier keys. So if you install Karabiner, you can pick some key that isn't normally present on a Mac keyboard (like Ins), use Vial to map Ins to the key on the FSSK that you want to act as Mac Fn/Globe, and then use Karabiner to remap Ins to Fn. In my testing, it works perfectly.

Karabiner is a well-respected open source Mac project that has existed for eons. I would not hesitate to recommend that others install and use it. It will solve your problem in under 5 minutes, no keyboard firmware update necessary. That said, I'll still continue to look into how to bring native Mac Fn support to the firmware, but the lack of that support should not prevent you from using your keyboard the way that you would like.

Irving wrote: 26 Feb 2025, 19:46Your proposal to have QMK and Vial in lockstep sounds great to me [...] I have not even thought of it and never looked beyond QMK.

Just to be clear (since the naming can be a bit ambiguous), Vial can refer to either the keyboard firmware, or the app that runs on your PC that you use to configure your keyboard. Vial (the firmware) itself is based on QMK. So QMK still sits underneath the covers, constituting the bulk of the firmware. But you do need to have Vial support added to the QMK firmware, in order for the Vial *app* that runs on your computer to be able to talk to the keyboard and program it. The Vial app cannot just work with any old QMK-based keyboard.

I don't maintain the stock QMK source tree for xwhatsit-based keyboards...that's other people's wheelhouse. And frankly I have no interest in maintaining QMK-only firmwares. So any work I release will be Vial-enabled. But I am also not the *author* of Vial, and have no control over the development of the app itself. So my point was that there are a lot of components at play that are outside of my control...I can "propose" whatever I want, but I can't dictate anything, and my proposals might just end up being no more than hot wind. If the Vial app is not aware that the Mac Fn key is even a thing, then it is going to be cumbersome at best to use Vial to configure that key, and worst-case it might not be possible to harmonize the two until both the QMK project allocates an official QMK keycode enum for the Mac Fn key, and then afterward the Vial project separately adds the chosen keycode to an updated version of their app. I could theoretically take the Vial app's source code, and patch in support for a new keycode myself. But then I am forced to maintain a separate fork of the Vial app, too! ...and I'd rather not, if I can avoid it.

As far as using base QMK on your FSSK, it might be possible, depending on what version of FSSK you have. If you have the first version that came out with the "modern" styling, then non-Vial-ified QMK firmware does exist for that. But if you have the second version that came out with the classic Model M styling, as far as I am aware, my Vial firmware is the only firmware that exists for that keyboard model. Nothing's stopping anybody from taking my work and backporting it to plain-jane QMK, but I don't believe anybody has yet, and I'm not going to take the time to do so myself.

Irving wrote: 26 Feb 2025, 20:02 Under ‚Private Messages‘, all old PM are under ‚Sent messages‘ whereas newer ones are under ‚Outbox‘. Do they not even get sent any more?

If a PM that you sent is sitting in Outbox, that means the recipient hasn't opened it yet. If it's in Sent Messages, then they have.

Irving wrote: 10 Feb 2025, 18:27 I recently got my FSSK (ISO-DE) delivered and now I am facing reality in the sense that macOS is not fully supported in the sense that it acts as if it were an Apple keyboard. [...] ...the latest firmware is based on NathanA's work (I tried to reach him, but to no avail.):

I haven't been able to hang out here on a regular basis recently, and Ellipse alerted me to your post. I was going to ask how you tried to contact me -- you couldn't have tried very hard! -- since I scoured my email inbox for any messages from anybody related to this, as well as all of my spam traps, and found nothing. Upon logging in to DT to reply here, I found you had PM'd me only once, back on Jan 7. FYI, the email notifications of PMs feature on the DT forums has been broken for a couple of months now at least, so the only way I would ever know that I had been sent a PM is if I happen to visit DT and log into my account. A PM no longer reliably generates an email notice to me. I do publish a direct contact email address on the website, but seemingly nobody ever thinks to use it...

Aaaaanyway... </digression>

I had read up on the history and technical details of the Apple "Fn" (and now Globe) key in the past. I went back and refreshed my memory on all of that, and also see that quite a bunch of new stuff has transpired since.

In short...it's a royal mess. (Thanks, Apple.)

I hate to be the bearer of bad news, but my read on the situation, without going into too much detail, is that there is still no good solution to this problem. Every option that's been presented has compromises, which will cause just about any third-party keyboard's Fn/Globe to not work quite the same way as it would on a genuine Apple keyboard. And if the problem were approached the way that the first blog author did, adding it as a user/custom code like that bring along its own limitations (cannot be used in combination with certain other QMK features like Layer Tap/LT(x), which may break expectations).

The firmware can certainly be patched to add it as a native keycode rather than as a user/custom one, which removes those limitations, and others have done exactly that. But they have only done it to raw QMK, not to VIA/Vial firmwares. The Vial app is not going to know anything about that keycode even if the firmware itself does, and in fact we would have to be very careful with picking a free code to use, or else Vial key mappings could get screwed up. Ideally, the QMK authors and the author of the Vial app would both officially update their respective projects, so that there was agreement on both sides. But that hasn't happened to-date, and will probably take a while.

I will look into this and play around to see if I can come up with the "best" set of compromises, but it won't happen overnight and will probably take a little while to sort out and test.

It sounds like, though, that there are alternative ways to approach this problem that don't require any firmware modifications to the keyboard...have you looked into any of those?:

• Assuming you don't need a Caps Lock...apparently MacOS will let you remap the Caps Lock key to act as Fn/Globe. So you can use Vial to map Caps Lock to the key that you want the Globe key to be on, and then ask MacOS to use Caps Lock as Fn/Globe, and ta-da! (And if you do need Caps Lock functionality, this could probably be simulated by crafting a Macro that auto-applies QMK Key Lock to the 'Shift' key. Though it looks like KEY_LOCK needs to be enabled during compile time, I haven't been doing that, and I don't know if there is enough room left to squeeze that feature in or not...hrm...)
• Although I have not yet found solid confirmation of this, it sounds like Karabiner-Elements can arbitrarily remap virtually any key to Fn/Globe. So, you could install Karabiner, pick a key you will never use (e.g., Insert), map that keycode to the key on your 'board you want to use as Globe, configure Karabiner to treat that key as Globe, et voila. Only downside is you have to install a third-party app on your Mac.

Either of these options seem to me like they are going to get you the quickest results.

One final note: regardless of what you do or how the firmware changes in the future, one thing that you canNOT do is to overload the function of the QMK/Vial layer-switching key that we have colloquially called "Fn" around here, to ALSO act as the Mac-native Fn. "Fn" in the context of QMK/Vial means something wholly different than what Apple means by it. To retain both features, you will need to dedicate two separate keys: one for each. I suppose it could be argued, though, that on either a FSSK or F104 board, there are enough physical keys that you can probably get away with only a single QMK layer, and be perfectly content, in which case you can just re-use whatever key is currently programmed to be QMK "Fn" (layer 1) to instead be the Apple Fn key...

Oelmuvun thank you for sharing your experiences with the problem.

The reason I went for the dry lube was to try to avoid any grease/oil attracting dirt/debris since the surfaces are exposed to the outside of the switch mechanism (unlike the rails inside an MX style switch). Have you found either of the lubes you tried to have attracted any gunk? How long have you used it for? I lubed my Alps white switches' sliders and rails with the same (literally) dry lube 9 years ago now and it has been flawless despite heavy use to the point where the keycaps are worn and shiny and nearly a decade of dust, crumbs and cat hair. I'm hoping lubing the stems yields similar results with the F104 unless there is a new batch of deburred barrels released or someone else discovers a better solution. Without doing anything to address the problem, I wouldn't be surprised if the keyboard felt too shitty and annoying to use after 5 years (which would be a damn shame).

I guess I did contradict myself between the two posts. It all depends on the situation I suppose. What I'm trying to say is if the scratchy feeling is going to follow anything, it seems to follow the stems once they are damaged. Once a stem is scratchy, it seems to be scratchy in its original barrel, always. Moving it to another barrel is SOMETIMES scratchy, sometimes not. Moving a "fresh" stem into a previously scratchy barrel never seems to be scratchy. (after I wrote the rest of this post, it makes even more sense).

I'm unclear on the intent of why the keys need to be washed. Is this to wash off possible debris that you think may be causing the issue, to wash off the dry lube, or just that you don't want someone's "dirty" key caps back? There are only 10 keycaps on the board with dry lube applied, most of which are from the icon set anyway. I'd rather take the time to wash not wash every other keycap on the board if I don't have to. I'm not trying to be argumentative, I'm just not sure what we're getting at here.

I checked the springs and the end of the springs point towards the top of the keyboard.

All that being said though I think this is all a moot point after discovering some more information.

Because the mechanism is impossible to see once it is in the keyboard, I pulled out a spare barrel from the first aid kit to line up the wear pattern on the stem with the position on the barrel, and I think I am almost positive now as to what is causing the wear. There is a burr inside the bottom of the barrel damaging the stems.

The wear pattern on the stems aligns precisely with the "rim" of where the barrel meets into the bottom ~1mm thick part just before it opens up to the area where the flipper goes. You can see it (and feel it) from the underside easier. There is an ever ever so slight ridge/burr protruding out. To the naked eye it looks like a sharp edge, but with the tip of a needle or screwdriver, you can feel the fine point catch on the burr. When the stem is pressed slightly skewed in the barrel, it rubs against this protruding ridge and this is what causes the wear on the parts that touch, the corner.

To try to draw a diagram of it, it would look something like this:
This correlates with what I'm seeing inside my F104 barrels, you can see right along that that ridge looks even more white, with what looks like worn stem material around it. If anything needs to be deburred, it would be the barrel itself, but that is impossible without removing every single barrel and doing something with that edge.

Take a look at the original IBM F122 barrel, that burr is not there. The shape is the same, the barrel goes down, there is a bevel, and then it goes straight down again to the bottom of the barrel, but the transition is perfectly smooth with no burrs. Pull out a few new barrels and compare them against any old Model Fs you may have. Check and feel that edge, I bet you'll find that it is there on the newly manufactured parts.

To me this seems to be a pretty definitive explanation to what is going on here. The burr on the harder fiberglass reinforced barrels (or whatever they are) is scraping against the softer PBT causing the scratchiness and wear. If the barrels are damaging the stems, that also explains why replacing replacing the keys just causes the problem to come back over time. It also explains why the "scratchy stems" that have been damaged don't "scratch" in the IBM barrels, because that burr isn't there for them to rub on.

Only you know exactly how the factory manufactures those barrels, but I would take a close look at whatever tooling or process determines what how the bottom edge of the barrel is manufactured and see if something can be done about it. The end result is definitely different than the way the original IBM part though.

Your earlier post noted "The biggest difference I noticed between the keys though seemed to be in the "corners" of the barrels. I'm really not sure what we're looking at there though. Is that is wear in the barrel plastic or stem material that has rubbed onto the barrel?. Dust? Nothing? Swapping in a new stem "fixes" the problem (until it would wear again) though. After swapping my scroll lock and T keys, the T barrel with the scroll lock key is perfectly smooth, while scroll lock with the worn T stem feels the exact same, scratchy. So in other words, the feeling follows the stem."

If the feeling follows the stem in different barrels, would that be different from what you noted above?

Again, without replacing the keys we won't know if it is an issue of some bad keys from the injection molds or some other issue.

It seems like letting the keys soak in warm soapy water for an hour, then rinsing them two or three times with non-soapy cold water may be ok. The lube may have to be removed with a hand towel.

For now I recommend that I just replace the main US key set. In a little while from now we will know if it wears the same with your particular usage.

Also have you checked that all springs match the updated photo in the manual, with the keyboard held vertically, space bar end up? Maybe springs in the non-ideal position may contribute to keys not working well over time.

Have you tried burnishing the back stem area with your fingernail with the process noted in the manual? That does smooth out keys in my experience.

Ellipse wrote: 06 Feb 2025, 21:32 After reviewing your additional notes and photos, I can't really diagnose whether there is an issue from here so it would be great if you could wash all the keys (soaked in water, not only wiped by hand) and send the full set back to me for testing. The parts must have all lube and debris removed. (Ultrasonic bath cleaning with dish soap would be best, if you have it)

I will send you a new set as a replacement. Please email me the tracking number after you have sent it back, and let me know if you have any questions with the above steps.

I also suggest cleaning out the barrels with a cotton swab / Q tip and 99% rubbing alcohol before reinstalling the new set. I have found that this helps Model M keyboards as well to improve smoothness.

However I do think that this is just something to do with the typing style you noted, pressing on the edges of the keys, that is more evident with the new production keys which may have slightly tighter tolerances in terms of space between key stem and barrel compared to your particular F122. As I mentioned, the keys should be more difficult to press on the edges than in the center, based on IBM's design, and I'd imagine everyone else has not noticed such an issue because they don't type near the edges of the keys. From what you have described, this is either an issue with one bad key set or due to abnormal usage patterns, so when I receive your set back I should know one way or the other.

Ellipse,

This is going to be complicated. I don't have an ultrasonic cleaner, so I can't ultrasonic clean them before sending them back. Secondly, at least initially, the keys only seem to be scratchy in their usual barrels, so I doubt you'll feel the same thing i you install them in a keyboard and try it out. However, you would be able to get a better look at them. Swapping the keys around or moving them to my F122 does not cause the same scratchiness. Third, my keyboard is a combination of standard keys, Icon keys, keys with lube (large keys around the edges) and keys without. Instead of trying to explain it, here's a picture of the current state .

With that in mind, tell me what you do and don't want back. I highly doubt though a new key set will behave any differently over time though if they are ultimately all the same. The same issue will likely just resurface after a few months of daily use.

Regarding some more photos and some more information on the wear:

Upon close inspection of the keys, there are 4 distinct spots that are wearing on all 4 key corners. The wear is occurring on the "corners" of the stem from the bottom of the step up to several mm from the bottom of the keycap. You can see that outlined the photo below in red on a bad key, green on a good key.
While the keys with the dry lube are not scratchy at all, you can see that same contact/wear patch happening easier on them because of where it rubs off. I would be very curious to hear from any other new Model F owners see this same wear pattern on their heavily used keys. If you can, please pop a few keys off and have a look for me. Surely it can't just be me. I double checked and found that this wear pattern is not occurring at all on my F122 keys even after 40 years despite looking nearly identical. Blaming it on my "typing style" is as absurd sounding as Apple telling people they were holding their iPhone 4s wrong back in 2010. If there is a manufacturing problem, materials issue, or tolerance problem, then that is what it is is and should be addressed as such. And if there is one, then it should affect everyone with that same batch of components equally. Whether they notice it or not is inconsequential. Either the wear is occurring prematurely or it isn't.

I type on old Alps keyboards, Model Ms, Original Model Fs, A handful of different Cherry MX switches all the same way and this seems to be the only keyboard that has a problem with it. I type quickly (110wpm) but surely my typing style along with resonators's isn't all that unique compared to everyone else. And to be clear, I'm not typing at the edges of the keys all day. The issue is that the heavily used keys become scratchy and have more resistance than the non-scratchy keys. When typing quickly it causes missed letters because those keys bind and don't press down all the way. That's obviously not the case with ctrl, alt, tab, those just feel bad when pressed slowly. The feeling is exactly like typing on a dirty Alps keyboard on those keys when they are pressed in just the right way. Chryrosran22 knows what I'm talking about.

I think you misunderstood about the teflon. I applied it to some keys because they were sticking after some time, that was not the cause of it. The keys right now that are not sticking are: the ones that are infrequently used and the ones that are frequently used with the Teflon. What that does to them does over time remains to be seen I suppose, but what I can say for certain is that keys without lube, exactly as they came from the factory, start to get scratchy over time on this F104.

I don't have a microscope to examine them very closely, but i took some keys off and looked at them today under a bright light and tried to photograph them. Honestly though, there isn't much to look at. From what I can tell, good keys don't really look different than bad keys. And the new F104 doesn't really look any different than the old F122, but they feel different for sure.

The biggest difference I noticed between the keys though seemed to be in the "corners" of the barrels. I'm really not sure what we're looking at there though. Is that is wear in the barrel plastic or stem material that has rubbed onto the barrel?. Dust? Nothing? Swapping in a new stem "fixes" the problem (until it would wear again) though. After swapping my scroll lock and T keys, the T barrel with the scroll lock key is perfectly smooth, while scroll lock with the worn T stem feels the exact same, scratchy. So in other words, the feeling follows the stem.

There are some marks on the sides of the stems, but I think that's just a red herring. That doesn't actually contact anything. I'm sure you understand the mechanism better than I do, but it seems that the 4 corners of the stem are what actually contacts the "corners" of the barrel and that's where the wear is occurring, particularly when keys are pressed off-axis.

Since I was swapping things, I swapped some keys between my old and new boards with some interesting results:
Swapping the scratchy F104 "T" key into my original F122 barrel is still super smooth (surprisingly).
Swapping the original F122 T key into my F104 barrel is also smooth.

Also of note, the old F122 barrel doesn't visually look at different that I can see, yet the interaction between the stems and barrels is somehow different. Despite the F122 being heavily used, dusty, and old, none of that seems to adversely affect the key feel in any way, it still manages to be smoother.

Just a heads up, the bending trick outlined in the manual and the video solved my problem. Thanks to all that helped. After, I started to have issues where the Command button (I am on a Mac) would only work sometimes. I reset and reset over and over and same issue. The key sounded fine too. Now I normally have my keyboard plugged into the back of my monitor which then connect vis USB C to my MacBook. I instead plugged it directly into the MacBook and the Mac then asks if I would accept this new device. Yet, I have been using this keyboard for well over a year.. Odd that it asked me that. Anyhow, I accepted and all keys work perfectly.

Hello everyone

I have an F77 and recently the spacebar is sticking. It sticks if I push it on the left side. If I push on the middle or the right, no issues. I had this problem during the summer and it went away on it's own.. But it appears to be back again.

Any ideas what I can do to fix this ?

Thanks
J

resonator wrote: 03 Jan 2025, 11:29 A few pages back I was trying to resolve issues with scratchy keys. I tried all sorts of experiments but I couldn't fix it. I improved it mildly by swaping caps, but I had no solution.

Today my replacement caps and barrels arrived so I replaced the stratchy keys, which of course fixed it. Inspecting the inside of the barrels showed a fair amount of wear and visible scraping on the inside (even those which I didn't experiement on). I can't say the cause, but I think those keys were always scratchy and worsened over time.

I'm not saying it'll always be the case, but I think the only remedy to my problem was to replace the barrels. I replaced the caps as well, but it's possible that wasn't needed.

I had the same thing as you resonator on the scratchy keys, especially on left ctrl, left tab and somewhat on caps lock. I found that just a drop of Teflon Dry lube on the outside of the stems helps immensely much like on Alps stems. I posted this in a comment on Chyrosran22's video. Ellipse mentioned in the comments there that he has had no reports of this and to follow the setup guide's steps to fix this. Nothing in there helped me resolve this though. No amount of burnishing, wiggling, bending, or breaking in made it any better. Like you said, it just got worse over time

Just wanted to share the dry lube tip since it helped me.

A few pages back I was trying to resolve issues with scratchy keys. I tried all sorts of experiments but I couldn't fix it. I improved it mildly by swaping caps, but I had no solution.

Today my replacement caps and barrels arrived so I replaced the stratchy keys, which of course fixed it. Inspecting the inside of the barrels showed a fair amount of wear and visible scraping on the inside (even those which I didn't experiement on). I can't say the cause, but I think those keys were always scratchy and worsened over time.

I'm not saying it'll always be the case, but I think the only remedy to my problem was to replace the barrels. I replaced the caps as well, but it's possible that wasn't needed.

wobbled wrote: 30 Dec 2024, 17:31

tron wrote: 30 Dec 2024, 17:11

xxhellfirexx wrote: 29 Dec 2024, 22:12 This means all knowledge in recreating these keyboards will be lost to history once again.

IIRC Ellipse had mentioned years ago that when these projects are complete he would write a book detailing the production process and lessons learned from the success and failures (I don't know if this is still on the table but recall a post about it). I think it's only fair to point out the risk involved with taking on a project like this, Ellipse has put in countless hours to setup a sustainable resource network to service these boards into the future and used profits to get parts dialed in via trial and error. There is always risk with group buys, it's common to have some minor imperfections that need touch up with boards in this price range. Also, the used market is quite expensive and IMO carries more risk when buying an older F or Beam from Ebay.

Except I don’t think these projects will ever be complete while he still profits from them and i don’t recall him ever saying this. Again, I have no issues with the dude making a profit, ellipse deserves something for the years he has put into this.
But that doesn’t mean he should cut corners to save himself money when it comes to customer support - if you put out a shit case with dodgy paint, either offer to replace them or partial refund them to pay for a proper paint job - dont just recommend a fkn paint pen which will never match a powdercoat job.
When he changed the paint for future cases, original customers should’ve been offered the newer cases either severely discounted or free. If there wasn’t a problem, the formula wouldn’t have been changed.
Once again, for this project to have even happened, he relied on valuable resources from other DT members. Now that these DT memebers need his support, his answer is to PM them claiming harassment and effectively telling them to F off.
If you can’t see through his assholery then please go get your eyes checked.
I also can’t stress enough that I’ve bought thousands of dollars worth from ellipse in the past both repro and original keyboards, so it’s not like I’m blindly hating on the guy here, I’ve just seen way too many valid complaints be shut down by him in the past, all with the aim for him to save a buck or two. Its disgusting.

Of course it'll never end. It's been "get your orders in by end of $currentMonth!" for years. It will never change, and it creates a false sense of urgency so that people who are on the fence just end up screwing themselves.

tron wrote: 30 Dec 2024, 17:11

xxhellfirexx wrote: 29 Dec 2024, 22:12 This means all knowledge in recreating these keyboards will be lost to history once again.

IIRC Ellipse had mentioned years ago that when these projects are complete he would write a book detailing the production process and lessons learned from the success and failures (I don't know if this is still on the table but recall a post about it). I think it's only fair to point out the risk involved with taking on a project like this, Ellipse has put in countless hours to setup a sustainable resource network to service these boards into the future and used profits to get parts dialed in via trial and error. There is always risk with group buys, it's common to have some minor imperfections that need touch up with boards in this price range. Also, the used market is quite expensive and IMO carries more risk when buying an older F or Beam from Ebay.

Except I don’t think these projects will ever be complete while he still profits from them and i don’t recall him ever saying this. Again, I have no issues with the dude making a profit, ellipse deserves something for the years he has put into this.
But that doesn’t mean he should cut corners to save himself money when it comes to customer support - if you put out a shit case with dodgy paint, either offer to replace them or partial refund them to pay for a proper paint job - dont just recommend a fkn paint pen which will never match a powdercoat job.
When he changed the paint for future cases, original customers should’ve been offered the newer cases either severely discounted or free. If there wasn’t a problem, the formula wouldn’t have been changed.
Once again, for this project to have even happened, he relied on valuable resources from other DT members. Now that these DT memebers need his support, his answer is to PM them claiming harassment and effectively telling them to F off.
If you can’t see through his assholery then please go get your eyes checked.
I also can’t stress enough that I’ve bought thousands of dollars worth from ellipse in the past both repro and original keyboards, so it’s not like I’m blindly hating on the guy here, I’ve just seen way too many valid complaints be shut down by him in the past, all with the aim for him to save a buck or two. Its disgusting.

genericusername57 wrote: 30 Dec 2024, 02:03
Again, given how many people that sit on the knowledge that Ellipse used to build these boards, how easy it is to coordinate chinese factories to make them, and how much money Ellipse is making doing this I just don't understand why you're not selling the same product but cheaper and with proper customer support? You're clearly upset enough about this to keep spamming the thread asking for answers you know you'll never get so why not just undercut him at his own game and deliver a better experience? Everyone would love it, and you'd be rich. If everything Ellipse knows has been handed to him by someone else on this forum, just ask that person? I don't understand what the issue is.

Am I really the only one who strongly suspects that the reason Ellipse is discounting his old stock so heavily is because he's earning fuck all from this and is deep into the red trying to make this a profitable business? If that would be the case, it's not really surprising that he's keeping whatever "secrets" he's got tight to his chest.

It's as if you multiply the number of sold keyboards and subtract the material cost and assume everything else has been free, such as loads of prototypes being sent across the world and rejected. Do you think that was pro bono from China?

I might be wrong, obviously, but I'd bet a lot of money that your view of Ellipse as the guy becoming a millionaire from the work of others and refusing to give back is false. I don't think his obsession with asking for reviews and trying to spread the word is to increase profits, I think it's to create them in the first place. That's why I accept some, let's say "non-european" business practices and still pay money to receive the product knowing I can't just hand it back to someone like logitech if a key is not seated correctly when I receive the thing. I think this is a fleeting moment where you're able to buy these keyboards factory new.

I'd be happy to buy these products from anyone really, but somehow only one person is actually selling them...

A very interesting reasoning...

You may remember the original phrasing of my question (focus on the "why"):

(Give and take) why don't you please share the information you hold (CAD, springs specs/source, and how to purchase from China) with the community to help in producing the flippers in the same way the community has done with the critical components you have (and still are) using in your products?

A valid answer would be: because it would put at risk "Model F Labs LLC".

This would have clarified that this post is not a Group Buy. It is a marketing/support/maintenance/information... in summary... a commercial thread. And my point is that DT should be a place to share, not a place to provide free information and support to businesses. "Model F Labs LLC" has already a website for this purpose: https://www.modelfkeyboards.com/

Once again ... give and take vs. sell an take.

I have no problem with companies making profit. I have problems from companies which just make profit.
My moral dictates that if you take something from a free place, you should try to offer something back for the benefit of the community. Coming back to the original phrasing: "in the same way"

dcopellino wrote: 30 Dec 2024, 01:59

idollar wrote: 29 Dec 2024, 10:52 Dear DT,


So this time I am not addressing you, Ellipse. I address all other DT members with my question:

Do you share my view ?
I mean are you fine with the way the forum has evolved by one single member ?
Are you find with a forum in which members take information, use the forum as their marketing and service platform without giving back in the same way that they have taken ?

I definitively don't agree with this approach.
I personally feel unfair to use DT to take information and work from others and give nothing.

Let's make it clear: I have no problem with the profit of 6000 keyboards. That's perfectly fine with me.
My point is and has always been sharing information and misusing (in my opinion) DT.

-- PD : I am sure that a long post will follow hiding this behind

Like many other DT users, I am also following this discussion, which apparently has very personal and direct tones. At first, I watched it from the sidelines without intervening, although I feel involved in some way, having been greatly inspired by the valuable contributions of i$ to this community on capacitive keyboards, of which I am a great enthusiast. It’s true! I played with transforming a Model M into a Model F, following the tutorials posted by i$ called FEXT and FSSK. Inevitably, on a microscopic and individual scale, I put myself in Ellipse’s shoes, playing with reproducing the legendary Model F but equipping it with a modern layout typical of the Model M. I have inevitably drawn enormous benefits from the numerous contributions that many users of this community have provided on both the hardware and software levels to make this project something concrete, without which everything would have remained just a theoretical dream.

Hello dcopellino,

A very good introduction. I really like your summary.

Let me comment on the bold text above: "which apparently has very personal and direct tones".
The conversation may have become somehow very personal and direct because the information which I (we) needed was kept by one single member. Also, to my best knowledge, this DT user is the only one which is using the forum as a "personal" marketing & support platform.

dcopellino wrote: 30 Dec 2024, 01:59 The actors involved in this project are several, and I am sure that, even having followed all their directions step by step, I will forget to cite all sources, starting from the masterful work of Tom Wong-Cornall, who laid the groundwork that many others then further expanded and updated, starting with wcass, pandrew, kmonov, listofoptions, dma, up to the latest developments from NathanA, Rico, and many others whom I will not list here due to my fault and ignorance.

Given this premise, as a simple user of this community, I wonder and ask - to the directly involved protagonists, both cited and not: - have you economically benefited from your valuable and necessary contributions? Otherwise, I would be quite shocked to learn that there are people capable of economically exploiting the work of others without recognizing the fair compensation. Please respond. After all, you are the only ones entitled to give meaning to this discussion, shedding light on the moral value of the people involved regarding the dilemma of take and give back prompted by I$, who started the thread. Thank you.

I will break the ice with your very interesting question:

The term "FSSK" and "FEXT", were proposed by myself some years ago. DT accepted the term which was never registered.
"Model F Labs LLC" used it to market their new product without any asking DT (nor myself).
"Model F Labs LLC" did not mention the root for the term until I complained.
"Model F Labs LLC" has never payed any royalties

In my personal case, there is no need to share any profit. Not at all.
To be 100% direct and honest: I complained because I did not like to see the FSSK term used as it creates confusion.

From the different posts which I read I can guess that the "Model F Labs LLC" is not sharing their profit.
I may be wrong ... "Model F Labs LLC" may want to contribute further.

idollar wrote: 29 Dec 2024, 10:52 Dear DT,


So this time I am not addressing you, Ellipse. I address all other DT members with my question:

Do you share my view ?
I mean are you fine with the way the forum has evolved by one single member ?
Are you find with a forum in which members take information, use the forum as their marketing and service platform without giving back in the same way that they have taken ?

I definitively don't agree with this approach.
I personally feel unfair to use DT to take information and work from others and give nothing.

Let's make it clear: I have no problem with the profit of 6000 keyboards. That's perfectly fine with me.
My point is and has always been sharing information and misusing (in my opinion) DT.

-- PD : I am sure that a long post will follow hiding this behind

Like many other DT users, I am also following this discussion, which apparently has very personal and direct tones. At first, I watched it from the sidelines without intervening, although I feel involved in some way, having been greatly inspired by the valuable contributions of i$ to this community on capacitive keyboards, of which I am a great enthusiast. It’s true! I played with transforming a Model M into a Model F, following the tutorials posted by i$ called FEXT and FSSK. Inevitably, on a microscopic and individual scale, I put myself in Ellipse’s shoes, playing with reproducing the legendary Model F but equipping it with a modern layout typical of the Model M. I have inevitably drawn enormous benefits from the numerous contributions that many users of this community have provided on both the hardware and software levels to make this project something concrete, without which everything would have remained just a theoretical dream.

The actors involved in this project are several, and I am sure that, even having followed all their directions step by step, I will forget to cite all sources, starting from the masterful work of Tom Wong-Cornall, who laid the groundwork that many others then further expanded and updated, starting with wcass, pandrew, kmonov, listofoptions, dma, up to the latest developments from NathanA, Rico, and many others whom I will not list here due to my fault and ignorance.

Given this premise, as a simple user of this community, I wonder and ask - to the directly involved protagonists, both cited and not: - have you economically benefited from your valuable and necessary contributions? Otherwise, I would be quite shocked to learn that there are people capable of economically exploiting the work of others without recognizing the fair compensation. Please respond. After all, you are the only ones entitled to give meaning to this discussion, shedding light on the moral value of the people involved regarding the dilemma of take and give back prompted by I$, who started the thread. Thank you.

Dear DT,

We already have an answer from Ellipse in the form of a Private Message ...

[attachment deleted by mod]

I am sorry if Ellipse felt harassed. It was not my intention. I do not think I have ever harassed anyone, but just in case I publicly apologise in case someone believes I did. As I said, it has never been my intention.
I just wanted to progress on a personal project and in DT people is common to ask for help. The point is that I want to built on top of the knowledge shared in DT, as we have always done. Ellipse has the knowledge and being an active DT member, I thought that he could help.

Now we know the answer: Ellipse is not going to share information. He will keep on asking for information (take) but he does not intends to give back.
An example of this attitude can be found in a recent post (https://www.deskthority.net/viewtopic.p ... 21#p521821)

The above was posted in the thread in which I asked for his help. Yes, he would only write to take.

So this time I am not addressing you, Ellipse. I address all other DT members with my question:

Do you share my view ?
I mean are you fine with the way the forum has evolved by one single member ?
Are you find with a forum in which members take information, use the forum as their marketing and service platform without giving back in the same way that they have taken ?

I definitively don't agree with this approach.
I personally feel unfair to use DT to take information and work from others and give nothing.

Let's make it clear: I have no problem with the profit of 6000 keyboards. That's perfectly fine with me.
My point is and has always been sharing information and misusing (in my opinion) DT.

-- PD : I am sure that a long post will follow hiding this behind

The FSSK space bar is not aligned with the case and the key row above due to one tab (right one from the user's perspective) being offset by about one mm towards the '.' key.

This causes the space bar to move out of the way once the '.' key gets pressed because the space bar's stabiliser wire makes room for the '.' key. In addition, the space bar sometimes jams.
I have read the section E1 of the manual's section 1 pertaining the space bar.
I do not dare to bend the space bar's stabiliser wire that much because the problem is the right tab's offset underneath the '.' key. I would need the wire to be further away from the back of the tab. I hope you get what I mean by looking at the photo attached.

How would you fix that?

The only issues that Ive seen that are notable are:

- Not replacing parts that are physically damaged and just telling the customer to use a PAINT PEN. That's just insane to me. This product is $400+ and we're told to use a paint pen on a part that arrived to us damaged.
- Not including enough keycaps to fill the B122 keyboard in the keycap set you have to PAY extra for. And the solution he comes up with for me is to buy relegendable key caps off the website that do not match the SA profile caps he sells for the beamspring boards on the website. The missing keys are also coincidentally keycaps that are nearly impossible to find online in any profile, let alone SA profile and a matching color. He could've found and worked with a manufacturer that could have made a key cap set that included all the necessary keycaps to fill a product that he sells for nearly $800.

That being said, I'm still pretty excited to get the B122 in. Luckily I found a solution for the keycap problem, but I still think it's unacceptable to not provide keycaps to fill the keyboard that you sell, especially when they're as hard to find as F13-F24 and M1-M10.

Thank you Ellipse!

To be honest the biggest help for a Linux noob like me would be someone to look at the Linux build of the Leyden Jar diagnostic Tool.

This proposal for solenoid and UART at the same time is perfectly legit.
I see that someone carefully studied the Leyden Jar schematics and read the RP2040 datasheet to do this proposal.
Now there is the problem of firmware fragmentation with two different PCB revisions to support, but I may have a solution for that :p

I have gone through the manual, and I might have missed this part on the key seating/troubleshooting, but I am having a "timing" issue.

Got my brand new Model F FSSK. Set it up, got the keys installed, tested, they all work. However, I am having a problem with letters getting typed "out of order". For instance, if I type "as" or "th" or "in" quickly (or normally actually) there is ~50% chance that they are going to be transposed.

If I type "was" about half the time it will be "wsa" or "interested" will be "niterested"

I have checked and tested it several times. I am POSITIVE I am hitting them in the "as", "th" or "in" order but they will be reversed annoyingly often. I fell and hear the click of the "a" before the "s" but on the screen it comes out as "sa"

I am plugged directly into the computer on a USB-A port. Did not have this issue with my F77 on the same computer and port. I have run the pandrew utility and everything looks good there.

Any ideas on what is going on with the slow response to keypresses?

Ellipse wrote: 21 Nov 2024, 20:36 As noted before, the main reason for the switch was because of the expected discontinuation of the mass availability of the Atmel chips, especially felt during the chip shortage several years ago. Without Rico's great controller we may not have been able to build Model F and Beam Spring controller replacements to keep these keyboards running long into the future.

Ellipse wrote: 21 Nov 2024, 20:36 The real-world functionality and performance are identical for these controllers in my testing, so please do not think you will notice anything different. The big advantage of the RP2040-based Leyden Jar is that it has more memory and the potential for advanced functionality such as PS/2 support and split keyboard communication between controllers, though the xwhatsit supports at least 3 layers which is plenty for the needs of almost all new Model F users.

I disagree.

The first thing that I noticed when installing (Rico's) RP2040 Leyden Jar was its speed. It is noticeble faster.
Rico has used RP2040 PIO ... (reference to simple explanation on how PIO works here https://tutoduino.fr/en/pio-rp2040-en/) which takes the matrix scanning out from the main processor. A very beautiful solution which contributes to speeding up the controller response.

Ellipse wrote: 21 Nov 2024, 20:36 NathanA is correct that only recently did the Leyden Jar controller finish prototype testing and was considered ready for production, and that we are waiting for the xwhatsit supply to be done before switching everything to Leyden Jar.

A nice way to state that it is stock related

Ellipse wrote: 21 Nov 2024, 20:36 Also, as of today the Leyden Jar still does not have firmware for these other keyboards, so the Leyden Jar controller is not yet ready for these other keyboards. I did test an older Rev2 Leyden Jar that worked fine with the new Model M classic style F104, so I do not foresee any issues expanding support to additional keyboards.

The original FSSK was not supported by the Leyden Jar firmware. This did not stopped me to compile my own. Moreover, I had to do it twice as I have daily use two different FSSK versions: the original v0r1 prototype and the group but v1r0.
I do not see any problem in doing it for other keyboards. Actually as I said I plan to migrate my other Model Fs to the Leyden Jar.

BTW: do you ship your keyboards Firmware VIAL ? My xwhatsit ended up without memory when I tried to compile VIAL on them.

I keep thinking that the Leyden Jar is a game changer. This is why I was surprised to read that AVR-xwhatsit keyboards are sold at the end of 2024.

Well done Rico. Respect !

pilcher wrote: 20 Nov 2024, 20:26

RainehDaze wrote: 20 Nov 2024, 19:59

pilcher wrote: 20 Nov 2024, 19:19

Have you used pandrew's utility to check the signal levels?

Somewhat related problem, I can't seem to get the keyboard to show up in it. Which is strange, I seem to have all the requirements installed to run the make script.

Since you mention make, are you on Linux?

If so, you may already be aware of this, but I wrote this doc that describes the steps I took to build the utility on Fedora 40.

https://gist.github.com/ipilcher/9a14d8 ... 71a282fa74

Yeah, I'll give that a shot; I'm guessing there's something that hasn't set up right and it's just failing invisibly, since it still opens well enough. Just... no keyboard detection. And I can see you've got quite a few more steps than just doing the build in the utils folder.

RainehDaze wrote: 20 Nov 2024, 19:59

pilcher wrote: 20 Nov 2024, 19:19

RainehDaze wrote: 20 Nov 2024, 17:48 Hmm, having a bit of a weird one with this, that I'm not quite sure of the cause. It sometimes seems that the H key is registering slightly behind everything else? So, I'll try to type the or that or this and it'll come out as htat. I can mitigate it by typing slower, but… it's not like the h actually has trouble registering, so I'm a bit confused.

Have you used pandrew's utility to check the signal levels?

Somewhat related problem, I can't seem to get the keyboard to show up in it. Which is strange, I seem to have all the requirements installed to run the make script.

Since you mention make, are you on Linux?

If so, you may already be aware of this, but I wrote this doc that describes the steps I took to build the utility on Fedora 40.

https://gist.github.com/ipilcher/9a14d8 ... 71a282fa74

pilcher wrote: 20 Nov 2024, 19:19

RainehDaze wrote: 20 Nov 2024, 17:48 Hmm, having a bit of a weird one with this, that I'm not quite sure of the cause. It sometimes seems that the H key is registering slightly behind everything else? So, I'll try to type the or that or this and it'll come out as htat. I can mitigate it by typing slower, but… it's not like the h actually has trouble registering, so I'm a bit confused.

Have you used pandrew's utility to check the signal levels?

Somewhat related problem, I can't seem to get the keyboard to show up in it. Which is strange, I seem to have all the requirements installed to run the make script.