deskthority

SOLVED 04/18/2025: Short somewhere in the controller, and loosing the screw just right will fix the issue. Also a bad solder on one of the 6 solders located right under where controler is mounted by 1 of the 2 screws. All ribbon solders look great, but the 6 solders right below the screw look to have been soldered badly as there is not solder all the way thru the pin and hole. I may have to re-solder those 6 pins right under the screw at some point because I never seen a soldier job that looked like this before where there is a visual space between the hole and the pin..... and it is likely this is causing the issue. There is also burn marks all around the 6 pins on the motherboard where loosening the screw fixes it, so the short must be arking somewhere and causing the controller to overheat at the 6 pins.

History of this keyboard:

This is my third Model F, which I received about two weeks ago. Yesterday (04/09/2025), it started typing "\162cg" randomly, and today (04/10/2025), the issue escalated—it’s now typing uncontrollably to the point where I can’t use the keyboard at all. I primarily use it for work, so this has become a big problem.

This is the only Model F that has ever given me issues. Interestingly, it's also the only one I've used straight out of the box without re-flashing the firmware. (not sure which firmware this one came with by default, but it worked with VIAL directly by default so I never had to re-flash it before).

Questions:
What steps should I take to diagnose the issue? Could this be a hardware failure? If so, how can I pinpoint which part is failing?

For context, I’ve opened up and worked on multiple Model M keyboards, including bolt-modding them. I’ve also repaired membranes using a conductive pen, and I've done soldering work on Cherry MX keyboards. So I feel confident in my ability to fix this, but I need guidance on how to identify the root cause.

Note:
I’d prefer not to send the keyboard back if I can avoid it. I’m really attached to the serial number ("666"), and my last two Model F keyboards arrived with significant damage to the packaging— and both had a large hole in the box. This makes me hesitant to ship it back, so I’d like to troubleshoot and repair it myself if possible.
Updates:
04/14/2025: Issue still persists as it has done it again after extensive use; but I was not in a position to pull up FLIP to check what exactly the keyboard was doing during the time it was occuring and unplugged the keyboard to swapped in a different F-104 due to the urgency of needing a functional keyboard immediately. Hopefully it will do the issue again when I am in a position to where I can actually check it as the issue happens intermittenly
04/13/2025: So far 48 hours has went by without the issues re-occuring (which is the longest it has been since the issues started happening). Will be using the keyboard extensively this week; so we'll if the issue comes back. Joe has also reached out to me personally via e-mail, and recommended for me to check the voltages; which will be my next thing to check if the issue comes back as I did not think about chercking the voltages when the issues were occuring, but I have pandrew utility installed and will pull it up to see what the "Signal level monitor" looks like the next time it gives me the issue.
04/12/2025: The reflash has worked for 24 hours now.
04/11/2025: It has gotten worst and now vial matrix tester is showing that all the keys are pressed at onced in addition to the issues i described above. It also appears that it sometimes fixes itself when the Ctrl or Windows key seems to be stuck and those keys get pressed down and fixes the issue where all of the keys are registered as being pressed down. All of these issues only come up intermittenly so far so it is difficult to figure out what is wrong as unplugging it and plugging it back in may fix the issue for the entire day.
I have swapped out this F104 to another PC and it does the same issues, So there is an issue with the keyboard for sure.
I went ahead and installed FLIP and erased and re-flashed the firmware to see if it fixes it for now. And if this continues I may open up it up and see if there are any loose connections.
04/10/2025:started happening again, and unplugging and replugging it back in fixes it temporarely but now the issue is not fixed the entire day but will come back.
04/09/2025: initial date where the issue happened, and a simple unplugging and plugging it back in seemed to fixed the issue for the entire day without it coming back.

I recommend reviewing the manual from beginning to end, as this specific issue of constantly pressed keys is addressed there. Everyone's first step should be reviewing the manual instead of posting about these kinds of errors as I feel that every error can be addressed by following all the steps in the manual, with the last steps being to contact me directly if the error cannot be resolved and posting on the forums if I am stumped.

Likely the springs and keys need to be adjusted and/or reinstalled or replaced because they are currently in the pressed down state. Sometimes a flipper gets stuck inside the barrel so you should make sure the spring can move freely, monitoring the signal level monitor utility screen for voltage changes when the spring is moved.

As you mentioned due to rough shipping things do move out of place, especially the springs.

This could all be avoided if QC existed and was coupled with spending more than chump change on shipping materials lol

Ellipse wrote: 12 Apr 2025, 03:52 Likely the springs and keys need to be adjusted and/or reinstalled or replaced because they are currently in the pressed down state. Sometimes a flipper gets stuck inside the barrel so you should make sure the spring can move freely, monitoring the signal level monitor utility screen for voltage changes when the spring is moved.

As you mentioned due to rough shipping things do move out of place, especially the springs.

I pulled up FLIP and the signals for each key are almost identical to my other F-104, so when it is working it appears fine. Because the issue happens intermittently it is hard to catch it during a time when it is occuring for me to pull up the app to check what exactly is going on.

It did have a rough shipping package but no visual damage on the keyboard it self and the keyboard worked fine for about 2 weeks since I have received it and it started having issues intermittently on 09Apr2025 to present.

wobbled wrote: 12 Apr 2025, 15:24 This could all be avoided if QC existed and was coupled with spending more than chump change on shipping materials lol

You would have to use the keyboard for an hour or longer before the issue may appear as it only occurs intermittently. Although my 2 (Two) F-104's came with shipping package damage; the keyboard itself appears to be undamaged and the accessories luckly were on the opposite side of where my holes were when I received it, so there are no missing parts (that I am aware of) when I received them. Here are some pics of all of my Model F box's you are curious:

F77 box was larger and actually "thiccc"er compared to the F104 boxes: Here is a comparison of the thickness of the F77 box with the F104 box: and a close up pic of the 2box's compared:

As noted in the manual, intermittent issues are sometimes due to dust or debris inside the keyboard, if replacing the springs and correctly reinstalling them does not fix the issue. The quickest way to fix this issue permanently would be to remove the keys and replace the springs and flippers with other ones, and also to wipe clean the PCB, especially in the areas of the affected key pads, as there may also be caked on debris inside. My guess is that the flippers are fine but it may be an issue with the spring installation to the flipper, or debris on the flipper and/or PCB for those keys.

Reflashing firmware will not often fix such an issue, unless you made any adjustments to the firmware.

Ellipse wrote: 15 Apr 2025, 00:31 As noted in the manual, intermittent issues are sometimes due to dust or debris inside the keyboard, if replacing the springs and correctly reinstalling them does not fix the issue. The quickest way to fix this issue permanently would be to remove the keys and replace the springs and flippers with other ones, and also to wipe clean the PCB, especially in the areas of the affected key pads, as there may also be caked on debris inside. My guess is that the flippers are fine but it may be an issue with the spring installation to the flipper, or debris on the flipper and/or PCB for those keys.

Reflashing firmware will not often fix such an issue, unless you made any adjustments to the firmware.

The best way to fix this is to send the broken shit back to you for a replacement.

Ellipse wrote: 15 Apr 2025, 00:31 As noted in the manual, intermittent issues are sometimes due to dust or debris inside the keyboard, if replacing the springs and correctly reinstalling them does not fix the issue. The quickest way to fix this issue permanently would be to remove the keys and replace the springs and flippers with other ones, and also to wipe clean the PCB, especially in the areas of the affected key pads, as there may also be caked on debris inside. My guess is that the flippers are fine but it may be an issue with the spring installation to the flipper, or debris on the flipper and/or PCB for those keys.

Reflashing firmware will not often fix such an issue, unless you made any adjustments to the firmware.

Hopefully, it won't come to opening it up that far. However, I’ve reviewed the video and will try to find some time to open it up and check the solder points. It doesn’t seem to have issues when used infrequently, but under heavy usage—such as in an office setting where it’s constantly in use—it eventually starts to show the issues

As much as I like Joe/Ellipse (and I do), his reproductions are not "Model F" keyboards. IMO, the only Model F keyboards are (were) the ones made by IBM and their affiliates. To my mind, in the best case, they are "homage" keyboards, in the way a lookalike but fake Rolex wristwatch would be called a "homage" timepiece.

Phil_Goodman there should be no issues, regardless of how much it is used. If there is any issue, following the manual fully will result in a complete fix. As a last resort I can help, if you are willing to lose using the board for a month or so (shipping back and forth and QC time) but all I would do is open up the board and wipe down whatever debris or dust entered the keyboard, and replace the flippers and springs for any affected keys. Outside of taking off keys which may take 10 minutes with a key puller, the rest of the steps should not take more than 10 minutes.

Another issue to check could be the grounding of the controller. Make sure both bolts are tight. I've seen odd behavior when these bolts are not tight enough.

Ellipse wrote: 18 Apr 2025, 05:16 Another issue to check could be the grounding of the controller. Make sure both bolts are tight. I've seen odd behavior when these bolts are not tight enough.

I think we figured it out.... you are close, but doing the opposite ended up fixing it: loosening the 2 screws on the controller will cause the symptoms to go in and out.. so there is also a bad solder on the controller somewhere; All of the solders on the ribbon's appear to look good. The solder in question is one of the 6 solders right under the screw that holds the controller in place; I bet that is the weak point and the reason why loosening the screw fixes it as it unflexes the board and reconnects the short somewhere.

I have a really good solder sucker so I should be able to redo this without an issue.

HOPEFULLY this fixes it. It took me a minute to figure it out because 1 of the screws holding the controller was actually alittle bit loose, and now i see why. And when I was putting it back together for the first time I screwed the controller in so it was flush (which wasn't even that tight.. maybe finger tight at best). And after i plugged it in and started loosing the screws the keyboard worked again and I realized that screw couldn't be on too tight or too loose.

modelf wrote: 17 Apr 2025, 22:02 As much as I like Joe/Ellipse (and I do), his reproductions are not "Model F" keyboards. IMO, the only Model F keyboards are (were) the ones made by IBM and their affiliates. To my mind, in the best case, they are "homage" keyboards, in the way a lookalike but fake Rolex wristwatch would be called a "homage" timepiece.

Your analogy isn't wrong, but I think it's more like comparing the original IBM Model M to the Unicomp Model M. I've yet to own a brand-new Model M from Unicomp that has lasted more than a year—except for the Mini M, which has held up for several years, though I don't use it as much.

Even so, I consider this a much better product compared to what Unicomp currently offers. Over the past few years, I've gone through around five or six Unicomp keyboards, and their quality and reliability might be some of the worst I've seen. Despite that, this keyboard far surpasses Unicomp's offerings by a large margin.

If I wanted something reliable and durable, I'd go with Cherry MX. But there's just something special about the design of Model M and Model F keyboards, despite how finicky and unreliable they can be.

Happy to hear that you fixed the issue! This was such an unusual error!

As I mentioned in the manual, on some keyboards the controller was not aligned correctly with the bottom inner assembly's ears and so it produced a short. The manual shows how you can push the controller back into place so that the ground plane square on the controller lines up with the square of the bottom inner assembly ears.

Ellipse wrote: 20 Apr 2025, 05:01 Happy to hear that you fixed the issue! This was such an unusual error!

As I mentioned in the manual, on some keyboards the controller was not aligned correctly with the bottom inner assembly's ears and so it produced a short. The manual shows how you can push the controller back into place so that the ground plane square on the controller lines up with the square of the bottom inner assembly ears.

I did the right bottom hand corner already before I saw your message to take pictures, but the bottom right hand corner originally looked to not have enough solder in it before i added more to it: However it may have not been the problem because it was still having the same issues even after i put solder into it: I am not 100% sure which one was causing the issue so I just redid the solder to all 6 of those pins and cleaned it up with some alcohol afterwards and it seems to have done trick with fixing the issues to a point to where it didn't matter how tight or loose I screwed in the controller.
So it might be fixed for good now (hopefully):