Miss Read Key Presses?

[Ronald Wells]

Is anyone else seeing a situation where if you're typing on the touch pad sometimes one of the settings buttons gets activated?  

I.E. you're typing in a message and randomly the modell, rotor or ring selector lights up and you have to press it to turn it off so you can keep typing?

I think it might have to do with "fat fingering" and not quite hitting the letter directly so maybe two keys are activated at the same time or something?

[Jeff Spears]

I have. It seems to be due to 'fat fingers' -- hitting two keys at once. Just as you say.

[Jürgen]

It's technically possible for that to occur. The keyboard uses a matrix arrangement where each key activates a row and a column. But the STM32 chip was running out of capacitive sensor connections, so the mode keys (and letters A and S) use "column with no row" signaling.

I have tweaked the sensing thresholds quite a bit, so it should be difficult to activate a column signal without activating the row signal when hitting a regular letter key. But it may still be possible to get a misinterpreted key if you touch it key far to its left (where the column electrodes are) and press down hard enough (to get the key registered).

Apologies if this becomes a nuisance! If you notice that this tends to happen on specific keys, please let me know which ones -- maybe I can adjust the threshold further for these. Not all keys are created equal, due to the wiring topology.

Regards,

Jürgen

[Ronald Wells]

pressing between these keys result in:

E & R -> modell

R & T -> modell

U & I -> Rotor

I & O -> Ring

D & F -> modell

H & J -> Rotor

J & K -> Ring

X & C -> Modell

C & V -> Modell

N & M -> Rotor

M & L -> Ring

This seems to match up with what you're saying about column but no row sensing, so it seems this is by design and I just need to be more diligent about hitting the keys correctly :)  If I notice any particular keys acting up more than others I'll shout it out.

My main concern was that I might have some kind of defect, but looks like it's all good!

Thanks so much for this awesome replica!

Ron

[Jürgen]

I think I can improve the key detection a bit further: Increase the sensitivity for detecting a row touch, and make sure I have the right one by looking for the strongest row signal. The changed code works well for me, but it would be great to test this with a few different users (and fingers!).

I am about to release a beta version of the new firmware 4.10 (with USB keyboard functionality and enhanced USB serial communication). Should be ready this weekend; I am just completing the user manual updates. I will post the beta firmware in a new thread here and would appreciate if many of you could test and provide feedback -- on the false key presses as well as the other functional changes. 

(There is no risk, you can always go back to the prior firmware. Chapter 5 of the user manual describes how to update the firmware. Both, the user manual and the current firmware 4.03, are available for download at https://e-basteln.de/computing/enigma/enigma-files/#user-information.)

Thanks in advance!
Jürgen

[Ronald Wells]

I'll be happy to test it out once it's available!

[Jürgen]

The new beta firmware is ready! See the separate thread, https://groups.google.com/g/enigma-touch/c/-UlJpIBC82c

[Bart Wessel]

Nope, never had that happening.

Bart.

Op zaterdag 9 augustus 2025 om 11:16:43 UTC+2 schreef Jürgen: