BTC 5339 circuit question

[light655]

I just bought a BTC 5339R (old new stock) and found out one of the keys aren't responding. So I took it apart and stuck a new piece of foil (cut out from a crisps' bag) on that key, and that was fixed. I've looked at the circuit board in the process of fixing it, and I found that besides the keyboard controller IC, there are two CD4044 latch ICs on the circuit board. I just wonder why those two are added into the circuit?

^ Here's the circuit board.

^ This is what I used to repair the key, in case you are interested.

[Muirium]

Latches? Caps / Num / Scroll lock lights by any chance? Or are they multiplexers for the matrix?

The PCB is very sparse. Looks like a more modern capsense board, besides those resistors and jump wires or whatever they're called.

[light655]

There a total of eight latches in the two ICs, that wouldn't make a lot of sense as there are only three LEDs to switch.
I thought the multiplexer is built in the controller IC, at least there weren't any ICs other than the cotroller on my other keyboards.

[Muirium]

Well, I'm no electrical engineer! All I know is multiplexers are often used when you need to scan a bigger matrix and don't want to increase the cost of your controller. It's all about saving pins. They're external address extenders, which are inherently cheap.

[light655]

I think that's quite likely, maybe the multiplexer got built in the controller later.

[Wazrach]

I know this doesn't pertain to the PCB, but how do you find the switches? I'm curious and would like to try one of these out sometime.

[light655]

I bought this keyboard on local shopping website in Taiwan.

[Rayndalf]

I bought this keyboard on local shopping website in Taiwan.

How do the switches feel?

[light655]

In short, rather awkward. Mine is the "R" version, the return force element is a rubber dome, I know some variants use springs instead. The top part of the key press (before triggering the keystroke) feels pretty much like rubber dome because it is a rubber dome. The lower part of the key press, on the other hand, is really weird. The force will increase a lot after the keystroke is registered due to the foam pressing onto the PCB. The rubber domes provide a tactile feel at the top of the key press, and the foam gives another increase of force, but because of the foam there isn't a strong feel of bottom out. I don't think a lot of people will like it, and it will take some time to get used to!

[micmil]

Latches? Caps / Num / Scroll lock lights by any chance? Or are they multiplexers for the matrix?

The PCB is very sparse. Looks like a more modern capsense board, besides those resistors and jump wires or whatever they're called.

https://toshiba.semicon-storage.com/inf ... e=TC4044BP

Quad NAND circuit. Likely some type of multiplexing.

[Xses]

I like that keyboard and how it feels and I believe it is a very decent keyboard.

On the ICs side - I've looked at the schematics for the linear version:
https://www.schematicsforfree.com/archi ... ypical.pdf
They are a bit different from R version, however in all cases, those 4044s are not used for CapsLock, ScrollLock, etc. In our versions, it is probably done just by the microcontroller itself (I/O registers). 4044 ICs are part of the main keys matrix. Microcontroller can only do 1 thing at the time, so it can only send an impulse to keys matrix, or check which key conducted, so it needs some sort of memory which remembers which key conducted the impulse and for that they used 4044s, so microcontroller does following routine - sends an impulse to the matrix, checks latch and reads which key was pressed, resets latch, sends impulse for next set of key, checks next latch and so on.