The series resistors on the FPGA side of the level shifter would slow down the edge rates as well commensurate with the 10-17kHz clock rate for less over/undershoot on the long keyboard cable.
I remember many moons ago I implemented a PS/2 port on a PIC MCU. And I found that I got quite a few parity errors. So having the ability to request the keyboard to re-send the last keycode is really important. Even if you don't want to control any of the LEDs and stuff, you still want to have the ability to send bytes out to the keyboard. Now you might think "doesn't matter, I'll just ignore erroneous keycodes". But keep in mind, you might receive the " Push" Code for a key, but then get a parity error for the release code. And then you end up with a stuck key.
🤔 Makes me wonder if there were other issues at play too. I'd expect a very low error rate... I guess I'll find out... Maybe leave a hammer laying on my keyboard for a week...
The series resistors on the FPGA side of the level shifter would slow down the edge rates as well commensurate with the 10-17kHz clock rate for less over/undershoot on the long keyboard cable.
I remember many moons ago I implemented a PS/2 port on a PIC MCU. And I found that I got quite a few parity errors. So having the ability to request the keyboard to re-send the last keycode is really important.
Even if you don't want to control any of the LEDs and stuff, you still want to have the ability to send bytes out to the keyboard.
Now you might think "doesn't matter, I'll just ignore erroneous keycodes". But keep in mind, you might receive the " Push" Code for a key, but then get a parity error for the release code. And then you end up with a stuck key.
🤔 Makes me wonder if there were other issues at play too. I'd expect a very low error rate... I guess I'll find out... Maybe leave a hammer laying on my keyboard for a week...
@@JohnsBasement it's possible that it was just crappy wiring on my part, or maybe I had a particularly crappy keyboard.
@@TSteffi Regardless, it is good to know!
Considering that up to 3 LED’s may be strongly illuminated, fusing for the 100mA recommended on that website might prove more reliable.
@@ChrisSmith-tc4df yeah, that is a schematic in fairly early design. I ought to go with 500 since it will be for both the keyboard & mouse ports.
I think it’s prudent to install some ESD protection diodes for the signal leads.
I've been thinking about that. Not sure what is most appropriate. What do you suggest?
Have a look the the Nexperia ESD Application Handbook. I have a hard copy but it’s downloadable as a pdf.
Did anybody notice the code for Z on the Xilinx scan codes page? 14:52
@@vasileceteras 😂