Announce: Slimmer

[kingswindsor]

Without pull-ups (internal or external) it won't work, there will be too much noise.

I took a quick look at the pikeyd source. As I could see the -pu option will turn on the internal pull-ups for both rotary encoder pins. So I would try that in your case. Just stop slimmer and pikeyd with their init.d scripts, and run pikeyd from the command line (foreground) with the -pu switch with sudo. You can use evtest from an other terminal to check if it works. It should.

Yes, I have external pull-up resistors (10k) on the encoder's output pins. I have no schematics, these are just simple pull-ups.

Thanks, I am away for the weekend but will try next week and report back.

[kingswindsor]

I took a quick look at the pikeyd source. As I could see the -pu option will turn on the internal pull-ups for both rotary encoder pins. So I would try that in your case. Just stop slimmer and pikeyd with their init.d scripts, and run pikeyd from the command line (foreground) with the -pu switch with sudo. You can use evtest from an other terminal to check if it works. It should.

Sorry if I am being dim but what exactly is the command to run pikeyd from the command line? When I try running it with the -pu switch it says there is a syntax error - only start stop or status with no options. I've stopped both Slimmer and Pikeyd.

[kingswindsor]

Sorry if I am being dim but what exactly is the command to run pikeyd from the command line? When I try running it with the -pu switch it says there is a syntax error - only start stop or status with no options. I've stopped both Slimmer and Pikeyd.

OK I was being dim. Now have progressed but still not getting any evtest reports from the rotary encoder. Any advice gratefully received. The steps I have been through are:-
- reboot
- stop slimmer and pikey
- checked using top
- sudo pikeyd -pu

Config file is /etc/pikeyd.conf
Polling 2 GPIO pin(s).
Found 0 I/O expander(s).
Ready.
Joystick init OK.
Press ^C to exit.

- Open new terminal
- check running via top - OK
- tce-load il evtest
- evtest /dev/pikeyd

.....................
Event code 250 (?)
Event code 251 (?)
Event code 252 (?)
Event code 253 (?)
Event code 254 (?)
Event code 255 (?)
Properties:
Testing ... (interrupt to exit)

Twiddle rotary encoder - nothing......

Just as a reminder, the pikeyd.conf file contains
ROT 23 24 KEY_LEFT KEY_RIGHT

Any obvious mistakes?

[tbali]

Any obvious mistakes?

I see no mistakes. Sorry for the late answer, I was on holiday.

[N_J]

Very nice development Balázs,

I have been experimenting with this over the weekend, but was not able to get it to work unfortunately.
The issue seems to be with pikeyd.
I work on an RP 3B, and apparently the original pikeyd does not work on this board.
I'll post the link in my second post (since I am new to this forum and this is my first post, I am not allowed to share a URL yet :-), but it comes down to this:

On RPi2 and RPi3 the GPIO_PERI_BASE address has changed from 0x20000000 to 0x3f000000.
So just change #define GPIO_PERI_BASE in joy_RPi.c to 0x3f000000 and it will work like a charm on a RPi3 (couldn't test it on a RPi2)

Seems to be a simple fix, but unfortunately I lack knowledge and skills to adapt your pikeyd to work on the RP3 (and 2).
would you be able to incorporate that fix and share with the community?

Thanks, best regards,
Niels

[N_J]

And here's the link:
https://github.com/mmoller2k/pikeyd/issues/3

Best regards,
Niels

[tbali]

Seems to be a simple fix, but unfortunately I lack knowledge and skills to adapt your pikeyd to work on the RP3 (and 2).
would you be able to incorporate that fix and share with the community?

Hi!

I will try to do this in this week. Maybe with autodetection of the PI's version.

Best regards,
Balázs

[N_J]

Great! looking forward to test.

Best,
Niels

[N_J]

Hi Balázs,

As it seems also lcdproc-server has this issue. I replaced the hd44780.so driver in lcdproc-server.tcz with a patched version I found, and have the display running now.

Best regards,
Niels

[tbali]

As it seems also lcdproc-server has this issue. I replaced the hd44780.so driver in lcdproc-server.tcz with a patched version I found, and have the display running now.

Please provide some info about this patch, and I will look into that too. Thanks!