1-wire DS1820 - OpenWRT - dualUSB bifferboard

[Petr Kracik]

Hi,

I've tried Dallas's 1-wire DS1820 temperature sensor by this tutorial

https://sites.google.com/site/bifferboard/Home/desktop-linux-distributions/slackware/1-wire

But I have problem with modules in OpenWRT

when I load w1-gpio one it will Write "Killed" and in dmesg i have this http://pastebin.com/JjJK4KEm

I used GPIO 8  by insmod w1-gpio-custom bus0=0,8,0 command...

but I can see on that GPIO pin some activity,, about each 10sec there I can measure some data (I don't have oscilloscope so I can just measure by voltmeter).

Any idea what I can do bad...

I've tried also "normal" powering of that DS1820 also parasite one.. without success,, but if I will use Serial 1-wire cable (interface) over USB->Serial then I can see it (by digitemp), so I know that DS is not burned...

also

root@OpenWrt:/sys/bus/w1# uname -a

Linux OpenWrt 2.6.37.6 #10 Sun Oct 23 11:53:18 BST 2011 i486 GNU/Linux

Maybe I missing something.... Thanks for reply

Petr Kracik
email: pet...@gmail.com
ICQ: 136530439

[nicolas le falher]

Hi,

It's not your ds1820 who don't work, it's module who crash.
I've got some ds1820 at home, I'll test module after work.

Nicolas

[nicolas le falher]

Try with this patch :
https://dev.openwrt.org/attachment/ticket/8952/w1-gpio-custom.patch

With i can load w1-gpio after w1-gpio-custom without warning (and sysfs is populated)

Nicolas

[Petr Kracik]

Hi

Thanks for patch

that w1-gpio really was loaded without error / warning, but still I can't  see that DS1820 in that bus/w1 tree,, I will check cables at Friday or Saturday since I have it at home and have just SSH access to it now...

root@OpenWrt:/sys/bus/w1# find

.

./uevent

./devices

./devices/w1 bus master

./drivers

./drivers/w1_master_driver

./drivers/w1_master_driver/uevent

./drivers/w1_master_driver/unbind

./drivers/w1_master_driver/bind

./drivers/w1_master_driver/w1 bus master

./drivers/w1_slave_driver

./drivers/w1_slave_driver/uevent

./drivers/w1_slave_driver/unbind

./drivers/w1_slave_driver/bind

./drivers_probe

./drivers_autoprobe

root@OpenWrt:/sys/bus/w1# lsmod |grep w1

w1_therm                1320  0 

w1_gpio                  472  0 

w1_gpio_custom           668  0 

wire                    9132  2 w1_therm,w1_gpio

root@OpenWrt:/sys/bus/w1# 

Regards Petr

Petr Kracik
email: pet...@gmail.com
ICQ: 136530439

2012/1/24 nicolas le falher <n.lef...@gmail.com>

Nicolas

--
To unsubscribe send email to bifferboard...@googlegroups.com

[Petr Kracik]

Hi,

So I've done little research I found this...

Each about 10 sec there is some communication (voltage goes lower), but nothing in "/sys/devices/w1 bus master" was not populated. 

If I use add "by hand" by

root@OpenWrt:/sys/devices/w1 bus master# cat w1_master_add

write device id xx-xxxxxxxxxxxx to add slave

then it's populated but of course I don't know right ID... but when I doing 

cat w1_slave

I can see some communication at that bus (each tam when I cat it that voltage goes lower).

also output is for example this:

root@OpenWrt:/sys/devices/w1 bus master/22-c10a01080032# time cat w1_slave 

ff ff ff ff ff ff ff ff ff : crc=c9 NO

00 00 00 00 00 00 00 00 00 t=-62

real    0m 0.80s

user    0m 0.00s

sys     0m 0.02s

if I will disconnect that DS1820 from interface then I've got different output

root@OpenWrt:/sys/devices/w1 bus master/22-c10a01080032# time cat w1_slave 

00 00 00 00 00 00 00 00 00 : crc=00 YES

00 00 00 00 00 00 00 00 00 t=0

real    0m 0.10s

user    0m 0.00s

sys     0m 0.01s

also in 1st output (with -t62) that took longer (as you can see at time output) so some communication there must be (I hope that w1-gpio works)...

I just wonder why auto search will not find it.

Also when I try Serial interface then I can see it (at server, not bifferboard)

Turning off all DS2409 Couplers

.

Searching the 1-Wire LAN

10DFC10A01080032 : DS1820/DS18S20/DS1920 Temperature Sensor

ROM #0 : 10DFC10A01080032

It looks like some problem with DS1820 version like DS18S20 and DS18B20 etc..

Any ideas ?

Regards Petr

Petr Kracik
email: pet...@gmail.com
ICQ: 136530439

2012/1/24 Petr Kracik <pet...@gmail.com>

[nicolas le falher]

>Each about 10 sec there is some communication (voltage goes lower)

For my part there are reset on the bus :/. (http://www.imagup.com/data/1142365590.html)
I did not manage to have something else for the moment.

Nicolas

[nicolas le falher]

Damn, by reading documentation it works better.

In "make menuconfig" add :
Kernel modules --->
W1 support --->
<*> kmod-w1
<*> kmod-w1-gpio-custom
<*> kmod-w1-slave-therm

Extract of w1-gpio-custom.c :
* <id> ID to used as device_id for the corresponding bus (required)
* <sda> GPIO pin ID of data pin (required)
* <od> Pin is configured as open drain.

So, here we go with open-drain :
# echo "w1-gpio-custom bus0=0,11,1" > /etc/modules.d/55-gpio-custom
# reboot

On reboot you will have
w1_master_driver w1 bus master: Family 28 for 28.0000034bc0b7.2e is
not registered.

# cat /sys/bus/w1/devices/28-0000034bc0b7/w1_slave
60 01 4b 46 7f ff 10 10 b5 : crc=b5 YES
60 01 4b 46 7f ff 10 10 b5 t=22000

On different boot, I've got different states with non-parasite mode :
- w1 slave not detected
- w1 slavec detected, but can't read temperature ( 18S20 doesn't
respond to CONVERT_TEMP )
- all works :)

With parasite mode, all works :)

Nicolas

[Petr Kracik]

Ok,, there is no diff if i will use ,0 or ,1

still slave is not detected and I i will add it by hand it can't convert temp

I attached scheme if I'm right at hardware base (as long I understood that from text how-to on wiki)

I meanwhile got format of that ID so I've got different output (-1250 instead -62)

Petr Kracik
email: pet...@gmail.com
ICQ: 136530439

2012/1/28 nicolas le falher <n.lef...@gmail.com>

Nicolas

[nicolas le falher]

I use 5v from Biff power supply with a 4.7k pull up.
With 3.3v (from jtag port) and 1.5k no problem for detection too.

But, with ds18b20 pin vdd not connected, I've got different value
between two refresh :
# cat /sys/bus/w1/devices/28-0000034bc0b7/w1_slave
ff 07 4b 46 7f ff 01 10 2f : crc=2f YES
ff 07 4b 46 7f ff 01 10 2f t=127937
# cat /sys/bus/w1/devices/28-0000034bc0b7/w1_slave
50 05 4b 46 7f ff 0c 10 1c : crc=1c YES
50 05 4b 46 7f ff 0c 10 1c t=85000
# cat /sys/bus/w1/devices/28-0000034bc0b7/w1_slave
ff 07 4b 46 7f ff 01 10 2f : crc=2f YES
ff 07 4b 46 7f ff 01 10 2f t=127937
# cat /sys/bus/w1/devices/28-0000034bc0b7/w1_slave
50 05 4b 46 7f ff 0c 10 1c : crc=1c YES
50 05 4b 46 7f ff 0c 10 1c t=85000

With vdd wired to gnd, value is stable :
# cat /sys/bus/w1/devices/28-0000034bc0b7/w1_slave
55 01 4b 46 7f ff 0b 10 d0 : crc=d0 YES
55 01 4b 46 7f ff 0b 10 d0 t=21312

But for me values are no correct.
I've got another ds18b20 wired to an usb9097 (usb<>w1) and I get ~17C.
Same with a multimeter :/

Nicolas