Re: [Gumstix-users] Humidity sensor to I2C interface

[Linus C. <li...@li...>]

I have seen the signals in the oscilloscope and the GPIO 34 doesn't work.

The commands I have used:
root@gumstix-custom-verdex:~$ echo \"GPIO out clear\" >/proc/gpio/GPIO39
root@gumstix-custom-verdex:~$ echo \"GPIO out set\" >/proc/gpio/GPIO39
root@gumstix-custom-verdex:~$ echo \"GPIO out clear\" >/proc/gpio/GPIO34
root@gumstix-custom-verdex:~$ echo \"GPIO out set\" >/proc/gpio/GPIO34

GPIO34 and GPIO39 are also FFUART.
So how can I use GPIO 34 as an input/ouput and not a serial. How can I
turn off ttyS0?

-----------

The FFUART (for Full-Function UART) serves as the default Linux
console on the Gumstix, and supports rates of up to 230kbaud. It is
exposed on nearly all 60-pin expansion boards.

The Gumstix Linux kernel maps the FFUART to /dev/ttyS0.
link: http://docwiki.gumstix.org/index.php/Gumstix_UARTs#FFUART

Thanks

On Tue, Dec 9, 2008 at 7:06 AM, Harry J Mason <hj...@ec...> wrote:
> On Mon, 8 Dec 2008, Linus Casassa wrote:
>
>> I connected a sht15 to the gumstix on the gumstix port in the robostix.
>> GPIO 39 and 34.
>
> The SHT15 has a different pinout to the SHT75 I used; check the datasheet
> carefully.
>
>> And I commented the GND VDD part in the main function:
>
> Are you providing 3.3v power some other way? No idea if it will work if the
> power is 5v and the data 3.3v.
>
>> The I compiled and run and I got error 4:
>>
>> root@gumstix-custom-verdex:~$ rayabot-sht15
>> temp: 65280 humi: 65280
>> Got error 4
>> temp: 65280 humi: 65280
>> Got error 4
>> temp: 65280 humi: 65280
>> Got error 4
>> temp: 65280 humi: 65280
>> Got error 4
>>
>> If I disconect the sensor and run again the program it returns the same
>> error.
>>
>> So I am guesing that the read and write of the GPIO are not working...
>
> "Error 4" means it tried to read 4 bytes and got an error (1 bit) every
> time.
> This happens when the data pin is always high on reading. When this happened
> to
> me it was because the sensor wasn't connected properly, so the input was
> always
> 1 due to the pullup resistor.
>
> You should be able to increase all the delays everywhere in the code without
> affecting the logic. Stick a scope or meter on the pins and see if it makes
> sense. If the clock and data levels change, it's likely to be a hardware
> rather
> than software problem.
>
> --
> |  Harry Mason  | .------------. |    .___,    |"Whatever you do will be |
> | University of | |  hjm200 @  | | ___('v')___ | insignificant. However, |
> |  Southampton  | | zepler.net | | `"-\._./-"' | it is vitally important |
> |    England    | '------------' | hjm ^ ^     | that you do it." Gandhi |
>



-- 
Linus Casassa
Estudiante Ingeniería Civil Electrónica
Fono: 56-9-97776941