Am 01.10.2015 um 10:10 schrieb Henry Pootel:
>
>> Do you run the onewire with 3.3V from the Raspberry's I/O pin? If so,
>> have you checked all your slaves are 3.3V compatible?
>>
>
> No, I use 5V for Rasberry and for w1 devices. It's one 5VDC power supply.
>
Henry, all the I/O pins on the pin header of the Raspberry Pi are 3.3V.
The Raspi board is a 3.3V device, 5V is only used for USB power and on
HDMI I²C, all other I/O pins are 3.3V. So if you connect a onewire bus
directly to GPIO4, your bus is run at 3.3V.
(Both the DS2450 and DS18B20 support a 3.3V bus, the DS2450 still needs
5V as supply voltage then, however.)
>
> And I've 4.7K resistor from the 1-wire to +5V.
>
Make it 1kOhms and ***tie it to 3.3V*** instead. The Raspberry Pi
doesn't tolerate 5V on its GPIO pins. The only reason you haven't fried
the CPU yet is the current is limited to ~1mA by the resistor. Still,
all kind of odd things can happen if you overvoltage the CPU I/O pins.
Mostly because you are opening protection diodes in the CPU input pin
which may cause timing issues.
>
> But I'll check a power line by scope. Thank you for an idea.
> However, I've got the same result with transformer and impulse power
> supplies (220VAC to 5VDC). I think the problem is not in bumps.
>
I meant bumps on the onewire busline, not on the power line. The
High->Low edges have to be rectangular. If you see bumps between 0V and
1V on the bus line near a High->low edge it means the host isn't able to
tie the bus to 0V correctly for some reason.
Kind regards
Jan
------------------------------------------------------------------------------
_______________________________________________
Owfs-developers mailing list
Owfs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/owfs-developers
