[Tinyos-devel] a DAC setting module/interface for T2
John Griessen
john at ecosensory.com
Tue Feb 12 14:09:48 PST 2008
Is anyone interested in discussing a future DAC interface for T2?
I will be making working code soon for a sensorboard that uses a DAC. Setting the DAC output is compatible with
a telosb mote expansion connector, so others could test the interface without needing the specifics of my
12 channel soil moisture sensorboard. My use of the DAC is to get calibratable
excitation voltage for analog capacitive moisture sensors. The DAC of a MSP430F1611
has a max current of +/- 500 uA at Vcc=2.2 Volts. It has 250 Ohms max output resistance, so loading it
with a multimeter is a reasonable suggestion for a demo app. I imagine most DACs can be read
with a multimeter without loading them down much, so the HIL function could be to set a voltage, and
suggest confirming it with a high impedance multimeter.
On the MSP430F1611 DAC output pins are shared with ADC, so does that mean a conflict?
I think not -- the DAC can probably be set independently of ADC operation. The datasheet doesn't say much.
Presuming there's no resource to manage related to the DAC/ADC shared pins, the two DACs have
a control register and a data register each, so what does an interface need?
Writing data to the data register and outputting can be one command called Write in the case of one
16 bit word being output, but DMA can be used, so another command WriteStream would be used to
start outputting words from DMA as some timer interrupt happens. WriteStream would use
a DMA buffer pointer and DMA buffer length as parameters.
I just need a single word output for my applications, so I think I will put off the HPL and HAL level
DMA interface for a while, and just implement the commands Dac12.Write() Dac12.Config().
Comments? Parts I forgot?
---------------------------------
Philip Levis wrote:
> On Feb 10, 2008, at 2:52 PM, John Griessen wrote:
>> So, is a DAC setting module a candidate for a TEP? I think it is a natural
>> for sensor networks, since it corresponds to sensor excitation source
>> for a sensor mostly done in code with analog resources (spare op amps,
>> comparators), increasingly found on microcontrollers,
>> plus a handful of passive parts.
>
> It is. TEPs have to have an associated/sponsoring working group, but
> their authors need not need to be a member of the WG.
>
> Phil
------------------------------
John Griessen
--
Ecosensory Austin TX
tinyOS devel on: ubuntu Linux; tinyOS v2.0.2; telosb ecosens1
More information about the Tinyos-devel
mailing list