Now, is it possible that the accelerometer is "warming up".
Previously you said that you were _sure_ the the value was
stable when the first measurement was made. Can you try it
with a truly fixed voltage on the A0,1 input? The fact that
the InternalVoltage is OK would indicate that initial S/H
time might not be the culprit...

MS

Dimas Abreu Dutra wrote:
> Hello Everyone,
> 
> Sorry for taking so long to answer. But here are my comments:
> 
> * I noticed the error on the first readings when using the external 
> channels (A0 and A1). It happens on both tmote invent that I have access 
> to. I am using the tmote's own sensors (accelerometer), and I've done no 
> hardware modifications.
> 
> * When I used the Msp430InternalVoltageC the first reading was OK.
> 
> * I believe on Michael's hypothesis, which is that the sampling time of 
> the sample and hold is not correctly set on the first run. The ADC 
> counts I get are like if there was not enough time to charge the 
> capacitor, they are proportional to the input voltage, but much smaller.
> 
> ~Dimas~
> 
> 
> On Thu, Jun 19, 2008 at 8:45 AM, Jan Hauer <[EMAIL PROTECTED] 
> <mailto:[EMAIL PROTECTED]>> wrote:
> 
>     What ADC channel are you using, external (A0-A7) or internal (>7)?
>     does the conversion require stable reference voltage (if so, is your
>     application wiring to AdcReadClientC or Msp430Adc12ClientAutoRVGC()?).
>     Do you see the same result when you use
>     tinyos-2.x/tos/chips/msp430/sensors/Msp430InternalVoltageC?
> 
>     Jan
> 
> 
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help

Reply via email to