I haven't looked at the battery monitor on the tmote. There appears to be an "InternalVoltage" module which may be the battery voltage. However I would have to spend the rest of the day tracing through schematics, code, and the MSP manual to figure out what they think they are doing, which could well be different from what they actually end up doing.
I'm not that familiar with the controller so it may be that the MSP has some sort of stable ADC reference. But if that is not the case then measuring a resistor ladder as described isn't going to give any meaningful data since the battery, the reference, and the ladder values will change in sync. So I guess you're on your own unless some tmote guru feels your pain. As for the accuracy, if you look at any ADC results you will see that the last couple bits are "wobbly" due to various types of noise, internal, circuit board, and from the signal itself. And if you look at the power supply line with a scope I'll bet you'll see 10-100mv of junk. Your analysis: 1.5/4096 is (probably) wrong because the max conversion value is still the reference at 3v, so the highest value you can get from the equal-resistor ladder would be 2048. But you made up for it by doubling it later... sorry I'm not that much use MS Flavio Pompermaier wrote: > Since I'm working on Tmote Sky (12 bit sampling resolution) I had a look > on the relative datasheet at page 23. > It says that Vcc= (sampledData/4096)*Vref*(2R/R) and sampling is done in > the middle of 2 equal resistors between the Vcc and ground. > So, from what I could understand Vref (a stable voltage level taken from > somewhere) is equals to 1,5V in this case. > So, the sampled value has an accuracy of > 1,5/4096=366microvolts...However, when measuring the Vcc this error is > doubled since the sampled voltage is relative to just an half of Vcc. > Thus, the error rises to 732microvolts, right? This, if Vref ,Vcc are > stable and there's no error with sampling. > As the battery get consumed, the Vcc decreases, but Vref should stay > stable, isn'tit? > Software components tries to keep Vref constant..Is that you were > saying? Then there's some sampling instability due to some strange > effect such as recovery of charges or stuff like that I imagine..that > should be the greater source of error I think, since for sampling > resolution I would have an error of just 1mV.. > Have I misunderstood something...? > > > --- *Gio 1/5/08, Michael Schippling /<[EMAIL PROTECTED]>/* ha scritto: > > Da: Michael Schippling <[EMAIL PROTECTED]> > Oggetto: Re: [Tinyos-help] Voltage measurament accuracy > A: "Flavio Pompermaier" <[EMAIL PROTECTED]> > Cc: "tinyos-help@millennium.berkeley.edu" > <tinyos-help@millennium.berkeley.edu> > Data: Giovedì 1 maggio 2008, 01:06 > > I think the user guides for the various Motes have some information > about the battery voltage measurement. However the Battery components > seem to vary across platforms and TOS versions. I haven't payed much > attention > to them so I can't help with that more. Try searching > this list for battery measurement info. > > One rather arcane thing you should know is that, on the mica's > at least, the "battery" measurement made by ADC7 is actually from > a fixed voltage reference chip and the reference value (the max > value you can convert) is the actual battery voltage. So the value > you get RISES as the battery ref drops. I believe the Battery software > components adjust for this, which makes them very hard to understand. > > If you're confused about the conversion resolution I gave below > let me know where you got lost and I'll try to make it clearer. > I just guessed at the accuracy based on the fact that the bottom > couple of bits from the onboard converters are usually not very stable. > > MS > > Flavio Pompermaier wrote: > > thank for the reply. However is there any documentation about this > calculation? I have not well understood the > part regarding the noise and the > conversion to millivolts.. > > > > Michael Schippling ha scritto: > >> The ADCs used are either 10 or 12 bits, depending on the controller. > >> That's either 3/1024 volts for the micas or 3/4096 volts for the > tmotes. > >> I think some of the Battery components convert the value to > milli-volts, > >> but there's some noise involved so the actual accuracy is likely > ~10 mV. > >> MS > >> Flavio Pompermaier wrote: > >>> Hi to all, > >>> is there anyone that knows the accuracy of batteries voltage > measurement? > >>> Is it in the order of millivolts or microvolts? Or what? > >>> > >>> Best regards, > >>> Flavio > >>> > >>> > >>> > ------------------------------------------------------------------------ > >>> Tante idee per la salvaguardia del pianeta > su > >>> Yahoo! for good > >>> > > <http://us.rd.yahoo.com/mailuk/taglines/isp/control/*http://us.rd.yahoo.com/evt=52436/*http://it.promotions.yahoo.com/forgood/environment.html>. > > >>> > >>> > >>> > >>> > ------------------------------------------------------------------------ > >>> > >>> _______________________________________________ > >>> Tinyos-help mailing list > >>> Tinyos-help@millennium.berkeley.edu > >>> > https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help > >> -- > >> Platform: WinXP/Cygwin > >> TinyOS version: 1.x, Boomerang > >> Programmer: MIB510 > >> Device(s): Mica2, MicaZ, Tmote > >> Sensor board: homebrew > > > > > > > > Tante idee per la salvaguardia del nostro Pianeta su Yahoo! For > Good > > > > http://it.promotions.yahoo.com/forgood/environment.html > > -- > Platform: WinXP/Cygwin > TinyOS version: 1.x, Boomerang > Programmer: MIB510 > Device(s): Mica2, MicaZ, Tmote > Sensor board: homebrew > > > ------------------------------------------------------------------------ > Inviato da Yahoo! Mail > <http://us.rd.yahoo.com/mailuk/taglines/isp/control/*http://us.rd.yahoo.com/evt=52437/*http://it.docs.yahoo.com/mail/overview/index.html>. > > > La casella di posta intelligente. -- Platform: WinXP/Cygwin TinyOS version: 1.x, Boomerang Programmer: MIB510 Device(s): Mica2, MicaZ, Tmote Sensor board: homebrew _______________________________________________ Tinyos-help mailing list Tinyos-help@millennium.berkeley.edu https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
