Hi Miklos

Without using ActiveMessageC the current was 4.5mA, using ActiveMessageC the
current was 2.8mA

Charles

On Thu, Jan 29, 2009 at 3:47 AM, Miklos Maroti <[email protected]>wrote:

> Hi Charles,
>
> Does it make any difference if you have the ActiveMessage included or
> not? How much does it draw without the ActiveMessage?
>
> Miklos
>
> On Wed, Jan 28, 2009 at 9:02 PM, Charles Elliott
> <[email protected]> wrote:
> > Hi Janos
> >
> > I tried taking out those lines an was still getting 3mA draw.
> >
> > I've switched to the Null application adding the flowing line to
> > NullAppC.nc:
> > "47  components ActiveMessageC;"
> >
> > No other changes were made and there is still 3mA of current flowing.
> >
> > Charles
> >
> > On Fri, Jan 23, 2009 at 5:15 PM, Janos Sallai <
> [email protected]>
> > wrote:
> >>
> >> Charles:
> >>
> >> As I wrote in my previous mail, to make sure that the RF230 radio is
> >> turned off after bootup, just add "components ActiveMessageC;" to any of
> >> your configuration files. You don't need to wire anything. On bootup,
> the
> >> RF230 stack will automatically take the radio to deep sleep. Of course,
> if
> >> your app starts using the radio, it will turn on...
> >>
> >> In your code (TestSchedulerC.nc), you _explicitly_ start the radio:
> >>
> >> 129:    if (call RadioControl.start() != SUCCESS)
> >> 130:      call Leds.led0Toggle();
> >>
> >> That's why the mote is drawing 3mA.
> >>
> >> As a sidenote, if you want to use low power listening (disabled by
> default
> >> on all platforms), you have to add -DLOW_POWER_LISTENING to PFLAGS in
> your
> >> makefile.
> >>
> >> Janos
> >>
> >>
> >> On Fri, Jan 23, 2009 at 3:57 PM, Charles Elliott
> >> <[email protected]> wrote:
> >>>
> >>> Thanks for the ideas but we are still seeing ~3mA when running the
> >>> application.
> >>> I've attached the source and makefile to this email.
> >>> Thanks
> >>> Charles
> >>>
> >>> On Wed, Jan 21, 2009 at 3:38 PM, Janos Sallai
> >>> <[email protected]> wrote:
> >>>>
> >>>> Charles,
> >>>>
> >>>> Just add "components ActiveMessageC;" to any of your configuration
> >>>> files. You don't need to wire anything. On bootup, the RF230 stack
> will
> >>>> automatically take the radio to deep sleep. Of course, if your app
> starts
> >>>> using the radio, it will turn on...
> >>>>
> >>>> Janos
> >>>>
> >>>> On Wed, Jan 21, 2009 at 2:13 PM, Charles Elliott
> >>>> <[email protected]> wrote:
> >>>>>
> >>>>> Thanks for the quick reply Miklos.
> >>>>> Using the program from \tinyos-2.x\apps\tests\rf230\RF230Test as a
> >>>>> guide we added in RF230LayerC and RF230ActiveMessageC to the wiring
> file.
> >>>>> To TestSchedulerC we added RadioState.turnOff() command and all of
> the other
> >>>>> required wiring and include statements.
> >>>>> After making these changes the motes drew 1mA less than before.
> >>>>> What are we missing?
> >>>>> Thanks
> >>>>> Charles
> >>>>> On Tue, Jan 20, 2009 at 12:17 PM, Miklos Maroti
> >>>>> <[email protected]> wrote:
> >>>>>>
> >>>>>> I think you need to include the RF230LayerC component to have to
> radio
> >>>>>> chip initialized and to move it into low power state. Miklos
> >>>>>>
> >>>>>> On Tue, Jan 20, 2009 at 11:11 AM, Charles Elliott
> >>>>>> <[email protected]> wrote:
> >>>>>> > Any help for the following question:
> >>>>>> >
> >>>>>> > "I have a question about the minimum power consumption in Iris
> >>>>>> > module and
> >>>>>> > telosb module when both of them are in sleep state and all of
> their
> >>>>>> > external
> >>>>>> > devices are inactive. According to my measurement, the current
> >>>>>> > consumed is
> >>>>>> > only 7uA by telosb, but is 5mA by Iirs, which is quite high and is
> >>>>>> > contradict with the specification given by Crossbow.
> >>>>>> >
> >>>>>> >
> >>>>>> >
> >>>>>> > I did experiments on these two kinds of motes using the program
> >>>>>> > TestSchedulerC. It is a program under the tinyos2-x/apps/tests
> >>>>>> > directory. In
> >>>>>> > the program, after the mote is booted, it posts three tasks to
> >>>>>> > toggle the
> >>>>>> > three led in the mote. According to the component wiring
> >>>>>> > relationship, when
> >>>>>> > the task queue is empty, the scheduler will finally invoke
> mcuSleepC
> >>>>>> > module
> >>>>>> > to get the microcontroller into low power state. I installed the
> >>>>>> > program on
> >>>>>> > both Iris and telosb, and used a multi-meter to measure the
> >>>>>> > electricity
> >>>>>> > current.
> >>>>>> >
> >>>>>> >
> >>>>>> >
> >>>>>> > Can anyone give me some suggestions? Any help will be greatly
> >>>>>> > appreciated!
> >>>>>> >
> >>>>>> >
> >>>>>> >
> >>>>>> > Jinfu Zheng"
> >>>>>> >
> >>>>>> > We are using the current tinyos source tree from the CVS
> >>>>>> >
> >>>>>> > Thanks
> >>>>>> > Charles
> >>>>>> >
> >>>>>> > _______________________________________________
> >>>>>> > Tinyos-help mailing list
> >>>>>> > [email protected]
> >>>>>> >
> >>>>>> >
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
> >>>>>> >
> >>>>>
> >>>>>
> >>>>> _______________________________________________
> >>>>> Tinyos-help mailing list
> >>>>> [email protected]
> >>>>>
> >>>>>
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
> >>>>
> >>>
> >>
> >
> >
>
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help

Reply via email to