[Tinyos-devel] more bugging about current consumption on MicaZ in tinyos-2.x

[Philip Levis]

On Jul 18, 2006, at 10:12 AM, Razvan Musaloiu-E. wrote:

> Hi!
>
> On Tue, 18 Jul 2006, Philip Levis wrote:
>
>> On Jul 18, 2006, at 12:07 AM, Joe Polastre wrote:
>>
>>> The ATMega128L used on the MicaZ has a very wide sleep current.   
>>> Some
>>> customers have reported as low as 5uA sleep current while others  
>>> have
>>> reported as high as 50-500uA sleep current.  The processor uses a  
>>> very
>>> archaic design relative to the new ATMega's and the superior TI  
>>> MSP430
>>> processors.
>>
>> Interesting. Which low power state? I have 20 or so micazs hanging
>> around my office; I'll try to test a bunch of them and gather some
>> data. I also have some Teloi, so I can do the same for them.
>
> I just tested some almost unused MicaZ motes and they show the  
> expected
> low power consumption (~2uA measured with the Extech). The one that  
> shows
> high consumption was used in a real deployment for several months.  
> I also
> tested 6 other motes from the same deployment and they also show a low
> power consumption.
>
> So it looks like some MicaZ consume significantly more even in  
> Power Save
> and Extended Standby. If would be interesting to know the any of  
> the 20
> you have also has this problem.
>
> I apologize for bugging everyone on a problem that seams to be very
> isolated.

I just tested 4, and measured values of

2.8 uA
2.7 uA
2.4 uA
2.0 uA

(a student is using the rest, and I didn't want to disrupt that  
unless these values suggested something funny).

It sounds like one of your motes was in some way damaged in  
deployment, and that micaZs generally show ~2uA when in the lowest  
power state. Measuring the other power states would be good: in 2.x,  
at least, this is pretty easy to do with the McuPowerOverride interface.

Phil