Hi Jens,

I know you can make things in 1,000 different ways but in the end, it
is powered by a battery which basically means your solution cannot use
more than say 5uA average current. There is not a lot you can do with
that :-) And secondly, the ideal solution would preferably fit within
1 cubic mm :-).

Michel



On Apr 21, 9:01 pm, jb-electronics <[email protected]>
wrote:
> Hi Michel,
>
> you are right, I did not assume it was so complicated. However, have you
> considered using frequency modulation of the IR signal? This way you
> could easily subtract daylight digitally.
>
> Jens
>
>
>
>
>
>
>
> > It really is a bit more complicated than that Jens. I generate a very
> > short but very strong IR pulse, I'd have to check the exact numbers
> > but it is somewhere around 250mA for 2 or 3 usec. The IR light
> > reflects from an object above the watch (a hand for example) onto an
> > IR photodiode. To make this work in all situations, the photodiode
> > should not saturate when there's a lot of light around (say you are
> > outside in the sunshine), so it's series resistor is relatively low
> > (which also helps making it very fast) and the signal received from
> > the powerful 250mA pulse is only like 40mV.
>
> > I first built that with some gates, capacitors and resistors. It
> > worked very well but used about 400uA average current if I am not
> > mistaken.
>
> > 10uA is pretty much what it should be according to the data sheets if
> > you add up all the currents. Honestly, a battery will last for ages
> > (about 8 years) with a 10uA load, so I am really not worried about
> > that. It's just that the processor supports deep sleep mode, so I'd
> > like to incorporate that as well
>
> > Michel
>
> > On Apr 21, 7:59 pm, jb-electronics<[email protected]>
> > wrote:
> >> Hello Michel,
>
> >>> Problem is that I need more components to be able to generate the
> >>> interrupt
> >> I would have used simple diodes that are OR'ed together to the external
> >> interrupt pin.
>
> >>>    and believe it or not, it actually used more current.
> >> Actually, I don't see that happening with diodes, do you? OK, if your
> >> sources have different voltage levels you need shifters and these
> >> require current, of course.
>
> >>> Standby mode means the processor is in sleep mode but wakes up about 8
> >>> times per second to check the sensor. The sensor only uses 2uA but the
> >>> processor uses more current in sleep than in deep sleep mode, so all
> >>> together it is 10uA.
> >> Have you played around with the configuration bits? I once realised that
> >> a silly enabled Brown-Out detect (which makes no sense with
> >> battery-powered devices) takes up almost all the current.
>
> >> Sorry if this has been mentioned before, but which processor do you use?
>
> >> Best regards,
> >> Jens
>
> >>> In deep sleep, the processor is completely powered down, only the RTCC
> >>> continues running. It can only be awakened by an external interrupt
> >>> (push button rather than motion sensor).
> >>> Michel
> >>> On Apr 21, 7:23 am, jb-electronics<[email protected]>
> >>> wrote:
> >>>> Looks very good!
> >>>>> Total circuit power is about 10uA during standby mode and 2uA during
> >>>>> deep sleep.
> >>>>> Deep sleep will be entered after a standby time-out to save battery
> >>>>> life. Clock will still run in deep-sleep mode.
> >>>> I am curious: Do you use an external interrupt to wake the processor in
> >>>> case of a motion etc? This might be useful and could avoid the "standby
> >>>> mode" altogether.
> >>>> Jens

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To post to this group, send an email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/neonixie-l?hl=en-GB.

Reply via email to