[time-nuts] Sidereal timekeeping

Bob Camp lists at rtty.us
Wed May 18 16:20:07 UTC 2011


Hi

Ok, say you randomly pick a PIC 16LF1826 as your target device. It typically
pulls 7 to 9 ua running with a 32KHz crystal as it's clock and 1.8 to 3.0 V
on the supply. Max supply on the low power part is 3.6V so two cells in
series is about right for it. 

An alkaline AA cell is up around 2700 maH according to the ever useful
Wikipedia. 10 ua would drain the cell in 11.4 years. The AA's I buy don't
last that long sitting on the shelf un-used. 

The only obvious issue is to figure out the motor drive side and see what
driving it straight rather than with a switcher does to the motor current.
That's going to involve hacking up a wall clock. Since the clock is likely a
single cell gizmo, the current probably will be a bit high.

There's always the LCD or LED version to think about....

Bob  

-----Original Message-----
From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
Behalf Of Bob Camp
Sent: Wednesday, May 18, 2011 8:09 AM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Sidereal timekeeping

Hi

The question can be easily worked out. A PIC  running a 32,768 Hz clock will
pull X ua doing this or that. I don't know X as I sit here, but it's a
number that comes off a data sheet. That much current off of two or three AA
cells will let you run for Y days. My guess is that Y is a pretty big
number. If it's not, put in C's or even D cells. At some point the number
gets big.

The drive for the motor is pretty simple, or it was last time I did all this
(1970's). Having the PIC drive the motor is not all that hard. The code
likely fits in a pretty small part. Weather you use a PIC or something else
is open to study. There probably are parts that pull less current running at
low speed than the PIC. I can think of a half dozen parts I'd check out. I'd
also look for something that's happy with less voltage than most of the
PIC's.

It would take more time to lay out the pc board than to write the code once
you had a target processor and motor in mind.   

Bob


On May 18, 2011, at 12:01 AM, Hal Murray wrote:

> 
> lists at rtty.us said:
>> In a full sized wall clock, most of the power is to the motor. On a wrist
>> watch - it depends on how well the watch is built. 
> 
> Thanks.
> 
> I think that means that it's not silly to generate a PPSS (Pulse per
Sidereal 
> Second) signal by counting to 364/365 of 32678 rather than letting a 15
bit 
> counter wrap around.
> 
> If "most" means 90% and we use another factor of 10 to implement the 
> compare/reset, that only drops the battery lifetime by a factor of 2.
> 
> Divide the target count by 2 and toggle one more FF if you need better 
> symmetry on the output.
> 
> -----------
> 
> Hacking the crystal adds another dimension to the
hardware/firmware/software 
> tower.
> 
> Is there a term for that?
> 
> ----------
> 
> The party line for something like this is that 1/2 the power goes into the

> bottom bit.  (assuming we are talking about the logic and not the motor)
> 
> I wonder what the power ratio (battery lifetime ratio) is for custom CMOS
vs 
> say, 4000 CMOS, or C, or HC, or low power CPLD or ???   The CPLD might be 
> interesting since you don't have to drive external pins/pads.  There is 
> probably some 4000/HC chip that includes a counter but would save power 
> because it doesn't bring the bottom bit or two out to a pin.
> 
> 
> 
> -- 
> These are my opinions, not necessarily my employer's.  I hate spam.
> 
> 
> 
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.


_______________________________________________
time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.




More information about the time-nuts mailing list