I resume the info suggested:

1) It is possible to use EEPROM and program FLASH memory for data storage
(though only for 1000 writes): however it is enough for this use and it a
nice stuff if the aim is a all-in-one logger.
2) If necessary, I can get by with 7 bits data and store two measurements
for instruction word (each 14 bits wide). I want at least 0.2 C. resolution
but the temperature range needed is very short (5 -15 C), perhaps 7 bit
value may be enough.
3) Isn't possible use a single 3V lithium cell (my PICs aren't LF type). But
with 32khz clock current consumption is very low, I can use  4 silver-oxide
button cells, 100 ma/h for about 2 months period.

I'm aware that this "all-in-one"  'F876 isn't the better choice for a
miniature and efficient datalogger, but I want an easy to build unit with
few parts: besides, I explain briefly my focus.

The purpose of this logger is check the water temp. of a little river  that
disappear in a cave. Water was traced with fluorescent dyes: it exit from a
large spring at about 8 miles from the entrance. The logger will sealed into
a plastic ball with ballast and will have to run through unknown submerged
passages. I haven't any assurance to recovery a single logger at the spring,
so I will make some of these units to have a sufficient probability to find
at least one of them!!

Many thanks to all people that have replied.
Marco








----- Original Message -----
From: "Jeff DeMaagd" <jeff@DEMAAGD.COM>
To: <PICLIST@MITVMA.MIT.EDU>
Sent: 7 marzo, 2002 16.30
Subject: Re: [PIC]: mini 16F876 datatogger


> The F87x series can write to its own memory.
>
> I bet that you might be able to write your software in a hundred
> instructions or so, leaving most of the rest of 8k words available for
> data.  If you can get by with 7 bit resolution, you can store two
> measurements per instruction location, as each instruction word is 14
> bits.  You could theoretically store 14,000+ samples with this.
>
> I don't know how many samples you want to take.  The method I suggest
could
> net 10 samples per hour for two months.
>
> The LF advice is a good one to heed.  If you do end up using the internal
> ADC, make sure you shut it off after every sample, as well as any other
> feature you use.
>
> I don't know if that one lithium cell is enough for this use, I've never
> tried this before.
>
> Jeff
>
> At 03:14 PM 3/7/02 +0100, you wrote:
> >A newbie question.
> >I would like to build  a very small, cheap and easy to-assembly
temperature
> >datalogger (about 1" x 2" sized) to use for a strange problem. Due to the
> >small size, I'm thiking to use only a single PIC chip (no external
EEPROM).
> >Probably a smaller size will be possible using an 8pin PIC and a serial
> >eeprom as 24LCxx, but I have some 16F876 and I want use them. These are
my
> >specs:
> >
> >-  Size 1" x 2" (about)
> >-  32Khz clock
> >-  Temp.measure with a thermistor.
> >-  one 3V lithium cell
> >-  >= 2 months operativity
> >-  >= 150 bytes data storage
> >
> >My doubts:
> >1) Is it really useful to store the temp. data in the 256 EEPROM bytes of
> >'876?
> >2) Is it possible to store the data also in an (unused) area of the
program
> >memory?
> >
> >Many thanks
> >
> >Marco
> >
> >--
> >http://www.piclist.com hint: To leave the PICList
> >mailto:piclist-unsubscribe-request@mitvma.mit.edu
>
> --
> http://www.piclist.com hint: To leave the PICList
> mailto:piclist-unsubscribe-request@mitvma.mit.edu
>
>

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads