I have been using a temperature sensor and opamp to give me a 0-4 volt
0-100oc temperature range. This fed into the ADC of an f877. This range
allowed me to easily display the temperature on an alphanumeric display
by a simple left shift of the 8 bit ADC value. I.E. 200 decimal would
equate to 100oC which was simply BCDed to the LCD display. The cost and
component count of the sensor and op-amp have got me looking for a
simpler method perhaps with a thermistor and another resistor.
Before I reinvent the wheel I thought I would ask here for your
suggestions and personal experience at making an almost zero component
count cheap 0-100oC temp sensor with a pic.
We've had good luck with an LM75 (National). It is an I2C interface, so it
interfaces to the I2C interface of an f877 or other PIC with just a couple
of pullup resistors. They tell me that we've been paying just over a dollar
in few thousand quantities.
It sleeps at a couple of uA via a command given over its bus, and reads out
directly in degrees C. For our application, its worst feature is that after
wakeup, it takes 100 mS for its delta-sigma converter to settle to an
accurate reading.
I am just learning about using an LM20, which is supposed to be able to
drive an ADC input directly. So far all I've gotten is noise but I think
that's because of the crudness of the prototype lashup I'm using.
Since it claims an output impedence of 160 ohms, and 0-2.5V output range, it
should work directly connected to a PIC ADC input without any buffer amp.
It is supposed to run on 10uA or less, but for super-low sleep current, you
need to supply its VCC from a PIC output so that it can be turned off when
sleeping.
Hi, have you tries using the Dallas DS1820 / 1, it is a 3 pin device, it
uses 1 wire communication protocol. With it being digital, there is no
calibration required.
>Hi, have you tries using the Dallas DS1820 / 1, it is a 3 pin device, it
>uses 1 wire communication protocol. With it being digital, there is no
>calibration required.
Carefull with the assumption hat because these sensor a direct to digital
they require no calibration. Most of the DS1820 we use register a value
that's incorrect by -2 deg! An offset value is almost always required. In
high resolution mode we have to recalibrate theses sensor daily.
>
> Carefull with the assumption hat because these sensor a direct to digital
> they require no calibration. Most of the DS1820 we use register a value
> that's incorrect by -2 deg! An offset value is almost always required. In
> high resolution mode we have to recalibrate theses sensor daily.
>
> Gary
>
I have also required an offset (approx -2) and slope (though small) when
using the DS1820 in high resolution mode. This is easly dealt with.
A more important problem I have with either the DS1820 or my setup is
that the DS1820 does not seem to read below about -10 C (and some will not
even go that low).
Currently I am assuming it has something to do with my 1-wire bus
timing (bit banging on a 4.096MHz PIC16C56/84) at the lower temperatures,
but so far I have not found a solution in that direction.
When they are reading, all functions work correctly including writing
to the user bytes and comparing CRC values.
I am curious as to whether others have ecountered similar problems.
Jamie Lawson
Controlled Environment Systems Group
Plant Agricuture
University of Guelph
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
I am using a DS18S20 for my outdoor temperature indication and it is
currently at -3 degrees F. I have found it agrees quite well with an Hg
thermometer. I have found that supplying 5 volts to the positive pin solved
a lot of wierd, unexpected readings. I am using Belden 1588 , Cat 5 cable
so the third wire wasn't a problem.
MITVMA.MIT.EDU]On Behalf Of Steven Chaplin
