Starting Point System : Forum

You are not logged in.

#1 2009-11-03 20:23:33

cjtoff
New member

Temperature acquisition

I wish to use some of the chameleon analog inputs to acquire temperatures in range 0 to 100 C.
I intended to use a PT1000 device [ 0deg = 1000ohms - 100deg = 1385ohms]
Are there recommendations for the type of probe to use or the type of circuit to connect to input?
Thanks

Offline

 

#2 2009-11-11 03:49:36

phealt
New member

Re: Temperature acquisition

CJ  -  I'm building something similar,   the PT1000 is a thermistor so you will need to build a wheatstone resistor bridge with the PT1000 as R1.  You will probably have to mess around with the resistor till you get the range you want.  There are a bunch of circuits out there.

Found this link:  http://www.facstaff.bucknell.edu/mastas … sist4.html

Last edited by phealt (2009-11-12 17:55:46)

Offline

 

#3 2010-04-06 22:12:26

Greasehauler
New member

Re: Temperature acquisition

I'm using a LM35 device that gives you a direct temp in C. The problem is, if I have 5 volts divided between 255 "clicks", the accuracy is +/- 2C. That is too much of a variance for my app. Has anyone been able to solve the problem to be able to measure within 0.1C? I only care for temps between 80C and 110C.

Offline

 

#4 2010-04-09 13:15:40

gremlinc5
Member

Re: Temperature acquisition

You could use a SPI IC Temperature Sensor like the TC72 -55 to 125 ░C anda nice 0,25░C/bit

http://search.digikey.com/scripts/DkSea … -5.0MUA-ND

It's funny as digital part, and you dont need electronic math.

Connect MISO MOSI and SCK pins and use SPI swap to find out temperature from sensor... it's not too hard.

Last edited by gremlinc5 (2011-10-18 10:24:29)


ST7FLite39 daughter board
REva 2.10
RLink 3.4
"uChameleon" as: SPI Master for debug, Analog output for sensing emulation

Offline

 

#5 2010-04-16 10:06:52

gremlinc5
Member

Re: Temperature acquisition

Greasehauler wrote:

I'm using a LM35 device that gives you a direct temp in C. The problem is, if I have 5 volts divided between 255 "clicks", the accuracy is +/- 2C. That is too much of a variance for my app. Has anyone been able to solve the problem to be able to measure within 0.1C? I only care for temps between 80C and 110C.

You must normalize 0 to 5 volt (255 clicks) with 80░C and 110░C... you could use a referenced and amplified op-amp.

Have you solved the problem?

Last edited by gremlinc5 (2011-10-18 10:24:41)


ST7FLite39 daughter board
REva 2.10
RLink 3.4
"uChameleon" as: SPI Master for debug, Analog output for sensing emulation

Offline

 

Board footer

Powered by FluxBB

Home