RTD are among the most accurate temperature sensors there is, however the raw sensors output a very low millivolt output and require good low noise insturmentaion op amps to raise the voltage to be more suitable for an arduino to utilize. A better solution might be to look at digital one wire sensors. There accuracy are good enough for most applications and the overall costs will probably be much better.
Adafruit sell some nice ruggedized ones suitable for indoor/outdoor use, check them out.
They even have a link to some software help and pointers to use them on an arduino:
Thanks for the information,
seems very simple to use I was asking to you also about PT100 because I've one PT100 at home and I was asking to my self if there is the possibility to use it also and if yes which is the schema to use for connect it to arduino ... Looking the accurancy of "Dallas DS18B20" is not too much precise ... because have + - 2° C against + - 0.4 ° of PT100
The DS18B20 is more than accurate enough for your needs, and is rugged and reliable. I got some PT100s before I got an Arduino. I bought a DS18B20, simply because I thought I might get something to happen more quickly. It was the best decision I ever made. I have never used the PT100s, they simply fill a hole in the plumbing.
Hi Guys,
Because I'm curios to try BS12B20 and PT100 (because I've already at home).
For BS12B20 I've already download the library for Arduino.
for the PT100,I've bought LM324 for amplify the signal ... once that I will did the circuit in attachment, just read the analog value and convert it with the PT100 conversion table ?
which is the calculation to do ?
gnusso:
for the PT100,I've bought LM324 for amplify the signal ... once that I will did the circuit in attachment, just read the analog value and convert it with the PT100 conversion table ?
which is the calculation to do ?
The link mentioned in reply #5 is the best information I know of and I believe it has all you need to know.
It occurred to me under the shower last night that they are essentially the same. The docs allude to this. Just change the 10k in the divider to 1k, and it will be good for PT100s.
The ADC would be any of the six analogue pins but I'm afraid I can't comment further on this. I stick with DS18B20 in order to specifically avoid the problems you are encountering!
I think the guide of open energy describes a PT100 with two wires.
I have a PT100 with 3 wires. Does anyone know how to connect the 3-wire PT100 to this curcuit?
Hi, just to confuse things you can even get PT100 in 4-wire mode.
If the length of lead from PT100 to the measuring circuit is large, ie 10metres then for continued accuracy you need 3 or 4 wire units.
In the case of the 4 wire, two of the wires supply sense current to the PT100, the other 2 provide the measured volt drop across the PT100 at very low current back to the measuring device which has high impedance input.
This is to minimise the resistance of the long leads.
Google PT100 4wire or 3wire for better explanation.
If you are going to use an amplifier to add some gain to the volt drop across a PT100 you need to use what is called an instrumentation amplifier.
This is usually an array of 4 op-amps connected to amplify the voltage but set up so that the drift of their characteristics oppose each other.
This is for temperature and supply stability.
If all you are using PT100 is to measure inside and outside temp then its overkill, 0.5DegC resolution and 1 or 2 DegC accuracy should be plenty from the other devices mentioned.
If you want accuracy, then use a bowl of icewater and boiling kettle and calibrate the devices after you have it up and running.
I use PT100's in my work and they are fine, but my customers are prepared to pay for the industrial grade interfaces needed.
(I admire anyone to take up the challenge of using LM324 to do the job in instrumentation configuraton, the LM324 application notes has a suggested circuit. Did one as a University experiment.)
Tom
One thing I'm not understanding on all of the topics regarding PT100.
Everybody says: "I want to have very high accuracy, thats why I want to use a PT100".
What nobody accepts: Most of the amplifier circuits are bullshit - really they are! They will have high drifts and accuracy will be not even in near of a DS18B20. And secondly they use all the shity A/D Arduino interface which has a super shit build-in A/D converter and the super shit AD pcb layout.
I was asking to you also about PT100 because I've one PT100 at home and I was asking to my self if there is the possibility to use it also and if yes which is the schema to use for connect it to arduino ... Looking the accurancy of "Dallas DS18B20" is not too much precise ... because have + - 2° C against + - 0.4 ° of PT100
