Yahoo Answers is shutting down on May 4th, 2021 (Eastern Time) and beginning April 20th, 2021 (Eastern Time) the Yahoo Answers website will be in read-only mode. There will be no changes to other Yahoo properties or services, or your Yahoo account. You can find more information about the Yahoo Answers shutdown and how to download your data on this help page.
Trending News
Use 3 wire thermistor/RTD with Arduino?
I bought the following probe from Amazon.
http://www.amazon.com/gp/product/B005OUJTZI/ref=oh...
I'm wanting to use Thermistors for my project, as that's what others have had the best results with for the specific app I am building.
When I ordered it, I thought it was a Thermistor, but now i'm not sure if it's a Thermistor or RTD?
Assuming it is a Thermistor, how would I hook it into my arduino? I've spent hours and hours googling, but most examples are for 2 wire thermistors. There are a few diagrams for 3 wire thermistors and RTD's, but I still can't get it to work.
It has 2 red wires, and one white wire. Right now I've got the white wire hooked up to +5V, and one of the red wires hooked up to A0 (analog in), along with a resistor pulling it down to ground. When I hook the other red wire up to anything, my A0 reading goes to 0 or 1024. Even with the way I've got it wired, the difference between freezing water and boiling water is just about 5 steps on the 0 to 1024 scale.
Please Help!
3 Answers
- EckoLv 77 years agoFavorite Answer
Yes it is an RTD. These are generally more linear than a thermistor and cover a wider temperature range, but are not as sensitive. Higher grade ones are predictably accurate (interchangeable). Basically it is a spool of very fine wire with a resistance around 100 ohms at room temperature. When heated the resistance increases slightly, about 0.003925 ohm/ohm/°C (0.4%). This is the same as a piece of copper wire. That is why 3 wires are used. Two connect to the resistance, and one is common. The third wire connects to one end of the resistor too, but it does not carry the excitation current, so the copper wire doesn't come into the resistance current path. It just measures the voltage. This is a part kelvin connection. The typical maximum allowed current is <1mA, to minimize self heating. Using precision 1mA constant current excitation, that means the voltage at 100 ohms is 100mV and this changes by 0.4 ohms per degree, so up one degree is 100.4mV. Offset the 100mV and there is degrees left, which have to be amplified from 0.4mV per degree. Your ADC is probably 1024 steps in 5V so a single step is 4.88mV. This link shows the connections:
http://thermocouple-online.com/rtd-1G.html
The things to consider are temperature range and accuracy required and response time and self heating.
Generally a thermistor has a limited range because of its non linearity. However it has more like 4% sensitivity in its useful range. The basic concept is the same, they are both resistance changes, but a thermistor reduces with a temperature increase. With a thermistor a half bridge is possible as the lower temperature is partly offset itself by the non-linearity. Note that thermistors need to be for the temperature range involved. Typically a room temperature type is 5K at 25°. Working with 5V reference supply, a 5K resistor in series with the 5K thermistor which is grounded, and the junction goes to your ADC. At 25°C it will be ~2.5V = 512 counts, and a 1 degree increase results in ~2.4V (-0.1V or ~20 counts per degree). A full bridge works better though, and a constant current excitation source does too. You will need a good thermometer to calibrate this at many points.
The RTD circuit shown in another answer is a valid principle in theory, but the components need to be precision, otherwise they will also change with temperature. The op-amp circuit with 3 devices forms an instrumentation amplifier. These can be bought ready made as an IC, with internal precision resistors. The reference is used to offset the whole output, perhaps to zero volts at 0°C. Sometimes a bridge circuit is used with an instrumentation amplifier. The arms of the bridge have to be very accurate resistors, especially for an RTD.
Here is a circuit that will work if care is taken:
http://circuit-diagram.hqew.net/5V-Powered$2c-Line...
This version is linearized and has an output of 10mV/°C - the same as the very simple to use LM35.
http://circuit-diagram.hqew.net/Single-5V-Supply-L...
Frankly the RTD path is rewarding for high precision but difficult to achieve this.
There are simple approaches to temperature sensors, for example an LM35 which is very linear but will need calibration for high accuracy. There are various grades. Its output can be scaled up using a simple op-amp circuit. (non -inverting amplifier). This link has data sheets etc. It is widely available.
http://www.ti.com/product/lm35. There are also one wire temperature sensors, which are interfaced using a 1 wire serial bust to your arduino, not analogue.
I just saw the application is a smoker or barbecue in your other question, maybe the LM35 is too low at 125 degrees or so. The RTD with teflon leads would be good to about 200°C if you persevere. Search for RTD temperature transmitter. These may have 4-20mA output, easily adapted to a voltage output using a 200 ohm resistor giving 0.8 to 4V output for the AtoD..
- Anonymous7 years ago
It's an RTD and you really need a little extra circuitry but it's pretty straightforward and minimal.
This page is very clear about all the details: http://openenergymonitor.org/emon/buildingblocks/r...