Created Date: November 21,2009

Created by: Tinker

Can I use a NTC Thermistor with either the RTD or Thermocouple card? is there any way to use the NTC as an RTD and just build in a different range value for that sensor?

It would help to know what you are trying to acomplish, and why would you even want to do that. One thing that occured to me is that some things (some servo motors for example) already have thermisitors built in, and you may have spare channel on a RTD or thermocouple card that is already installed. While that senario makes sense, one would still want to know what you expect to acomplish, if you are trying to avoid burning up a motor, you probably don't need fractional decgree acraucy, being even tens of degrees off would probably not be too bad in that case (especialy if you put your alarm set point on the low side to start with) Or perhaps you just want to know if a temperature changes, and what the actual temperature is not so important as the fact that it is not what it was.

There are a number of probems though, first an RTD is a positive temperature coefficient device so a NTC thermisister will give reverse readings, also the coeffcient of a thermosister is generaly much higher that an RTD. And a thermocouple generates signals in the milivolt range while thermisistor is just a resistor, you would need to have a source of voltage to use a thermocouple input.

Still, if you combined your thermisistor with some fixed resisters and, for the thermocouple input, a voltage supply, you probably could do it, but there would be some seroius math or table lookup to get a value anthing like the actual temperature.

Oh, also a thermocouple inputs cold juntion compenstation will not be helping in this case, some thermocouple inputs have the option of switching to a linear milivolt range, it that is an option I'd use it.

Created Date: November 21,2009

Created by: DetroitSound

Thermistor input

I don't know if you 're familiar with the concept of a Wheatstone Bridge

http://upload.wikimedia.org/wikipedia/commons/thumb/9/93/Wheatstonebridge.svg/300px-Wheatstonebridge.svg.png

You would probably need to use a circuit like this to get it to work properly (for this you would use it with a voltage input like a thermocouple).

Your only real issue would be scaling. Even for thermocouples, different types (like J, K, E, T and so forth) all have different temperature / voltage curves (voltage at 0c versus 25c versus 50c versus 75c versus 100c and so forth). Even with RTDs there are two major types and they have different curves. In your case the thermistor would replace Rx, and R1, R2 and R3 would have to be chosen to get the desired result. Vg is the voltage output and the choice of resistors and reference voltage will make the difference in how it operates.

But - I am not an engineer, I never claimed to be. I like to use the simplest design that works, and for me, if you have something that is capable of an RTD or thermocouple input and you can get one of them that suits the needs of the application, I would just use that and not mess around.

Unless you have an important reason that you HAVE to use the thermistor...

Created Date: November 21,2009

Created by: DetroitSound

With this circuit, assume that you want the voltage to rise with the temperature. The resistors would be chosen to give a zero (or small) voltage across Vg, D is positive and B will get more negative as the temperature increases. R1 and R2 will have the biggest impact on the curve while R3 will affect the null (lowest voltage).

Another thing to be aware of - termistor is a contraction of two words - THERMal resISTOR. Any voltage across a resistor will create current (that's how its resistance is measured) and voltage times current equals watts. (simple Ohm's law). So simply the current through it can cause the thermistor to heat and change its apparent value. This is not always a bad thing; some circuits use this to provide protection. But when it is used for measuring purposes the currents have to be kept very low (often under 100 microamps) to keep it from heating and affecting the reading. Under the right conditions a NTC thermistor can go into "thermal runaway ", meaning that as it gets hot, its resistance drops and its current increases which causes it to dissipate more power, which causes it to get hotter, its resistance drops more, current goes up, power dissipation goes up and so forth. An RTD uses a PTC and it would tend to reduce this problem inherently.

Which is another good reason RTDs have widespread industrial use.The resistance is much lower and they have much less of a problem with internal heating. Three and four wire RTDs make it easy to compensate for resistance in the leads. Plus they tend to be more rugged.