remote temperature measurement using RCTIME and a thermistor

GuellingerGuellinger Posts: 2
edited February 2009 in BASIC Stamp Vote Up0Vote Down
Hi everyone,
Scott Edwards (Nuts and Volts) writes that the temperature measurement described in No 7 of Basic Stamp 1 Application Notes (using a thermistor and the RCTIME/POT command) is not suitable for remote use (i.e. not more than a foot between the thermistor and the Stamp). Can anyone tell me why that is?
Looking forward hop.gif to any clues.
Benji

Comments

  • 10 Comments sorted by Date Added Votes
  • JonathanJonathan Posts: 993
    edited February 2009 Vote Up0Vote Down
    Thermistor output very low voltages and longer leads will degrade the accuracy. If you want greater range, there are lots of other temp. sensors. It depends on what the temp. range you need is.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    www.madlabs.info - Home of the Hydrogen Fuel Cell Robot
  • GuellingerGuellinger Posts: 2
    edited February 2009 Vote Up0Vote Down
    Thanks for your reply.

    My measurement range is 10C to 140C, with lead lengths of up to 10m, and an accuracy of +-0.5K or better if poss.

    What attracted me to the RCTIME solution is that one STAMP could handle 16 RTCs, without needing many other components. Ideally I would like to use PT1000s, because they are widely used in my industry (solar thermal systems). With a 1Kohm resistor I would need a 1000uF capacitor (or larger) for the RCTIME set-up to work, and I'm not sure how accurate the results would be with a capacitor that size.

    I guess the solution for me is to use a PT1000 and an ADC?

    Benji
  • Tracy AllenTracy Allen Posts: 6,286
    edited February 2009 Vote Up0Vote Down
    Thermistors have a relatively high resistance, and in the case of the #7 app note you referred to, it is 10kOhms at 25 degrees Celsius, increasing at lower temperatures. When that resistance is out at the end of a pair of wires, the wires act as an antenna to pick up noise voltages that add to the thermistor signal. At the same time, the RCTIME command is very sensitive and responds to the instantaneous voltage including the noise component at the input pin relative to the Stamp Vss pin. In order to exact the best performance, it is possible to shield the twisted pair wires and to take care to avoid ground loops by grounding as close as possible to the Stamp Vss pin.

    The reading will bounce around with any noise that is picked up and added to the signal. In fact, you can make a nice random number generator by running the RCTIME command on purpose with a high-ohms resistor and big loop of wire.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Tracy Allen
    www.emesystems.com
  • Tracy AllenTracy Allen Posts: 6,286
    edited February 2009 Vote Up0Vote Down
    A platinum RTD is another matter. The resistance is 10 or 100 times lower than a thermistor, so the circuit is less subject to noise pickup. On the other hand, the resistance change with temperature is quite small. You are correct in your assessment to use a PT1000 with an ADC. If you don't want to get into op amps, the ADC should either be differential, so that it can work with a bridge circuit, and/or it should be high accuracy (16 to 24 bits) so that it can sense small voltage changes.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Tracy Allen
    www.emesystems.com
  • I have been searching for 3 weeks daily for a circuit I could build to acquire temperature, through use of an RTD, over a long distance. The wheatstone bridge circuit with DAC sounds like the answer. However I am struggling with this. I understand the concept of one of the resistors being variable in order to balance the bridge due to the varying resistance of the RTD. What I can't wrap my head a round is who wants a circuit where one needs to manually adjust a pot in order to balance the bridge every time they need a temperature reading and then determine the resistance of that pot. I have 8 different points I want measured. It would be more practical to walk the distance than balancing the bridge.

    Seems to me one should be able to send the voltage difference to a DAC and manipulate this value in determining the temperature.

    I must be missing something, please help.
    Steve
  • Sdyke4898 wrote: »
    I have been searching for 3 weeks daily for a circuit I could build to acquire temperature, through use of an RTD, over a long distance. The wheatstone bridge circuit with DAC sounds like the answer. However I am struggling with this. I understand the concept of one of the resistors being variable in order to balance the bridge due to the varying resistance of the RTD. What I can't wrap my head a round is who wants a circuit where one needs to manually adjust a pot in order to balance the bridge every time they need a temperature reading and then determine the resistance of that pot. I have 8 different points I want measured. It would be more practical to walk the distance than balancing the bridge.

    Seems to me one should be able to send the voltage difference to a DAC and manipulate this value in determining the temperature.

    I must be missing something, please help.
    Steve

    I am assuming you mean a Temperature Dependent Resistor (TDR) by RTD. No potentiometer required, and unless high accuracy is required just the TDR and another resistor in a series. What you need is an ADC (Analog to Digital Converter), not a DAC (Digital to Analog Converter) for this. Then you get a numeric value that is proportional to the temperature.

    Another option is to use a 555 timer with the TDR in the circuit to produce pulses, measure the resulting pulse width or frequency, and convert that to the temperature. Should work well for your application.
    In science there is no authority. There is only experiment.
    Life is unpredictable. Eat dessert first.
  • Sdyke4898 wrote: »
    I have been searching for 3 weeks daily for a circuit I could build to acquire temperature, through use of an RTD, over a long distance. The wheatstone bridge circuit with DAC sounds like the answer. However I am struggling with this. I understand the concept of one of the resistors being variable in order to balance the bridge due to the varying resistance of the RTD. What I can't wrap my head a round is who wants a circuit where one needs to manually adjust a pot in order to balance the bridge every time they need a temperature reading and then determine the resistance of that pot. I have 8 different points I want measured. It would be more practical to walk the distance than balancing the bridge.

    Seems to me one should be able to send the voltage difference to a DAC and manipulate this value in determining the temperature.

    I must be missing something, please help.
    Steve

    Your original post in "General Discussion" has been removed as it is is a duplicate to this question.
    Infernal Machine
  • You have two threads running with the same question. They will have to be combined into one thread. Which do you want to pursue?
    Infernal Machine
  • Publishing,
    My apologies. I didn't see them as necessarily the same question. The one above was seeking specific information on a wheatstone bridge with 3 fixed resistors for which I haven't received any info on yet, the other was meant to be more on the generic end for which I have received, what I consider, quite valuable information.
    Can you help or cut me some slack?
  • Kwinn, you got me on the DAC; that was a typo, but not on the RTD. We are both correct. It is a Temperature Dependent Resistor, but industry refers to them as Resistive Temperature Detector.
    Thank you for the advice, I really appreciate it.
Sign In or Register to comment.