Advertisement

Blog

Temperature Measurement, Part 5

In concluding this series on temperature measurement, I thought I should mention a few gotchas that got me. Thermocouples can have all kinds of packaging around the junction to make them easier to handle and to affix them to the heat source, like a metal clip. Sometimes the junction contacts the metal (intentionally), so when it is connected to some metal like a boiler, the thermocouple is grounded. The thermocouple will still work, but if either of the connections will not tolerate a ground connection — beware. I used a full wave rectifier to generate DC power on my board, which meant that its 0V was a diode drop from ground. It released the magic smoke from the diode bridge.

A thermocouple junction encapsulated by a mounting tab. The yellow connector indicates a K type thermocouple and will fit into any thermocouple thermometer. The connector will be black for a J type, blue for a T type, purple for an E type, and so on.

A thermocouple junction encapsulated by a mounting tab. The yellow connector indicates a K type thermocouple and will fit into any thermocouple thermometer. The connector will be black for a J type, blue for a T type, purple for an E type, and so on.

Thermocouple wires are colored to indicate the type, with the negative wire of the pair always marked counterintuitively as red. A lot of instrumentation uses thermocouple connectors (see figure above) where the pins are made from the same alloy as the thermocouple wires. They are also colored for the thermocouple type, but if you read the Agilent application note (and many others), it becomes apparent that the actual connector is not necessary if you are designing your own equipment. It will transition to copper somewhere, and the resulting thermocouples will null out. And always make sure you have a resistor from the negative line of the thermocouple to 0 V of your design. Frequently, op-amps/instrumentation amps won't give you the desired result without it.

Beware of what exactly you are sensing, as in the case of the DS18S20. Sometimes it helps to bond the sensor to the object with thermally conductive epoxy. This is relevant to measuring the cold junction temperature. Since the cold junction sensor should be measuring the temperature of the actual connector contacts of the thermocouple wires, just having the sensor on the board next to the connector may contribute to an error due to thermal air currents and board conduction (to say nothing of self-heating of the sensor if it is a thermistor).

Finally, we all need to consider calibration. In designing the electronics, it is preferable to have some kind of source that you can use to simulate the temperature and take all the way through its range. As a hobbyist, you could make a simulator using resistors and pots (for thermistors and RTDs) and a resistor divider and pots for a thermocouple (and a very good DVM). If you are doing this professionally, you really need a calibrator. There are dedicated calibrators like those made by Altek. I must say that one of the most essential pieces of equipment I have is the model 741 general-purpose calibrator from Fluke.

Having said that, there is a bigger problem in the calibration of an onboard temperature sensor, irrespective of sensor technology. When you build the device, how do you know what the temperature is at the sensor? Do you work on ambient? How accurate is the thermometer you are using? And on and on. You can get a flavor of the results from Max Maxfield's blog on EE Times (a sister site), I’m becoming hot and bothered….

My question to all the readers: What devices do you use to measure temperature? What advantages do you perceive in those particular ones?

Related posts:

34 comments on “Temperature Measurement, Part 5

  1. amrutah
    April 28, 2014

    Aubrey:  “What devices do you use to measure temperature?”

      Mercury thermometers for measuring the body temperatures and digital thermostats (I think its a thermistor) for measuring the room temperatures.

      The link of Max Maxfield's blog that you shared, is the exact case I found in my home temp control system.  Though digital measurement that I see now is different from the one I had on the analog counter-part, but looks more accurate as i feel comfortable as well as my energy consumption has gone down.

     

  2. samicksha
    April 29, 2014

    I guess mercury thermometers are the most commonly used meters used to measure body temperature but parallel to same thermometers based on the equation of state of a gas are complex to handle.

  3. amrutah
    April 29, 2014

    @Samsicksha: Yes the mercury based analog thermometers have been used since many days, but many countries are banning on the use of mercury based devices due to its toxic nature.  The gas based thermometers might be too complex for day-to-day temperature measurements, but might be very useful for very high pressure, temperature and volume applications.  More importantly, I don't understand how are these calibrated.

    The digital thermometer which are based on the bi-metal, thermocouplers, thermistors or integrated tempsensors are comparitively easy to manufacture and calibrate.

  4. SunitaT
    April 29, 2014

    This has been a detailed series of five articles on temperature measurements and you should not stop there. I use Resistance temperature detectors. These are the kind of temperature sensors that work by its resistor changing resistance value with temperature increase. I consider them better than thermistor sensors because of their accuracy .From the previous temperature measurement part 4; accuracy is considered one of the main parameters to be considered. It also has high temperature ranges, high stability and has good inter-changeability.

  5. SunitaT
    April 29, 2014

    The few drawbacks that may arise from temperature measurement make it a risk. The problem with the onboard temperature sensor calibration is one of the major drawbacks that need to be looked into. This is because this problem brings a lot of uncertainly when it comes to establishing the accurate temperature readings at the sensor. To make this worse, this is the same case regardless of the sensor technology that is applied. There surely needs to be addressed so as to make readings better for accurate conclusions and calculations. 

  6. amrutah
    April 29, 2014

    @SunitaT0: Linearity is also another important parameter along with the accuracy.  As you said, thermistors can be used for the large range of temperature, but most of the thermistors have a non-linear characteristics w.r.t. to the temperature for major part of the range, which can affect the accuracy of measurement.  Some kind of calibration has to be built-in to compensate this.

  7. RedDerek
    April 30, 2014

    Nice set of articles on temperature measurment. One does have to look at the desired temperature range, packaging, accuracy and amount of integration in order to get a good idea as to how to approach the best method.

    I use thermocouples mostly. I have played with the earlier semiconductor devices, and somewhere around the home lab, I have a USB plug-in temp sensor.

  8. yalanand
    April 30, 2014

    After following all your series on temperature measurement, I have learnt a lot and thank you very much Aubrey. I also would like to say that gas thermometers should be used more and that companies that are manufacturing mercury based thermometers should stop and adopt the gas one or any other thermometer that does not a threat to the environment like digital thermometers. Mercury is known to be toxic in nature and it is not good for the environment.

  9. geek
    April 30, 2014

    “I use thermocouples mostly. I have played with the earlier semiconductor devices, and somewhere around the home lab, I have a USB plug-in temp sensor.”

    @RedDerek: Is that part of a personal hobby that you have? Or do you use these as part of your work tools on a typical day?

  10. geek
    April 30, 2014

    “From the previous temperature measurement part 4; accuracy is considered one of the main parameters to be considered. It also has high temperature ranges, high stability and has good inter-changeability.”

    @SunitaT0: Isn't accuracy always a subjective assessment? I mean that you can't have one criteria or threshold for accuracy for a sensor throughout. The desired accuracy with the same sensor can vary across different applications that the sensor is used in.

  11. RedDerek
    April 30, 2014

    @tzubair: I do use the thermocouples occasionally. Most of my applications are not critical to accuracy, but cost is key. Also, I do have a few for hobby activities.

  12. Davidled
    May 1, 2014

    I wonder how often temperature sensor would be calibrated once electronic board is packaged. Well, how much is the deviation between calibrated and uncalibrated sensor?

  13. antedeluvian
    May 1, 2014

    DaeJ

     

    I wonder how often temperature sensor would be calibrated once electronic board is packaged.

    This depends entirely on the application. I doubt that a home thermostat is ever calibrated (certainly mine isn't). But then what is the problem if it is a little off. You just adjut the thermostat until you feel warm. In industry though they can calibrate periodically between every 3 months and a year especially where the temperature is critical to the process.

    Well, how much is the deviation between calibrated and uncalibrated sensor?

    That is a very difficult question to answer. Irrespective of the technology there is whatever the manufacturer specifies, but that is in very specific circumstances. Some semiconductor sensors offer 0.5degC accuracy or better. But then you have tyo consider the environment- what is there is a power transistor on the board and it only gets warm in some circumstances? What exactly is doing the sensing- as I mentioned some semiconductor sensors measure the temperature conducted in on one of their pins.

    I think the bottom line is that if you can't guarantee the measurement that you promise, you need to calibrate.

  14. fasmicro
    May 3, 2014

    >> Mercury thermometers for measuring the body temperatures and digital thermostats (I think its a thermistor) for measuring the room temperatures.

    Mercury thermometer is the industry default for most applications. Yet, the best idea is to make your circuits to work across different temperature range by using better process technology and circuit strategy. The concept of temp measurement and calibration especially in the consumer electronics industry is not very popular anymore.

  15. fasmicro
    May 3, 2014

    >> Yes the mercury based analog thermometers have been used since many days, but many countries are banning on the use of mercury based devices due to its toxic nature.

    Which countries? Norway, Switzerland? Just curious because that will cause trouble as there seems to be no major alternative.

  16. fasmicro
    May 3, 2014

    >>  I have played with the earlier semiconductor devices, and somewhere around the home lab, I have a USB plug-in temp sensor.

    Very interesting. I have not been dealing with a lot of temp measurements in my designs. I work in the MEMS analog front-end for gyroscopes and XLs and I do not encounter these issues a lot.  Meanwhile, any link for the temp sensor with USB plug-in?

  17. fasmicro
    May 3, 2014

    >> I wonder how often temperature sensor would be calibrated once electronic board is packaged. Well, how much is the deviation between calibrated and uncalibrated sensor?

    For most applications, that is not really an issue. I have a really affordable laptop that shuts down once it starts heating up. That is the normal calibration that Gateway has put there. Yes, if they have done otherwise, they could have prevented the shutdown by managing the temperature. But it may cost me more than the $267 I bought the laptop. The conclusion is these things are driven by cost.

  18. goafrit2
    May 3, 2014

    Well, how much is the deviation between calibrated and uncalibrated sensor?

    That is a general question. It all depends on the product or the design. The most important is what you can tolerate in your product. That is why it is good to do a lot of sims and run so many temp, voltage etc variations and see how well you do across those ranges. Some do Monte Carlo to track coners and get all the big picture.

  19. Netcrawl
    May 4, 2014

    @fasmicro sensor can be calibrated by comparing them to another one, or by calibartion at a fixed points, comparison calibration is the mnost commonly used method where the thermometer under test is compared to the temperature display of a portable heat source. 

    Gateway has put that to prevent heat build-up, but for me a fan could do better, fan is a very effective in managing temperature. 

  20. Victor Lorenzo
    May 4, 2014

    @fasmicro >> Which countries? Norway, Switzerland? Just curious because that will cause trouble as there seems to be no major alternative .

    Not only those countries, in fact the whole EU is in the process, and not only mercury thermometers, the vast majority of products containing mercury will be banned in a mid-term period according to World Health Organization (one fact sheet here).

    There is also the Minamata Convention on Mercury (pdf here) and mercury thermometer are explicitely listed.

    Yes, there are non electronic alternatives for mercury thermometers, less accurate but also less aggressives for health and the environment. One is a gallium-indium-tin alloy called galinstan. All (to my knowledge) non electronic/non alcohol thermometers you can buy in (at least) Spain are mercury free. I'm not 100%, but as it comes from an EU directive it should apply to all EU members.

  21. fasmicro
    May 4, 2014

    >> Gateway has put that to prevent heat build-up, but for me a fan could do better, fan is a very effective in managing temperature. 

    You are very correct on the fan issue. What happens though is that sooner or later the fan loses control and the laptop shuts down. Nevertheless, it is a very easy way of managing temperature issues.

  22. fasmicro
    May 4, 2014

    >> All (to my knowledge) non electronic/non alcohol thermometers you can buy in (at least) Spain are mercury free. I'm not 100%, but as it comes from an EU directive it should apply to all EU members.

    Thank you. I suspected so with my potential names. I know that EU will lead in this area. The problem though is that without EU and China where most of these things are made, the world will not see a radical change in usage. There are many chemicals banned in EU which are freely used in U.S. and China and around the world thereby diminishing the impact. There are few good things from mercury. They should also ban lead.

  23. Davidled
    May 4, 2014

    First, Electronic device has high definition temperature control that stores the pre-defined value. In other words, the device implements the capability for pre-calibration. Hence, similar to self-healing, self-calibration would be processed while monitoring temperature sensor. Second, there are three types thermocouple such as type J, K, and T. Instead of changing different types of thermocouples, sensor of device adjusts the sensor output based on pre-tuning value of memory.

  24. amrutah
    May 6, 2014

    @fasmicro:

    The best idea is to make a circuit work with 6-sigma complaince across different PV and T so that we are 99.9% assured that the we have a good temperature monitoring device.

        The consumer electronics like the AC system which has a thermostat uses bi-metal mechanical sensors or thermistors and I do think they need correct calibration.

  25. amrutah
    May 6, 2014

    @Victor: The liquid alloy “galistan” has most of its features matching the mercury but has its own set of problems, but as you said it can eventually replace mercury from thermometers.

    Thanks.

  26. amrutah
    May 6, 2014

    @fasmicro:  You may want to check the following links about the ban

    Phase out

    Countries and their opinions

       There are many alternatives being planned out, its upon us to open our eyes and embrace one.  Many governments are actively taking part in eliminating the mercury based thermometers.

    Few alternatives to mercury based thermometers.

  27. amrutah
    May 6, 2014

    @yalanand: “also would like to say that gas thermometers should be used more”

       I agree that we need to eliminate the usage of mercury based thermometers, but the gas based thermometers are complex, may not be good for small temperature ranges like -20 to 85 degC, portability etc.

  28. amrutah
    May 6, 2014

    @redDerek: “I have a USB plug-in temp sensor.”

       This is interesting and thanks for sharing.  What kind of application is this for? and after interfacing this USB temp-sense to your PC, what do you do? 

    I am unable to gauge its usage.

  29. RedDerek
    May 6, 2014

    If I recall, it was either a Maxim or National Semi demo board from about 8 years ago. Not sure which box thte bard is in right now.

  30. goafrit2
    May 8, 2014

    >> Hence, similar to self-healing, self-calibration would be processed while monitoring temperature sensor.

     

    Good point but doing that for very cheap consumer electronics systems will not help any company stay competitive. Self-calibration is not free and viable for fairly high quality products. But when it means commoditized products, that becomes a burden.

  31. goafrit2
    May 8, 2014

    >> The best idea is to make a circuit work with 6-sigma complaince across different PV and T so that we are 99.9% assured that the we have a good temperature monitoring device.

     

    That is certainly the best strategy. Solve the problem at the level of circuit. Make the strategy to be driven by circuit and do not depend on these extra corrections and calibrations. If designers invest time on simulations and models, we can design-out some of the issues that require expensive calibrations etc.

  32. goafrit2
    May 8, 2014

    >> The liquid alloy “galistan” has most of its features matching the mercury but has its own set of problems

    Just getting to know this galistan for the first time. Googling it does not make it seem like a really good alternative to mercury.

  33. goafrit2
    May 8, 2014

    This seems like a global movement. Unfortunately the right people are not listening. Even the new startups like Nest all depend on mercury thermometers for their wonders.  Nothing will happen until the U.S. govt buys out the makers as they did to those making tungsten bulbs for migration to LED bulbs. You need to buy them out and then they will move to newer safer thing.

  34. goafrit2
    May 8, 2014

    >> @redDerek: “I have a USB plug-in temp sensor.”

    Many has asked the same question. How can this work? What is the use? And how can one get a piece of this tool. Just thinking about it is cool but if the USB gives you the temp, you still need a feedback to actually make things happen, i.e. control the temp.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.