Sometimes you need to know the temperature of something from a ways away. That might be a smoker, a barbecue, or even a rabbit hutch. This project from [Discreet Mayor] might just be what you’re looking for.
It consists of a MAX31855 thermocouple amplifier, designed for working with commonly-available K-type thermocouples. This is hooked up to a Texas Instruments CC1312 microcontroller, which sends the thermal measurements out over the 802.15.4 protocol, the same which underlies technologies like Zigbee and Thread. It’s able to send radio messages over long distances without using a lot of power, allowing the project to run off a CR2023 coin cell battery. Combined with firmware that sleeps the system when it’s not taking measurements, [Discreet Mayor] expects the project to run up to several years on a single battery.
The messages are picked up and logged in a Grafana setup, where they can readily be graphed. For extra utility, any temperatures outside a preset range will trigger a smartphone alert via IFTTT.
Keeping a close eye on temperatures is a key to making good food with a smoker, so this project should serve [Discreet Mayor] well. For anyone else looking to monitor temperatures remotely with a minimum of fuss, it should also do well!
Typo: CR2023 = CR2032
Rabbit hutch and smoker, IFTTT?
B^)
The most badass version of this would use the thermocouple itself to charge a capacitor and power the transmitter….
My starting point for your idea would probably be to read the 1968 documentation of the research into materials done by RCA for NASA as to what should be used inside a RTG * (The power source used in the 1977 Voyager space probes would have have had their origins here).
One thing to remember is that if you want to use a thermocouple to measure something, then for high precision and high accuracy** ideally you need to cause no current to flow (Or almost no current).
But if you want a junction to provide power, then you need to draw as much current as you can whilst optimising for maximum power which will be less than maximum voltage (there will be an additional reduction in the voltage drop across the junction and a drop across the connecting wires because they have resistance the more current you draw, and that resistance will also change with temperature – and the more current will increase that temperature ).
I’m wondering if a quick and dirty two dimensional table could be created where the current and voltage are both measured and a temperature was also measured. And then that lookup table is utilised with just a current and voltage measurement instead of operating in a power generation mode, a quiescent mode and a temperature measurement mode.
* https://www.osti.gov/biblio/4716190/
** https://en.wikipedia.org/wiki/Accuracy_and_precision