I stumbled across your HeaterMeter project sometime back when making my own wifi bbq monitor and I found the information helpful.
I used the ET-732/733 temperature probes in my project and wish to provide feedback on your wiki comments regarding these probes where you graded them as only "acceptable" and noted "mysterious errors... making the temperatures fluctuate for no apparent reason" and a poor response curve with the HeaterMeter standard 10 kohm bias resistor.
I believe the fundamental issue you had with these probes is a mischaracterisation of their electrical properties. I wrote to Maverick technical support and they were kind enough to send me the datasheet. These probes use a Semitec 103GT-2 thermistor with a nominal resistance of 1 MΩ at 25°C which is significantly higher than the more common 10k or 100k thermistors.
Given this high impedance, it is little surprise that the response curve is poor for 10 kΩ bias resistors. You are correct to recommend a higher bias resistor although I eventually found a bias resistor in the range of 150 kΩ is better to optimise readings across the full cooking temperature range.
High-impedance 1 MΩ thermistors are specialized for high-heat BBQ environments because they maintain resolution at high temperatures and prevent self-heating. However, their high resistance makes the signal fragile, rendering it extremely vulnerable to electrical noise (EMI) and moisture leakage, which can cause erratic readings. Additionally, ADCs have an internal "sampling capacitor" that must charge to take a reading. For 1 MΩ high source impedance, the ADC internal capacitor may not charge fast enough before the ADC triggers. Ideally, to ensure accuracy, the circuit should use a high-impedance Op-Amp buffer to prevent the ADC from "loading" the signal, along with robust filtering to stabilise the voltage.
Maverick ET-732 - 733 (Semitec 105GT-2 Thermistor) datasheet.pdf
I stumbled across your HeaterMeter project sometime back when making my own wifi bbq monitor and I found the information helpful.
I used the ET-732/733 temperature probes in my project and wish to provide feedback on your wiki comments regarding these probes where you graded them as only "acceptable" and noted "mysterious errors... making the temperatures fluctuate for no apparent reason" and a poor response curve with the HeaterMeter standard 10 kohm bias resistor.
I believe the fundamental issue you had with these probes is a mischaracterisation of their electrical properties. I wrote to Maverick technical support and they were kind enough to send me the datasheet. These probes use a Semitec 103GT-2 thermistor with a nominal resistance of 1 MΩ at 25°C which is significantly higher than the more common 10k or 100k thermistors.
Given this high impedance, it is little surprise that the response curve is poor for 10 kΩ bias resistors. You are correct to recommend a higher bias resistor although I eventually found a bias resistor in the range of 150 kΩ is better to optimise readings across the full cooking temperature range.
High-impedance 1 MΩ thermistors are specialized for high-heat BBQ environments because they maintain resolution at high temperatures and prevent self-heating. However, their high resistance makes the signal fragile, rendering it extremely vulnerable to electrical noise (EMI) and moisture leakage, which can cause erratic readings. Additionally, ADCs have an internal "sampling capacitor" that must charge to take a reading. For 1 MΩ high source impedance, the ADC internal capacitor may not charge fast enough before the ADC triggers. Ideally, to ensure accuracy, the circuit should use a high-impedance Op-Amp buffer to prevent the ADC from "loading" the signal, along with robust filtering to stabilise the voltage.
Maverick ET-732 - 733 (Semitec 105GT-2 Thermistor) datasheet.pdf