Other Blogs in this project
My original proposal for the soldering iron monitor using a k-type thermocouple mounted on a ceramic washer that was screwed into the soldering iron stand with a ceramic screw connected to a STM-332U Temperature Sensor Module is shown in Figure 1. Testing and blogging about the STM-332U revealed that using a k-type thermocouple without a thermocouple amplifier would not be practical. The output voltage from a k-type thermal couple would only produce about 4 mV at 100 °C. The analog inputs of the STM-332U are 8 and 10 bits, so sensitivity would not be very good . I also could not find a thermocouple amplifier that would operate on 1.8 V that is produced by the sensor module for external circuits.
Therefore, the k-thermocouple was replaced with a 10k Ohm NTC Thermistor and a 10k Ohm resistor connected together to form a resistor divider. Testing also showed that a stand-off in proximity to a hot soldering iron did not get hot enough to clearly indicate that the iron was on. A 14 gauge copper wire connected to the thermistor was added and configured to touch the soldering iron near the handle. The copper wire would conduct the heat from the soldering iron to the thermistor and provide a robust indication that the iron was on. The modifications to my original proposal are shown in Figure 2.
Figure 3 shows the thermistor mounting hardware in detail. I tried to get the thermistor as close to the soldering iron as possible using a combination of screws, nuts, stand-offs, and washers that were in stock at Newark.com. I used the smallest hardware available to keep the “thermal mass” as low as possible so that temperature measurement would be as quick as possible. It turned out that M3 hardware was the most available at the time of my order. I used 14 gauge copper wire that I stripped out of some home electrical wire I had on hand for the wire to contact the soldering iron.
A longer stand-off would have made the assembly easier as there was not much thread depth for the screws on each side. I added a nut to keep the stand-off tight, and another one to keep the thermistor and wire attached to a lug tight.
I selected an EPCOS NTC Thermistor part number B57703M103G40 because it had a lug for mounting and had a 10k Ohm resistance at 25 °C  . Also, EPOS makes very high quality components (past experience with some of inductors). Figure 4 shows the resistance verses temperature (blue trace) and the digital value that the STM-332U would generate from the resistor divider at the analog input ADIO0 for a given temperature (green trace). NTC thermistors have an exponential relation with temperature  so the plot for resistance is semi-log to provide more detail at higher temperatures.
Figure 5 shows the modifications to the left and right soldering stands.
Temperature Sensor Configuration
Two STM-332U sensors were configured as temperature sensor, set point, and occupancy control using DolphinStudio and the EOP-350 Universal Programming Board (UPB), as shown in Figure 6.
Temperature Sensor Modifications
These STM-332U sensors were modified to add the resistor and thermistor to form a resistor divider between ground and the switch power output SWPWR. The center tap of the resistor divider is connected to analog input ADIO0. An illustration of these connections is shown in Figure 7.
I was planning on using a mating connector (Samtec SFMC-110-01-L-D) to attach the resistor and thermistor to, but the connectors have not arrived yet so I soldered them directly to the boards. The spacing between ground and ADIO0 is just about the length of a 1206 size resistor. So I solder a 10k Ohm 1206 size resistor between these pins at the base of the connector (see Figure 8). Fortunately, I had some from another hobby project.
I cut up some female-female jumpers and used them to make an impromptu connector for the thermistor. The two wire assemblies were soldered to ADIO0 on top of one of the 10k Ohm resistor pads and to the SWPWR pad. I used hot glue to hold the wires in place and provide some strain relief. These modifications would still allow the STM-332U to be plugged in to the EOP-350 UPB, if they require new or further configuring.
Figure 9 shows the modified STM-332U sensors connected to the thermistor on the soldering stand. The top photo in Figure 9 shows the soldering iron and how it contacts the copper wire.
Figure 10 shows the modified STM-332U sensors set in place and the FHEM output can be seen showing the temperature (19 °C) and the set point of 109.
Soldering Iron Monitor Operation
Figure 11 shows the whole setup with both soldering irons and stands, the modified temperature monitors, and a DMM with a thermocouple to measure the thermistors before the soldering irons are turned on. FHEM shows that both sensors are reporting 19 °C and the set point of 110. The 110 reading corresponds to about 20 °C from the green plot in Figure 4.
The left soldering iron is turned on, and after a while the thermistor reaches about 60 °C and the sensor shows the ADC output of 213 (see Figure 12). This value corresponds to about 65 °C from the green plot in Figure 4.
The left soldering iron was unplugged and the right one was plugged in. The right side soldering reached a temperature of about 48 °C and the right side sensors shows an ADC reading of 194. This corresponds to a temperature of about 53 °C from the green plot in Figure 4. The left side station was cooling while the right side heated to be values shown and reached an ADC reading of 167 or about 38 °C using the green plot in Figure 4.
There seems to be a 5 °C or so error between what the sensors report and what I measured. The thermocouple and DMM measurement were slow and it took a while for the thermocouple to measure a temperature. It also kept slipping off the thermistor, so this may have contributed to the error. The plot in Figure 4 uses resistance verses temperature from the datasheet and does not take into account component tolerance variations, which would also contribute to some of the error measured. Even with these errors, the monitor show that the irons are operating. An effective alarm could be built in FEM or OpenHAB using rising and falling temperature slopes, room temperature iron temperature differentials, and a threshold of about 50 °C.
 EPCOS, NTC Thermistors for temperature measurement, B57703M
 EPCOS, NTC Temperature Measurement Sensors