Comments on programming and temperature from DFM.

[rpratt20]

I have now programmed 3 DFM17 units with the new firmware. I found making connection to the programming pads a little difficult. My experience would indicate the wires must be short such as 3-4cm and near equal in length. I decided to use a thin plastic support inserted into the probe connection and route the wires out where the probe would normally be and making connection to the programmer at the exit.

Horus v2 has been easy to decode and all modes have transmitted as I set in the config. Decoding APRS is a challenge and I have not succeeded although the sound is correct to the ear. Each DFM has settled into a bit different frequency than programmed but after stabilizing they remain within a small range.

I have also used resistors and a probe to investigate the temperate function of the original sonde. It appears that a NTC thermistor is used.
DFM17 Temperature Resistance

-61.8 444000
-53.2 242000
-52.2 226000
-8.7 20000
-13.4 25000
-2.4 15050
7 10030
24.6 5050
21.6 5900 this is from 2 booms at room temp and not a transmitted temperature (some error)
36.4 3280 this is from 2 booms at elevated temp and not a transmitted temperature (some error)
76.9 987 resistor

Calibration curve would be 1/T = 0.0007340831 + (0.0003081634 * ln(R)) + (2.4666873E-009 * (ln(R)^3)) and this appears to be within 0.1 degree celcius. This link shows the thermistor in series to ground and third V719 U4 common channel pin switched by PB-5 of the STM. https://wiki.recessim.com/view/DFM-17_Radiosonde#/media/File:Stalk_pinout_mapping.png. V719 pins 4 and 6 remain unknown but reading a thermistor would normally involve a voltage divider. The switch could be for power saving or to reduce heating of the probe. Calibration may depend on individual boom.
Randy