AAG cloud sensor

From ObsVermes
Jump to: navigation, search


The meteorological condition under which AAG Cloud watcher was tested.

Current RTS2:

svn checkout https://rts-2.svn.sf.net/svnroot/rts-2/trunk/rts-2 rts-2

Current INDI:

  • AAG Cloud watcher INDI driver version 2009-04-19
    • heating strategy
      • simplified
      • works now stable above 90% humidity
      • heats in state IS_RAIN as long as rain_frequency does not drop more than -3. arbitrary units
    • Heater Impulse minimum triggers a manual heating (cut off by THRESHOLD_MAX or rain > (*target + impulse[IMP_DELTA]))


Graphs at the observatory

A cloudless half night and day

The data from Davis meteo station.
The sky temperatures, green the sensor, blue corrected.
No rain.
The sensor is mounted on the dome and points to South West.

Snow on the sky temperature sensor

Snow covers the whole surface.

If snow lies on the sky sensor the sky temperature (green curve) and the ambient (red) temperature are identical. 2009-12-20 was the coldest night, temperature dropped below -20 degeres, since we measure it.

The absolute sky temperature alone does not indicate a clear sky. In addition the difference between sky and ambient must be taken into accont.

Dense cirrus clouds in winter

Cirrus clouds.

The night was clear and during the late morning (before 12 CET) and afternoon high cirrus clouds appeared. The sensor seems to be capable to measure a temperature difference of abot -32 degrees.


by António Alberto Peres Gomes (AAG)


Heating algorithm, heat amount and target temperature (INDI only)

The heating algorithm is optimized for fast detection of the transition from DRY to WET. If the dryness (rain_frequency) is above THRESHOLD_MAX the PWM duty cycle is always zero. If it is between THRESHOLD_MAX and THRESHOLD_DRY the driver follows the policy defined in the above mentioned documentation. If it drops below THRESHOLD_DRY, either IS_WET or IS_RAIN, heating is started immediately for period given in impulse[IMP_DURATION_WET] resp. impulse[IMP_DURATION_RAIN]. If this period is exceeded and the rain sensor is in state IS_RAIN the heating continues for the time defined in IMP_CYCLE_RAIN in case the dryness (rain_frequency) exceeds a threshold defined in THRESHOLD_DRY_OUT. If that is not the case a stronger rain shower is assumed. Under these conditions the heater can not dry the sensor (and this would not be sensible at all). After the period defined either as impulse[IMP_CYCLE_RAIN] or 2. * impulse[IMP_DURATION_RAIN] the timers are reset (which ever occurs first).

If the state is IS_WET several heating cycles are necessary. Currently impulse[IMP_CYCLE_WET], corresponding more or less to the original Heat Impulse Cycle in section Setup->Device, is not used.

The amount of heat generated in the rain sensor depends reciprocally on the read cycle time (polling time, default value 10 seconds) and on the target temperature.

Electrical constants

Currently the value of the constant LDR pullup resitanance is read out as 50 kOhm, it should be 56 kOhm. Antonio says

The problem of the 56K and 50K is probably due to the following:
- The AAG_CloudWatcher only reads the electrical constants values from the unit once when it is installed for the very first time. It stores this information in the windows registry. It may happen that a previous version of the software has stored 56 K. I have checked the value from the string you have sent and it is 50 K.