No announcement yet.

AAG Weather station

  • Filter
  • Time
  • Show
Clear All
new posts

    AAG Weather station

    Just got my AAG weather station, hooked it up to HS, set up the McS plugin to accomodate it, and it is working great. The plugin assigned device codes without a hitch, even though I already had 8 temp sensors allocated. Only problem now is where to mount it.

    BTW - I got one of their [AAG] humidity sensors on a wire. Very nice job they did with it - the sensor is mounted on a piece of CAT5 terminated with an RJ11 plug. The sensor is protected by a vented metal shroud. Very rugged.

    Hi Ive just got a 1 wire weather station and then temp sensor plug in for my homeseer
    I cant seem to get it working (i cannot enter the com port in the setup GUI the box is low tally )
    i need it to be on com 2
    Sorry sure this is finger trouble
    ATB Mark


      I'm assuming you are using the DS9097 to interface with the weather station. The comm port is defined when you install the drivers for this interface and mcsTemperature finds that one, reports back to you which one it was, and uses it. I believe you can use the ibutton viewer which is a free download from DallasSemi (same reference as the TMEX drivers in the manual) and change the default if you need to. You should get the viewer anyway since it is a good debug tool.


        Thanks so much that has got me working just one oustanding problem ive got temp and windspeed working ok but direction is north at all times.
        also any pointers to getting the weather info in the UK
        Thanks so much for all you help


          The reliability of the wind direction sensor(s) appears to be very poor. Several, including, myself have defective units. The plugin has a debug mode enabled with a checkbox on the Interfac/Main tab. When it reads the wind direction it will return 4 analog values that should range between 0 and 4v. When the readings don't get above 2v then no valid wind direction is detected and north is the default. The actual calculation is based on a lookup table where the range of values for each of the 4 readings will determine the wind angle. I posted this table somewhere in this or the Temp05 forum, however, all you need to look at now is the 4 values and see if they exhibit sufficient dynamic range.

          Duncan Ellison is pretty active on this thread and I believe he obtains his data from MSNBC. I do not know what ACCID code he is using, but the manual describes the general approach for locating one in your area.


            Hi Michael
            thanks for info ive got temp and windspeed working fine but spent a week trying to get the direction stuff working here is the log data

            6/19/2003 10:40:46 AM~!~mcsTemperature Plugin Debug~!~WindDirection Voltage 0 =0
            6/19/2003 10:40:46 AM~!~mcsTemperature Plugin Debug~!~WindDirection Voltage 1 =0
            6/19/2003 10:40:46 AM~!~mcsTemperature Plugin Debug~!~WindDirection Voltage 2 =3.00004577706569E-02
            6/19/2003 10:40:47 AM~!~mcsTemperature Plugin Debug~!~WindDirection Voltage 3 =3.00004577706569E-02

            assume the data from the a/d converter is 3mv
            ive loaded the software that came with the weather station and the software from weather and they work fine on all sensors so have to assume that the hardware is ok

            can see the data jumping up to 2.28 volts in the java app

            should i rig up an external power supply ?

            any thoughts thanks for the help


              might this be the trouble as i am using a V3

              The first one is that a DS18S20 is used instead of a DS1820 for temperature measurement. The DS18S20 is an upgraded version of the DS1820 IC.
              The second difference is that a DS2450 A/D converter is used instead of a DS2407+(7)DS2401 for wind direction sensing. That was made to minimize the load on the 1-wire network.


                I believe I only developed DS9097 code for the V3 of the instrument. I'll look again at the DS2450 code to see if I can find something that may be wrong with it. In my case I get the same results if I connect my hardware up to the Temp05 or to the DS9097 with the plugin. I'll need to test with the iButton viewer as well. I do not remember doing this, but I did run their weather executable and did get "random" wind directions.


                  Thanks so much for your help, would be fantastic if i could get this fixed .


                    I connected my AAG weather station up to ibutton viewer and could only see voltages of 0 or 2.4 volts on any of the 4 A/D channels. This is the same readings I obtained from mcsTemperature. I obtained the lookup table from DalSemi's weather.exe C source code and readings of 0, 2, and 4 volts are expected. I confirmed with Mitch (Midon Design) on a thread on this board that I should be seeing 4V. His opinion what that my unit was defective units.

                    What I do not mind doing if you are willing, is to take some readings for your wind direction vane to see if we can come up with another lookup table that will provide acceptable results. If I recall when I did something like this in the past I could resolve some, but not all 16 positions. If you do this leg work, then I will make the lookup table an .ini setting and you can edit it.


                      Thanks for the kind offer, ive done a little more web searching and have found some good information,that i hope other people might find usefull as well.

                      ive found a circuit diag of the weather station ( at the bottom of the page .

                      I would also assume the max voltage would depend upon cable distance (data level to be rectified etc) so not quite sure of the max level,ive sent a mail to for some infoon this .

                      From this i would make these assumptions

                      with all sensors off i would assume max + voltage on all a/d

                      for other positions

                      1 on a/d C half of + v (47k resistor in series with 47K resistor)all rest at +v

                      3 on a/d B half of + v (47k resistor in series with 47K resistor)all rest at +v

                      5 on a/d A half of + v (47k resistor in series with 47K resistor)all rest at +v

                      15 on a/d D half of + v (47k resistor in series with 47K resistor)all rest at +v

                      7 on a/d D 0v (end of reed connected to earth)all rest at +v

                      9 on a/d C 0v (end of reed connected to earth)all rest at +v

                      11 on a/d B 0v (end of reed connected to earth)all rest at +v

                      13 on a/d A 0v (end of reed connected to earth)all rest at +v

                      for any reading between the reeds i would assume reeds next to each other come on giving voltage diff on 2 a/d converters

                      i have asked aag to confirm this all

                      as i am not seeing this +v on the sensors i would a assume that the unit is faulty as you have previously said. Cant quite see how the other software packages worked ??

                      Thanks for all your kind help ,i shall keep you informed of progress



                        Hi Michael,
                        just had a thought, as the sensors run of a power supply derived from the data, i was just thinking that problems could be caused if not much data is present on the 1 wire only send request data every so many seconds, if so would it be worth sending the request mainy times ,this would provide a lot of data that would make sure the capaacitor gets charged .
                        what do u think ??? am i mad ???
                        ATB Mark


                          You should be able to try this yourself. There is a switch polling interval that can go as fast as about 300 milliseconds. Even if you do not have switches, the bus will be pinged looking for whatever is out there and this should provide you with some bit movement. I did try it and saw no difference in the 2450 response.


                            The definitions of the various voltages can be found in the article here.





                              I found the problem with the DS9097 handling of the wind direction and V4.1.2 is posted. I was scaling to 2.5 volts rather than 5.