Announcement

Collapse
No announcement yet.

xapmcs1Wire - 1 Wire (DS9097U / DS9490D) xAP Node

Collapse
This topic is closed.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    #61
    fixed my problem 0n 1.6.1

    I deleted mcsXap1Wire.ini in the config folder and restarted the app. It seems to be ok now. Not sure what went wrong with the app and/or java runtime.

    Comment


      #62
      I have the gypsum at the level of the topsoil depth that is about 3 or 4 inches. The weather has been very consistent for the past week with totally clear skies and no rain. At the start of the period I totally soaked the soil around the sensor. This would equate to the maximum moistue content it would ever experience. After a week it can be seen that the sensor is drying out with a change of about 100 points on my scale. The modulation due to temperture was about 50 points. A usable sensor reading may be able to be achieved if the data was filtered and compensated for temperature. It would take quite a bit more data to develop an algorithm that would account for soil type, sensor depth, and temperature.
      Attached Files

      Comment


        #63
        There is also the effects of humidity that correlate well with the soil moisture I would not think the air humidity would have that big an effect at 3 or 4 inches, but I did not expect that much of an effect due to temperature at that depth.
        Attached Files

        Comment


          #64
          Michael,

          I could see a temperature effect at 3 or 4 inches but like you I wouldn't have figured that humidity would have an effect.

          Eric

          Comment


            #65
            I did some analysis with about 3 weeks of data. While the air temperature did warm the soil with some thermal delay the correlation was not consistent enough to obtain much better hourly readings. What does track reasonably well is the average daily readings. Over time these will likely be affected by average temperature. I'm not going to do any more with the device, but this will provide some information to those that may want to use it in some application.

            The attached is the average daily sensor reading which in my case is the average of 288 samples from midnight to midnight. It shows the reading is totally saturated soil, a trend as it drys out, the effect of 0.3" of rainfall, and the continued drying trend. Note that it was about 5 days after the 0.3" of rainfall that the additional moisture was lost from the soil.
            Attached Files

            Comment


              #66
              DS2409 - Hub

              The DS2409 switch was added to provide the functionality of a 1-wire Hub. Each DS2409 will show up as two sensors of type HubSwitch. The application will look for sensor on the main line and the branch of each HubSwitch device.

              If a hub is used then I recommend that all sensors are connected through the hub rather than on the main line. This is because a sensor on the main line will show up in each polling cycle and a polling cycle will occur for the main line plus each active hub branch.

              When nothing is connected to a hub branch then I suggest that the sensor type of unused branches be changed to "switch". This will prevent the software from activating the branch and reduce the total processing time.

              No support is provided for nested hubs, but multiple hubs are fine. There was also no special provisions made for reset pulses through the hub.

              Comment


                #67
                Selectable Min/Max resolution for Temp sensors

                A column has been added to the browser page that allows a temperature sensor to be checkboxed as low-resolution. No check indicates high resolution. This will result in faster conversion times. Depending upon sensor type the conversions willl be 200 ms for low resolution and up to about 700 ms for high resolution.

                Comment


                  #68
                  Wind Speed Problems?

                  Michael,

                  I am just getting going on HS2 and upgraded the mcs1-wire to ver. 1.8.8.
                  The wind speed indication seems very low and sluggish.
                  I flipped back to my early ver 1.5 and it was fine. It gave me a fairly true speed around 10-12 mph, the newer ver. was only reading 4-6 mph.
                  All my other devices are correctly read.
                  any ideas?

                  Comment


                    #69
                    There is a calibration setting as part of the setup that was added to support other wind gauges. When I use the the published algorithm for the AAG it seems that it yields speeds that are slow. I do not have a calibrated wind source so I'm not certain what the correct value is. It is simply a counter with counts taken over a period of time. I believe the wind gust is max counts per 15 second interval of the update period and wind speed is the counts per the user-specified update interval. My memory is that the gauge produces 2 counts per revolution. The algorithm used to convert revolutions per minute to miles per hour is

                    ' mph = (counterNow - CounterLast) / 60# / 2# * 2.453 / CSng(iSampleInterval)
                    mph = (counterNow(0) - Val(.Value(3))) * 1.2265 / CSng(iSampleInterval) * nWindCalibration

                    Comment


                      #70
                      I've tried setting different calibration settings but it doesn't seem to change very much, maybe 1 mph or 2.

                      You said
                      "wind speed is the counts per the user-specified update interval"


                      So does this mean if I take updates sooner than 60 seconds that the counts will be wrong for mcs1wire?

                      I have tried a 60 second interval, but the reading still is wrong no matter what calibration is set at.

                      Comment


                        #71
                        The wind gust will not be generated if the sample rate is under 30 seconds. The wind speed will be computed at any sample rate. The validity of the data becomes less as the sample time is decreased due to the small sample size that will be available.

                        Enable the debug and there will be a wind speed counter line of output for each time the wind speed is computed. It will show the counter, the counter at the prior sample and the interval between the two. Those, plus the calibration constant are the variables in the equation posted above.

                        Comment


                          #72
                          I looked in the IO window and saw figures I had seen in the browser. I took a screen cap of it.


                          This is the wind speed sensor. What I don't get is why it was put in as "Water Flow and Water Flow Rate". These are the same numbers I see in the IO window for the counter. I monitored this and found that no matter how much the difference between the two readings were, the wind Speed never really changes.
                          I don't know if this wil help but this is my calibration settings.

                          Comment


                            #73
                            Setup looks ok. The DS2423 has 2 counters and the second one defaults to water flow. Collect some data with the debug so we can see what is happening.

                            Comment


                              #74
                              debug?

                              Comment


                                #75
                                I guess I did not bring the setting to the user interface. Edit the ini file and add DebugLog=true in the section where the other general settings are located. Start the application to allow it to read this key. There is also a DebugLog2 key, but I do not know if it does anything at this time.

                                The output will be Homeseer.Event/Event.Log messages. You may want to run the attached to make it easier to view the log as the messages are being generated.
                                Attached Files

                                Comment

                                Working...
                                X