Announcement

Collapse
No announcement yet.

mcsTemp stops recording Temp08 readings

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    #16
    The type mismatch makes no sense since it is simply an assignment statement of one string variable to another string variable. Perhaps it was an earlier version that had different line numbers. We will just need to see if it still occurs with current versions.

    The triggers message is also not something I would expect, but I know its trace does not suffer potential line renumbering. In this case a call is made to HS to see how many events are defined. I then iterate through each event in increasing event number and get the event object. The error occurs when an attempt is made to use this event object. It is a different event (#5, 8, 9, or 10) that generates this error. I would not think there would be event indexes that did not have an associated event object. For the next build I will iterate through the event collection and check to make certain the event object is valid before using it.
    Last edited by Michael McSharry; January 30, 2005, 09:54 PM.

    Comment


      #17
      Calibration of humidity sensor not working consistently with 4.32.6.

      Attached is the IO stream (one clean read, one not so clean), the ini "Devices" section showing bias I entered for humidity (4,4,2) and the trace entries matching the IO Stream. The proper bias was not added to the devices (R1: 2 was added instead of 4, R2: 4 was added okay; R3: 2 was not added) - assuming I'm, reading the trace info right -

      A few reads later I noticed the bias was added properly to R1 and R2, but still not to R3. Added the IO Stream and trace after the other two in attached file
      Attached Files

      Comment


        #18
        New error reported after installation at end of boot up of 4.32.6.

        31/01/2005 10:15:19 PM~!~mcsTemperature~!~Version 4.32.6 Registered with Homeseer
        31/01/2005 10:15:20 PM~!~mcsTemperature~!~Temp05/08 Comm Port 5 Opened
        31/01/2005 10:15:21 PM~!~mcsTemperature~!~Last temperature sample recorded at 31/01/2005 10:13:24 PM
        31/01/2005 10:15:21 PM~!~mcsTemperature~!~Last forecast sample recorded at 31/01/2005 9:26:59 PM
        31/01/2005 10:15:29 PM~!~mcsTemperature~!~ TemperatureReceiveVersion5: on line 680 Type mismatch
        3

        Comment


          #19
          The [Devices] section tells the story. R1, R2, and R3 are identified as Humidity index 3, 5, and 6 respectively. They should be 1, 2, 3. Note that the Temp reading from these devices are reported as Temp sensors #3, 5, and 6.

          My initial look at the code did not identify why this is happening. But at least it looks as if the reason for the calibration problem actually has nothing to do with calibration.

          [Devices]
          410000004FEE3A26H="R1,3,4,0,0"
          800000004FE5AA26H="R2,5,4,0,0"
          820000004FFCE126H="R3,6,2,0,0"
          D60000004FFF2626="R4,1,-1.5,0,0"
          3D000800412B7A10="R5,2,-0.4,0,0"
          410000004FEE3A26="R6,3,,0,0"
          DF00080046A0E810="R7,4,1,0,0"
          800000004FE5AA26="R8,5,1,0,0"
          820000004FFCE126="R9,6,0.45,0,0"
          7D00080048F1CB10="R10,7,,0,0"

          Comment


            #20
            So what do I do at my end? The [Devices] section isn't something I created or manually edited. Now if I change R1, R2 and R3 to 1, 2, 3, what happens to R4, R5 and R6, which are currently 1, 2, 3?

            Perhaps this mix up occurred (a while ago) when I changed my humidity sensors to temp sensors in the hopes of getting the calibration to work (then changed them back because, if I recall right, it didn't work plus I lost the minimum humidity setting)

            Comment


              #21
              I do not believe anything you did with configuring your Humidity sensors should cause this problem.

              Your main role is to pray. You could also help by collecting data at the time of Homeseer startup with the General Debug checkbox set. The first cycle of data collection is where the index is collected. If you look in the IO Window you will see a VER, TMP, and DIS command send to the TEMP08. After the DIS command results are returned no additional debug information is needed. This sequence should take no more than about a minute. All debug info will go to the HS log.

              Comment


                #22
                have attached some bedtime reading...
                Attached Files

                Comment


                  #23
                  I go to bed when my brain dies. Reading is not in the cards for that time of day.

                  The caption below if from your HS log when the indexes are determined. This shows that the TEMP08 returned two bad readings (R1 and R3). This means that the indexes for these devices is not learned and they should revert back to the value last stored in the ini file. Since the indexes are out of range in the ini file for humidity sensors, they all get mapped to R1.

                  01/02/2005 8:31:40 PM~!~mcsTemperature~!~Temp05/08 Data: Reading Sensors...
                  01/02/2005 8:31:40 PM~!~mcsTemperature~!~Temp05/08 Data: Humidity #01[410000004FEE3A26]= ???
                  01/02/2005 8:31:40 PM~!~mcsTemperature Debug~!~Database Timer=5
                  01/02/2005 8:31:40 PM~!~mcsTemperature~!~Temp05/08 Data: Humidity #02[800000004FE5AA26]=64%
                  01/02/2005 8:31:40 PM~!~mcsTemperature Debug~!~Record ID Info=02/800000004FE5AA26, Type = 6, Preferred Device=, Sensor Known=10, SetupOpen=False, DB Created=True
                  01/02/2005 8:31:41 PM~!~mcsTemperature Debug~!~Record ID Info Sensor Present=True, with Device Code=R2 at index 1
                  01/02/2005 8:32:02 PM~!~mcsTemperature~!~Temp05/08 Data: Humidity #03[820000004FFCE126]= ???

                  Three plans of attack on this one.

                  1) I changed the plugin logic to accept the bad reading if it is trying to capture indexes, but not use the value returned. You will need to keep your eye on it to make certain this logic is working correctly.

                  2) Edit your ini file to restore the default indexes to 1, 2, and 3. This actually should get rid of the problem even if I did not update the pluign. The plugin update will be needed when you change your TEMP08 sensor configuration with the INI command.

                  3) Add a power wire to your Humidity sensors if they do not already have them. It looks like you have reliability problems with reading humidity sensors.
                  Attached Files

                  Comment


                    #24
                    Here's my [devices] section now with new version installed. Looks like the 1,2,3 worked this startup.

                    [Devices]
                    410000004FEE3A26H="R1,01,4,0,0"
                    800000004FE5AA26H="R2,02,4,0,0"
                    820000004FFCE126H="R3,03,2,0,0"
                    D60000004FFF2626="R4,01,-1.5,0,0"
                    3D000800412B7A10="R5,02,-0.4,0,0"
                    410000004FEE3A26="R6,03,,0,0"
                    DF00080046A0E810="R7,04,1,0,0"
                    800000004FE5AA26="R8,05,1,0,0"
                    820000004FFCE126="R9,06,0.45,0,0"
                    7D00080048F1CB10="R10,07,,0,0"

                    my humidity sensors are powered. I was told I need to expect the odd bad reading (specifically, the ~25% occurance I see is not uncommon) and I don't think a boot up reading is particularly immune to it. Maybe I was fed a line about acceptable frequency of errors, but I would think it's probably better that plug-in expect them to happen at boot up.

                    attached is trace for my latest startup.
                    Attached Files

                    Comment


                      #25
                      check out my devices file last night, before I restarted HS. Looked at it after I had changed my humidity offset (to blank) but it was not having any effect on my devices.

                      [Devices]
                      410000004FEE3A26H="R1,97,,,0"
                      800000004FE5AA26H="R2,97,,,0"
                      820000004FFCE126H="R3,97,,,0"
                      D60000004FFF2626="R4,97,-1.5,,0"
                      3D000800412B7A10="R5,97,-0.4,,0"
                      410000004FEE3A26="R6,97,,,0"
                      DF00080046A0E810="R7,97,1,,0"
                      800000004FE5AA26="R8,97,1,,0"
                      820000004FFCE126="R9,97,0.45,,0"
                      7D00080048F1CB10="R10,97,,,0"

                      This is before and after restart this morning:

                      [Devices]
                      410000004FEE3A26H="R1,3,,0,0"
                      800000004FE5AA26H="R2,5,,0,0"
                      820000004FFCE126H="R3,6,,0,0"
                      D60000004FFF2626="R4,1,-1.5,0,0"
                      3D000800412B7A10="R5,2,-0.4,0,0"
                      410000004FEE3A26="R6,3,,0,0"
                      DF00080046A0E810="R7,4,1,0,0"
                      800000004FE5AA26="R8,5,1,0,0"
                      820000004FFCE126="R9,6,0.45,0,0"
                      7D00080048F1CB10="R10,7,,0,0"

                      [Devices]
                      410000004FEE3A26H="R1,3,,0,0"
                      800000004FE5AA26H="R2,5,,0,0"
                      820000004FFCE126H="R3,6,,0,0"
                      D60000004FFF2626="R4,01,-1.5,0,0"
                      3D000800412B7A10="R5,02,-0.4,0,0"
                      410000004FEE3A26="R6,03,,0,0"
                      DF00080046A0E810="R7,04,1,0,0"
                      800000004FE5AA26="R8,05,1,0,0"
                      820000004FFCE126="R9,06,0.45,0,0"
                      7D00080048F1CB10="R10,07,,0,0"

                      Comment


                        #26
                        The first device reported by the TEMP08 after the DIS command is a DS2438 with a type identified as "T". I have never seen this. What do you have connected that only reports what assume is temperature?

                        The identification of the sensor indexes is something that has suffered from maintaining backward compatability with earlier versions. For some devices the index of interest is what is returned form the DIS command (e.g. DS2405's) For others it is the index from the TMP command (e.g. DS1820). At some point the wrong one of these two were saved to the ini file [Devices]. To avoid this problem I changed it to save both indexes so there should be no confusion internally.

                        The design intent of the sensor indexing / identification is to use the information from the ini file until the unit reports something different at time of startup. This means that a reading that is not delivered at startup will not matter unless a configuration change was just made.

                        The ??? readings from the TEMP08 likely mean that the device did not response when its address was placed on the 1-wire bus. At this point in the stream of events it does not matter what devices is connects such as a DS2438 for Humidity or DS18B20 for temperature. It essentially means that the pulses for thte address were not recognized by the sensor or the sensor's response not recognized by the TEMP08. This usually is a wiring topology problem which can being either reflections that interfere with the pulse stream or capacitance that suck up the pulses. Mitch should be able to provide all the conditions that cause a ??? to be resported. It may be case that the humidity sensor saturates and gives excessively high readings until it dries out, but they will be readings. I have 2 humidity devices and have response reliabiltiy the same as all my 1-wire devices. There is a reported problem where the device reports 35 which if filtered by the SW. The temp sesnors have a similar singularity at 185F/85C.

                        I do not understand what you are trying to indicate with the 3 versions of the [Devices] section. I suspect it is all related to the indentification method that I just changed.

                        Be suspect when running this initially since it did involve quite a bit of code change. I did reconnect my Temp05 to my HS computer to test with this version so I know it generally works, but my environment is different than yours. The [Devices] entry will now have two number for the index with a colon delimiter. 05:03 means DIS index 5 and TMP index 3.
                        Attached Files

                        Comment


                          #27
                          Michael,

                          The "T" type is the as-shipped type for an on-board DS2438 on TEMP08. Any DS2438 can also be used this way. DS2438's can be designated as Temperature, Voltage, Humidity, or Barometer.

                          The "???" readings are indicators of bad readings from any device, where bad could mean that the device is not present, or that the device returned a reading that did not match the CRC, or that the reading did not match what was expected - in the case of a humidity sensor this could be humidity > 100%.

                          Mitch
                          http://www.midondesign.com

                          Comment


                            #28
                            The [devices] section I reported showed what I interpreted to be strange behaviour in what you indicated was the index field and relavent to the problem in my situation. Sometimes there are 2 digits, sometimes only one, one time it shows 1,2,3 for my humidity sensors, the next it shows other values, including 97 for all sensors at one point. If this okay with you, it's okay with me, but valid (i.e. not ???) temp08 readings were not being recorded in my situation when I reported this. If you need to see something else, please mention it in your replies so I'll know what to provide you.

                            Speaking of my situation, am I the only one using mcsTemp with Midon's Temp08 and MD3020E humidity sensors?

                            Comment


                              #29
                              I suspect you are the first one that has multiple sensors of some type (Humidity) other than temperature sensors. This is where the problem you uncovered exists. In my case I have 2 humidity, but 1 is connected to a DS9097U. Your feedback has been very valuable and especially perceptive of the information that may be needed to help resolve the problem. Thank you for sticking with it and helping get through it.

                              The 97 is used to flag this sensor is believed to be associated with a DS9097 adapter. It becomes the default and was activated when a colon was not found for a sensor in the D:I column of the Analog tab when the setup page was saved. Error counts and configuration settings are saved at different times. One is from the formatted info in the setup page which contains leading zeros and the other is an unformatted number. I think I coverred all my bases with the latest posting

                              Comment

                              Working...
                              X