An update and good news!
After the communications drop outs that I encountered back with rain on Aug 6, I worked with Fine Offset directly to try and resolve. My initial gut reaction was an issue with the ObserverIP software as I knew that the Fine Offset had been attempting to get better range between the Osprey unit and the ObserverIP but due to issues with the indoor unit, had dropped back to an earlier version of the RF code in ObserverIP firmware 4.2.1. Then when the sporadic drop outs continued AFTER the rain ceased and in fact took many hours to finally return to solid transmissions, I began to suspect moisture intrusion into the Osprey since immediately after the rain, we returned to hot, sunny, dry weather which would have eventually evaporated any internal moisture.
So, on Aug 9, I pulled the Osprey down and tore into it. Did not find any evidence of moisture still present, but I really did NOT expect such given the weather in the days between the issue and then. What I did find was a questionable seal around the solar panel which lays almost horizontal near the center of the Osprey. There was a line of RTV, short and partial, on 2 of the sides, but only a fraction of the entire perimeter on the backside was sealed with RTV. It was easy enough to run a continuous bead around the back side of the panel. I also decided to run RTV into the lap joints where the sensors attach to the arms of the unit just to ensure that rain could not seep into the interior. Then I buttoned everything up. I also took the opportunity to install my home brew aspirator on the solar rad shield. Then we had to wait for more rain.
Well, we got rain in spades starting the afternoon of Aug 12. LOTS of rain, downpours of rain with rates at one point of 5.4"/hour! The good news, NO drop outs. The rain has continued thru this morning, Aug 14, and still rock solid comm between the Osprey and the ObserverIP. AW initially wanted to blame the ObserverIP location and I did humor the suggestion long enough to verify that I had a solid connection with it in an alternative location, then I moved it back to where it had resided for a year and the comms remained solid, so it was NEVER an issue with the ObserverIP.
FO and AW were both really appreciative of my digging into the Osprey and finding the lack of a total perimeter RTV seal on the panel, which is the MOST likely point for moisture intrusion and also sits immediately adjacent to the circuit board.
In hindsight, I wished I had setup the WS-2909 display to feed AW's network in the beginning as that would have given us a record of whether the display was ALSO loosing comms but given that I had a year of history with my ObserverIP and several months with the Meteobridge, I really did not see the need when I installed the WS-2902. That is now corrected. I have 3 feeds into the AW network as they run off the MAC address and since the display, ObserverIP and the Meteobridge all have different MAC addresses, no problem sending all 3 for comparison.
Still very impressed with the Osprey despite this bump in the road.
After the communications drop outs that I encountered back with rain on Aug 6, I worked with Fine Offset directly to try and resolve. My initial gut reaction was an issue with the ObserverIP software as I knew that the Fine Offset had been attempting to get better range between the Osprey unit and the ObserverIP but due to issues with the indoor unit, had dropped back to an earlier version of the RF code in ObserverIP firmware 4.2.1. Then when the sporadic drop outs continued AFTER the rain ceased and in fact took many hours to finally return to solid transmissions, I began to suspect moisture intrusion into the Osprey since immediately after the rain, we returned to hot, sunny, dry weather which would have eventually evaporated any internal moisture.
So, on Aug 9, I pulled the Osprey down and tore into it. Did not find any evidence of moisture still present, but I really did NOT expect such given the weather in the days between the issue and then. What I did find was a questionable seal around the solar panel which lays almost horizontal near the center of the Osprey. There was a line of RTV, short and partial, on 2 of the sides, but only a fraction of the entire perimeter on the backside was sealed with RTV. It was easy enough to run a continuous bead around the back side of the panel. I also decided to run RTV into the lap joints where the sensors attach to the arms of the unit just to ensure that rain could not seep into the interior. Then I buttoned everything up. I also took the opportunity to install my home brew aspirator on the solar rad shield. Then we had to wait for more rain.
Well, we got rain in spades starting the afternoon of Aug 12. LOTS of rain, downpours of rain with rates at one point of 5.4"/hour! The good news, NO drop outs. The rain has continued thru this morning, Aug 14, and still rock solid comm between the Osprey and the ObserverIP. AW initially wanted to blame the ObserverIP location and I did humor the suggestion long enough to verify that I had a solid connection with it in an alternative location, then I moved it back to where it had resided for a year and the comms remained solid, so it was NEVER an issue with the ObserverIP.
FO and AW were both really appreciative of my digging into the Osprey and finding the lack of a total perimeter RTV seal on the panel, which is the MOST likely point for moisture intrusion and also sits immediately adjacent to the circuit board.
In hindsight, I wished I had setup the WS-2909 display to feed AW's network in the beginning as that would have given us a record of whether the display was ALSO loosing comms but given that I had a year of history with my ObserverIP and several months with the Meteobridge, I really did not see the need when I installed the WS-2902. That is now corrected. I have 3 feeds into the AW network as they run off the MAC address and since the display, ObserverIP and the Meteobridge all have different MAC addresses, no problem sending all 3 for comparison.
Still very impressed with the Osprey despite this bump in the road.