Announcement

Collapse
No announcement yet.

Firmware V42 and comm failures?

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

    Firmware V42 and comm failures?

    I've noticed communications stats for only two devices in my setup showing abnormally high communications failures and both are V42 firmware - these two are the only V42 devices in the network. One device is an OutletLinc and the other is an outdoor ON/OFF module. A "stress test" event that cycles a target device every 15 seconds for 30 minutes turns up errors only with these two V42 devices. Also tested using the same event and on the same circuit were two ON/OFF plug-in modules (V46 and V48), one I/O Linc (V41) and two Micro ON/OFF modules (V43) - all passed with zero comm failures. Testing was also done on multiple circuits across all three phases and the result was the same. In all tests, no load was connected to the devices. Nothing is showing in the detailed log as to why these two devices are failing. Any idea what would be wonky with V42 firmware?

    #2
    can you post a simple table of the stats for the V42 devices and some of the others under the same conditions: circuit, etc
    Mark

    HS3 Pro 3.0.0.534
    Hardware: Insteon Serial PLM | AD2USB for Vista Alarm | HAI Omnistat2 | 1-Wire HA7E | RFXrec433 | Dahua Cameras | LiftMaster Internet Gateway
    Plugins: Insteon (mine) | Vista Alarm (mine) | Omnistat 3 (by Kirby) | Ultra1Wire3 | RFXCOM | NetCAM | MyQ | BLRadar | BLDenon | Jon00 Charting
    Platform: HP h8-1360t, Windows Server 2012 R2, i7-3.4GHz, 16GB memory

    Comment


      #3
      Once I narrowed it down the the two V42 devices, I tested on the same circuit at a receptacle about 32" directly upstream of the PLM. This circuit only powers these two receptacles and no other items were plugged in. This evening I will make some changes to the cycling event so that all test devices are subject to the exact number of cycles and retest. Will post the results then.

      Also noticed something interesting with a V44 SwitchLinc dimmer. If the device is ON and a subsequent ON command is sent from the plug-in, a comm failure is recorded. The same for subsequent OFF commands if the device is OFF. Doesn't seem to cause a problem but just strange.

      Originally posted by mnsandler View Post
      can you post a simple table of the stats for the V42 devices and some of the others under the same conditions: circuit, etc

      Comment


        #4
        Originally posted by Zach View Post
        ... I will make some changes to the cycling event so that all test devices are subject to the exact number of cycles and retest....
        Have you considered using the Test Comm button on the Insteon Manage Devices page?

        Comment


          #5
          claude,

          I should have included the fact that I did use the Test Comm function in the plug-in in my original post. I used the function in the device programming page that issues 20 pings. Each device was tested ten times; 200 pings per device. The OutletLinc returned 198 at one hop and 2 at two hops - no failures. The Outdoor ON/OFF returned 196 at one hop and 4 at two hops - no failures. Only when actually operating the two devices are comm failures noted. When running a full-blown operation test, the successful comm rate hovers in the 72-77% range. Historically, the ping testing has been a nearly perfect indicator of actual operating reliability but with these two V42 devices, not so.

          Originally posted by claude View Post
          Have you considered using the Test Comm button on the Insteon Manage Devices page?

          Comment


            #6
            Did you try the ping test with them on and with them off, or with them only in one state? I find that some of my loads generate noise, so that I get some failures with the load on and usually none with the load off.

            Comment


              #7
              Once I discovered the problem, I disconnected the loads and then tested. After seeing your reply, I connected the V42 devices to a double-conversion UPS in order to force the devices to communicate over the air with the PLM. Much to my surprise most of the comm failures stopped; there are a few but the success rate is >95%. There has to be some noise somewhere on the system that's causing an issue with these two devices. Being a commercial/industrial electrical contractor, I've done the entire system in conduit with steel boxes and SS covers so it's not exactly the type of system Insteon has designed their products for.

              So......at the end of the day we have two V42 devices that respond to pings just fine but they fail pretty often when actually issued a command to change state - polling does work though. Location doesn't seem to matter. The same two devices play nicely if not allowed to use powerline communications. V43 and V44 SwitchLinc devices work perfectly in the locations where these devices used to be. I guess it's one for the 'go figure' file.

              Thanks to all for the ideas and assistance!!

              Originally posted by joegr View Post
              Did you try the ping test with them on and with them off, or with them only in one state? I find that some of my loads generate noise, so that I get some failures with the load on and usually none with the load off.

              Comment

              Working...
              X