Can't re-link 2477S to ISY

[swbrains]

I had a 2477S that was installed for months and worked properly.  It was the master controlling 2 light fixtures (LED) and linked to a slave 2477S on the other side of the room.   A couple of days ago, it started blinking red after both on and off actions, but still worked and still controlled the slave properly (although sometimes there was a delay in the slave responding in sync).

 

I replaced it with another 2477S (new) and it failed to link to the ISY.  

 

I finally replaced it with an old 2476 ISY and it linked but quickly showed similar symptoms of not always controlling the slave.  The slave would blink red when it was used to turn the light on/off.  There was often a delay between each switch responding to the other's action.

 

I re-installed the original and now I can even link it to the ISY.  Rebooted ISY and PLM.  Factory reset switch multiple times. I also updated the ISY to the latest stable version 4.5.4

 

I unplugged Christmas lights (which were recently installed) and unplugged nearby devices.  No other new devices on this circuit.  

 

Below is the log file while trying to link the original 2477S.  I have tried adding the device selecting the 2477S from the device list as well as auto-detect.  Both failed with the same error: "cannot determine device link table address".  On other attempts, I get "cannot determine Insteon device type".

Thu 12/08/2016 09:47:17 PM : [40 F4 C1    ] Added to list of devices to link to ISY
Thu 12/08/2016 09:47:17 PM : [INST-TX-I1  ] 02 62 40 F4 C1 0F 0D 00
Thu 12/08/2016 09:47:17 PM : [INST-ACK    ] 02 62 40.F4.C1 0F 0D 00 06                 (00)
Thu 12/08/2016 09:47:19 PM : [INST-SRX    ] 02 50 40.F4.C1 30.6B.4C 23 0D 02           (02)
Thu 12/08/2016 09:47:19 PM : [Std-Direct Ack] 40.F4.C1-->ISY/PLM Group=0, Max Hops=3, Hops Left=0
Thu 12/08/2016 09:47:19 PM : [INST-TX-I1  ] 02 62 40 F4 C1 0F 10 00
Thu 12/08/2016 09:47:19 PM : [INST-ACK    ] 02 62 40.F4.C1 0F 10 00 06          ID-REQ (00)
Thu 12/08/2016 09:47:28 PM : [INST-TX-I1  ] 02 62 40 F4 C1 0F 10 00
Thu 12/08/2016 09:47:28 PM : [INST-ACK    ] 02 62 40.F4.C1 0F 10 00 06          ID-REQ (00)
Thu 12/08/2016 09:47:37 PM : [INST-TX-I1  ] 02 62 40 F4 C1 0F 10 00
Thu 12/08/2016 09:47:37 PM : [INST-ACK    ] 02 62 40.F4.C1 0F 10 00 06          ID-REQ (00)
Thu 12/08/2016 09:47:41 PM : [40 F4 C1 0  ] Calibrating engine version
Thu 12/08/2016 09:47:41 PM : [INST-TX-I2CS] 02 62 40 F4 C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0
Thu 12/08/2016 09:47:41 PM : [INST-ACK    ] 02 62 40.F4.C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0 06        (00)
Thu 12/08/2016 09:47:50 PM : [INST-TX-I2CS] 02 62 40 F4 C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0
Thu 12/08/2016 09:47:50 PM : [INST-ACK    ] 02 62 40.F4.C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0 06        (00)
Thu 12/08/2016 09:47:59 PM : [INST-TX-I2CS] 02 62 40 F4 C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0
Thu 12/08/2016 09:47:59 PM : [INST-ACK    ] 02 62 40.F4.C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0 06        (00)
Thu 12/08/2016 09:48:03 PM : [40 F4 C1 0  ] Failed to add device, reason 9
Thu 12/08/2016 09:48:03 PM : [All         ] Writing 0 bytes to devices

I'm not sure what to do now as I can't even re-link this device to the ISY....

 

 

 

 

[stusviews]

Disconnect and cap the load (red) wire and try again to add the 2477.

[swbrains]

Thanks for the quick reply.  Will have to try again tomorrow as the room is occupied now for sleeping...

[swbrains]

FWIW, I tried plugging the PLM into various outlets around the house using a 50 ft heavy duty extension cord.  It worked slightly better in that it seemed to be able to detect the device more quickly, but still couldn't successfully add it, usually resulting in "cannot determine device link table address".  Eventually, after several attempts plugged into a particular outlet, the PLM was able to add the switch but left it showing in a state needing updates.  Several attempts at "write updates to device" or "restore device" would appear to succeed (no error) but the device update icon still appeared next to the device name.  Again, after several more attempts at both actions, it finally removed the icon indicating it had a pending update.

 

But now the master switch controls the light (it is connected to the load) but it doesn't control the slave switch.  Both devices are linked as controllers and responders to each other according to the admin console.  Oddly, the slave switch will control the master switch/light fixture (and no red blinking from either device), but the slave switch does not respond (no click or LED status change) when the master switch is used.  So the slave doesn't appear to act as a responder, even though it shows up as a responder to the master switch and the scene in the console.

 

Since I still had a lot of trouble adding the device even with the PLM plugged into different outlets/circuits, I'm guessing there is something somewhere that is sucking signal before it gets to the switch(es).  Since both devices are on different circuits, I'm guessing the power line signal simply isn't making it back to the panel and then back to the other device (I do have a hardwired signal bridge at the panel).  

 

What's confusing me, though, is that both devices are relatively new dual-band 2477S switches (both v.45 according to the ISY) and they are about 15 feet directly apart from each other at opposite ends of a bathroom counter from each other with no walls or other obstructions in between.  They couldn't be more directly in line with each other.    I unplugged everything in the bathroom last night while trying to add the switch but nothing seemed to help (even though all these devices had been plugged in before when it was working properly).  It seemed like pure luck that it finally added it, but still missed something in the linkage as the slave switch doesn't respond to the master (but does control the master properly).  I thought the dual-band part of these devices was supposed to compensate for power-line signal errors and make the connection "over the air" if the PL signals didn't make it through, but perhaps I'm misunderstanding the role of the dual-band functionality.

 

Anyway, today I'll try removing the load from the master to see if it makes it easier to link.  Even though everything is working right now except for the slave being controlled by the master, there's obviously something still amiss.

[swbrains]

FYI - I disconnected the red load wire from the master switch, but still can't write updates or restore the 2477S.  It is having the same symptoms as when the load was attached -- the admin console doesn't give an error after the update/restore process, but it leaves the green pending writes icon on that device.  When I tried "write updates", here is the log of events for that action (again, device left in pending writes state after this update):

 

Fri 12/09/2016 11:32:12 AM : [iNST-TX-I2CS] 02 62 40 F4 C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0

Fri 12/09/2016 11:32:12 AM : [iNST-ACK    ] 02 62 40.F4.C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0 06        (00)

Fri 12/09/2016 11:32:12 AM : [iNST-SRX    ] 02 50 40.F4.C1 30.6B.4C 2B 2F 00           (00)

Fri 12/09/2016 11:32:12 AM : [std-Direct Ack] 40.F4.C1-->ISY/PLM Group=0, Max Hops=3, Hops Left=2

Fri 12/09/2016 11:32:21 AM : [iNST-TX-I2CS] 02 62 40 F4 C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0

Fri 12/09/2016 11:32:21 AM : [iNST-ACK    ] 02 62 40.F4.C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0 06        (00)

Fri 12/09/2016 11:32:22 AM : [iNST-SRX    ] 02 50 40.F4.C1 30.6B.4C 2B 2F 00           (00)

Fri 12/09/2016 11:32:22 AM : [std-Direct Ack] 40.F4.C1-->ISY/PLM Group=0, Max Hops=3, Hops Left=2

Fri 12/09/2016 11:32:31 AM : [iNST-TX-I2CS] 02 62 40 F4 C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0

Fri 12/09/2016 11:32:31 AM : [iNST-ACK    ] 02 62 40.F4.C1 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0 06        (00)

Fri 12/09/2016 11:32:33 AM : [iNST-SRX    ] 02 50 40.F4.C1 30.6B.4C 2F 2F 00           (00)

Fri 12/09/2016 11:32:33 AM : [std-Direct Ack] 40.F4.C1-->ISY/PLM Group=0, Max Hops=3, Hops Left=3

[swbrains]

Update:  I went around the house seeking power strips.  One in the living room had all my A/V equipment plugged into it but already had a FilterLinc on it.  I unplugged it anyway wondering if the FilterLinc had failed but even with the entire thing unplugged I couldn't get it to update the switch in the bathroom (no error, just pending writes).  I finally found a power strip on the same circuit as the swtichlinc behind a nightstand.  It has been there for some time and I believe the same devices have been plugged into it.  I unplugged it and was able to restore the 2477 immediately several times.  

 

The odd part is that I plugged it back in a minute later and was still able to restore the 2477 successfully.  Then, a few minutes later, it failed to update again.  So I unplugged the power strip again and the update worked immediately.  I added a FilterLinc and plugged it back in and was able to restore the 2477 again.  It really seems it was that power strip, but it's very odd that after plugging it back in without the FilterLinc, I was able to restore the 2477.  It would seem like a red herring, but other than that, it was very consistent.  Every time it would fail to restore, I would unplug the strip and it immediately succeeded to update.  

 

So, I'll keep an eye on the switches and see of the red blinking LED resurfaces. If so, I can unplug that power strip and see if the red LED goes away.

[Teken]

I thought the dual-band part of these devices was supposed to compensate for power-line signal errors and make the connection "over the air" if the PL signals didn't make it through, but perhaps I'm misunderstanding the role of the dual-band functionality.

 

 

 

From personal experience devices which are dual band but don't offer the ability to turn off or select RF vs Power Line. I have observed both communication paths are integrally linked to one another. I know lots of the comments and feed back are anecdotal in nature and doesn't offer imperial hard data.

 

But first hand experience is normally one of the measurement criteria where facts are obtained.

 

As such in my personal experience those devices such as the following: On-Off Relay, Fan Linc, Dual Outlet, 1-10 Ballast, and possibly the DIN? All offer the ability to select and turn on-off the RF vs Power Line.

 

These devices in firmware and hardware appear to me at least in many installations mimic the RF only devices like Z-Wave / ZigBee. My hopes are the next generation of Insteon products incorporates the very same in all of the hardware line. I'm not really sure why Smartlabs hasn't done this given several hardware revisions have been seen mid stream and in following years?!?!

 

Lastly, my hopes are they will continue to push the RF output as has been seen on several products like the HUB II, Dual Outlet, and RE.

 

All of these products exceed the standard 150 open air RF TX/RX as seen in the standard issue product line. Given, none of these devices even come close to the FCC guidelines for low output RF devices. Meaning there is still plenty of room for increased RF output to increase the reliability, distance, and speed of which a signal is seen and acted upon.

 

Glad you figured it out it was a noise maker / signal sucker in the home . . .