ISY Not Detecting Button Activity

[justin.cool]

ISY not detecting KPL button activity        

 

Lee, thanks for your prompt response to my other thread, Device Restore Failure, but while trouble shooting per your advice it appears to have grown into something bigger, possibly a PLM failure. Thus, the new thread. 

 

Configuration  Dec 2014

 

ISY-994i IR/PRO, running 4.0.5 and UI 4.0.11

PLM:  2412S  (Have 2413S, but not yet installed)

Approximately 120 Insteon devices:

• Mainly Switchlincs, Lamplincs, Outletlincs, On/Off Adapters, etc etc
  10 KPL’s – about 15% are dual band units
  Triggerlincs (2)
  RemoteLincs (3)
  RemoteLinc2 (4)
  Controllincs (2)
  Access Points (4) – 2 on each phase from each subpanel
  Dualband LampLinc dimmers (4)
  Dual Band 240V load controller (2) (Normal Open, dual band)
  Signal Linc (2)
  Venstar Thermostats (3) with V2 RF interface
  EZRain/EZFlora Irrigation Controller v.92
   

House is about 4000 sq feet, two floors, but spread out.

• Two primary subpanels, one secondary.  Secondary is dedicated to dryer, pool filter and water feature, and some outdoor outlets on the patio. 
• Both primary panels each have a signal linc installed in the panels. 
• Both primary panels have Leviton 51110-1 120/240 120/240 Volt Single-Phase Panel surge protectors. 
• All power strips in use in the house are either filtered via X-10 Pro 20 A in line or 15A plug in filters, or do NOT have any surge suppression or noise filtering.
• No UPS in the system.
• Numerous other devices are filtered via one of the above or a simple X-10 5A plug in filter or a Leviton in line 5A filter.
• All thermostat dongles (T1, T2 and T3) are Rev 2.2R

 

Issue

 

Problems began with failed Device Restore for a KPL.  While troubleshooting that problem, several other problems appeared. 

Please Note:  Lee, I was not able to follow your suggestion exactly, ie “Move the KPL to the PLM location with a patch cord and see if comms fail well into the restore.”  However, I was able to take one of the new KPL’s with a patch cord and plug it into the same outlet as the PLM.  See below. 

 

• Configured a new replacement, non-dual band KPL with a power cord so I could plug it in, verify it works and then do a “Replace Device” so I don’t have to regenerate all the links for this KPL-8. 
• Successfully attached the new KPL to the ISY / network.  However, when doing a simple check, like pressing buttons on the KPL and watching the ISY detect the button press, the ISY (through the Admin console) did not see / detect any button presses. 
• I then configured a second new non-dual band KPL the same way,   Successfully attached it, and the same problem:  the ISY (through the Admin console) did not see / detect any button presses. 
• I went to the Event Viewer, set up Level 3 “Device Communications Events” and watched while I pressed buttons on this second new KPL.  NOTHING, no traffic and no detection by the ISY of any button presses.
• I then toggled the overhead light switch, (a non-dual band switchlinc with no known issues), and again, Nothing on the event viewer, no detection by the ISY.  (As you would expect, the light did go on).  At this point, I began to suspect a more fundamental problem, like the PLM maybe? 
• I cycled power on the PLM (2412S installed in 2008). 
• The event viewer now showed traffic for the overhead light switch, but still did NOT detect the button presses on the KPL. 
• I then plugged the 2^nd new KPL into the same outlet as the PLM.  When the buttons were pressed, NO event viewer activity, (still at level 3 on the event viewer).  No change in the status of the ISY admin console.  However, the lights / LEDs on the sides of the PLM and piggy-backed Access Point flashed with every button press. 
• NOTE 1:  I should state here that numerous times throughout this endeavor, I was able to kick off a Query from the ISY admin console and the ISY detected and updated the proper status of the device /button being queried.  This was the case for the new KPL’s and the overhead light switchlinc.  Thus, some level of valid communication appeared to be taking place.  
• NOTE 2:  About 1 month ago a number of flakey things with multiple Insteon devices popped up, but it was truly flakey.  In other words, sometimes things would work fine, others would only work once out of two or three presses of a button.  The KPL Device Restore issue I spoke of initially (in the other thread) is the only one that was pretty much a solid failure….and appears to have now devolved into a broader problem.   

 

Question:

1. Are these strong indications of a PLM issue?  Do I need to troubleshoot further to verify?
2. Is there something else I need to look at?

 

[LeeG]

The difference in command response from the KPL was like ...

 

an appliance had been turned On that was absorbing Insteon message activity,

or the PLM stopped sending clear messages,

or the KPL started to fail at that point.    

 

Rarely does a failure show so clearly in a trace, from a consistent Hops Left=2 (as good as it gets) to Hops Left=1 and mostly Hops Left=0 (as bad as it gets and still function).  The problem is from the trace it clearly shows when the issue started but does not identify the source of the failure.   When it was one device (the KPL) I assumed (not proven) the KPL failed.  Now that there are symptoms across the Insteon network it more points to something generating noise (or absorbing signal) or the 2412S PLM.

 

I would replace the PLM first since you have a 2413S available.  There is a section in the Wiki (6.1 if memory is correct) that describes replacing the PLM.  Every device in the system must be updated so it could take a few hours, particularly with older devices that have to be updated one byte at a time. 

[justin.cool]

Lee,

 

As always, thanks very much.  I will take your advice and change out the PLM first. 

 

Next, I will replace the KPL that is not able to operate reliably.  It is worth noting that this KPL is mounted in a box with another, newer KPL that is not showing any problems / issues whatsoever.  While that doesn't rule out the noise / signal sucking scenario, I am inclined to think the KPL has actually failed. 

 

Third, I will watch closely to see if the new PLM eliminates / changes any of the other "flakiness" that I have been seeing across the system.  Ie, guest bathroom fan has an ISY s/w timer to shut off after 20 minutes, and about one in 5 times the ISY does not detect it being turned ON and hence does not turn it off. 

 

I did recently discover that several devices in my new workshop are signal suckers and that appeared to be consistent with the start of the "flakiness" issues.  I installed filters on them all.  In particular, these signal suckers included:

• Makita 18v battery charger
• Emergency lighting units
• The plug-in-module Eco-Lite flashlight battery chargers

If you could give me your thoughts on one more question I would appreciate it.  The time it took to do a device links read from the suspect KPL seemed particularly long compared to what it has been in the past.  Should that "long" time have made me suspicious of anything, like noise, signal sucker, or failing KPL? 

[LeeG]

Not that I saw.   It takes a minimum of 17 commands to write a link record * 32 link records before the point where comm dropped off.  That is at least 544 commands which does take time to execute.  

 

The latest I2CS KPLs use Extended commands so it takes one command per link record.