LeeG

Cannot verify the Leak Sensor will never false. Mine has not but it is an electronic device. To get a single email issue a Disable for/in the Program sending the email. It will be necessary to manually Enable the Program. Or a Wait can be coded after the Notify for say 24 hours and then issue an Enable for the Program to again get emails after 24 hours if the Leak Sensor continues to signal Wet.

FYI - if a new device is ever needed there is a "Replace xxxx with ..." function. Install the new device, add it to the ISY under a dummy name. Right click the old/failed node entry, select the "Replace" function. A list of replacement candidates will be displayed. The new dummy name is selected. All Scene and Program references will be adjusted to used the new device address. The old node name continues with the new Insteon address, the dummy entry representing the new device is automatically deleted. The old Insteon address is never used by the "Replace" function so the new device can be installed in the old device location before doing the Replace. The device types have to be the same. A dimmer replacing a dimmer, a relay replacing a relay, etc.

Every device has links back to the PLM. That is why the Restore Modem (PLM) takes time when the PLM has been replaced as all the link records in all the devices have to have the PLM address changed. If two PLMs existed all the devices would have to have links to the backup PLM and would be receiving all the messages in parallel. That would increase the overall Insteon traffic and expose the backup PLM to the same traffic load as the primary PLM as well as increase the device link management cost to maintain duplicate PLM link records. I suspect maintaining two PLMs would cause more problems than it solves considering how long a PLM normally lasts. Also when a PLM failed all the links to the defective PLM would have to be removed to avoid the comm failures trying to talk to a device that is not responding.

The battery powered devices do have to be done one at a time. Establishing manual linking mode to keep the device awake is a real linking mode. Putting a second battery device into linking mode at the same time links the two devices. If the SwitchLinc is defective the only solution is to replace it. Try pulling the air gap switch for a minute or two to power cycle the device. Can also try a factory reset. If neither work either the comm to that location has a problem or the device has failed. I keep an Appliance cord for testing devices in question. Allows plugging the device into the same plug point as the PLM which eliminates the comm. question.

What is the objective regarding the notification? Send an email every hour, once a day, need to know the requirement. I suspect the unfiltered water would provide a more solid Wet detection. I have an outside water connection that is supplied before the filtering system. No need to filter water to wash the car. It does cause the Leak Sensor to cycle slightly less but it is not absolute. The initial 'damp' condition with the unfiltered water still allows the Leak Sensor to cycle. We actually have good well water unfiltered. A state University runs tests periodically. I would think high iron content would make the water more conductive. Regardless of the water situation there is no latching function in the Leak Sensor so any smoothing out of the Wet/Dry indications have to be done with ISY Programming.

The Leak Sensor has no latching mode. If the conductivity of the leak is sufficient the Leak Sensor will remain in Wet mode. My water comes from a community well filtered through a whole house filtering system. It is common for the Leak Sensor to cycle in this water until sitting in about 1/16" of water. Even at that depth it does cycle infrequently. From the number of users who report similar situations it seems this is "normal". I suspect the more pure the water the less the conductivity factor and thus more cycling is possible. The effect of the cycling can be reduced by using a Wait in the Program monitoring the Leak Sensor.

Blinking Red means the second Access Point is receiving the RF test message from the first Access Point and both Access Points are on the same phase. If the point of the Access Point in the garage is to simply relay the message over RF it is working fine. If the point of the Access Point in the garage is to couple the phases then that is not happening. Unchecking the Trigger Reverse option would not affect the ISY posting the state of the Sensor. It means that Trigger Reverse was causing confusion regarding what commands the Sensor is sending versus what the Query status is reporting which is the direct opposite to the commands being sent when the Sensor changes state. Now that Trigger Reverse is not being used the commands the Sensor sends on a state change match the state that Query returns.

Remember that a Query will change the Sensor state because Trigger Reverse is being used. This happens when the I/O Linc comm is working correctly. When Trigger Reverse is being used and Query does not change the Sensor state that indicates a comm problem. If the event trace is taken open and close the door along with issuing a Query after each door movement.

Post the event trace running at LEVEL 3. When some messages are received but the ISY does not change status it sounds like some of the messages are missing. This could be a comm issue or it could be a link record problem. The 4 tap Set button test does not involve using the Set button on the second Access Point. 4 taps on one of the Access Points Set button and observe the LED on the other Access Point. The second Access Point should blink green (or Red if on same phase) because it is receiving the test RF message.

"the "heartbeat" signal is a "node" in the ISY, but it will NEVER be OFF, right?" This is not correct. The Heartbeat node can be On or Off. The Heartbeat message contains an On or Off command based on the Wet/Dry state of the Leak Sensor. This is my Heartbeat Program If Control 'Leak Sensor-Heartbeat' is switched On Or Control 'Leak Sensor-Heartbeat' is switched Off Then Wait 25 hours Send X10 'A10/All Lights On (5)' Else - No Actions - (To add one, press 'Action')

The 'Is Load' is for your information.

SetMonkey13 The idea is correct but the Adjust Scene needs a slight change. Using the Scene Name in the 'In Scene' parameter affects use of the Scene Name in a Program or through the Admin Console. To affect the Motion Sensor use the Motion Sensor - Sensor node name in the 'In Scene' parameter. The Motion Sensor - Sensor is the Controller and it is the Controller node name that is needed in the 'In Scene' parameter.

It is 15% off when purchasing $500 worth of equipment.

A simple approach from a Programming perspective On Only Mode: unchecked - (On commands only) Sensing Mode: checked - (Also known as Occupancy Mode) If Control 'Motion Sensor I2CS-Sensor' is switched On Then Wait 30 seconds Set 'SwitchLinc Relay' Off Else - No Actions - (To add one, press 'Action') Each time the Motion Sensor detects motion it sends an On command. No minimum 30 seconds when in Occupancy Mode. Only if the Wait expires does the device turn Off. If motion continues to be sensed within the Wait 30 seconds a new Wait 30 seconds is established.

For the posted Program the Folder condition changes while the Program is in a Wait. This stops the Program execution because the Folder condition is no longer True. What is trying to trigger other Programs while the lights are flashing. There is a limited amount of bandwidth on the powerline. It is probable that other Insteon activity cannot get on the powerline while lights are blinking if the other Programs are being triggered by Insteon activity. Insteon allows a single Scene to be running at once. It will cancel a Scene when other Insteon activity occurs. Need more information about what the Program that is blinking lights is doing along with what is being used to trigger the other Programs.

The Program is setting the "Local" On Levels correctly. These are the values that affect what the individual switch does when its local paddle is pressed. The Master switch that is controlling the load should go to the set Local On Level value when its paddle is pressed On. The Slave switch status LEDs will go to the set Local On Level value when its paddle is pressed. The Local On Level value for the Slave switch has no other affect besides controlling the status LEDs on the Slave switch. There are lots of On Levels in play here so let’s go through them. The Local On Level has been covered above. The Scene Name Responder On Levels affect what each Scene Responder does when the Scene Name is turned On with a Program statement or turned On with the Admin Console. The ISY PLM is the Controller when the Scene Name is used. There are two additional Controllers defined for the Scene, the Master switch and the Slave switch. The Responders to these Controllers have their own Responder On Level settings. That is the 50% Responder On level that is displayed for the Responder Slave switch when Master switch node name as Controller is selected. When the Slave switch node name as Controller is selected there will be a Master switch Responder On Level. To summarize in Insteon every Scene controller has a unique set of On Level values for its Responders. In addition to the Scene Responder On Levels the Local On Levels are in effect for the switch where the paddle is being pressed. This means that the time based Program that is setting the Local On Level settings correctly has several more Responder On Levels to set if all the Responder On Levels are to be adjusted as the Local On Levels are now. An Adjust Scene statement that specifies the Controller name in the 'In Scene' parameter and Responder name in the 'Set' parameter must be added to the Program. I believe that will be three additional Adjust Scene statements in the Then clause and three additional Adjust Scene statements in the Else clause.

Insteon requires a Scene if both the KPL button and the FanLinc Light should come on together. Turning the KPL button On from the Admin Console or the phone app does not cause the KPL to turn On its Responders of the ON button. Insteon does not work that way. MobiLinc will turn a Scene On when the ON button Icon is clicked if the ON button is a Controller of an ISY Scene.

That would be about right. The Motion Sensor being an RF device sends more messages than a wired device. A trace of the motion sensor that triggers a Program to turn On the master bath light takes about 1.5 to 2 seconds. To avoid the delay have the motion sensor turn the light on directly and use the Program to control when the light turns Off. Sun 09/01/2013 01:56:08 PM : [iNST-SRX ] 02 50 13.F2.00 00.00.01 CB 11 01 LTONRR (01) Sun 09/01/2013 01:56:08 PM : [std-Group ] 13.F2.00-->Group=1, Max Hops=3, Hops Left=2 Sun 09/01/2013 01:56:08 PM : [ 13 F2 0 1] DON 1 Sun 09/01/2013 01:56:09 PM : [ 13 F2 0 1] ST 255 Sun 09/01/2013 01:56:09 PM : [iNST-SRX ] 02 50 13.F2.00 00.00.01 CB 11 01 LTONRR (01) Sun 09/01/2013 01:56:09 PM : [std-Group ] 13.F2.00-->Group=1, Max Hops=3, Hops Left=2 Sun 09/01/2013 01:56:09 PM : [iNST-DUP ] Previous message ignored. Sun 09/01/2013 01:56:09 PM : [iNST-SRX ] 02 50 13.F2.00 19.82.4F 41 11 01 LTONRR (01) Sun 09/01/2013 01:56:09 PM : [std-Cleanup ] 13.F2.00-->ISY/PLM Group=1, Max Hops=1, Hops Left=0 Sun 09/01/2013 01:56:09 PM : [iNST-DUP ] Previous message ignored. Sun 09/01/2013 01:56:09 PM : [iNST-TX-I1 ] 02 62 12 B7 52 0F 11 7F Sun 09/01/2013 01:56:09 PM : [iNST-ACK ] 02 62 12.B7.52 0F 11 7F 06 LTONRR (7F) Sun 09/01/2013 01:56:10 PM : [iNST-SRX ] 02 50 12.B7.52 19.82.4F 2B 11 7F LTONRR (7F) Sun 09/01/2013 01:56:10 PM : [std-Direct Ack] 12.B7.52-->ISY/PLM Group=0, Max Hops=3, Hops Left=2 Sun 09/01/2013 01:56:10 PM : [ 12 B7 52 1] ST 127

I took the button caps off when I ran the latest series of tests to better observe the LED levels. There remains a predictable minimum difference but it looks like the On level can be adjusted over a range of 16 increments. That is, when the Off level is set at 2 the On level can be set in a range of 0-15. The 0 On level appears to be 8 increments above the Off level but can be increased in increments of 1 to a value of 15. The increase in the visual On level is a very small amount for each +1. I could not see the difference with the opaque caps in place but is visible in a dark room with the caps off. I tried to replicate the test I ran a year ago taking the button caps off but found that KPL had died during a storm and was replaced in the test bed. Not all KPLs seem to increase the On level at the same exact rate but that can easily be the difference in components in any given KPL. The increase in visual On level for each +1 is very small. This does not allow the Off level to be set independent of the On Level. For a 0 Off level the button LEDs are off when the button is Off. The max On level with a 0 Off level is nowhere the max On level with a 7 Off level. I do not know the reasoning behind that design and SmartLabs will never provide the rationale behind that design. It would appear that using the color change kit for selected buttons only was not anticipated.

That's great. Thanks for taking the time post back the results.

To restore the KeypadLinc link database right click the Primary node, select Restore Device. To restore the PLM link database select File | Restore Modem (PLM).

If no X10 traffic in the event viewer watch the Programs | Summary tab. Is the Program with the Or triggered at all (Last Run Time) and if triggered what was the If evaluation (Status=true/false).

When a KPL button is pressed at least two messages are sent from the KPL. inbound Group Broadcast - triggers program inbound Group Cleanup Direct (this message is sent for each device linked to the KPL button) When the Program works I think this is the sequence that is occurring inbound Group Broadcast - triggers program outbound X10 inbound Group Cleanup Direct When the Program If is more complex I think what is happening is this inbound Group Broadcast - triggers program outbound X10 at the same time the KPL is sending the inbound Group Cleanup Direct the outbound and inbound messages interfere with each other The solution is to add a Wait of 2-3 seconds before the X10 message. More Wait time could be required depending on the number of devices the KPL button is linked to. Anticipating the response that the Wait was tried and it did not solve the problem, the Wait is needed. That fact that it did not solved the symptom before does not mean it is not needed.

Run Tools | Diagnostics | Event Viewer at LEVEL 3. Tap the button that works. When that command activity completes tap the button that does not work. Post the event trace. Could have a bad tac switch for the button but more likely a link record for the button has been lost either in the KPL or the PLM.

Right click Program name, select Copy to Clipboard and paste Program that does not work.