LeeG

Thanks for the additional motion sensor traces. I’ll sum up what I see in the hopes it will generate some ideas from others. The symptom is a SwitchLinc physically turns On coincident with the SwitchLinc sending a Group Cleanup message to the PLM as though the paddle was pressed (there was no paddle press). This has occurred on three SwitchLincs at different firmware revisions (v27 & v39). Although only two failure traces, one from ISY and one from PowerHome, both occur at the same point in a motion sensor state change Group message sequence. One was a motion On and one was a motion Off. There have been rumors about a new command but to this point nothing has been documented that the motion sensor could send to the SwitchLinc to cause it to change from a Responder to a Controller and send the Group Cleanup message. In both traces the Group Broadcast from the SwitchLinc is not present. It is not a false message generated by the PLM to the ISY/PowerHome as that would not explain the SwitchLinc physically turning On. Group Broadcast from motion sensor – Group On and Off Group Broadcast from motion sensor = Group On and Off Group Broadcast from motion sensor – Group On and Off Group Cleanup from motion sensor to PLM – Group On and Off Group Cleanup from SwitchLinc to PLM – Group On To summarize, this is not a symptom that can be explained within the Insteon command protocol I am familiar with. The actual failure appears to be the SwitchLinc tact switch but it has occurred with three different SwitchLincs at two different firmware levels and so far always coincident with a motion sensor Group On or Off message sequence. Thoughts from others would be appreciated.

The link being written is for button 2 to Control 15.06.13. This Extended message Thu 12/16/2010 08:14:22 PM : [iNST-ERX ] 02 51 0A 30 F3 12 9F E5 11 41 BD B2 36 FC 0C 3F 84 00 01 80 06 B0 03 40 00 is garbage. The expected message is a simple SD message (normal small Insteon message in response to the previous Set MSB which is the first command when writing a byte of RemoteLinc memory. Need full trace of Scene create to be sure the above message was not the result of something earlier. Remove batteries for 2 minutes. Replace with new batteries. Be sure the Link Management | Advanced Options | Automatic is option in effect. Post full Event Viewer trace with Device Communications events of writing Scene with RemoteLinc button as Controller of Scene. Identify RemoteLinc Insteon address and Scene Responder Insteon address. This is the expected Set MSB command and response. Fri 12/17/2010 04:54:23 AM : [iNST-ACK ] 02 62 11.AD.CF 0F 28 0F 06 SET-MSB(0F) Fri 12/17/2010 04:54:23 AM : [iNST-SRX ] 02 50 11.AD.CF 12.9F.E4 27 28 0F SET-MSB(0F) If RemoteLinc continues to respond with Extended message after a Set MSB, with new batteries, it is defective. A Factory reset and Restore Device should be attempted before replacing.

Red nodes represent Controllers from the device perspective. Such as the buttons on a Controlinc. The device sends messages to the Red nodes. The Blue nodes represent Responders from the device perspective. Such as the output relay(s) on an I/O Linc or EZIOxx device. These device nodes receive messages from other Controllers (such as the ISY/PLM). KeypadLinc buttons appear in Red. They can be a Controller or a Responder from the device perspective. Messages come to the Red KPL nodes when a KPL button is pressed. Because these nodes are also Responders, messages can be sent to the Red KPL nodes to turn the KPL LED On and Off. The color scheme should be consistent across all the nodes in the My Lighting tree.

Illusion More data on the unsolicited On message. I checked the trace on the PowerHome forum describing a similar situation. It is very close to your trace. In the Powerhome case the motion sensor is issuing an Off sequence. Immediately following the Group Cleanup with the max hop count of 1 from the motion sensor there is a Group Cleanup from another switch. No initial Group Broadcast message from the other switch which is the same as your situation. The only difference is on the PowerHome trace the motion sensor receives the ACK and does not retry the Group Cleanup with the max hop count 3. Two traces does not make the case for sure but it is looking like either an issue with the motion sensor, the switch firmware where it is reacting to the motion sensor sequence, or the PLM. I am asking the same questions on the PowerHome forum .... motion sensor version ? switch type and firmware revision ? PLM type and firmware revision ? Perhaps a commonality between the two configurations can be found. Lee

Illusion Can you post another motion On from that motion sensor. It would be good to know if there are always multiple Group Cleanup messages from the motion sensor or multiple Group Cleanups only when the failure happens. There may a marginal communication problem with that motion sensor most of the time or perhaps the retries are indicative of an intermittent RF problem. Also, what is the PLM revision level? Lee

The Motion Sensor signaled motion Tue 12/14/2010 02:19:54 PM : [iNST-SRX ] 02 50 0A.3E.E6 00.00.01 C7 11 01 LTONRR (01) After the expected two additional Group Broadcast messages an expected Group Cleanup is received from the motion sensor Tue 12/14/2010 02:19:54 PM : [iNST-SRX ] 02 50 0A.3E.E6 13.24.AA 41 11 01 LTONRR (01) Then the unexpected Group Cleanup from the vanity switch. This message looks like a valid Group Cleanup from a paddle press except the initial Group Broadcast that should have come first is missing Tue 12/14/2010 02:19:55 PM : [iNST-SRX ] 02 50 0C.FB.F5 13.24.AA 41 11 01 LTONRR (01) Then a second Group Cleanup comes from the motion sensor which is both unexpected and the max hop count is three. The max hop count going up would indicate the motion sensor did not get back the expected ACK to the first Group Cleanup and did a retry. Tue 12/14/2010 02:19:55 PM : [iNST-SRX ] 02 50 0A.3E.E6 13.24.AA 47 11 01 LTONRR (01) The problem here is the first retry would normally have a max hop count of 2 followed by another retry with a max hop count of 3. Much more analysis needs to be done here but the Group Cleanup with the vanity from address might be the motion sensor first retry attempt that was presented incorrectly. ONLY A GUESS AT THIS POINT. Since the vanity switch is new I don’t see this as a switch failure even though the message with the vanity address looks valid (except for missing Group Broadcast that should have preceded it). The ISY did what it should have because it received a message indicating the vanity turned on. It can be valid for a Group Broadcast not to arrive first. That is why Insteon follows up with the Group Cleanup as part of the Group protocol. This will take some time to think about the possibilities. The event trace should be continued to get another failure to see if the pattern is consistent.

Occupancy mode is controllable with the ISY through the same mechanism as the other Motion Sensor options. I don't have a v1 motion sensor but don't think Occupancy mode is available at that level. Smarthome does not mention much about Occupancy mode beyond the product description "Wireless INSTEON Motion / Occupancy Sensor". From a functional perspective when Occupancy mode is Off the Motion Sends an On command when motion is detected AND the timeout interval is not active. Only one On command even if more motion is detected during the timeout interval. When Occupancy mode is On the Motion Sensor sends an On command for each motion detected.

On Only mode is not the same as Occupancy Mode. On only mode sends On commands only, the Off command is suppressed. The timeout interval applies with the next On command coming only after the timeout interval has expired. When Occupancy Mode is set, On commands are sent for every motion detected whether the timeout value has expired or not.

mitch236 Glad it is working for you. Don't know why you saw the results you did when you created a Scene with a single KPL button as a Responder. That Scene should have been available as an option under Insteon. I just ran that scenario under 2.8.7 with a single KPL button as a Responder in a Scene. It was listed in the available devices/Scene pull down. No need to change anything now that you have a working Program. Just keep in the mind the next time you need to do it that a single KPL button as a responder should work. Lee

sorka On the Smarthome web site - ISY & PLM as a package ISY-99i PRO - $426.98 ISY-99i/IR PRO - $466.98 On the Universal-Devices web site - ISY Only (no PLM included) ISY99i/IR PRO - $369.00 Smarthome does not carry the ISY Only as an item. Always comes packaged with a PLM. The price difference is whether there is a PLM included with the package. Lee

mjtyson Do you have a pair of Access Points (or other Dual Band devices) installed to couple the two 120V legs. You could have a problem with Insteon signals passing from one 120V leg to the other. Another problem could be the device connected to the ApplianceLinc is causing the ApplianceLinc to turn On and Off based on the load sensing capability of the ApplianceLinc. If this is the case the ApplianceLinc does not have Controller capability. It does NOT have the ability to notify some other device (such as the ISY PLM) that is has been turned On or Off due to load sensing. If neither of the above applies please expand on circumstances. Thanks Lee

sorka The item from Smarthome is an ISY and PLM in one package. The other is for an ISY device alone (no PLM). Lee

sorka From the UDI web site ... "Standard ISY-99i is field upgradeable to ISY-99i PRO to support additional devices/scenes" I already have the PRO so I don't know if this upgrade is available under Help | Purchase Modules or something that is arranged through UDI itself. Lee

mitch236 KeypadLinc Secondary buttons cannot be turned On/Off with Insteon Direct commands. Define the KPL Secondary button as a Responder in an ISY Scene. Then turn the Scene On/Off just as you would a device . Lee

jeffwild1 If you do decide to go back to 2.8.7 DO NOT CHANGE ANYTHING ELSE. Leave everything as it is where you are happy with RF devices working and change only the ISY image. Takes less than two minutes to go to 2.8.7 and another two minutes to go back to 2.8.2 if RF regresses at 2.8.7. The blinking LEDs is the result of switching PLMs. Being back at the original PLM, it is likely the devices were updated with the other PLM address and have not been updated again with the original PLM address. Lee

tpolito A Time triggered ISY Program can Query whatever devices you want as frequently as you want. Querying too many too often can generate lots of Insteon traffic but it is easily done. Lee

Perhaps PLM support was planned at some point before Smartlabs starting shipping PLMs with three different memory configurations. Actually 4 if you count the PLM that has the larger memory but the PLM firmware using it as though it had the small memory configuration. Perhaps it is something planned for the future.

jeffwild1 It indicates there is something interfering with RF communication. Could be a neighbor’s 900 MHz spread spectrum cordless phone. I guess an Access Point failure such that it is sending false RF signals could also be a possibility but I have not seen any user post that type of Access Point failure. You could unplug all the Access Points except the one piggy backed on the PLM (I prefer plugged into the same circuit as the ISY PLM but not physically piggy backed on the PLM). Then see if the Motion Sensor can be added back. That would eliminate the other Access Points as a source of RF signal interference. The RF portion of the Insteon mess network cannot be configured or controlled by any Home Automation application which includes the ISY. There would seem to be no means for the ISY to be causing this. Might try repeating the initial Access Point Set button procedure for insuring the Access Points are on opposite 120V legs and are receiving each other’s RF signal. If this procedure does not result in reliable Access Point LED response then there is either an Access Point failure or RF interference. Lee

wwat I believe the email problem is resolved in 2.8.7. I’m pretty sure the release notes link off the 2.8.7 announce topic shows that as being corrected and there have been a few other posts detailing similar email issues in 2.8.6 which have been answered as resolved by 2.8.7. I jumped in here only because of the time difference between the East coast and the West coast. Lee

Although a PLM can be physically added to the My Lighting list it is not functional. It cannot be used in Scenes because the 254 predefined nodes representing each of the possible 254 Group numbers are not defined. Also there are at present PLMs with three different memory configurations in the field. To correctly manage the PLM link database requires the application of the same algorithm used by the 2.8.x images to determine SHN device PLM memory size. Today that algorithm is applied against SHN devices only. Without knowing the PLM memory size the correct start location of the PLM link database is not known. One can do a Show Device Links Table but the Set MSB value will not be correct across the full line of possible PLMs that can be added. Lee

Teken Sometimes it helps if what looks like a complex Scene definition is broken down into the individual pieces the Insteon hardware sees. The ISY is great at making Scene definition easy from a user perspective. All the underlying mechanics needed by the Insteon hardware is done under the covers when the ISY Scene definition is moved out to the individual devices. Take a simple ISY Scene XXXX where there are 3 Switchlincs, one is physically controlling the load and the other two SwitchLincs are logical 3-4 way switches. Simply make all three SwitchLinc devices Controllers of the Scene XXXX. Scene XXXX SwitchLinc1 – Controller SwitchLinc2 – Controller SwitchLinc3 – Controller Under the covers what the ISY actually defines is this …. Scene 1 (arbitrary name for this example) SwitchLinc 1 – Controller SwitchLinc 2 – Responder – Bright level 100% Ramp rate 0.2 seconds SwitchLinc 3 – Responder Scene 2 (arbitrary name for this example) SwitchLinc2 – Controller SwitchLinc1 – Responder – Bright level 40% Ramp rate 5 seconds SwitchLinc3 – Responder Scene 3 (arbitrary name for this example) SwitchLinc3 – Controller SwitchLinc1 – Responder – Bright level 10% Ramp rate 2 seconds SwitchLinc2 – Responder - Bright level 50% Ramp rate 4 seconds Scene 4 (arbitrary name for this example) ISY/PLM – Controller SwitchLinc1 – Responder SwitchLinc2 – Responder SwitchLinc3 - Responder SwitchLinc2 as a Responder to Scene 1 and Scene 3 can have a different Bright level and Ramp rate for each Scene. Same with SwitchLinc1, it can have a different Bright level and Ramp rate as a Responder for Scene 2 and Scene 3. When the Scenes are examined from the perspective of what the Insteon hardware sees it may be easier to see why things work the way they do. Each Scene’s Responders operate according to the definition for that particular Scene. When Scene 2 is turned On and Off by pressing SwitchLinc2 paddle, Insteon takes no action regarding Scene 1 or Scene 3 responder relationships. That is why KPL buttons do not respond unless they are part of the specific Scene being controlled, even if some or all of their responders are changing state. Scene 4 is what is controlled when Scene XXXX is used in an ISY Program or by the Admin Console. In its most basic form, from an Insteon perspective, every Scene has one Controller and one or more Responders. If a device is NOT a Responder in a particular Scene it is unaffected by any action done by that Scene. Lee

PGME Congratulations. You have all the basic knowledge to set whatever KPL buttons On or Off you want to. You mentioned mutually exclusive KPL buttons (radio buttons) in an earlier post. That function works only for a button press within the KPL where the button is actually pressed. You can set up buttons such when button C is pressed On D and E turn Off but this works only when button B is physically pressed. Turning button B On from another device does not cause D and E to turn Off. Just the way KPL firmware works. Sometimes it is easier to define a Scene with a single Secondary button and control that Scene like you would an individual device. The problem with that approach is that you cannot get simultaneous device response as you get when you turn a Scene On and Off and all the devices follow simultaneously. Kind of pick your own poison. A Scene per KPL button may be easier to set up in some sense but the visual effect of devices turning On/Off in sequence rather than simultaneously may be to objectionable. Good Luck. Post back if something new comes up that has not been covered so far. Lee

PGME Yes, turning Off all the devices that are physically controlling loads has no effect of any device outside that group of devices. Each KPL button or SwitchLinc that is not directly controlling a physical load has to be added to the activity of the "All Off" button. Still unclear if the "All Off" button is driving a Scene which has all the devices that are controlling loads or it is triggering an ISY Program that is turning Off specific devices. Primary KPL buttons (the ones that can control a physical load whether a physical load is connected or not) can be turned On and OFf just like any other individual device. Secondary KPL buttons CANNOT be turned On and Off as individual devices. You can define a Scene used for All Off function and make all the KPL Secondary buttons in play as responders in that All Off Scene. Then the All Off button Program turns the All Off Scene Off along with all the individual devices. If the All Off button is driving a Scene that contains all the load controlling devices, simple add the Secondary KPL buttons as responders to that Scene. Lee

I believe that is true. The window can be minimized (I think) but the Admin Console is the application that is creating the Event Viewer trace based on information the ISY/PLM is passing out to the Admin Console. There are other things that can be done. Change the type of load being controlled by the switch. Disable any Program that has any interaction with that switch. Play with the responder link records in the switch to eliminate various control paths to the switch. Probably more things like that available if one thinks about it more. The only way to get some objective information is to trace the activity. The situation on the other forum where a trace was available, it showed a Group command sequence that can only happen when the paddle is physically pressed. I am assuming that is not the case here. Do you have video recording capability that can be shifted into that room when no one is home. Perhaps observing how the lights turn On, slow ramp up, immediate to full on, etc might provide more clues.

What is the "All Off" button? What is it doing? If that is a Scene and a Toggle mode KPL button does not turn Off that KPL button is not part of the "All Off" Scene. If the All Off button is simply turning Off all the devices that will not be reflected at the Scene level. The KPL button(s) has to be included in the devices that are being turned Off by "All Off"