oberkc

The program suggested by themlruts is how I would do it. " I just don't understand how I can loop and continue to query the sensor until closed." Why do you believe there is any need to loop anything. When the sensor changes, it will tell the ISY. I also don't understand you purpose in the one second cycle of your relay. Is the relay connected to anything?

What makes you think the 9pm program did not run? Is it disabled? If you check program status after 9 and before midnight, does it show TRUE? What evidence is there that the 11pm program did not run? Are you basing this upon the fact that the backlight levels remain unchanged?

The brute-force method would simply be to add a statement at the end of your program to turn off the kpl button. If this button is a secondary button, you may need to create a new scene, containing only that button, and turn the scene off by program. An alternative, depending on your needs, is to create a scene with the ROG keypad controller and chandelier as responder, and modify the program: If CONTROL 'ROG Other.... Then Set moms bedroom kpl on Wait 5 minutes Set new scene off Else Nothing

Also, not all plms are dual band. Is yours? Are you plugging the hub into the same outlet as the plm?

That is quite interesting. Most of the time, with my dedicated tablets that never leave the house, I dont percieve any latency beyond about a second. Perhaps mobilinc on these units dont go through an "initial refresh"? Keep us informed what you come up with. What I would like, but have not yet figured out, is a widget that changes (dark and light, perhaps), depending on state of an insteon device or scene. In other words, I would like a widget that simulates a keypadlinc.

Because of my problems revealed when I upgraded to android 5.0 (thread created in the mobilinc forum), I have suddenly become keenly aware of sluggishness when connected via https connections. In my case, I dont use my phone much around the house, and use other tablets connected solely via local http connection, which is quite fast. With the phone (both cellular and wifi data) and mobilinc, I have yet to Establish a good understanding of how mobilinc transitions from cellular (https) data and wifi (http) data, but I can say that tablets are alway on http connections, and they are fast. Cell phones, when on https connection, have a response lag that can be near 30 seconds in my case. I also believe that if I am home, the phone on wifi, and I continue to see those delays,that mobilinc has not made the transition to http non-secure connection. My point...if you are seeing delays when on wifi, consider the possibility that you are still on https secure connection and have not transitioned to http as your settings suggest should have happened.

Mwareman, Thinking a little further about my approach and your comments, I believe you are correct about one problem with my approach. Unfortunately, I use these toggle widgets sparingly, and in situations where they are the only mechanism that can control the scene. To avoid hav7ng the toggle widget get out of sync, I might choose instead to pick one de ice in a scene as an indicator for the scene state. Then I would use the indicator device as a condition in a program. If Status Indicator device is off then Turn scene on Else turn scene off

The program I posted i disabled to avoid having it otherwise react to changes in status. The only way it can activate is if called by mobilinc. I hav never noticed any problems with mobilinc sync operations. Your proposed approach to the fan problem is near exactly how I would approach it. Like the first, I would expect these would need disabled.

I do toggle tasks with tasker widgets, if that is what you seek. First, i create program (disabled) such as: If This program is false Then Turn light on Else Turn light off I then create a tasker widget, through the mobilinc plug-in, that executes the program if path.

In some cases, I have become a fan of the harmony hub. This would require no line of site between remote and ISY, so long as one of your hub blasters is pointed at the ISY. Alternatively, I see no reason it could not be used to transmit IR commands to the ISY through an IRLinc. It has been a while since I programmed the IRLinc, but I recall that it could learn pretty much any IR command and execute a program from that command as a trigger. I think the IRLinc directly supports scenes, as well.

I would besurprised if moving the parentheses was the solution to this specific problem. After your second description, I have a different understanding of your symptoms. If the door waited the five minutes before trying to close, began to close, then reversed back open, consider the possibility that something interrupted the door safety sensors.

Good news on the communication front. I remain mystified as to the cause of the original problem...why your door would immediately trigger when opened. Have you seen this again?

My quick review of your programs makes me thing that this is NOT a problem, but I just wanted to point out that when you turn the ALL scene off, it could trigger an evaluation of program 1 as a result of a status change of the mudroom icon switch. I notice that this program can be disabled at times by program 2, and that if program 1 is triggered by a change in status, it will likely evaluate false, which is pretty harmless. But you are most familiar with your programs and intentions. Double check that there are no unintended consequences there.

If you are using the KPL to indicate door status and control door, then the two scenes look about right. In the scene with the relay, I assume the KPL is controller. In the scene with the sensor, I assume the KPL is responder. Sometimes, scenes tests fail when there is a program triggered by one of the devices in the tested scene. Try temporarily disabling any such programs, and re-run the scene tests. Also, run several times. In general, I have come to understand that when you have zero remaining "hops", this can be an indication of less-than-optimum communication.

No...not to my knowledge. As I understand, programs run simultaneously. The midnight program should run normally, even if another program is also running. Question...is 'Island (Mudroom Door ICON)' part of the scene 'All'? If so, when the midnight program runs, it could trigger an evaluation of program 1. I cannot help but be concerned that this may be a problem if it happens at the wrong time. I haven't given it a lot of thought, however. Perhaps your disable commands take care of this problem. Regardless, I don't think that is a factor in your original problem.

No, you understand it just fine. In my mind, if the door is closed (sensor off) and is opened between the prescribed hours, I would expect the timer to start, and initiate a scene command 5 minutes later. IN retrospect, you may need a re-allocation of parentheses in your program, but I don't believe that this is a factor here: If ( On Sun, Mon, Tue, Wed, Thu From 10:30:00PM To 6:00:00AM (next day) Or On Sat, Fri From 11:00:00PM To 7:00:00AM (next day) ) And Status '19.5C.D1-Sensor' is On Then Wait 5 minutes Set Scene 'Garage Door Relay' On Else - No Actions - (To add one, press 'Action') I am having trouble seeing why this program would have immediately sent a command to the door when opened. I, too, expect the wait period to complete before sending the command. I don't believe the parentheses issue would cause this, either. I would be looking for scenes created that include both relay and sensor, or other programs that may be impacted by relay or sensor.

I see nothing that starts any 15 minute countdown or ANY reason that the first (second?) program would run. Are there other things you have going on here not shown? If your second program turned off ALL lights, and the island failed to turn off, then there is likely a scene definition error, or a comm error.

The sensor, possibly, transitioned from OFF >>> ON, triggering the program? Perhaps it good that you step back and define exactly what you want to happen. What are you trying to achieve? Are you, simply trying to create a program that closes the door, if open, at 10:00? Do you want a program which closes any open door between the hours of 10:00 and 7:00? Something else?

Hard to say without knowing the definition of "work". Since you have no ELSE statements, there is probably no reason to have to/from times. It would be simpler to simply have "time is". In my mind, this program, as is, will turn ON a relay scene "garage door relay" at 10:30 on some days and 11:00 on others, but only if a sensor is ON at these times. This program will do nothing at 6:00am or 7:00 am, or any other time.

"What is the reason for this?" I might do that from time-to-time if I were using the TRUE or FALSE status of that program as a condition of another program. For example, if I wanted a program to be true only while executing a THEN clause, then turn false immediate after completion of the THEN clause. Then I could use the status of that program to know whether it was currently executing.

Xathros...do you have any experience changing properties of a scene specifically including a battery powered device? I was wondering if the motion sensor need be put into linking mode for the changes to take effect.

The only difference that I understand between direct linked devices and those created by the ISY would be the presence of the PLM in the device link records. If you have a device that was otherwise added to the ISY, I cannot imagine why a scene that was later created directly would be any faster. If you noticea difference in scene responses, I would dig a little deeper into that problem. Something strange is going on here beyond the scene creation mechanics, in my estimation. Like the others, I find it bad practice to have scenes created outside the ISY.

What is a "direct linked scene"? Are you creating scenes directly from the devices, themselves? For the ISY to work best, all scenes should be created through the ISY. I expect xathros suggestion to be doable, but until we see your programs and scenes, generalities are all we have to discuss.

The options can be a bit broader than this, but if you have a program condition who's status will be changed by the program execution, then, yes, it is often best and easiest to break these into two programs. If your execution has no WAIT statements or REPEATs, then it matters less to break them into two programs.

That is, in the end, how I see it as well. I remain very satisfied with my insteon house. The continued introduction of new-and-improved insteon devices makes me think that things are working out for smarthome. I have no intention of wholesale changes to z-wave, but if I see a z-wave device on a clearance table somewhere, it is nice being able to take advantage of that savings. Other than that, it will be insteon for me.