Another ALL ON Event

[Teken]

I've only had one ALL ON event and that took place when there was no other Insteon activity, but it also took place about a month before the power supply circuit in my (22 month old) PLM failed.

 

I'm wondering if AC voltage fluctuations could be a factor in triggering an action in the PLM.   Just curious if anyone has put a digital voltmeter on the AC line to monitor line voltage, especially during evening hours when the loads are high.

 

It's a long shot but being that there's no answer yet I thought I would throw this out.

 

This is something I have expressed several times but not too many have the time or tools to monitor long term power fluctuation with in their home.

 

Well, besides me of course . . .

 

Ironic since I have not seen a ALL ON / ALL OFF event but can say during a Alpha firmware trial I did see quite a few random issues taking place.

 

I had one instance where the ISY reported the GDO was closed when it never opened to start with. I have come to expect to see the closed state upon a ISY reboot. But, the ISY was not rebooted at the time but this false flag message came anyways??

 

I have seen massive KWH consumed in the ISY Console also which appeared out of no where. This is a current screen capture of what is being displayed now in the UI.

 

There are many power, energy, voltage data loggers in my home which monitor all facets of the homes power. This includes every independent circuit along with (isolated) devices which are monitored separately for long term tracking.

 

So it would be great if people could track and monitor their incoming power supply or determine if this is being caused by some kind of start up noise whether it be a sump, fridge, HVAC, well pump, what ever. We have all seen how noise on the line impacts X10 devices and can also make Insteon devices behave oddly.

 

Since there appears to be a wall of silence and sharing of information lots of this is best guess. To repro any sort of problem takes time, patience, and the willingness to provide factual information from all sides. To date no one has answered why this problem persists with the new firmware?? 

 

I have also noted there haven't been any users with the 99 controller reporting this problem??? If there are users who are using the 99 and see this same ALL ON / ALL OFF issue it would be good to know if this is with a replacement PLM??

[Jimbo.Automates]

It is interesting that you mention a power fluctuation, because my wife mentioned that she noticed a light get dim, when it started happening, not sure if it was before or during because I assumed it dimmed when everything was being turned on, which is a lot of lights and 3 GDO's. Also, noticeable power fluctuations don't happen in our house, and I don't have the equipment to monitor it.

 

I am mostly bummed because I've never had the issue for over two years, and replaced my plm just before the two years to avoid the capacitor failure issue, and to get the firmware fix for the all on :'( But I have recently added a few scenes and programs to manage large groups of lights, so I really think that has something to do with it. I also accidentally added an MS as a controller of a scene that already had a program turning on the scene from the MS,so that has been fixed.

 

UDI has followed up to my email to get more specifics and they are looking into it.

 

Sent from my Nexus 7 using Tapatalk

[Michel Kohanim]

Hi stusviews,

 

... the only reason I asked about the RemoteLinc/ControLinc is that they are the only devices with All On/All Off buttons. Did everything in your house turn on/off or just a few? If a few, is it possible to see whether or not they are linked to any of the RLs? Also, do you recall anything that might have happened just prior to the event.

 

Power fluctuations are indeed quite interesting. Now that Dwight's situation was NOT related to sensors, I have to revisit my hypothesis. This said, does any of you have something like:

1. A controller for which there's a program to do something ... i.e sensor switched on, then ... 

2. The same controller as part of a scene as a controller

 

With kind regards,

Michel

[Jimbo.Automates]

Michel, just to clarify, I did have 1 and 2.

 

Sent from my Nexus 7 using Tapatalk

[Michel Kohanim]

Hi Jimbo, thank you.

 

With kind regards,

Michel

[Techman]

Regarding AC power fluctuations:  if someone feels real bold and has a rheostat that they can use to control the AC power to a PLM it might be an interesting experiment. Seems that the common denominator to the all on issue remains elusive.

 

We know that the capacitors in the previous releases of the PLM were marginal at best, makes one wonder what other components in the PLM are sitting on the edge.  Even though the all on function was disabled in the PLM firmware does it still remain as a viable option in the hardware, if so can it be triggered by a false signal?

[Teken]

Regarding AC power fluctuations:  if someone feels real bold and has a rheostat that they can use to control the AC power to a PLM it might be an interesting experiment. Seems that the common denominator to the all on issue remains elusive.

 

We know that the capacitors in the previous releases of the PLM were marginal at best, makes one wonder what other components in the PLM are sitting on the edge.  Even though the all on function was disabled in the PLM firmware does it still remain as a viable option in the hardware, if so can it be triggered by a false signal?

 

I would not recommend going that route unless you have a method to monitor the current draw on the device. There are many times a (regulated) power supply will fail to operate as designed. Meaning, when the voltage goes low current goes up.

 

The end result is increased amps / current flowing through the circuit board, traces, wires, components. Thus resulting in a over current condition and meltdown.

 

Despite popular belief of the general populace a *brown out / voltage sag* is more harmful to electronics than a voltage rise / surge event. Manufactures have spent so much time ensuring their products are able to sustain high voltage events most of them have forgotten to address the low voltage events that are more prevalent in many parts of the world.

 

Just some food for thought as I wouldn't want to see anyone have a crispy cream PLM 

[Techman]

I would not recommend going that route unless you have a method to monitor the current draw on the device. There are many times a (regulated) power supply will fail to operate as designed. Meaning, when the voltage goes low current goes up.

 

The end result is increased amps / current flowing through the circuit board, traces, wires, components. Thus resulting in a over current condition and meltdown.

 

Despite popular belief of the general populace a *brown out / voltage sag* is more harmful to electronics than a voltage rise / surge event. Manufactures have spent so much time ensuring their products are able to sustain high voltage events most of them have forgotten to address the low voltage events that are more prevalent in many parts of the world.

 

Just some food for thought as I wouldn't want to see anyone have a crispy cream PLM 

 

In this case a PLM failure would be a good thing. When a capacitor starts to fail it will usually leak current before it becomes an open circuit, We need someone with an expendable PLM and some test equipment, anyone out there?

[Teken]

In this case a PLM failure would be a good thing. When a capacitor starts to fail it will usually leak current before it becomes an open circuit, We need someone with an expendable PLM and some test equipment, anyone out there?

 

LOL  . . . 

[stusviews]

Michel,

 

Most of the RemotLincs do not control any lighting. Of the two that do control lighting, one RemoteLinc controls lights in only one room, the other controls the backyard lights. But, all the interior lights and the front yard lit up also, so I've discounted those as the All On trigger.

 

Thanks for suggesting possible possible culprits.

 

Stu

[Michel Kohanim]

Hi Stu,

 

Thank you. Do you recall any abnormality right before the event?

 

With kind regards,

Michel

[stusviews]

Hi Michel,

 

Log emailed. From more than 5 minutes before the event until turn on started.

 

Stu

[Michel Kohanim]

Thanks Stu,

 

Got it and responded.

 

With kind regards,

Michel

[dstanley]

Darn. I just had another All On Event tonight at 7:46 pm on 01-11-2015

 

This one had nothing to do with my garage doors though (even though they both opened!).

 

I think it might have been triggered by a RemoteLinc this time that I used to turn on my Office Light but I can't really be sure. My log is enclosed that shows the garage door sensors activate. i had to go around the house and turn everything off.

 

This is still using the 'test' PLM that went into service on Labor Day and had been bug free until now.

 

Dwight

All On-Sun 2015-01-11 07_46_33 PM.zip

[WayneW]

Darn. I just had another All On Event tonight at 7:46 pm on 01-11-2015

 

This one had nothing to do with my garage doors though (even though they both opened!).

 

I think it might have been triggered by a RemoteLinc this time that I used to turn on my Office Light but I can't really be sure. My log is enclosed that shows the garage door sensors activate. i had to go around the house and turn everything off.

 

This is still using the 'test' PLM that went into service on Labor Day and had been bug free until now.

 

Dwight

 

For anybody possibly affected by the "All On" issue, I suggest creating a power fail recovery program.  See http://wiki.universal-devices.com/index.php?title=Power_Failure

It isn't an ideal recovery method, but I find it much easier to manually run the "query all" program (to make the log interesting) and kick off the power recovery program than it is to run around the house fixing everything.  Especially if I am at work and she isn't  In a worst case scenario where you cannot run programs, somebody could also then reboot the ISY to fix most lights.

[Michel Kohanim]

Hi dstanley,

 

I am so sorry to hear. Do you have any programs where the condition includes devices that are in scenes activated be the same program?

 

With kind regards,

Michel

[dstanley]

Do you have any programs where the condition includes devices that are in scenes activated be the same program?

 

Michel,

No - not that I am aware of in this instance - I think this happened when use my RemoteLinc to turn on a scene with just one light in it. No program was involved in this action.

There may have been a program (KitchenMotion,Off) running that fades off my kitchen counter light scene after a delay as my wife had just left to come downstairs where my office is located.

Other than that I don't think there was anything happening that could have caused the collision/flood that we believe contributes to these events.

 

I would be more than happy to provide access to my system to you/support should you like to look at anything ...

 

Dwight

[Teken]

Hello All,

 

I wanted to bring this post to the attention of the forum members: http://forum.universal-devices.com/topic/15136-wierd-problem-with-keypad-2334-222/

 

Can UDI / Smartlabs engage this customer to see what is the root cause. I truly believe this might be the missing piece people have been looking for with respect to the ALL ON / ALL OFF problem.

[LeeG]

Unfortunately this was caused by the PLM link table not being updated.  When a device is controllable with Direct commands but PLM does not see events from button/paddle press the PLM side of the links is always in question.   Problem solved when PLM restored.  That likely compressed out duplicate and no longer used link records making room for the new link records.   This is not associated with All On.

[Michel Kohanim]

Hi Dwight,

 

Thank you. Can you please send your loss to support@universal-devices.com with the time it happened.

 

With kind regards,

Michel

[dstanley]

Hi Dwight,

 

Thank you. Can you please send your loss to support@universal-devices.com with the time it happened.

 

With kind regards,

Michel

 

Done!

[cwagner2401]

Hi Michel,

 

I have had a related thread running since May 2014:    http://forum.universal-devices.com/topic/13152-devices-turn-on-or-off-unexpectedly/  

 

This week Techman pointed me over here.  

 

I've read this thread and until I read that dstanley saw the all-on after using the flashed PLM, I was getting hopeful.

 

2 questions for you:

      1) Even though dstanley still sees events that are closely aligned with what I'm seeing, do you think it's worth me changing out my PLM for 9E? (I have 9B)

      2) i see that you asked a couple times in this thread about ELK.  I'm using ELK rules for my lighting and garage door control.  Can you tell me why you asked about ELK?

 

Thanks!

[Michel Kohanim]

Hi cwagner2401,

 

I think it would be best to wait and see if we can get to the root cause. In most cases the issue is related to programs that turn on and off KPL sub buttons. Especially when the trigger itself is a controller for a scene.

 

ELK has an all on feature. Do you see ELK events in the log?

 

With kind regards,

Michel

[cwagner2401]

Hi Michel,

 

Agreed.  Root cause should be the goal and would likely be a lot easier to find if a passive Insteon "sniffer" existed.  But it doesn't so, subscribing to the theory that every once in a while even a blind squirrel finds a nut,  I continue to persevere.  

 

Yes.  ELK is an active player in my system. My ISY has the ELK module. (ELK HW: 0.13/FW: 5.3.0  ISY FW: 4.2.18

 

The issues I have can be described as two separate, spurious event types:

   1) Someone opens the garage door via the Liftmaster remote, eKeypad or wired button and then (rarely) an all-on occurs within 30 seconds (estimated: haven't measured timing)

          a) immediately after the event and before shutting off lights;  looking at the ISY console, it didn't show all lights status as on.  Only lights on were the ones ISY normally has on at that time of the day.

 

   2) At random times a garage door opens without any human or programmatic trigger 

          a) this hasn't happened since May 2014. this condition comes and goes at long term intervals.

           (rarely) an all-on occurs within 30 seconds (estimated: haven't measured timing)

 

 

ELK role with the garage doors:

Each door has a magnetic sensor wired to an ELK zone and each door's Liftmaster has an IOLinc attached to the control wires of the Liftmaster. ELK rules detect the zone violation when the door opens and do the following

-set output 208 on for 5 mins

-turn on garage interior lights

-turn on garage exterior lights

-turn the rear hall light on

-turn all KPL-E buttons on via "Garage Open Indicator" scene

-speak "A garage Door Is Open"  over speakers located throughout the house

 

when garage door is closed:

-turn off all KPL-E buttons via "Garage Open Indicator" scene

-check to see that all garage doors are closed and if so, speak "all garage doors are closed"

 

when output 208 turns off:

-turn off garage interior lights

-turn off garage exterior lights

-turn off rear hall light

 

Thanks for any insight you can offer.

[EricK]

I'm working today, and got a call from my wife this morning that every light in the house turned on. This was after pressing a secondary button on a fairly new 8 button KPL to turn on our dining scene. This all on event did close the elk water valve. There are multiple responders to the scene including buttons on that Kpl. The plm is just over a year old. The reason I use insteon is that our house had x10 when we moved in and the bedroom lights sometimes all turned on in the middle of the night.

I think I will try to get sh to replace the plm.