Leak Sensors - A (Suggested) Complete Program Package

[EricK]

Bump.

Still receiving a daily alert that the laundry variable control program is not running (variable=3). When I check in the admin console it shows that the program is running. Not sure what to do.

Eric

[Teken]

Reboot the ISY and report back.

 

 

Encrypted By: Phoenix Security Solutions

[EricK]

Teken,

I rebooted and the Laundry Leak sensor variable=1.  The variable control 2 program shows that it is running.  I did not get an alert yesterday, but I had triggered the HB of the leak sensor the day before.

We'll see what happens.

Thank you,

Eric

[Michel Kohanim]

Hello EricK,

 

I am so very sorry for missing this post. My first question:

When you noticed the issue first, could you trigger wet/dry (clicking the set button) in the original location or only after you moved it to the kitchen?

 

Also, please do not hesitate to contact our tech support.

 

With kind regards,

Michel

[belias]

Hi EricK,

 

Sorry for my delayed response as well, but it looks like you're getting some good suggestions.  If the VC2 program shows as "Running Then" in the status, I'm not sure why would get a notification email.  If it continues to happen after the ISY restart, check the value of the variable itself (something like sLeakLaundry) - it should be either 1 (normal) or 2 (heartbeat missed).  If it's 3 then at least we know why the email alert is being triggered...

[EricK]

Good morning.

It has been about 36 hours since I rebooted the ISY.  So far so good the leak sensor is back in position and working fine.  The variable = 1, not 3 like it was before.  The laundry VC program is running.  Michel, the leak sensor is behind the washing machine, but very close to some DB devices.  I brought it 10' into the kitchen where my laptop was with the admin console up to check it.  

Eric

[EricK]

Hello,

I thought things with my leak sensor were ok, I had not received a notification for about 2 weeks.  However, last night I received notification that the laundry leak sensor variable=3.  I just looked in the admin console and the variable is 3.   There was no entry for last run for the variable control 3 program.   

There is good communication with the leak sensor.  I kept it out of the laundry room for a while and had no problems.  I placed it back in the laundry room a few days ago, not quite behind the washing machine, with an AP plugged in a few feet away.  It is possible that this was three days ago, so by the time 26 hours passed for the variable to get to 3, then for the notification to come (I have it set to send at 6pm), that placing the leak sensor back in the laundry room coincided with the problem.

I am trying to understand the cause of this problem.  It seems that the only way for the variable =3 is for the variable control 3 program not to run for 26 hours, then to run to get the then to set the variable to 3.  Is there a communication drop with the device so the communication does not run.

I made a patch program to set the variable to 1.  

I have a spare leak sensor so I may swap devices.

Thanks,

Eric

[larryllix]

Hello,

I thought things with my leak sensor were ok, I had not received a notification for about 2 weeks.  However, last night I received notification that the laundry leak sensor variable=3.  I just looked in the admin console and the variable is 3.   There was no entry for last run for the variable control 3 program.   

There is good communication with the leak sensor.  I kept it out of the laundry room for a while and had no problems.  I placed it back in the laundry room a few days ago, not quite behind the washing machine, with an AP plugged in a few feet away.  It is possible that this was three days ago, so by the time 26 hours passed for the variable to get to 3, then for the notification to come (I have it set to send at 6pm), that placing the leak sensor back in the laundry room coincided with the problem.

I am trying to understand the cause of this problem.  It seems that the only way for the variable =3 is for the variable control 3 program not to run for 26 hours, then to run to get the then to set the variable to 3.  Is there a communication drop with the device so the communication does not run.

I made a patch program to set the variable to 1.  

I have a spare leak sensor so I may swap devices.

Thanks,

Eric

Status 3 indicates that the heartbeat monitor program is not running. not a leak sensor problem.

 

Without rehashing all the program  logic in my head IIRC the watchdog program has to be forced to run the first time by  another program that gets executed on power up of the ISY. Perhaps this option was not set to accomplish that.

[EricK]

Larry,

The program should be running at startup but I will verify. I had not received a notification for almost two weeks and it is just the one leak sensor. All others are fine. I did not receive a notification tonight, so far so good.

Eric

[belias]

EricK, et al.,

 

This has been bugging me for a while (why the VC 3 program would run and set the $Leak variable to 3).  I think I figured it out, and have a proposed solution.  

 

Here's the scenario for how it happens:

 

1.  Leak sensor goes offline for 25 hours

2.  The VC2 program sets $Leak = 2 and then stops running

3.  Sometime later, a heartbeat signal is received

4.  The VC2 program now runs and sets $Leak = 1

5.  ** Here's the problem:  In this instant, the $Leak variable = 1, BUT the "last run time" of VC2 hasn't been updated, so all the conditions of VC3 are met and it runs, setting the variable to 3

6.  So now, the $Leak variable is stuck in 3

 

 

So, here's my proposed fix:

 

Add a 5 second wait at the top of all VC3 programs:

Then
        Wait 5 seconds
        $sLeakKitchenSink  = 3

This should allow enough time for the "Last Run Time" to be updated, once the leak sensor comes back online and the VC3 program won't run.

 

 

If this works, please let me know and I'll update the original post.

 

 

Thanks,

 

- Brian

[EricK]

The problem persisted with my patch program off.  Heartbeat is on.  I carefully analyzed this program:

If
        $s.LeakSensorLaundry is 1
    And Time is Last Run Time for 'Leak Sensor - Laundry variable control 3' + 26 hours 

Then
        $s.LeakSensorLaundry  = 3

Else
   - No Actions - (To add one, press 'Action')

 

This program is wrong, it is a circle.  It should be:

If
$s.LeakSensorLaundry is 1
And Time is Last Run Time for 'Leak Sensor - Laundry variable control 2' + 26 hours

Now I am not sure why this problem with the notification just began after almost a year if the program was wrong from the beginning.  Unless I somehow altered the program while checking over things during the last few weeks.  All of my other leak sensor programs are correct.  

Now that I have this fixed we'll have to see what happens.  If I still get the notification then I will add the delay.  

Thank you very much.

Eric

[larryllix]

I have to wonder that a new firmware version caused the speed to change on something.

[EricK]

Larry, I did upgrade the firmware recently but not sure exactly when.

[raydoc]

Would I be able to use this program with the water sensors and the current sensors to work out a way to notify me not only if there is high water in my sumps but if the sump pumps are not actually running? The current sensors know if the pump is running and what I need to have is a notification that the pumps are running AND losing ground.

[EricK]

Raydoc, can you describe in more detail what kind of equipment you are using. What type of sensor is it that currently monitors the pump. You could use a float switch connected to an io linc to monitor if the water level is high. This would be set higher than the water level that triggers the pump, this way you know you are losing ground. If you also had an io linc connected to the pump then you could get have a message sent if both the float switch and pump are on. The leak sensor programs rely on the heartbeat signal that is sent out by the leak sensors. Io linc do not have a heartbeat. I use the leak sensors and also have a float switch in the koi pond.

Eric

[raydoc]

I use an Insteon current sensor to see if the pump is running. I have the Insteon "water sensors" (wireless open/close sensor with a wire and some contacts that trigger the event when they get wet and short).

[raydoc]

So, I think I have my 2842 open/close module enrolled with the ISY. It seems to get latched to whatever state it goes to  not being momentary like I thought it would be. 

Also, I've read through this thread but I am not sure I that I follow all of your changes. Is there a new "roll up" program that incorporates what you think is the current state of the art?

[EricK]

Ray doc,

There are some issues with the leak sensors that the programs in this thread take care of. I have considered the home made water sensor from the open close like you have.

I think you just need program like this:

If pump sensor is off

And water sensor is on

Then send notification

Make a custom notification "check pump, water level is high"

I am not sure if your open close sensor has a heartbeat function. If yes then you can incorporate some of the programs and variables in this thread to monitor it.

Regarding the "latching" I think there are some options. One would be to then disable the program in the then, after the notification. A second program would then be needed to enable the program.

If status sensor is off

Then enable program.

 

Another option might be to add a wait after the notification. Hopefully one of the real experts will chime in.

Eric

[stusviews]

So, I think I have my 2842 open/close module enrolled with the ISY. It seems to get latched to whatever state it goes to  not being momentary like I thought it would be. 

Also, I've read through this thread but I am not sure I that I follow all of your changes. Is there a new "roll up" program that incorporates what you think is the current state of the art?

 

A 2842 is  Motion Sensor. Do you mean a 2843 (Open/Close Module) which does not have a latching nor a momentary mode or a 2450 (I/O Linc) which does? Or something else?

[hart2hart]

EricK, et al.,

 

This has been bugging me for a while (why the VC 3 program would run and set the $Leak variable to 3). I think I figured it out, and have a proposed solution.

 

Here's the scenario for how it happens:

 

1. Leak sensor goes offline for 25 hours

2. The VC2 program sets $Leak = 2 and then stops running

3. Sometime later, a heartbeat signal is received

4. The VC2 program now runs and sets $Leak = 1

5. ** Here's the problem: In this instant, the $Leak variable = 1, BUT the "last run time" of VC2 hasn't been updated, so all the conditions of VC3 are met and it runs, setting the variable to 3

6. So now, the $Leak variable is stuck in 3

 

 

So, here's my proposed fix:

 

Add a 5 second wait at the top of all VC3 programs:

Then        Wait 5 seconds        $sLeakKitchenSink  = 3

This should allow enough time for the "Last Run Time" to be updated, once the leak sensor comes back online and the VC3 program won't run.

 

 

If this works, please let me know and I'll update the original post.

 

 

Thanks,

 

- Brian

 

Brian:

 

Thanks for well though out suite of Leak Sensor programs that account for leak sensor quirks. I looked at original post and did not see this update. Did you decide it worked?

 

Thanks,

 

Paul

Edited December 8, 2014 by hart2hart

[EricK]

Brian:

Thanks for well though out suite of Leak Sensor programs that account for leak sensor quirks. I looked at original post and did not see this update. Did you decide it worked?

Thanks,

Paul

So far I have not had any other problems with my leak sensors without the wait added in.

Eric

[stusviews]

Ditto.

[MikeD]

I also did not add the Wait to my VC3 programs and have not had a problem although I just should add the Wait since it could not hurt. With 5 leak sensors around the house everything has been running great. Heartbeats have been received so I know all the sensors are talking. I also can vouch for how this set of programs keep you informed of a leak. I have an area that is prone to get wet due to a small crack in the foundation wall and with all the rain we had this past Saturday the leak programs came into play. During the night one of my leak sensors triggered and I was notified via email of the wet condition. I received a follow-up email the next morning at 9 AM. Everything was back to normal once the area dried up. This was the first time I received a 'real world' notification since implementing these programs a year ago.

 

Thank you again Brian for sharing.

 

~Mike

[belias]

Hi All.  Glad the programs are working well, and apologies for my delay in getting back to this thread.  I have updated the original program post to include the 5 second wait in the Variable Control 3 program.  Thanks everyone for the feedback on this.

 

- Brian

[bmah]

Brian, just wanted to say (along with many other people) thanks for your work on this set of programs and for making them available to the community.  I can't report much yet because I just set them up for my deployment but so far so good.

 

Best,

 

Bruce.