Leak Sensor - Reflect Actual Status

[Teken]

I received another 2852-222 Leak Sensor which indicates a hardware production of 4314, Rev 2.3 and ISY firmware of v.43.

 

As others indicated this version of hardware does not seem to toggle the sensor from on to off when pressed? To confirm a few basics and expected behavior placing the sensor in a plate of water caused the device to immediately send a wet on.

 

The dry node indicated off in the Admin Console, as expected.

 

Watching the leaks sensor the LED flashed about every 15 - 17 seconds. Taking the unit out of the plate of water did not make the node change state which was also expected in the Admin Console.

 

While viewing the sensor in a dry state the LED indicator stopped flashing??

 

I have to assume that meant it was no longer wet.

 

Taking this test one step further I linked the leak sensor to a KPL status bar. Placing the sensor in a plate of water caused the expected KPL status to light up. As I reset the KPL status each time the leak sensor flashed its LED the KPL status also came on?

 

This repeated about 4-5 times and no other ON's were seen from the unit. As indicated above after drying the sensor pressing the reset button did not in fact change the nodes state from wet off, to dry on.

 

It did however continued to send another ON to the KPL status bar??

 

I am not sure why the engineers have decided to go this way but find this a little odd. I will be testing a couple more later this week and see if they react the same. Right when you thought you knew the expected behavior of how these things operate.

 

Surprise . . .

Edited August 26, 2015 by Teken

[jb642]

I received another 2852-222 Leak Sensor which indicates a hardware production of 4314, Rev 2.3 and ISY firmware of v.43.

 

As others indicated this version of hardware does not seem to toggle the sensor from on to off when pressed? To confirm a few basics and expected behavior placing the sensor in a plate of water caused the device to immediately send a wet on.

 

The dry node indicated off in the Admin Console, as expected.

 

Watching the leaks sensor the LED flashed about every 15 - 17 seconds. Taking the unit out of the plate of water did not make the node change state which was also expected in the Admin Console.

 

While viewing the sensor in a dry state the LED indicator stopped flashing??

 

I have to assume that meant it was no longer wet.

 

Taking this test one step further I linked the leak sensor to a KPL status bar. Placing the sensor in a plate of water caused the expected KPL status to light up. As I reset the KPL status each time the leak sensor flashed its LED the KPL status also came on?

 

This repeated about 4-5 times and no other ON's were seen from the unit. As indicated above after drying the sensor pressing the reset button did not in fact change the nodes state from wet off, to dry on.

 

It did however continued to send another ON to the KPL status bar??

 

I am not sure why the engineers have decided to go this way but find this a little odd. I will be testing a couple more later this week and see if they react the same. Right when you thought you knew the expected behavior of how these things operate.

 

Surprise . . .

Hi Teken -

 

Merry Christmas also!!

 

I have 4 new "leak sensors" that respond the same way....Have you heard anything about this "revision"?  My old sensors would immediately show "dry" once removed from water....

 

These new sensors continue to show "wet-ON" until i push the button on the device.  Then it goes to "wet-off" and "dry-on" returns.

 

Thanks -  John

[LeeG]

jb642

 

The v.43 Leak Sensor firmware has been around for more than a year.  As you indicate once Wet it stays Wet until physically Dry and the Set button is tapped.  Not sure why they changed the design, maybe a short Wet condition was being missed.  

 

EDIT: added Event Trace of v43 Leak Sensor showing series of Wet messages followed by Dry messages when Set button tapped to reset Wet state

V43LeakSensorWetDry.txt

Edited December 26, 2015 by LeeG

[Teken]

Hi Teken -

 

Merry Christmas also!!

 

I have 4 new "leak sensors" that respond the same way....Have you heard anything about this "revision"?  My old sensors would immediately show "dry" once removed from water....

 

These new sensors continue to show "wet-ON" until i push the button on the device.  Then it goes to "wet-off" and "dry-on" returns.

 

Thanks -  John

 

My understanding is this was done to ensure a wet condition was seen and locked in place. This would then necessitate the user to physically check the sensor and local area for a wet condition. I see value in both methods for those specific use cases so this really doesn't bother me.

 

It comes down to first knowing the expected behavior and next programming for such.

 

As the first generation leak sensors would simply flash and return to a dry state immediately upon wet detection. It also did not flash its LED in a cycle to indicate the wet condition either. Later iterations of the leak sensor were much better and those are the ones you and I have were it would return to a dry state once the wet condition resolved itself which required no user intervention.

 

The problem here is if a wet condition was restored to a dry state this incident may not be tracked by the user. So I can see value by Smartlabs in locking in the wet condition and forcing the user to physically inspect and manually reset the leak sensor from wet to dry.

 

My goal in this thread was to document a out of band behavior which was the sensor would repeatedly send a wet condition in excess of 4 times. Pressing the reset button had no effect on the unit to change it back to a dry state. As we all know battery operated devices send multiple signals when activated and gather this is done to ensure the signal is received by the Insteon network.

 

This was expected by me, but again what was not expected, is that it was broadcasting well past the three times. The unit also would not reset itself when manually pressed which confused and frustrated me to no end! After hard resetting this device several times and reviewing the level 3 error logs to confirm what I saw.

 

I didn't want to take a chance this was defective unit and sent it back to Smarthome. The RMA unit I received back operates as expected and its possible the one I had on hand was simply defective or not programmed correctly from the factory.

[jb642]

Thank you LeeG and Teken......

 

My "original" 4 leak sensors showed wet the entire time the contacts were "shorted" (wet) and went back to dry once the contacts became "dry".  I was surprised to see a different action when installing 4 additional new sensors (The first 4 sensors were V2213 Rev2.0 - the other 4 are V4314 Rev2.3)

 

JB