Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Mel99]

I might be going crazier faster than I thought.

I'm doing a small project to automate the lights and exhaust fan in the first floor half-bath, Powder Room.  I want to turn the lights and fan on when someone walks into the room during the day and evening, and turn on just the lights from 10 PM until 6AM. 

For sake of being complete, I'm using AHP 3.228, a CM15A, a WS12A, an XPS3, an MS14A and a macro with a condition and an else statement.

Everything works - the macro, the MS and both wall switches perform as expected.

The hitch in this little git-along is the placement of the MS.  Initially, I wanted to position it so that it would work, be inconspicuous and not be activated by someone just walking by the open door (maybe I should expand the project to sense the door being open and have a voice announcement say, "alright, already, close the door").  Various wall positions were unsatisfactory for a variety of reasons.

Longish story shortened a little.  One of the options tried was to set the MS on the sink just below the mirror.  This worked but was not permanent. The next experiment was to velcro the MS to the mirror.  At that point the setup stopped working.  I could see the red LED in the MS flash, but no RF event was reported in the AHP Activity Monitor.  My test control was another MS14A placed in the room with my AHP computer controlling my desk lamp.  The second MS continued to send RF messages during all my tests which were reported in the Activity Monitor.

I did a complete reset of the first MS - took batteries out, put them back in and reset the codes with the same results.  It worked while setting on the sink but not on the mirror.

I then searched these forums and found no mention of a mirror affecting MS's.  BTW, it's a regular rear-surface bathroom type of mirror.  Is it possible the the silver (I think) mirror backing absorbs the RF signal?  Another BTW, The MS on a wooden block 1 1/2" in front of the mirror works ok.

[Brian H]

I does sound like you nailed it.
The mirror must have a metal coating and we know motion sensors on metal can be problems.
Mounting on the wood spaced it away from the metal backing.

[dave w]

door").  Various wall positions were unsatisfactory for a variety of reasons.

I then searched these forums and found no mention of a mirror affecting MS's.  BTW, it's a regular rear-surface bathroom type of mirror.  Is it possible the the silver (I think) mirror backing absorbs the RF signal? 

Yes, Motion Sensors and Slim-Line switches can be influenced when mounted on a metal surface. It may be the metal blocking the signal but I suspected the inductance of the metal cause the transmitter circuit to move off frequency enough that the X10 receiver (TM751, CM15A, RR501) to miss the signal. Regardless, you have pin pointed the problem, the metalized paint of the mirror is causing the problem. and you have already discovered the solution.

What about mounting the MS on the top of the door frame pointing down?

[Mel99]

[

What about mounting the MS on the top of the door frame pointing down?

I've spent a lot of time in that small room looking around, analyzing sight lines and comparing them with X-10's recommendations for the best angle for people passing by the MS.  And, it never ocurred to me to do that.  Thanks

[dave w]

What you are trying to do always creates a challenge because of the time delay between the moment the MS sees a body, and the moment the lights finally come ON. So since each X10 command takes nearly 1 second, this means a person could be almost 2 seconds or more into the room before the light comes ON. So the sooner the MS can see the person the better, but then that opens the situation for the lights to come ON when the person was not actually going into the room. I have the same problem using MS to turn on stairway lighting to the basement. I am down at least four stairs before the lights come on. In my situation I could use a door switch to speed the process a bit, since the stairway door is normally kept closed. But I don't think there is a good answer to using X10 to turn on lights the instant the room is entered. 

I think Dave X10  (and others) is also grappling with this, maybe he has found a good solution.

[Mel99]

Yeah, I noticed the time delay.  It's usually in the 1 to 1 1/2 seconds after the red LED flashes.  It might not be that big of a problem in this case as, typically, when I go into the room the light from the hall illuminates the bathroom some.  As I'm simultaneously closing the door and reaching for the light switch, the lights come on.  And after a 1 second delay, the macro starts the fan.

This project is a not-so-gentle hint to some in my family who leave the door open, with the hall lights providing all of sole light, and don't turn on the fan.  I have threatened to install a door closer device if this continues.  I know I can get spring loaded hinges which will swing the door toward the closed position (leaving it somewhat ajar), but I don't think it will make it close all the way and latch.

I have yet to investigate an X-10 type of door closing device and any ideas would be welcome.  But, I would want the lights to come on before the door is totally closed anyway...

Ah, a thought and I don't want to re-do all of the above.  Is there a delay when using a door switch when normally closed door is opened?  Or, if it is in the 1-2 second range would just the act of opening and entering the door eat up that delay??

And, automating the lights on the stairs to the basement is already on my list.  That door is normally closed.

Thanks,

Mel

[dave w]

Each X10 command take about 700ms-800ms, so the motion sensor transmits it's ON command from the movement, that takes approx 700ms. AHP receives and processes the command and transmits the ON command back to the bathroom light, another 700ms or so. Might as well round this up to 2 seconds.  If AHP inserts DIM commands then the total time window is even more.

One way to speed the process is to eliminate AHP from having to turn the light on. Set the motion sensor to turn on the light directly. (MS House Code, Unit Code is same as the bathroom light HCUC. That will trim almost one second from the delay. Then use AHP to follow up; dim light, turn on fan, turn off light 5 minutes after movement, etc etc. The problem with this approach is the light will always come on since you are bypassing any conditional control (daytime - nighttime, etc) that AHP could render. Also, if the person in the room quits moving the MS will time out and turn off the light since it has direct control rather than AHP.

[Mel99]

I've been looking at putting a DS10A on the door and when it opens, triggering the macro to start.  Is it possible to use the DS10A directly with
AHP and a CM15A or does it need to go through OnAlert?

thanks,

Mel

[Brian H]

Has to go through On Alert.

There is another way. The WGL V572RF32 can learn 32 X10 security devices and translate the security signal to an X10 address. Like your security sensor could trigger lets say a D1 when activated.

http://www.hometoys.com/ezine/08.06/green/wgl.htm

[dbemowsk]

Is there any reason you couldn't use a powerflash module?  I do realize that the powerflash needs an outlet and the DS10A is battery powered, but it is set with an x10 address and can trigger a macro easily without the need to go through On Alert or have to buy a WGL device.  You can also use the same type of door sensors that the DS10A uses.

Dan B.

[HA Dave]

This may be a good time to (forget X10) and use a building supply motion sensor switch. They are fast and reliable and not even expense to buy ether. I use these kind of switches in the garage and the laundry room. If you don't need X10's control in this particular use.... it might be OK to use something else.

[Mel99]

Thanks for all the great suggestions.  What I have working now is the lights and fan come on from 6 AM to 10 PM and lights only after 10 and until 6.  The rationale for using a door switch was to cut down on the delay in getting the lights going.  You know, somebody opens the door and by the time they get into the room enough time has passed to get the lights on.

My next step is to have the lights come on dimmed during the 10 PM to 6 AM time period.