Add motion sensor keeping three way toggle switch

[gar]

160202-1926 EST

jasonsim:

I standby my original comment in post #2. I don't really know the precise specifictions of what you want to do.

You assume I know what these specifications are based on your limited description.

Presently you have some room with some lights that are not a large power load, and these lights are controlled by two 3-way switches. If the lights are left on and the room becomes vacant for some period of time the lights are to be turned off.

I believe what you want to do is manually be able to turn the lights on or off from either switch position, and if the room becomes vacant and the lights have been left on, then following a time period automatically turn the lights off. This auto off function senses every possible location in the room including both switch locations. Both of the present 3-way switches are to be retained. Possibly more than one motion sensor is to be allowed. Cost to be less than $100. This would mean the existing wiring is to be unchanged except some added series inhibiting function.

This does not tell me what is to happen if the switches are left in the on state, the sensor times out, and I return to the room. I will assume that if I return that the sensor(s) will detect me, and turn the lights on. But this can't happen if the sensors don't see the areas where the switches are located.

So my question is what parts of my above assumptions are in your specification?

Some more later on at least one motion sensor.

.

 

[GoldDigger]

I would hazard a guess that most of us did not think about putting the sensor ahead of the switch. To easy.

See post #4....

 

[gar]

160202-2119 EST

GoldDigger:

Your #4:

It is simple to wire a motion detector in series at either the load end or the other dead end of the switch leg (not between the two switches where there are two traveler wires.
The result will be that as long as the motion detector stays triggered the light will be controlled manually on and off using any of the three way switches. And when the motion detector times out the switches will not be able to turn on the lights, period.

I don't think that it is that simple.

Is the function that you describe what the OP wants? I don't know because there is not a complete description of the desired logic. The desired logic has to include how the motion sensors cover the area of interest.

If the logic is such that on timeout of the sensor(s) that you can not use the switches to turn on the lights, then how do you get the lights turned on? This will depend upon how the sensor(s) work and what action (person moving) is required to activate the sensor(s) if they can even be activated.

What happens is very much dependent on how a particular sensor works.

.

 

[GoldDigger]

160202-2119 EST

GoldDigger:

Your #4: I don't think that it is that simple.

Is the function that you describe what the OP wants? I don't know because there is not a complete description of the desired logic. The desired logic has to include how the motion sensors cover the area of interest.

If the logic is such that on timeout of the sensor(s) that you can not use the switches to turn on the lights, then how do you get the lights turned on? This will depend upon how the sensor(s) work and what action (person moving) is required to activate the sensor(s) if they can even be activated.

What happens is very much dependent on how a particular sensor works.

.

It may not be what the OP wants, but it does respond to "I wonder how many users would think of putting the sensor before the switches."

 

[gar]

160202-2144 EST

Some partial comments on how a particular motion sensor logically works.

The sensor is a Lutron MS-OPS5MH-AL. This is a single pole or 3-way device.

Somewhere it is stated that a neutral is not required. Superficially this would imply a two wire device. But it is really a three wire device, meaning it needs a neutral connection. The neutral connection is achieved via the green wire, EGC, connection to the EGC at the switch box. In the instructions it states the sensor will not operate without "grounding" the green wire. This can be proven by experiment that connection of the grenn wire to either the EGC or neuttral is required for the sensor to work. This direct or indirect neutral connection is required to provide power to the electronics. Just the series path thru the switched hot wires will not operate the electronics.

Is this requirement for a direct or indirect connection to neutral common to all motion sensors or not I don't know. Good design I think should require it.

This particular sensor has an electro-mechanical SPDT relay that I believe is bi-stable. I have not verified this. The unit that I bought today is defective. There is no significant continunity from the blue wire to the associated relay pole. Upon opening the sensor and testing various points it appears there is an open circuit between the blue wire solder point and the relay pole, somewhere in a multilayer board. The relay itself does not appear to be defective. Since I opened the unit the warranty is void, but there was no other way to determine the cause. The low information homeowner that would have bought this unit would have had a long day trying to figure out why his device did not work. I have verified that the relay is mechanically bi-stable. In either relay state I can power-down and then power-up with no relay change in state. The relay itself is the binary memory.

In 3-way mode of operation this unit must be paired with a mechanical 3-way switch per the instructions. Not sure of the reason at the moment.

It may be possible to operate this sensor in its 3-way mode with another manual 3-way, and then bridge all of the 3-way network with additional SPST motion sensors in parallel. This would be a strainge logical operation.

We need a definition of the required logic.

At a higher cost than $100 I might base a system on a GE RR series relay. This would allow for as many locations as desired for manual on-off control. Additional on control could be provided from a multitude of motion sensors, or on control could be provided only by the manual switches. Off control would be from both the manual controls and the last off motion sensor.

.

 

[Fulthrotl]

There are two doors, a switch by each door that opens to a short hallway. There is an open door way into the room.
The switches turn on all lights in the hallway area and room.

http://www.wattstopper.com/products...w-ultrasonic-ceiling-sensor.aspx#.VrGZvjYrKuo

they make this in a hallway model.
get one of them, and feed the switchleg to provide power
to the threeways as discussed.

if the switches are "off" nothing will light up when you enter
the space. i'd just use the motion sensor, and set the timout
up nice and high to prevent nuisance turn offs when the occupancy
is in use, but not getting a lot of foot traffic.

you can use a second similar model, set up for areas instead of
hallways, and both of them in parallel feeding the 3 ways
will get you there. will it be over $100? yep.

you want under $100? a two hour wind up timer feeding the
three way. $35.

cheap.
fast.
good.

pick any two.

 

[ptonsparky]

It may not be what the OP wants, but it does respond to "I wonder how many users would think of putting the sensor before the switches."

The cheap POS motion sensors I originally envisioned, that may fill the OPs budget limitations, are typically at the fixture.

 

[gar]

160203-0834 EST

jasonsim:

What purpose do your 3-way switches have once you add motion (occupancy sensors)?

.

 

[ptonsparky]

160203-0834 EST

jasonsim:

What purpose do your 3-way switches have once you add motion (occupancy sensors)?

.

Yes, why have switches at all?

 

[Strathead]

Yes, why have switches at all?

This is very common here, and when you think about it, wall sensors are built with switching in them. It would depend on the location, like in a public bathroom, hallway or large open office switching is probably not wise or needed. In a single occupant office, conference room, special use room etc. There are many reasons why a person may want the lights off even though they are in the room. Also, while not described here dual level switching that shuts off some of the lights to control level can be applicable to almost any situation.

 

[gar]

160203-1042 EST

So that I don't leave false information on how the Lutron works I have to say I don't know how the blue wire functions. It is possible the relay is not a SPDT contact arrangement, and that the blue wire is used as some sort of trigger.

I have put the unit back together and I am done playing. A Craftsman Torx 41652 17x1 1/2 driver is required to remove or insert the screws holding the cover halfs together.

.

 

[gar]

160203-1055 EST

Strathead:

Your comment is why there needs to be a clear accurate definition of the logic required of the switching system.

.

 

[sparkyrick]

160203-1042 EST

So that I don't leave false information on how the Lutron works I have to say I don't know how the blue wire functions. It is possible the relay is not a SPDT contact arrangement, and that the blue wire is used as some sort of trigger.

I have put the unit back together and I am done playing. A Craftsman Torx 41652 17x1 1/2 driver is required to remove or insert the screws holding the cover halfs together.

.

The blue wire is for three way switching. http://www.lutron.com/TechnicalDocumentLibrary/0301619b.pdf

 

[gar]

160203-1319 EST

sparkyrick:

Thanks for the link. It is somewhat more readable than the copy shipped with the sensor.

For 3-way operation the blue wire, per the instructions, gets connected one of the traveler wires. One of the two black wires gets gets connected to the other traveler wire. If the switch doesn't work, then you are instructed to switch the black wires. Why one of the black wires was not made red as the marking on the printed circuit board implies as a possibility is unknown. Why have a trial and error process?

My initial assumption was that the blue wire went to one of the two fixed contacts of a SPDT set of relay contacts. Now it looks like the relay contact set is only SPST. One other observation is that under some circumstances that with the green wire connected to neutral and 120 applied to a black wire that the sensor relay changed state when connecting a 10 megohm voltmeter between green and blue.

At the moment I don't plan to play with sensor any more.

.

 

[gar]

160203-1636 EST

Update;

Previously I did not closely look at the details for installation of the Lutron in a 3-way application. I made some incorrect assumptions.

The mechanical 3-way switch is changed to operate as a SPST switch. The traveler wire that is connected to the Lutron black output now goes to the load (alternatively to the hot supply) and to one side of the SPST wall switch. The other terminal of the SPST switch goes to the traveler to the Lutron blue wire. Thus, the blue wire is a signalling wire and not a load wire.

.

 

[Chamuit]

Lutron

Lutron

Have you taken a look at the Lutron Maestro Wireless devices? I put a set up in a 3-car garage for a guy whose kids left the lights on all the time.