Recessed lighting controls


(Joepicco) #1

i currently have recessed lighting through out my home and i control them via lutron dimmable wall switches that I control with a logitech remote. i am looking at upgrading to Smartthings setup but wanted to get some advise on wall controls that are compatible to the smarthings system.


(Chrisb) #2

@joepicco

Generally speaking, if it’s a z-wave switch, then it will be compatible with SmartThings. SmartThings does not, at this time, work with any remotes that I’m aware of. The philosophy of SmartThings appears to be to automate things as much as possible so that you do not need a remote. Using motions or other sensors and programs controls to turn on or off lighting with as minimal user interfacing as possible. That said, you can always use your smartphone as a remote for nearly all your direct control needs.

Now the fine print:

  1. The Android app is still officially in Beta and doesn’t have everything that the iOS app has. In particular when talking about lighting and dimming, the Android app does not currently have the ability to control the dim level of lights. On and off works just fine for all switches (dimming as well as regular), but it can’t control the dim level right now. This will be added in a future update, we just don’t know for sure when.

  2. Beware of switches that aren’t switches. Some companies (I know Levitron, not sure about others) make a product that looks exactly like a z-wave switch, but it is really a secondary controller or 1-scene controller. Usually not an issue much but if you’re ever unsure, ask someone here. Homeseer.com is also a pretty good resource.

  3. Speaking of Lutron… they control a patent covering what information a switch can send to a controller or hub when the switch is physically pressed. In order to have their switch able to report back to the hub it’s current state (called “Instant Status”) a company has to license this patent. Some do (Levitron, Cooper, others??), some don’t (GE/Jasco, Evolve, Intermatic). For those switches that don’t there are some issues. SmartThings has found a way to get around most of the problems and a firmware updates planned in a few months should solve even more, but in short:

  • Dimmers will not report as off right away when physically turned off with the switch.
  • An Aux switch in a three way setup will not report on or off right away when physically turning on or off the AUX switch.
  • If a switch (dimmer or on/off) is not in direct communication with the Hub (ie, it’s using the mesh network), it will not report on or off right away.
  1. It is important to note that most Z-wave switches require a neutral. Many of us with old home with older wiring don’t always have neutral available to use. You may want to double check your wiring before investing too much. (I would guess with your existing switches that you’re probably be okay, but obviously that’s just a guess.) There are some dimmers (such as the GE/Jasco 45612) that do not require a neutral, but these only can control incandescent lights, not CFLs or LEDs.

Sorry if I’m overloading you with info… just trying to pass on what I’ve discovered over the last few months myself. :slight_smile:


(Coolcatiger) #3

are there any switches which flawlessly work with ST ?


(Chrisb) #4

Should work flawlessly:

I don’t own any, so I can’t say for certain, but I believe that the Levitron and Cooper Aspire switches should do everything and report instantly when turned on, off, or dimmed. These are all “associating” switch (ie, they license the patent from Lutron) so the switch can tell the hub immediately: “I’ve been turned on!” or “I’ve been dimmed to xx%.” This communication should work regardless of whether the switch is directly communicating to the hub or communicating through the mesh network.

Works flawlessly when close to the hub:

On/Off switches (not dimmers!) from “non-associating” companies (like GE/Jasoc, Evolve, Intermatic) work flawlessly when close to the hub. Here’s why: when these switches are pressed, they can’t communicate to the hub that they were pressed, but they do send out a sort of node broadcast message. ST sees this broadcast and assumes something happened, so it does a poll of the switch, reads the current state, and updates the information on the switch. BUT… this “node broadcast” is not transmitted through the mesh network. So if the switch is not directly communicating to the hub, the hub will never see the node broadcast and will not know to poll the switch to check it’s state.

Now, SmartThings does poll devices on about a 5 minute bases, so the state of the switch will eventually be updated. But this is way too slow to use for programming purposes.

(The firmware update I eluded to earlier will let the Hub initiate regular polling updates on the order of seconds instead of minutes, which make these report in very near-real time. Not Flawless, but pretty dang close. At least, that’s the hope.)

    I personally have many of these devices from GE/Jasco. The on/off switches update at very, very near real time. In fact, I have one setup where when I press a switch on, it turns on a different switch, which turns on a light. When I press the switch the light is on within less than a second.

Works when close to the Hub, but with problems:

Dimmer switches from non-associating companies (like GE/Jasoc, Evolve, Intermatic) have an added wrinkle. When in direct communication with the hub, turning them on will work the same as the on/off switches: They send a node broadcast, the hub picks that up, runs a poll, and updates status. No problems there.

But when you turn off a dimmer, it doesn’t just power off. Instead it slowly (over a second or two) dims to off. And that delay is what causes the issue. This is what happens when you hit that switch:

  1. Press button to turn off.
  2. Light starts to dim
  3. Node broadcast sent
  4. Light is dimmed to ~60%
  5. Hub see broadcast, polls switch.
  6. Light is dimmed to ~30%
  7. Switch reports that is it on (granted, it is turning off, but right now it is on… only at 30% on, but still on!)
  8. Light is dimmed to 0%

So you see the problem. Because that communication between the hub and switch is so fast (normally a good thing!) the light didn’t finish turning off until the communication was done.

As discussed above, this will be updated within 5 minutes from normal device polling, but again, makes it useless for programming purposes.

    I personally have a few of these devices. On reports in very, very near real time. But when turning off it takes a long time before the hub reports it as off.

Doesn’t work:

There are two areas there:

A.) Aux switches. Aux switches are used in three-way setups and are connected back to the master or primary switch. Physically pressing an Aux switch does NOT send a node broadcast and therefore the Hub does poll and switch isn’t updated until the normal periodic update happens. I believe that associating devices don’t have this problem, but can’t say for certain.

    I have two of these. SmartThings only updates from periodically polls.

B.) Non-associating devices not in direct communication with Hub (on/off or dimmer). As discussed above the node broadcast is not “passed on” along the mesh network so if the switch is not in direct communication with the hub the hub will not update until it does the periodic poll.

    I have four of these. SmartThings only updates from periodically polls.

I apologize again for the very long post, but as you can see it isn’t an easy yes or no answer. There’s LOTS of grey in between.


(Joepicco) #5

Chris, thanks for the information. But if my lutron wall switches work off of infrared how is it seen by the smartthings hub? Thanks for the info on the z-wave.


(Chrisb) #6

I think you might have mis-understood what I was saying in point 4 above:

4. It is important to note that most Z-wave switches require a neutral. Many of us with old home with older wiring don’t always have neutral available to use. You may want to double check your wiring before investing too much. (I would guess with your existing switches that you’re probably be okay, but obviously that’s just a guess.) There are some dimmers (such as the GE/Jasco 45612) that do not require a neutral, but these only can control incandescent lights, not CFLs or LEDs.

I’m talking specifically about wiring here, not your actual switches.

Most “dumb” switches are nothing more than a way to open or close an electrical loop. They just interrupt one wire (usually the hot or line wire). Not complicated at all. Because of this, especially in older homes, there isn’t always a neutral line in the box where the dumb switch was.

Smart switches (such as Z-wave) require a neutral**. They do more than just interrupt an electrical loop, they also use a small amount of power themselves in order to be ready to respond to wireless signals. Because they are always using electric they need a neutral in order to allow a complete loop without running power thru the lights.

Now, I suspect that your existing switches are going to be similar as they need to always be ready to receive an IR remote signal. That’s why I suspect you already have neutral for these switches meaning it will be easier to replace these with z-wave switches. But if I haven’t made that clear so far, this is only what I suspect. You’ll want to double check for sure before you start buying switches.

Chris

** And of course because nothing in the z-wave world is ever black and white: While it’s true that nearly all Z-wave switches require neutral, there are some that do not. For example, the GE/Jasco 45612 dimmer. This one does not require a neutral line to operate. However there are some draw backs. First, it can ONLY be used with incandescent lights. CFLs will flicker and flash. LEDs may flicker or flash and many not turn off all the way. Second, when it’s off, the 45612 does NOT act as a z-wave mesh repeater. (This generally won’t be a huge issue unless you use 45612s exclusively and have a large house with no other z-wave products that work as repeaters.) Third, the 45612 can be used as a 3-way, but the Auxiliary switch (GE/Jasco 45610) still requires a neutral even though the master switch doesn’t!


(ActionTiles.com co-founder Terry @ActionTiles; GitHub: @cosmicpuppy) #7

Good info, but couple points/questions:

  1. I've had decent success on a limited scale using incandescent dimmers with "dimmable" LED based bulbs. I am not sure the industry standards on this, but I'm guessing that major manufacturers that list their bulbs as "dimmable" realize that consumers will think these will work with common dimmer switches already installed in their homes. This carries through to incandescent compatible Z-Waves.
  2. Regarding the need for a "neutral" in light switches ... How did X10 get away with this? Is the neutral only required in 3-way situations (i.e., I used half-dozen cheap X10 in-wall dimmers in old home with no neutrals).** -- Oh... you addressed this in your "**" footnote: Switches feeding through incandescent are able to keep a trickle current running in order to power themselves.
  3. I wonder if there is a (safe) "hack" for the neutral problem. Really need to brainstorm on this. A battery pack -- difficult to swap out, but perfect if we can recharge it seamlessly). A trickle current** -- but not possible with non incandescent?? or ?

(Chrisb) #8

@tguchat Addressing your points:

1. (and 2.): I’ve heard this as well. As you indicated basically what’s going on with “incandescent-only dimmers” that do not have a neutral is this: Because the switch needs to have some power in order to “listen” for z-wave commands, these switches “leak” a little power through the circuit. It’s low enough that incandescent bulbs don’t light up. Because LEDs require so much less energy than incandescents that current is sometimes enough to light them up. But especially if you’re in a setup with three or four bulbs it’s sometimes enough resistance to not light up the LEDs.

3: Someone else suggested that you might use the ground line as this is pretty low amount of electricity. Obviously this isn’t up to code and there is potential for problems I’d guess, but it might work too. I certainly wouldn’t advocate it but in a pinch it’s a possibility. Unfortunately for those of us in old homes, ground isn’t always available either.

RE: Hack— What if you hacked the switch and remove it from the regular power instead powering it via a rechargeable battery? Then set it up so that when the light was one the part of the power was diverted to charge the battery. Of course, to do this means that you’d have to have a much larger than normal hole in the wall to fit all this stuff. Additionally the cost involved in doing this (time and materials to hack the switch and swap out power lines, rechargeable battery, electronics to control the recharging process) might be more expensive than having an electrician put in new wiring.