Future of HA lighting

[iostream212]

So the future of lighting automation seems to be the networked bulb. So what are your thoughts regarding this advancement? For me it would replace a sizeable investment in automated switches. What are the benefits and disadvantages to this? May be perfect for a fresh install. What about for an established system? Seems like another layer of integration. Lifx and hue look great, but I can't throw away thousands of dollars in investments for colored LEDs. I guess my issue is the great enthusiasm for this technology when it seems like it is going to nullify established investments.

 

[jaymartin]

Disadvantages: change the wiring in your house so switches don't actually control the load but rather just send signals to something (the bulb or a central controller). I suppose this is practical, and it is (sorta) the way the old hard-wired systems worked, but I'm not convinced this is the future. I don't see much new construction going this direction in the near-term either, particularly given that each of these things has it's own API and ecosystem.
 
I'd be hesitant to confuse "enthusiasm" for "hype". These things are getting lots of hype at the moment - it remains to be seen if there is real enthusiasm from consumers to take this approach.

 

[wkearney99]

Agreed, without a drastic change to the electrical code and everyone's wiring you're not going to see controlled-bulbs taking anything over any time soon.
 
Besides, they just raise the cost of the bulb and the complexity of the setup.  Whereas wall switches just work and stay working, regardless of the bulb type you put in the fixture.
 
That and you'd take a hit at resale for having non-standard lighting.  Automation generally does more hard than good when it comes to selling the house later.

 

[az1324]

LIFX is working on control from standard dimmers/switches.  I also brought this up to Philips on their forum recently.  You should be able to control them with a normal switch by one or more quick flicks off/on.  A standard decora can be converted to a normally on/momentary off switch by putting a pen spring under the paddle.  You would think Leviton would start doing this at the factory when demand becomes relevant and it should not increase the cost of the switch over a standard decora ($1-2).  Of course some NEC rule will probably make an air gap or off-lock a requirement.  These colored bulbs are still a luxury item but when standard white LED bulbs with a radio are available for a few dollars more than without it will be a no brainer.

 

[Westcojack]

I also agree that this won't go far unless there is a very easy to set up app that can control rooms separately and also together. As far as changing code, there is no need, just remove the switches, connect 2 wires and install a blank plate.
 
Also, if the lighting control is done right and is very easy to use, in my opinion it make the home worth more. No large banks of switches.

 

[jhdale]

I don't see any real advantage in making the automation smarts disposable. It's a good solution for someone who is playing with a few lights, but that's about it. If you have to replace/modify your switches, it makes more sense to me to replace them with automated switches.

 

[wkearney99]

Yeah, well, I've had the 'one or two flicks' stuff for outside motion sensing spotlights and it's rubbish.  It just flies in the face of consumer expectations on how a light switch should work.  
 
It'd also be pretty hard to escape the completely valid rules about air gaps on switches.  Otherwise it's just far too easy for someone to electrocute themselves thinking the socket's dead just because the light isn't shining.
 
There is the technique of hiding dimmers in nearby locations, or centralizing all of them in one place, but it's a pretty high-end setup that's going to buy into that much effort and expense.  For them, yeah, maybe it'd be clever to have keypad control over smart bulbs.  But I doubt the market's large enough to justify the effort.
 
One clever approach we've planned to use is Lutron's hybrid keypad.  It has 6 buttons but also supports one light load (whereas past keypads where just buttons only).  Put something like this on the circuit and have it support sending RF commands to bulbs directly and you start to make progress.  The existing light circuit behaves 'normally' but now you've got physical controls that can speak to other devices wirelessly.  Right now that's other Lutron dimmers, but barring the usual proprietary nonsense (both lutron and philips) there's nothing from an engineering perspective that would prevent it from working.  Here you'd bridge between established wiring/code requirements and new tech.  Trouble is, I don't think Philips is prepared to do the switches, nor Lutron the bulbs.  And it would be highly unexpected to have either of them cooperate with the other.
 
But eventually we come back around to costs.  It just doesn't strike me as cost effective and any tangents, to put the brains in the bulb.  Sure, it's clever and all, and a few situations could make use of it, but I doubt it'll go mainstream.

 

[jhdale]

One clever approach we've planned to use is Lutron's hybrid keypad.  It has 6 buttons but also supports one light load (whereas past keypads where just buttons only).  Put something like this on the circuit and have it support sending RF commands to bulbs directly and you start to make progress.  The existing light circuit behaves 'normally' but now you've got physical controls that can speak to other devices wirelessly.  Right now that's other Lutron dimmers, but barring the usual proprietary nonsense (both lutron and philips) there's nothing from an engineering perspective that would prevent it from working.  Here you'd bridge between established wiring/code requirements and new tech.

Insteon KeypadLincs are the same way. One load, plus either 4 or 7 buttons that control other things. But that only solves half the problem: how to control devices for which there is no switch. It doesn't solve the other problem, which is dealing with the existing switch in the wiring to the existing fixture.


I should also note that, in addition to the disposable smarts issue, smart bulbs also don't make much sense for multi-bulb fixtures. Suddenly you need to buy the smarts 3-4 times per fixture instead of once.

 

[wkearney99]

Well, if you've got the physical controls capable of controlling via RF then you're half-way there.  Assuming, of course, that the various vendors agree to support something standard in this fashion.  As Yoda might say, holding my breath I am not.  
 
It can solve the existing switch problem, assuming the neutral is coming to that switch location.  Then it'd be trivial to just run that same load to the fixtures without it being switched, and have the keypad controlling them via RF.  Again, assumptions about compatibility and all that...  The device could still 'switch' the load, it would just default to it always being on.  Otherwise I suspect it'd run afoul of code issues by changing a previously switched receptacle to always-live.
 
Well, it could make some sense in a multi-bulb fixutre as a means to provide varied lighting, either by dimming or even color combinations.  Disco chandelier anyone? 
 
I'm not saying there's no place for smart bulbs, just that the ones being sold currently don't have enough going for them to work for much more than a hobbyist-sized audience.  Get some more standard integration going and that could change.

 

[az1324]

Yeah, well, I've had the 'one or two flicks' stuff for outside motion sensing spotlights and it's rubbish.  It just flies in the face of consumer expectations on how a light switch should work.  
 
It'd also be pretty hard to escape the completely valid rules about air gaps on switches.  Otherwise it's just far too easy for someone to electrocute themselves thinking the socket's dead just because the light isn't shining.
 
There is the technique of hiding dimmers in nearby locations, or centralizing all of them in one place, but it's a pretty high-end setup that's going to buy into that much effort and expense.  For them, yeah, maybe it'd be clever to have keypad control over smart bulbs.  But I doubt the market's large enough to justify the effort.
 
One clever approach we've planned to use is Lutron's hybrid keypad.  It has 6 buttons but also supports one light load (whereas past keypads where just buttons only).  Put something like this on the circuit and have it support sending RF commands to bulbs directly and you start to make progress.  The existing light circuit behaves 'normally' but now you've got physical controls that can speak to other devices wirelessly.  Right now that's other Lutron dimmers, but barring the usual proprietary nonsense (both lutron and philips) there's nothing from an engineering perspective that would prevent it from working.  Here you'd bridge between established wiring/code requirements and new tech.  Trouble is, I don't think Philips is prepared to do the switches, nor Lutron the bulbs.  And it would be highly unexpected to have either of them cooperate with the other.
 
But eventually we come back around to costs.  It just doesn't strike me as cost effective and any tangents, to put the brains in the bulb.  Sure, it's clever and all, and a few situations could make use of it, but I doubt it'll go mainstream.

 
Yeah well those motion sensor lights are cheap and dumb whereas these new smart bulbs have good microprocessors in them.  A normally on switch where the paddle always indicated up/on may get around the rules.  If not a small latch to lock the paddle down would be simple.  In large volumes smart bulbs are only marginally more expensive than dumb bulbs. 
 

Insteon KeypadLincs are the same way. One load, plus either 4 or 7 buttons that control other things. But that only solves half the problem: how to control devices for which there is no switch. It doesn't solve the other problem, which is dealing with the existing switch in the wiring to the existing fixture.


I should also note that, in addition to the disposable smarts issue, smart bulbs also don't make much sense for multi-bulb fixtures. Suddenly you need to buy the smarts 3-4 times per fixture instead of once.

 


Being able to control bulbs separately in overhead light groups is a benefit.  Large volumes make the smarts very low cost.  Over 10 years disposable is ok.
 
 
It doesn't matter what makes more sense from the user's perspective since it seems far more likely that companies are going to scale production of smart bulbs and drive the prices down faster than the price of smart switches.

 

[iostream212]

Really why automate 5 bulbs when I can automate one switch? And while the allure is strong to color my rooms with light it doesn't seen strong enough to justify the cost or headache of integration. Hue was not impressive from what I saw on YouTube. Bad lagging and took several seconds for a scene to realize. I am sure I can find a place or two for a few networked bulbs, but I agree with the consensus of the majority voiced here.

 

[NewZealandHome]

Here is my view on the past and future of Home Automated Lighting...

How did the tech get here so fast... it actually didn't happen overnight... take a look at the 5 year timespan of two industries that coincided..
Super Increased Technology + Super Decreased Price. Each one below could not have happened without the preceding tech. advancements.. reminds me of playing Civilization III 
ARM->Apple->Android->Realtek WiFi->LED smart bulbs.
LED Torches->LED Bike Lights->LED Car lights->LED panels->LED Street lights->LED tubes->LED Bulbs->LED smart bulbs.
Key companies: ARM Holdings, Apple, Google, Samsung, Realtek, SamsungLED.  (all of the latest new tech can be directly attributed to their success and is shown in their overwhelming profits)

Future of HA Lighting...
SmartThings (EM357 ARM Cortex-M3)  <- yep ARM Holdings again. so cheap, so powerful like you wouldn't believe.
LiFX / LimitlessLED Wifi Enabled Lighting (Toshiba / Realtek / SamsungLED)  <- Realtek and other new RF/Mesh technology chips, alongside new LED tech. 
Android wall light switch controllers (Full color, Full touch)

It just happens that LED and ARM and other chips run at similar voltage levels... how convenient
 
So what does that mean for the future of HA lighting?

It means that you can place a small motion sensor chip in every room.   An example is that as you get up in the morning, each chip knows that motion is moving from the bedroom outwards, and is not likely an intruder. And things start automatically, for example our coffee machine turns on, your kitchen lights turn on,  your lights in your bathroom change color to the weather report for that day.. so you know if its going to be bright and sunny or raining as you get dressed. If your motion sensors sense that you are in a good mood and dancing or whistling, then cool music will come on and start playing.  The possibilities will be limited only by our imaginations.

By the end of this year Lighting will be full color, full brightness, full cheap, and full wireless. Downlights, Tubes, Bulbs, Spotlights. Lighting will be directly controlled by 2.4Ghz RF remotes, 2.4Ghz RF coin battery operated wall switches, iPhones, Androids, Android touch panel light switches, all operating a WiFi and a mixture of Mesh and Point to Point technologies.


 

 

[NewZealandHome]

Really why automate 5 bulbs when I can automate one switch? Hue was not impressive from what I saw on YouTube. Bad lagging and took several seconds for a scene to realize.

here is the reality:  replacing a light switch is too scary for 99% of people.  the user interface of a light switch, umm hello, apple already invented the mouse, and now invented the touch UI to redefined how we interact with others.  Light switches are *NOT* the future User Interface for lighting. The future UI is geolocation presence keyring sensors, phones, smartwatches, and the lightswitch will not even be hardwired to the lights in the future. only power for rf and full color touch devices will be at the wall.  There will just be velcro-like light switches on the wall that can control anything in the home. Place them anywhere you like.. stick them to the sidetable next to your bed. even stick the light switch to the sideboard of your bed.  then you don't have to worry about where your bed is positioned in relation to the headboard and light switches.

It is definitely feasible for people upgrading to LED for savings. reality is that people will realize, why spend $10 for a non-smart LED bulb, when I can spend $16 for a smart LED bulb.

I've never used Hue, because its too expensive, and if its as bad as you say it is.  All smart LEDs I have seen, respond to any control in around about 0.001 seconds.  The LEDs even feel like they change before I've even fully pressed my finger on the iphone capacitive touch screen.

Cheers,
Hamish.
 

 

[picta]

Here is my view on the past and future of Home Automated Lighting...

By the end of this year Lighting will be full color, full brightness, full cheap, and full wireless. Downlights, Tubes, Bulbs, Spotlights. Lighting will be directly controlled by 2.4Ghz RF remotes, 2.4Ghz RF coin battery operated wall switches, iPhones, Androids, Android touch panel light switches, all operating a WiFi and a mixture of Mesh and Point to Point technologies.


 

It is good to have a dream   I would be happy if I could find a decent quality regular LED bulbs that are not regular medium base. Most LED bulbs currently on the market are for the Edison base, but majority of the light fixtures use other types of bulbs. Less than 10% of my bulbs are Edison, the rest are MR15, G9, PAR 38 and Candelabra, besides the fluorescent fixtures that were required by code. You'll be hard pressed to find even the floor and table lamps that use medium size bulbs any more. I was actually looking for one to try out the RGB remote lights and found very few options. So until the LED bulbs will be made in all types of bases and connector types, I don't see it gaining mainstream use.

 

[jhdale]

here is the reality:  replacing a light switch is too scary for 99% of people.  the user interface of a light switch, umm hello, apple already invented the mouse, and now invented the touch UI to redefined how we interact with others.  Light switches are *NOT* the future User Interface for lighting. The future UI is geolocation presence keyring sensors, phones, smartwatches, and the lightswitch will not even be hardwired to the lights in the future. only power for rf and full color touch devices will be at the wall.  There will just be velcro-like light switches on the wall that can control anything in the home. Place them anywhere you like.. stick them to the sidetable next to your bed. even stick the light switch to the sideboard of your bed.  then you don't have to worry about where your bed is positioned in relation to the headboard and light switches.

This scenario of yours is still not plug-and-play. People don't want to replace a light switch? Then this already fails because their existing switches don't work with smart bulbs.

Now suppose you design a house's electrical system around this. You have your RF switches and your smartbulbs. Sounds cool. Now the user needs to replace a bulb. Unscrew old bulb, screw in new bulb, all the switches linked to the old bulb are now non-functional. The user needs to tell the software that they have performed a replacement. The UI problem is not solved. But if you had smart switches instead of smart bulbs, everything would be fine...