New Type of Lighting Automation

This looks like a cool home automation Kickstarter project, that could potentially make home automation lighting control cheaper and better. So instead of controlling lights by changing out the switch, now you could just potentially switch out the bulb. One bulb connects to Wi-Fi and the other ones connect to it via ZigBee. Its just not real clear which bulbs have Wi-Fi or if some type of Wi-Fi to ZigBee bridge is required. Looks interesting.

http://www.kickstarter.com/projects/limemouse/lifx-the-light-bulb-reinvented?ref=live

Very cool concept, but I have the hardest time getting behind this concept of smart-light bulbs. I guess it's a cool idea to be able to partition all your lights any way you want, but how frustrating would it be to fire up your smart phone to set the mood, just to realize that you left the mechanical switch off? All it takes is a guest or a kid to hit the switches and screw everything up.

The other thought I have is the fact that, as LED lighting evolves, there's no reason to have such an easily replaceable bulb style as the Edison base. If a light is supposed to last 10+ years, why constrain ourselves to those designs? Instead I expect to see new fixtures start flooding into the market that are whole replacements - and when some LED start to fail, you'll replace the entire fixture. Maybe that technology will even evolve to where a simple little LED module can pop out.

Again - cool concept though - so time will tell where it goes. The only way I see something like this working though is if the light switch is also converted to something compatible - ideally that can access the different bulb colors, banks of bulbs, etc. Something that doesn't directly power the circuit but instead taps inline so the lights get reconfigured to be always on, and now the light-switch connects to the same mesh network to have control.

I'm tempted to order one, I really could use this, but mostly concerned about interoperability (is it ZigBee HA, or some custom stuff on top of the basic ZigBee infrastructure). This Reuters article has many people scared now, but it is full of errors.

I'm very sure its not ZigBee HA compatible. In fact I don't think it could be because HA only specifies on/off and diming, not which color. Its a good example of how even a new standard can become obsolete pretty fast.

Its amazing how technology can evolve. So now for about $50 you can not only control this bulb from anywhere in the world on your phone, you can even change what color it is. I don't know about you, but my home automation system cost much more to do that and i can't even remotely change the color. Hey, just a white LED bulb costs $25 today. But you can extend this some more. Why do you need home automation software running on a PC that is on all the time? Why can't it run in the cloud, and you access it via a web browser? The software can access your light bulb (or locks, fans and the pool pump) via the Internet and Wi-Fi. Hmmm, gets you thinking.

This looks similiar to the one Phillips was displaying at this years CES show (though that wasn't controlled via an iPhone and used its own controller). You can also see the YouTube demo of the video (pretty cool).

It's definitely interesting and I hope companies like Philips are taking notice that there is demand for this. I would assume their original plan didn't call for the official Zigbee HA stack but with all the money they have raised they should really join the alliance and use it. They also have the opportunity to make their "master bulb" a real Zigbee HA controller. I am concerned about their ability to make a high output, long-lasting, robust, RGB bulb with the resources they have, though the level of funding should give them access to additional resources. I hope they use them for a production line with high quality control, reliable components like only grade A LED modules, over-engineering the power supply, over-engineering the heatsink, and rigorous testing.

I have been wanting a light switch with a Wi-Fi stack for the longest time (controllable via simple ASCII commands), but no one wants/is allowed to build it (I think Control4 had something similar at one point, but it was proprietary).

I'm not sure if this product needs to be UL certified/tested (if it does, I'd expect that to be a show stopper, ignoring the many potential IP/patent issues). This is definitely close to what I want, and being able to pick colors based on notifications would be awesome (I currently already use the flashing lights trick to notify of certain events).

Looking forward to their detailed FAQ/specs, definitely need to see some answers.

az1324 said:

It's definitely interesting and I hope companies like Philips are taking notice that there is demand for this. I would assume their original plan didn't call for the official Zigbee HA stack but with all the money they have raised they should really join the alliance and use it. They also have the opportunity to make their "master bulb" a real Zigbee HA controller. I am concerned about their ability to make a high output, long-lasting, robust, RGB bulb with the resources they have, though the level of funding should give them access to additional resources. I hope they use them for a production line with high quality control, reliable components like only grade A LED modules, over-engineering the power supply, over-engineering the heatsink, and rigorous testing.

Click to expand...

There actually is a ZigBee standard just for controlling LED lighting, called ZigBee Light Link. The ZigBee Alliance has a webinar about it here:
http://www4.gotomeeting.com/register/890271775

Its free, but you do have to register.

Typically devices with two different profiles, eg. HA and Light Link, would not know how to communicate with each other, but a "bridge" could make this happen.

Basically my impression from the Kickstarter group is that they are generally pretty inexperienced, but they also have lots of money now and they are trying to do everything right. In addition, they have gained lots of press, especially for how fast money is coming in, and that has caused experts to reach out to them. I think they can do it, but it likely will be delivered late.

UPDATE: I wrote the developer of this bulb and asked him what would be supported. He wrote back and clarified. The bulbs do support 802.15.4 which is the physical layer that ZigBee uses, they DON'T support ZigBee. Their intention is to just release APIs for various platforms to control the lights. I don't fully agree with that approach, but I guess they didn't want to pay the ZigBee fees.

Since they are giving each bulb an IP address they are probably running 6LoWPAN and actually that's probably better in the long run than the Zigbee stack.

Smarthome recently came out with an Insteon-controlled dimmable LED bulb. It works with either power-line signaling or Insteon wireless. I got one to play with, and it works fine. The light output is supposedly equivalent to a 60-watt bulb. I shipped it off to my daughter in college, along with a wireless Insteon switch, so she could have a switch by the door to turn on something other than the awful flourescent ceiling light in her room.

I suspect it wont be long before there are wifi enabled switches. They provide a lot of information on their site and I would bet that once they get through their kickster project an investor will step in and get them going further. They have a far way to go, but given they raised $1.3mil so far you can bet their idea wont go away any time soon. I could see them putting in BT proximity into the bulbs in a version 2 of the devices, then wifi switches next.

I could really see these in the big box stores within a year from now.

Anandtech.com had an interesting HA article that mentioned there will be a forthcoming wifi standard (years away) focused on lower power for devices such as thermostats. That will definitely be interesting once the manufacturers get the associated wifi chips for such devices.

David