Use this space to cheer the creator along, and talk to your fellow backers.
Have a question?
Fair enough Chris! Now that we are getting down to the home stretch I am going to try to get more frequent updates out. Working on the next update right now!
I appreciate the comments here, but would like to get an email with this information even if its in a rough form. I have about 6 projects which are in similar states, and i like to get a email one a week or so to help follow the progress.
Thanks for the information here, and i hope to see a box in the mail in the next month or two.
I planned to do a more detailed update by today but we are still collecting more information on remaining hurdles and timelines. We should still be close to the new schedule, but nonetheless we expect to be going into May. More exact updated timeline coming very soon!
As always, thank you for your continuing patience!
Hi Greg and everyone,
Sorry to leave you hanging. We just received word on Wednesday that UL approved the design changes and new test units are presently being manufactured for the physical tests. I'll do a more detailed general update tomorrow, but for now please accept my apologies for being behind on the Luminodes and slow to give updates.
Believe me, I want to get these out to everyone as soon as possible! We are getting really close and we're working really hard to finish them. We still believe end of April is possible, but it will be close and again thank you for your patience! And in particular you Greg. I recall our conversation very well and you were one of the first large supporters, thank you! I will do everything I can to make sure that you and all the Kickstarter backers have a great experience and great support with your Luminodes and starter kits as soon as we can get them to you.
Also, despite (or perhaps causing) the slow updates is that with a lot of hard work, a lot of pieces are falling into place. We plan to show a demo very soon demonstrating the power of our system with the Luminodes. We hope you will be as excited about it as we are.
Thank you again!
So where are we at? Have you submitted the modified units to UL yet? Have they approved them yet? You are two months behind now. Once you get UL approval how long to produce the switches?
Something that may not be readily apparent about the Luminode is that it is enabling technology for my home automation system integration learning platform which I have been working on for many years now. I have designed the learning platform in such a way as to integrate with a wide variety hardware, protocols and systems. So although the Luminode uses a newly inveted network protocol (which I do plan to publish), I intend to integrate my learning system running on a Rasperry Pi with many different systems such as Insteon, the Ubi, security panels, home entertainment systems as well as the SmartThings puck so that Luminodes can interact with a wide variety of devices.
Do you plan to have SmartThings certify your switches compatible with their controller? SmartThings is another Kickstarter project for Home Automation.
Hi Greg, sorry to leave you and everyone hanging. Although 3 weeks have passed we are still expecting to hear very soon about FCC and I'd rather not send two updates back to back in the event we do hear about FCC this week. Whether or not we hear back I will send an update by the end of this week since there are a couple of other new developments I want to report on... both good!
How about an update. Its been three weeks since the last. Did you pass FCC and UL testing?
This first version will only be for 120V. We are already looking into how to design international versions, but we are not far enough to let you know details such as when.
I'm from europe - I hope you'll have 230V support out from the box.
You have some excellent legitimate questions. I'll do my best to answer all of them.
As for Linux vs. Windows I see three common circumstances.
1) You have your own Windows custom or semi-custom system that you want to integrate with the Luminode. The USB dongle has drivers for Windows, Linux, and Mac OS, so integration can be straight forward on any of those OS's.
2) You already have a Windows box running 24/7 and don't want to have a second box up 24/7. Our goal is to make the Linux version light weight enough to run on a very lightweight, lowpower and inexpensive plug computer along these lines (http://www.engadget.com/2011/02/03/dreamplug-is-the-low-powered-lilliputian-pc-for-people-with-rea/) or similar making it reasonable to have it running 24/7.
3) You don't have a machine/controller running 24/7 and don't want one. The Luminodes are programmed to work together even without a separate computer. You wouldn't of course be able to access them with your phone or integrate them with other systems but they would still learn scenes and work.
As for the video playback/pause scenario in the video, my plan is to make that behavior learnable by our back end learning system. Ideally the way that would work is that your video playback device (such as an automatable DVD player, media PC or DVR) would need to be “instrumented” in some way such that the pause/play events would get observed by the back end. Then the back end would look for clusters of events. So if you pause your movie, followed by turning on your kitchen light and your bathroom light, the system would begin to recognize that cluster or grouping of events belong together and turn the video pause event into a trigger that also turns on the kitchen and bathroom lights.
There are of course a variety of ways to achieve the same behavior, such as through manually creating a macro, or eventually I plan to make a simple universal RF remote that interfaces directly with the back end system thus allowing it to learn grouping of actions.
I hope that addresses your concerns and I hope that you decide to continue to back the Luminode!
I'm a Windows user. I wouldn't want to have a separate Linux box to maintain (and which might be on 24/7). I've been considering backing out of my pledge, but If the back end will run on Windows, I'm more likely to pursue this and see what I can do with it.
Regarding the video demo: You've explained how getting off the couch is sensed. What does the back end system need to know about the DVR/DVD/Other and how does it know it? Is the Luminode system reading the IR signal from the pause button to know that the video has been paused?
A frequent scenario at my house is that we'll pause a DVD (using the remote) and go to the kitchen for a drink refill and maybe to the bathroom. I would want the actions of pausing the DVD and getting up from the couch to turn on a light in the kitchen and bathroom and then turn those off after we sit down on the couch and hit play on the remote. Could you please explain how that would be done and what hardware is required to accomplish it?
In answer to your questions:
1) We do have a prototype smart phone application, but it was not the one pictured in the video. The one in the video is an example of both what could be done and along the lines of what we plan to do.
1)a) The Luminodes do not connect directly to the Internet so some sort of gateway system or computer is necessary to access them remotely.
1)b,c,d) The dongle supports drivers for Windows, Linux, and Mac OS. However, we have developed the back end system mostly on Windows and are working on porting to Linux. We would rather support the back end exclusively on Linux for manageability of development, but will consider supporting both Linux and Windows if there is enough interest to do so.
2) Yes indeed! The internal LED is controlled via XML just as the load is. Integrating with Twine would be very straight forward.
PS- Thank you! I appreciate you holding out for the Luminode. I have firsthand experience trying to get the most out of Insteon dimmers. The protocol is odd but works for infrequent signals, and in fact I have no interest in making a sprinkler module or a multi-button controller since Insteon works fine for those. However, my learning system involves bursts of lighting commands which the Insteon protocol does not handle to my liking, resulting in noticeable delays between lights participating in the same scene.
1) In the videos you show a smart phone application, is this a simple example of what could be done, or does it actually exist?
-followup how hard is it to gain control over these remotely (myself) given - a) no always-on computer in the house, b) always-on windows machine in the house, c) windows web server in house, d) linux web server in house
2) can the internal led be controlled via xml? (you suggested traffic - but perhaps a twine sensor for temp data or other generic information?)
PS this looks great, I've been looking at Insteon because they are almost simple+cheap enough for the hobbyist - this seems to bridge that gap. (I am a hobbist level tech geek and expect to program some simple web interfaces to communicate with the devices -thanks for this)
Wayne, thanks for your great question. We made an early design decision to use a leading edge triac for simplicity and cost savings. This is a design decision I have revisited quite a bit and in hind sight I wish I would have gone with a trailing edge solution right away even with the additional cost. With that said, I do trigger the triac with long sustained pulses meaning that the Luminode does work quite well with most newer dimmable compact florescents and dimmable LED bulbs. Down the road I would like to produce a version that does trailing edge or even something more sophisticated to not only work with inductive loads but get better performance out of dimmable florescents and LEDs probably using an IGBT. But I want to do more research on it first.
Do you support both "Leading Edge (Triac)" and "Trailing edge (for Magnetic transformers)" Dimming ?
Dave, thanks for your question and your pledge! I posted an answer to your question in the FAQ section of the project. I'm pasting it here for your convenience.
From FAQ: I already have Z-Wave, Insteon, or other automated switches installed. Can Luminodes be integrated with these other lighting controls?
Yes. The wireless protocols are different, however they can co-exist and be integrated by connecting the Luminode dongle to the same machine that talks to the other switches. We already have developed Insteon drivers to translate their protocol to XML so that their devices work with our leaning software. We intend to do the same for as many other systems and protocols as we can.
And here I just sunk a bunch of money into Z-Wave dimmers and outlets, and still haven't gotten the system running!
In the interest of preserving my investment in Z-Wave, can you comment on whether or not it is planned, or even possible for there to be some way for the two technologies to interoperate?
What I really like about the NetworX security panel is that it doesn't try to do too much, its communication protocol is straight forward, and it's cheap!
As for integrating with other security panels we have not had much opportunity to explore which ones are easily integrated with and which aren't. The main thing to look for is their communication protocol, usually over an RS232 connection. If you tell me what security panel you have I could do a quick investigation for you.
It looks like my first comment response was not formatted very well, and I don't know how to replace my comment. So I reformatted the same response into the FAQ... Next comment coming shortly about security panels.
Thanks for your questions Rick! I'll start with the scenario in our video.
The beauty of our learning system is that it does what is sometimes referred to as sensor fusion. So he example in the video where the system knows whether you're on the couch involves a simple contact sensor in the couch connected to a NetworX security panel. In addition the back end system can either learn what you want it to do based on actions that tend to happen close to each other in both time and location or behaviors can be hard coded like most other systems.
So getting back to the example in the video. The way that works is that the back end system is configured to know that there is a contact sensor in the couch. It knows that the DVR or other video device as well as Luminodes are all in the same room. After observing behavior such as pausing a movie, getting up off the couch, and adjusting the lights from a Luminode switch, the system begins to recognize that these actions can be "clustered". So then the system will begin actuating multiple actions in the "cluster" when one of those actions happens. In that way the system can know to pause what you're watching and adjust the lights based on getting up from the couch if that is what you normally do.
So in that way, all the devices that generate events become a type of sensor. So not only does the couch sensor function as a sensor, but also the Luminodes and the DVR.
In the video, how does Luminode detect when the person exits and enters the room and then adjusts the lights? Are there other sensors in the room that are on the same network? It seems like it would have to be more than just an occupancy sensor to function the way it did in the video. More generally, what other sensors are available that function with Luminode?
What specific characteristics of the NetworX NX-8 system allow it to be used in the system so that I can find out if my current security system could be used at some point