Slashdot Mirror
ZigBee Pro, the New Home Automation Standard?
An anonymous reader writes "Echelon, Microsoft, Intel, Sun and the Electronic Industries Alliance have been trying to create a home automation standard for two decades — to no avail. Now the ZigBee Alliance, proprietor of a low-rate two-way wireless mesh networking technology, says it will prevail. In six weeks, automation vendor Control4, which has about one million ZigBee nodes installed, will flip the switch on the new ZigBee Pro, which promises interoperability among light switches, thermostats, door locks, motorized shades, security systems, remote controls and some 36 million electric meters."
I love the idea of home automation, then I realize that my light switch isn't that far away.
light switches, thermostats, door locks, motorized shades, security systems, remote controls and some 36 million electric meters.
But I'd really prefer if my locks remain off any kind of network and have my security system talk over good old-fashioned copper.
I've been hearing about ZigBee and Z-Wave for years. But if you look at what's out there available to you, it's crap. Poor selection, limited capability, and a high price.
Meanwhile, Smarthome and their INSTEON protocol have a broad selection of very powerful and flexible components, available today at a good price. For a DIY home-automation job, there's no contest.
Personally, I think INSTEON will become the de-facto standard that takes over from X-10. The others are just not competitive in the ways that matter.
I sound like a shill, I know. Sorry. I just like Smarthome stuff. But I wish they wouldn't embarrass me by hawking pseudo-science crap like electromagnetic water softeners.
Take off every 'ZIGBEE'!!
"Liberty may be endangered by the abuses of liberty as well as the abuses of power." -- James Madison
...ZigBee? Who the hell came up with that name?
The main problem with advanced home automation is the cost, inoperability between brands (which works into the cost since you have to buy everything from the same company) and basic problems with those networks. They mostly work in the 2.4GHz band (where the average microwave oven and just about any wireless device operates) which causes random issues with connectivity and synchronization. And then they have the most awful interfaces to program it. They mostly work in Windows and crash at random are difficult to decipher and if you're lucky enough to get a web interface you're stuck with ActiveX controls. And then if you want to make it work with other things, there is no scripting language for it.
Custom electronics and digital signage for your business: www.evcircuits.com
I like the idea of home control, lights that turn on and off, and I've been doing it with X10 for about 20 years. But I realize it has problems, poor reliability, requires neutral in the switch box for most installations, switches and outlets that actually stop functioning after 2 years, limited availability, poor selection of switch types and colors, and extremely high prices.
So Insteon comes out and solves the first problem, and nothing else. Hey, dig that light switch for $45 plus shipping! (http://www.smarthome.com/2476S/SwitchLinc-Relay-INSTEON-Remote-Control-On-Off-Switch-Non-Dimming-White/p.aspx). A standard switch costs all of $1.
And ZigBee doesn't even have interoperability on it's side? And I'm guessing we're not going to see remote switches for $1. I'd even settle for $5-10. I'm guessing the switches will cost $70. It's like they aim at the high-end of the market to get a little traction, then settle comfortably into selling $45 light switches.
It's been many years, and I guess the market isn't there, because everything we have now is overpriced and underperforming.
You were mistaken. Which is odd, since memory shouldn't be a problem for you
Does it run on Linux?
Very often, people confuse simple with simplistic. The nuance is lost on most. - Clement Mok
As someone who earns his money in industrial automation it amazes me how limited these home automation firms think. They want me to buy multiple sensors each with only one I/O point on them??? They want me to buy plastic toy-like stuff that breaks if you push the contact a few thousand times? And then there is the matter of future-proofing: in 5 years time nobody will be able to read the sensors anymore that you bought because "everybody" is on the new standard. What about spare parts for existing stuff, are they expecting me to rewire the house each time they come up with a new platform? Not a chance.
Then there is software: Windows XP, maybe with .net, was a valid choice for building the interface when the company designed it a few years ago but I expect my light switch to last at least 25 years.
These days you can run an oil refinery with a touch of a button and keep it running for 20 years with available spare parts. And you can get data in & out of that system in any format you want. Show me the same on a scaled down version for my home and I'll start installing it...
I've invested in relays and dimmers from National Control Devices and have run Cat 5e all over my apartment, even into my light switches and where I expect to put sensors in the future. It's hard-wired, hence secure and safe from interference, and speaks via simple ASCII to a serial port which available on nearly any embedded controller. The great thing about serial is that you can add a dirt cheap serial-usb or serial-ethernet interface.
I'm not really interested in a proprietary interface like Zigbee. What is needed is a HA API. That way you can write a driver for all the proprietary protocols such as this, as well as things like ProXR, Dallas 1-wire, DMX, and many more.
Some ideas for a back-end to the API can be taken from the aging Perl app Mister House. What would then be a REALLY nice addition is a MythTV module front-end so you can control the whole house via your television.
Phillip.
Property for sale in Nice, France
The nice thing about X-10 is that the protocol is simple and there are lots of devices that work with it, most of which are relatively inexpensive. It's also friendly to the home hobbyist, and the hacker, since you can buy interfaces that will hook up to your PC via a serial port and write your own scripts, or download free software like Misterhouse.
If I can't turn my outside lights on at sunset via a script, then turn that script into a cron job, don't even talk to me about it. I'll write the interface myself, just give me a clean API I can code to.
We hate it when Microsoft or Apple take the attitude of "No, we won't open up our API and play nice with the open source crowd. At best we will make you join our developer program and sign an NDA. At worst we won't talk to you at all."
When the home automation vendors do it, they're no better. They don't deserve our respect or our help.
Zigbee pro is not new. They may have a newer version but it has been around for years. The main issue with any zigbee chip is its high current draw. You need at least 55mA for receiving data and if you run a mesh network, you cannot use power saving built into the chip. Older chips are not compatible with newer versions even thought they are labeled the same, and there are numerous unexplained problems with them. Unless you are running from the mains and hence don't need to conserve power, give them a miss. Plus they run at 2.4GHz, so you get interference issues. They also don't use a standard pin pitch so you have to make a breakout board or spend extra for SIL sockets (x2). Been there, done that. Sure, if you have unlimited budget you can play with them, but they are not as good as they like to make out. Plus in the EU you are limited to 10dB output, when the pro versions are capable of 100dB.
The chips are known as XBee, the protocol is zigbee. They promise long battery life - probably true if you run off a car battery ! Try running them off AAs or PP3. You need at least 2500MaH to last a few days if they are set up to listen for data. And that includes power saving produced by hacks. Transmit I can make last for 1.5 months, if it is intermittent (ie, on an alarm condition), but the receive always has to be ready, hence the high power requirement.
Add personal vibrators (you know what I mean) to the list and we'll have a winner...
It must have been something you assimilated. . . .
I Am A Zigbee Programmer.
Some of the points being raised are a bit.. underinformed.
Interoperability: the Zigbee Cluster Library includes standard APIs for many kinds of devices, including lightswitches, HVAC, home security, etc. Devices that are certified to conform to the specification are fully interoperable. The Zigbee APIs are publicly available at zigbee.org.
Battery life: battery powered devices may last for several years if they "sleep" between transmissions. Their "parent" node in the network stores messages destined for the sleeping node. These so-called "sleepy" nodes cannot route for other nodes though, so if you have a physically large network, you'll probably want some non-sleepy devices in there running on building power. This is one of the most important features of Zigbee, and in spite of some of the other commenters here, this is actually real.
Price: this is the key reason why Zigbee will succeed: it is cheaper to retrofit a building with Zigbee devices than to knock out walls and run new wires. It's far more expensive than installing tradition switches in a new building, but that's not the a "use case." My company's clients are all looking at retrofitting HVAC systems on existing buildings and are finding some decent cost savings.
Interference: Zigbee does use the 2.4 GHz band as a lot of other devices, but there are various mechanisms (link-level acks and retries, and some other things I don't understand) built into Zigbee to mitigate this. In our tests, interference has not been an issue. Metal objects such as doors and filing cabinets have been a much bigger problem in our tests.
- Dave
as i recently tried to use Zigbee Pro in a development project and ended up throwing it away in favor of a custom 900mhz star network, let me say this... ZigBee is poorly designed for almost any application. its a 250k/bit link that at most can pass data at 20kbit, the rest is eaten up in the overhead of the mesh, theres also no support at the protocol level for transmitting information larger than 1 packet, which, if your using encryption, is 70 bytes. Zigbee should stay where it was originally designed for, industrial sensors and hvac controls, for home automation its not enough, light switches, fine...anything that has to listen and take action (window shades) is going to need power anyway and might as well use X10 at that point.
"... does a light dimmer use something like a potentiometer?"
A light dimmer uses a Triac, a semiconductor device that can be turned on at some point during one half-cycle of alternating current. If it is turned on late, the light is dim. If it is turned on earlier, the light is brighter. When the alternating current passes through zero voltage, the device turns itself off.
A Triac is a kind of Thyristor.
(Disclosure/Insight: My company, MMB Research http://mmbresearch.com/, makes ZigBee Smart Energy hardware and software to help people integrate this kind of technology into products, and I've been involved with ZigBee for a number of years.)
A lot of commenters here seem to be comparing the various features of competing home automation technologies, which is certainly appropriate, but you also have to look at the bigger, future picture.
ZigBee - and specifically the ZigBee Smart Energy profile is becoming the standard of choice for in-home networks that will exist on the Smart Grid.
So it's one thing to compare ZigBee to Z-wave or X10 on a merits basis (though I believe it's far superior based on years of real-world experience), but when you consider your utility is going to put a ZigBee Smart Energy enabled meter/gateway in your house, and that hundreds of OEMs are going to be integrating it into wide variety of appliances that can hop on that network, you're going to see drastic reductions in cost, and increases in choice and quality.
In a few years, there might be a handful of WiFi or Z-Wave thermostats (or pool pumps, or light switches), but there will be dozens of ZigBee ones, because the installed user base will be there.
Now, Control4 is talking about ZigBee Pro and the Home Automation profile, which isn't technically part of the Smart Energy profile, but they can coexist, and many developers of Smart Energy products/solutions - including ourselves - have implemented both, opening up the HAN (home area network) to a variety of devices and controls.
..and BTW I agree with you, will be to institute selling their electricity to you the same way they have to get it, at variable rates. When they have to go get some peak power juice, the costs go way up, real fast. They might just decide to pass those fees on to you with smarter meters, maybe even down to per minute pricing. You decide to run your heaviest loads during peak power costs for them, be prepared to pay a lot more for it. It could happen! Take them a bit to get it pushed through local legislatures and PUCs, but I am sure they have well trained lobbyists for these tasks.
Zigbee is a mesh networking protocol that runs on top of the 802.15.4 standard. Not an improvement, a higher level protocol. As you point out, there are better, freer, alternatives. Zigbee itself is actually quite expensive.
1.) non-critical applications: TV, light, shades
2.) critical application: pumps, heaters, etc...
3.) security items: door openers, intrusion detection, fire detection, etc...
for 1 IR control, maybe wireless possible, but not exclusively
for 2 a constant sensor feedback/monitoring by the central control system is required, with auto-shutdown in case of failure. the power-lines should be the communication medium.
for 3 only a wired system is feasible
- I am dreaming of a cheap downgraded Ethernet with a TP bus and CSMS/CD protocol at 5 or 2 Mbit allowing cheap twisted pair wiring and cheap press-fit connectors. ( If I would get advice on how to make an Ethernet interface with a FPGA I might attempt to do it myself ).
The home controller needs to be able to communicate with all media for 1,2 and 3 as well as with the local home-LAN, the INternet and maybe even get a phone connection, wireless or POTS for alarm calls.
todays Home-control is too expensive: It must cost no more than $ 10 per switched device and it must provide normal operation on the switch location - like the old manual switch additional to remote/central/automatic control. It should directly replace manual switches in the electroc home installation. X10's idea is not bad - but far too expensive and it lacks a built-in back-channel.
zigbee was fine in certain circumstances, but has largely been superceded by IPv6 over Low power WPAN aka 6lowpan Two major advantages of 6lowpan are that it is more or less regular Internet (TCP/IP) the other is that, as a result, more secure and almost infinitely more scalable.
Additionally, zigbee is not a standard, 6lowpan is. That difference has important repercussions for long term planning of projects. The IETF has a good track record for standard maintenance. There are also GPL tools for 6lowpan devices.
6lowpan is more flexible. Unlike zigbee, which is fine in some contexts, 6lowpan works with a variety of wired and wireless, low-power, low-bitrate transmissions.
The Internet is where things happen nowadays. 6lowpan is part of that.
Beta is broken and the link to classic doesn't work. Stop wasting our time or there won't be anybody left here.