Let’s talk Z-Wave...


Active Member
Alright, to start this off here’s my list of hardware/software:
HAC07A Master Remote
Intermatic HA06C Switches
Intermatic HA22 USB Interface
TZ16 Z-Wave Thermostat

First off, the TZ16 apparently can’t be controlled by the Master Controller. I can add the Thermostat to the network fine, but it won’t let me assign it to a channel. Essentially I’ve spent $250 to replace a $15 thermostat that does the exact same thing with no added benefit. Since I can’t get my Master Controller to copy to CQC (more later), I’d be content to control the Thermostat with the HAC07A’s built in timer functions but I can’t even do something as simple as that.

Secondly, despite CQC supporting an older release of Z-Wave’s SDK, newer Z-Wave hardware does not seem compatible with that SDK. In this case, I can not copy the network from the HAC07A via the HA22. CQC can see the HA22 and recognizes it as a Z-Wave interface but it just isn’t able to understand this Master Controller. I’ve reset the MC, I’ve tried copying from different distances, nothing. I also tried an older ACT serial PC interface and I had the same problem.

Now I could understand this if I was running very old hardware or software. It isn’t reasonable to expect backwards compatibility with things that are years out of date but let’s be reasonable here. Z-Wave is hardly old enough to be having issues like this. This stuff has been around for what, 1-2 years at the most? Am I really expected to pay out the nose every time I buy a new device to update all my devices so that they can talk to each other? I understand the dangers of being an early adopter but this is just poor planning through and through.

In addition to the hardware not supporting each other, the barriers in updating software are absurd. $4,500 was the last price I heard for the full blown Z-Wave SDK. Does Zensys really expect to make money on the SDK? Wouldn’t it be smarter to sell the SDK for under $1000 to expand the amount of software that supports Z-Wave and opening it up to a larger user base? Razor companies use the exact same strategy to get people to buy new razors in that they give away the handle for free and make their money on selling replacement heads. This isn’t rocket science, I imagine most people find a software package (CQC, Homeseer, etc.) that does what they want to do and as they expand into other areas of interest they purchase the hardware that is supported by their software package. The amount of money I’ve spent on CQC dictates that my hardware purchases hinge on what CQC supports, not vice-versa. I’m not dumping CQC to find something else that supports Z-Wave when CQC does everything I need it to with the exception of supporting Z-Wave. Give the software away for dirt cheap and the hardware will follow. No software support for Z-Wave means no hardware purchases from people in the same boat I’m in.

Ultimately, I’m just peeved off that I have all these things that I want to do and should be able to do but can’t because Z-Wave can’t talk to itself. I have a thermostat I can’t control, and software that is prohibitively expensive to upgrade.

At the rate I’m going I thinking about dumping Z-Wave altogether and going with UPB despite the price difference. After all, wouldn’t it be better to spend more on hardware that actually works than spend less on hardware that may possibly, if I’m really, really luck, work in the future? I’m just really angry at how difficult this seems to be when coming from something like X-10 that was pretty straight forward. Z-Wave was supposed to be an easier, more reliable alternative. In my case, it has been neither.

So now that I’ve got that off my chest (although I don’t feel any better), let’s hear some Z-Wave success stories. Or horror stories for that matter.
Wow. Sorry to hear about your problems.

I am using a very simple Z-Wave network right now. I have seven wired in wall dimmer switches and five plug-in dimmer modules. I also have six of the plug-in appliance modules to control loads that killed my INSTEON appliance modules. They all work great for me.

The controllers that I am using are the HA07 (x2), the HA09 (x3) and one lonely ZTH100 as the master. I also have a HA22 with some ControlThink beta software that I play with.

I have a few ZIR000 PIR detectors that I have not thoroughly played with yet.

So far, I would say I have no problems with Z-Wave (I'm the guy that also has 9 months experience with INSTEON hardware and their SDK, that baby hurt coming out).



Edit added:

I paid $49 for my ControlThink SDk, I believe that they are about $69 now, with the USB stick. I do not know how that $4,500 SDK price thing started, but I can tell you now, as an actual fact, that a very good Z-Wave SDK is almost 'ready for prime time' at about $70.
The $4500, from my understanding, is to access the actual Z-Wave protocol. The $70 SDK, again from my understanding, that is available with the USB stick is a .net black box that you can interact with but doesn't allow you to access the Z-Wave hardware directly.

In order to use that SDK with CQC a translator has to be written that sits between CQC and the .net SDK. I was reading old posts on the CQC board about this and read some back and forth between Dean and Chris and what I walked away with was that Dean isn't willing to write a half-assed driver that can't directly control the Z-Wave hardware, which to avoid costs $4500, and Chris isn't willing to (and truthfully, I don't know if this is in his power or not) to offer a break on the price of the SDK protocol.

At one point Dean even offered to write a C version of the SDK, under an NDA, in exchange for free access to the protocol. I don't know what a C version of the protocol offers that the .net doesn't but since Dean had Z-Wave hardware at the time it's obviously something that would help tremendously in getting full Z-Wave support into CQC.

I can turn my bedroom light on and off with the MC and I have my front outside light turning on and off automatically with the MC but I was doing the exact same thing with X-10, reliably, for a fraction of the cost. What's the point in making the switch to a more expensive solution that does the exact same thing?

Is the software you're using a Z-Wave Certified (i.e. Z-Wave interoperable) product? If it is (which you can tell by the Z-Wave logo and interoperability statement that is required to accompany it), then it should work for you. It may or may not be compatible with all the USB sticks out there, but when used with a compatible USB stick it would be able to talk to new hardware just as well as old. The core Z-Wave messaging and routing system is and remains fully backwards and forwards compatible.

In our development efforts, we test against hardware using 100-series (1st gen) and 200-series (2nd gen) chips, as well as hardware using the old Z-Wave emulation chips. And with battery-powered devices. And with the ACT USB controllers. And with the newer USB and SDIO controllers. And things seem to work very, very nicely.

With the Z-Wave PC SDK (and the Compact SDK for CE, and the Micro SDK), we take care of many, many requirements (such as understanding how to build a well-functioning RF device, automatically utilizing new features in newer firmware versions, understanding lots of nuances, etc.) behind the scenes. It wraps up a very expensive development effort into an object model that is very simple to use, but very powerful--and is backed up by lots of testing in real-world installations.

All of that said, I think we may have the perfect solution for the issue at hand. Watch for announcements at EHX, and I think you will be very pleased.


Isn't your software the only z-wave certified software available at this time?

I have been talking to Micah in private and I completely understand what is problems are. CQC simply does not have the complete support for z-wave that is required to make it worth using. This is mainly due to the cost required to develope software for z-wave.

I am looking at two alternatives. The first is to gather a few CQC members together and purchase a developers kit so that Dean can write in the low level support that he prefers. The second is to use the upcoming xPL Z-Wave connector which uses your z-wave SDK the foundation. The xPL SDK is starting to look like the better option simply because of cost.

I understand the amount of time, money, and effort put forward by your company to create such a stable platform for z-wave software. I really wish you and Dean could have worked something out where he could re-write the SDK in a C++ version. It seemed like a win win situation to me but of course I only see it from the consumer side and not the buisness side.

So Micha is in the situation where he needs to decide to wait for the Official release of the ControlThink SDK so that the developer of the xPL connector can release his software or wait until tax time and go in with me to purchase the software SDK for Dean so that he can write his own software or simply dump z-wave altogether and move to UPB which seems to be a nice alternative to z-wave.

I have to admit that I would not blame him for moving to UPB givin all of his frustrations. The fact is that CQC is a fast growing professional home automation product and is now comparable in popularity to other software such as HomeSeer. It just seems to me that it would make more since for manfactures of hardware devices such as ControlThink/Intermatic USB Stick to want to assist low budget developers in supporting their hardware.

It is obvious that Micha has not even considered droping CQC and moving to other software. CQC is one of the best applications available for just about everything except z-wave. It just baffles me why you would not be willing to work with Dean (an genius software developer) and allow him to write a C++ version of the SDK for you at no cost to you and he would even turn the rights back over to you making it your product.

Z-Wave is fully backward compatible. We (HomeSeer) have been doing Z-Wave since ACT first put products out on the market. What you do not get if you have an older controller is the ability to do some of the new things, but operating devices works just fine, including thermostats.

The firmware that Zensys updates which is paired with the different versions of the hardware is where new features are added, such as the ability to force the network to rediscover its neighbors - useful if you have a node that has gone bad and you don't like the extra delay waiting for re-routing to happen. Another useful feature is the ability for a master controller to be replicated, and then the copy to become the new master controller. There are other features, but it is not important in this discussion...

The SDK gives you access to hardware as well as software, and the hardware aids software development as well as hardware development if you are a hardware person. If you only seek to support the various interfaces out there, then you only need to have your software understand the data in the various messages that are sent and received by the interface. In other words, the lack of support for a thermostat is likely a software issue, not a hardware issue. There is a "standard" API for interfacing with the Z-Wave interface, and the only additional layer needed is knowledge of how to work with a USB interface if that is what you are trying to support, but beyond that the RS-232 and USB interfaces are the same. What is about to happen, however, is that Leviton (and perhaps this is what ControlThink is doing) is about to come out with an interface that speaks a proprietary language, so you can add support for their interface using their API rather than the Zensys one.

So, I obviously cannot speak for Dean, but we have supported Z-Wave for a long time and the products are truly backward compatible. I have 44 nodes in my home and love the speed that Z-Wave operates at. With our new Z-Troller interface, it gets even better because you no longer have to have a separate master controller and secondary interface controller - they are one in the same.

I am also sorry there has been a bad experience with Z-Wave, but I felt I should defend the technology a little. There were some bad hardware products initially that gave it a bad rap, but the technology is pretty sound and is improving rapidly with feedback from manufacturers and users - much more rapidly than 30-year-old X-10 ever did! Hardware such as motion detectors that only gave you 10 seconds to run between the source and destination node when doing an association are fully and completely attributable to bad research and design by the manufacturer, not the technology. (And this issue was resolved by the manufacturer.)

I hope this information helps!
There’s a lot of good information in here that I’m sure is very helpful to people who understand these things more than me. Unfortunately, I am not experienced nor interested enough in understanding how all this works to really dig into what has been said.

I don’t know if CQC is Z-Wave certified. I know there is a driver for it but it doesn’t work with the HAC07A Master Controller.

This isn’t just a matter of Z-Wave not working with CQC, however. For the low, low price of $4,500 we could get Z-Wave to work with CQC. We can also wait for the xPL SDK to finish and then we’ll have support then.

This also has to do with my having spent $250 on a thermostat I can’t control with my HAC07A remote. All the hardware I have is, as far as I know, close to the newest hardware available yet the only thing I can do for all the money I’ve spent is turn my lights on and off.

Out of curiosity, I went hunting for the UPB SDK and was pleasantly surprised to find there is no SDK but the entire protocol is, from what I can tell, distributed for free. Convince me that spending $2,225 on the SDK is better spent than spending $2,225 on UPB hardware and trying to develop my own driver.

In regard to Z-Wave support for Homeseer, that is very good and I imagine that Homeseer users are very happy. Unfortunately that does me no good in regard to controlling my thermostat with my Master Controller or controlling my lighting and thermostat with CQC.

Chris, I appreciate what you’re doing. And while I’m curious as to what will happen at EHX but I’m not interested in future promises of wonderful things to come. It’s crude, but a favorite saying of mine is, “Wish in one hand and crap in the other. Tell me which gets full first.†Whatever announcement is going to be made at EHX does nothing to help me now. I’m not ready to jump ship just yet, in fact I’ve ordered a new ACT dimmer switch to see if I like it better than Intermatics (which is a whole different issue) but if something doesn’t change between now and when I’m ready to buy more hardware then this switch will probably be the last piece of Z-Wave hardware I purchase.
I did not fully appreciate (read) what you wrote earlier, so I did miss the crux of the matter for you.

Yes, devices added later in the Z-Wave life such as thermostats must be supported by the device you want to use to control them, such as the master controller in your case. Just to clarify the backward compatibility bit I mentioned earlier - the master controller (as was mentioned) can be used to add the thermostat node to the network, and that allows the node to talk with the other nodes and be a router in the network - all good - but it does not (obviously) take care of making sure the controller knows what it is and how to speak with it.

ACT's ZTH100 controller will support 1 thermostat, but if you have more than one then (at this point in time) you can use any controller to add it to the network, but something like HomeSeer would be the only way to control multiple thermostats.

I am not surprised that the UPB protocol is free, but does that include the various UPB interfaces as well? Either way, it is free because Marshall Lester was going up against X-10, Z-Wave, and some other technologies when he came up with it, and so it was in his best interest to make it free to try to get it used out there, and that is what has happened. Marshall is making money licensing the technology, Zensys is licensing/selling the Z-Wave chips - big difference!
I think Chris and his SDK are a good product and I use it for my purposes quite well. Its also very cheap and opened the door for Zwave for anyone to use. If CQC cannot afford to support ZWave its their problem. If you want to donate money to CQC go ahead. I have no need for the CQC software or Homeseer and am happy with Elk M1 and the SDK to control a few lights around the house.

I would like to know what this xPL SDK and what is so great about it. By the way, the .NET stuff will be built into Windows Vista so if your a developer like the CQC people and you cannot get over your C++ soapbox about how much tighter and better the code is it is time to start. If CQC was running on embedded systems with little processor or memory than yes lets worry about performance. I can buy a Dell computer for less money the CQC software and will perform well with any software. I think the .Net SDK from Controlthink is what others should jump on instead of buying the 4500 kit from Zensys.

One more thing (now I am on my soapbox) , we really need to start seeing these Z-wave products (the hardware) now and they need to be affordable. Too many delays will kill this product no matter who writes software to support it.

Tombo, I appreciate where you're coming from but I'm going to be very careful to tip toe around turning this into an all out flame fest.

If all I needed to do was turn lights on and off, then I could get away with just using the Master Controller and I wouldn't have a use for CQC or an Elk for that matter.

However, my intentions include home theater, weather tracking, sprinkler control, HVAC control, lighting control and who knows what else. While I could use the Elk to do most of these, CQC does all of them and I can make it look the way I want. So, yes, I will continue to "donate" to CQC.

However, CQC being unable to afford or unwilling to purchase the Z-Wave protocol is not CQC's problem. It's Z-Wave's and my problem. I've already said I'm not going to drop CQC so in essence, if Z-Wave wants to keep me and other CQC users as customers then they it's their problem.

Again, this is not just about CQC's inablility to control Z-Wave. This is also about Z-Wave's inablility to control other Z-Wave devices. My "Z-Wave" thermostat can not be controlled by my "Z-Wave Master Controller." My MC is the newest MC you can purchase, I believe, so unless it has an outdated version of the firmware (which I can't check) there's no reason why it shouldn't work.

This thread is also not just me blasting on Z-Wave, although I admit that's part of it. I'm also curious to hear what people who are happy with Z-Wave are doing with it, what their hardware is, and how they have it all setup. So starting with you, Tombo, since it sounds like you're very happy with Z-Wave, tell me about your setup.

The official v1.1 release of the Z-Wave PC SDK is up for download on our forums. I'll give the xPL guys a buzz. Also, there are a bunch of programs built on top of the Z-Wave PC SDK which are either shipping now, or will be shipping soon. They should all be Z-Wave Certified.

Regarding translating our Z-Wave foundation into C++... It was a nice offer for Dean, although unfortunately it's one of those situations where the numbers didn't add up. We can translate the software internally for less than 10% the cost of initial development--and then we would still have to QA all of it, maintain and support the code, etc. In the end, supporting .NET (and .NET CF, and .NET MF) gets us all the way from small 512k devices all the way up to mega-servers, and that seems to meet our needs nicely.

Micah, I very much understand your analogy :) The good news is that there are great options out there now, and more coming in the future.

Micah said:
However, my intentions include home theater, weather tracking, sprinkler control, HVAC control, lighting control and who knows what else. While I could use the Elk to do most of these, CQC does all of them and I can make it look the way I want. So, yes, I will continue to "donate" to CQC.

I hear what your saying about the more complex pieces. Elk M1 for me controls many things well but it does not have a good HMI or graphics engine to tell you status and more of what is going on. Especially for things like sprinklers or weather or temperatures. The touchscreen panel should have more of a user configured Graphic setup. I guess CQC does well in this area of Elks weakness. But does CQC really need to have this Zwave control?

I use Elk connected to my Thermostat (serial comm) . I like Zwave for many things, I just never saw the need for a wireless thermostat. I would like Zwave to keep on doing lighting and outlets etc but also have some of the capabilites that the GE Caddx wireless offers. I would like to add alot more motion sensors, electronic keys, electric dead bolts but there is nothing for these zwave controlled devices yet. Its nice to have the outside lights turn on when someone presses the front door bell or opens the garage door. Flashing the lights when the alarm goes off is also nice or turning all lights on when a smoke detector goes off is great too.

I guess alot of people do more with home theatre than I do. I either select an HD pay per view, sport event, or soon pop an HD DVD into my xbox 360 (soon). I never quite understood what automation does for someone here. (I also have a media center pc and it seems adequate for music and photos).

I would recommend to use zwave as a simple turn on/off/dim and motion/sensor detection device for now. The remotes are not really made for complex devices. The thermostat is the one that really screws up the basic device setup scenario and it is causing problems for many users. Zwave has worked for me but as others have mentioned it is not perfect yet. The self-healing routing is very important for proper operation.
Tombo, you raise a very good point when you ask whether or not CQC needs to support Z-Wave.

There is an actual solution that would work and that would be to purchase and install and Elk panel. CQC already has support for the Elk and can tell the Elk to manipulate the Z-Wave modules.

However, I would like to keep my install as simple as possible. CQC already has the ability to do everything the Elk can do in terms of scheduling events, reacting to changes in the enviroment, etc. To me, adding an Elk panel (while I admit would probably be more reliable than a PC over time) is a ~$400 investment in a system that's redundant to what I already have. CQC could do it, if it had a Z-Wave driver. What's standing in the way of getting that driver is either time, money or both.

I went with the Z-Wave thermostat because if I can get full Z-Wave support in CQC, I should be able to control any Z-Wave product that comes out in the future that conforms to that version of the protocol, no matter what it was. One of the reasons I went with Z-Wave was because of the expandibility of the Z-Wave technology. Because it's wireless I could see in the not so distant future things like battery operated Z-Wave curtain controllers, temp sensors, etc. which all conform to the same communication protocol simplifying the whole design of HA systems.

If all I was looking for was a protocol for lighting control, then UPB would probably be my first choice. The switches are sexy and extremely flexible. However, I wanted to avoid the, "I'm using X protocol for lighting, Y protocol for sensors, Z protocol for HVAC" situation by deciding on a single protocol that could potentially do everything.
While we're on the topic of zwave issues. The Elk Z-wave module is one of those 'early hardware units with problems' with no resolution yet. I would not recommend going with Zwave if you plan on using an Elk M1G to control your PLC network. Some day I am hoping....

The SDK chris's team has put together is very easy to use but does not give you direct access to the lower layers of the protocol you are probably craving for when experiencing problems (I don't think was their goal). Chris at ControlThink and the homeseer crew have been nothing but extremely helpful when I was first diagnosing problems with my Elk zwave controller.
I was really hoping to hear from more than just 1 or 2 satisfied z-wave users despite the negative opening post.

Unfortunately I'm going to add to the negativity a bit more. I've decided I'm finished with z-wave. I came home from a 3-week trip that was a whole lot of no fun to find 1 of my 2 installed switches has died. It's off and won't turn on or respond to any commands, either via the remote or physical paddle pushes. I will probably replace it with the switch I ordered last week but that will be the last z-wave piece of hardware I purchase.

Best of luck to Chris and the z-wave team in the future. You've got a great idea and there's a hell of a lot of potential, I just hope you can get past this rocky start and get these issues ironed out.