My reluctant head long dive back into HA

Here I go again. I went nuts two decades ago and bought a ton of X-10 stuff and loaded up my house. The house is long since gone and my desire to automate was in remission. Living for years with a direction sensing driveway alarm tied to a pair X-10 powerflash interfaces that sounded chimes and output via a pair of universal modules to my surveillance system which then sent emails and pictures. I was quite satisfied. One by one my powerflash units died and the market to replace dried up.

I start thinking what will I do next. We often forget to close garage or shop doors so I wanted something to alert me to their status if left open too long. After a long search I opted to buy an Elk M1. Found a good deal and ordered. Then the wait. Research on what the cool new Elk might do led me here. Reading about how others are using their systems and integrating with other automation systems I got the bug again. Mind you I'm still waiting on my Elk to show up.

Today I bought a VeraLite. I think I may be out of control. Money spent and nothing to play with. And now the hunt for the inevitable bank breaking sensors, switches, panels, etc. I am going to try not to spend too much before I actually have something to work (play) with.

Update: I think I jumped the gun on the VeraLite. I was just reading where the Elk plugin never made it out of alpha and development may not move for a while if ever. The code is being offered to the public, which is likely not a good sign.
 
If you want a supplemental device to the Elk look at the Universal Devices ISY.

http://www.universal-devices.com/residential/isy994i-series/

I don't own one but I'll probably get one soon. The Elk honestly does a LOT on its own.

Without any supplemental processor/controller my elk:
- Controls two dozen lights/fans/small appliances (serial connection to insteon modem)
- Controls my HVAC (serial connection to HAI serial capable thermostat)
- Controls my Motorized Shades (relays wired/soldered to remotes - cheaper and simpler than the expensive Somfy shade controller and works flawlessly)
- Controls my garage door (wired directly to the main relay on the elk)
- Interfaces with my BlueIris based security camera network (scripts on a computer that interface through the elk's ethernet expander)

I have a very full fledged home automation system that is run virtually 100% from the Elk itself. Maybe give it a fair shot at doing what you need before you buy an additional controller.


All of that said, my Insteon lighting system can do WAY more than the Elk's code for Insteon permits, so I'll probably get then ISY. The ISY controllers talk in real time with the Elk to benefit from everything the Elk knows, but enables far more advanced combinations of lighting devices and conditions under which they act.

Just my two cents. Welcome back to one of the world's most eternal hobbies.
 
Welcome back!

It's definitely a good idea to play with what you've ordered to see what it does/doesn't do... If you want to graduate from X10, UPB is kinda the next in line - similar in that it's powerline based but much more powerful and reliable; however, UPB is lacking in RF control. For that, I use a W800RF32 (WGL Has a couple variations of X10RF to UPB; this one outputs via Serial) to let me use a couple cheapie X10RF remotes with my UPB system and I use the DIY super-antenna listed in the site's articles for massively increased range. There's Z-Wave that has seen a lot of growth but I'm not sold on the stability of the whole environment; or there's Insteon which, coupled with an ISY, seems to be able to do about anything under the sun too.

I'll echo what's said above - I personally do 95% of my automation with the Elk, which includes everything from garage door, sprinkler, HVAC, lighting, and security - and because it's all one system, anything that can talk to Elk can talk to everything else. I also have a $99 intro version of Elve - it lets me run a few more advanced rules and because of the way it inherently keeps track of UPB, the Elk sees the updates fixing one of the issues with the Elk (it can't track lighting status of lights activated by a scene). There's a lot more Elve can do, but I haven't been able to motivate myself enough to do more since the Elk does things so well.

Good luck, and if you ever need more ideas, there are a few older threads lying around where people have done everything from starting their cars, automating their chicken coop doors, intelligently tracking mousetraps, etc - all from their HA systems. The options are limited only by your budget and creativity!
 
You guys make returning the VeraLite seem like a smart move. I honestly care more about monitoring than automating at this point, so the Elk looks like a good place to focus.
 
If you're sold on Z-Wave then I think the Vera is a smart choice; otherwise with Elk and UPB I don't really see its point. I bought one when they first came out and I never even plugged it in - it's still in my closet.
 
Can anyone elaborate on what Vera or the ISY can do in terms of lighting that Elk can't? I know that my currently installed dimmers don't report their status back to the Elk, so it's tough for it to keep track of whats on/off (particularly when used manually), but I understand that's a limitation of the dimmers I have - not the controller itself.

Regarding Vera vs. Elk, I believe that Vera can be used to manage the full set of user codes for door locks (30 of them I believe). Whereas with Elk, you are really limited to 2 user codes, since they don't do anything with the user codes... but this isn't exactly lighting ;)

I've currently got a number of lighting rules setup with Elk and haven't yet found a limitation for anything I've tried to do... just looking for benefits and/or advantages of using another controller for lighting.

For those that are familiar with setting up Groups/Areas with Elk and the M1XSLZW, please see the thread I have on setting up Areas in the Elk forum (here)... I've got some questions regarding setup ;)
 
I almost bought a Vera (the first version) when Micasa Verde had refurbished ones on ebay for cheap. The problem with the Vera is it's very limited compared to what a PC can do. Further, there's no easy way to interface it with a PC as they seem to have dropped development the virtual VRC0P plug-in.

The Elk M1G by itself is not a bad way to go, but adding PC control doesn't hurt anything as the Elk can still stand alone if the PC fails. I use a free HA program called Premise that's very powerful. It can do what the "pro" version CQC or Homeseer does, but for free (and arguably in a much better IDE that's highly versatile)!

Z-wave is not a bad solution and it is widely supported if you use Leviton's VRC0P v3 serial controller, but you have to spend A LOT of time learning how to properly set up a z-wave network. It's not plug and play. Leviton offers free online courses you can register and take:
http://www.leviton.com/OA_HTML/SectionDisplay.jsp?section=38998&minisite=10251

Also there's this guide;
http://communities.leviton.com/servlet/JiveServlet/previewBody/2212-102-1-3430/App%20Note%20Vizia%20RF%2B%20system%20guide.pdf

And other documents here:
http://communities.leviton.com/community/knowledgebaseforums/home_automation/application_notes?view=documents

Basically, if you don't have 8-10 hours to spend learning about z-wave before purchasing, don't bother. The z-wave stuff is too confusing and when you buy a z-wave device, its capabilities are not clear until you actually install it! You'd think information such as z-wave classes that are supported would be required to be stated on the item you purchase, but it's not.
 
If you're using UPB, hopefully you saw my post not too long ago - where just by having Elve running and connected to my UPB network with its own PIM, my Elk accurately tracks status now - because Elve watches for scene activations and then polls affected switches - and the Elk sees the status updates too... so with a $99 app I have accurate status AND I can do more advanced rules and even touchscreens one day.

The drawback to the Vera is that it has only the most limited support for the M1 - so you can't get all trick with rules and outputs to make the two talk back and forth; otherwise Vera was built to have the best Z-Wave compatibility with every device made.

The ISY as I understand has much better integration and will handle all your insteon light setup and management and has a much more powerful rules engine than the M1.
 
drvnbysound:

You'll find that a PC is much more versatile than a Vera. With an RS232 VRC0P, you can access the same z-wave classes the Vera is using (with the exception of some proprietary stuff that is not part of the zensys z-wave specification). Premise comes with an open-source VRC0P module, so if a device comes along, you can easily add it.

You can definitely do all kinds of neat things with Premise + VRC0P + z-wave door lock. You can track who unlocked the door and perform specific actions for that particular user (turn the TV to their favorite station, greet them, etc). Further, the kwisket z-wave locks are also smart enough to report how the door was locked or unlocked:
1. User code entered (to lock or unlock)
2. via key from outside or manual interior deadbolt knob

When they report this to Premise, different actions can be performed. This is nice because you only want to disarm the system if an actual code was entered. You want to arm stay mode if the deadbolt was locked form the inside, but arm away if a usercode was entered to lock from the outside.

Further, if you want to make modifications at the driver level, it's opensource and written in vbscript. Premise Builder makes use of vbscript or click and point. There's also an SDK that supports C++, .net or Java (while Vera uses only Lua and has nothing compared to Premise Builder's easy to use interface).
 
PS: Premise also has an opensource module for the Elk supporting every feature that Elk's ascii text protocol allows. You can trigger events in Premise via Elk's keypads, display text to the keypads, control lights, etc...
 
I almost bought a Vera (the first version) when Micasa Verde had refurbished ones on ebay for cheap. The problem with the Vera is it's very limited compared to what a PC can do. Further, there's no easy way to interface it with a PC as they seem to have dropped development the virtual VRC0P plug-in.

The Elk M1G by itself is not a bad way to go, but adding PC control doesn't hurt anything as the Elk can still stand alone if the PC fails. I use a free HA program called Premise that's very powerful. It can do what the "pro" version CQC or Homeseer does, but for free (and arguably in a much better IDE that's highly versatile)!

Z-wave is not a bad solution and it is widely supported if you use Leviton's VRC0P v3 serial controller, but you have to spend A LOT of time learning how to properly set up a z-wave network. It's not plug and play. Leviton offers free online courses you can register and take:
http://www.leviton.c...&minisite=10251

Also there's this guide;
http://communities.l...ystem guide.pdf

And other documents here:
http://communities.l...?view=documents

Basically, if you don't have 8-10 hours to spend learning about z-wave before purchasing, don't bother. The z-wave stuff is too confusing and when you buy a z-wave device, its capabilities are not clear until you actually install it! You'd think information such as z-wave classes that are supported would be required to be stated on the item you purchase, but it's not.

etc6849, I do agree (to some point) about Zwave - it's certainly a bit convoluted in the way that manufacturers are able to pick and choose what parts of the Zwave protocol that they want to implement for their various devices. I feel as if they (sales and R&D guys) end up sitting in a room and actually determine what features and/or classes they are going to add... Particularly if the implementation is going to take too long and thus cost too much money, and often end up with products that fall short in terms of features/capability. However, it is nice that the protocol/alliance allows a number of manufacturers to build their own devices, thus creating competition and moving the technology along... As opposed to UPB for example, where it seems that nothing is happening, and only a few manufacturers have UPB devices.

Due to varying devices having various levels of implementation to the Zwave protocol, I would agree that it's not exactly plug-and-play. However, it's not to say that it's difficult to continue to build a Zwave network and update your controller accordingly. Since I added my Zwave network to Elk (albeit only a few months ago), I've added a few dimmers and outlets since then, and simply updated the VRCOP accordingly.

I know that I wish I would have seen this spreadsheet before I bought my Zwave dimmers. I don't know that I would have actually done anything different, since I purchased a decent number of dimmers at a fraction of their regular sales cost but at least I may have been aware of their capability or lack thereof. The big ones for me are the support of the "beaming" class as well as Instant Status. Unfortunately, none of those dimmers can do both...
 
If you're using UPB, hopefully you saw my post not too long ago - where just by having Elve running and connected to my UPB network with its own PIM, my Elk accurately tracks status now - because Elve watches for scene activations and then polls affected switches - and the Elk sees the status updates too... so with a $99 app I have accurate status AND I can do more advanced rules and even touchscreens one day.

The drawback to the Vera is that it has only the most limited support for the M1 - so you can't get all trick with rules and outputs to make the two talk back and forth; otherwise Vera was built to have the best Z-Wave compatibility with every device made.

The ISY as I understand has much better integration and will handle all your insteon light setup and management and has a much more powerful rules engine than the M1.

I did see that, and I know it's got to be nice to have correct status all the time... However, I already had a decent amount invested in Zwave... not only with lighting, but with the Kwikset locks too, and possibly a thermostat soon. At one point I was really thinking about doing a hybrid solution with UPB dimmers (for arguably better reliability and possibly easier to achieve correct device status), but I found that I was going to have to keep a Zwave network anyway for my locks (and possibly thermostat), and maintaining both was just going to be more work than I wanted to deal with.
 
No, Z-Wave is complicated!

Imagine: your kitchen dimmer's lamp burns out. Now when your PC triggers your living room light, it takes longer to come on?!?! This is confusing if you don't study things in advance and have an HA system smart enough to trigger events to tell you to change the bulb (the Premise module does).

Of course, we know from studying things that most z-wave dimmers only have control power when the lamp's filament is not burned out, and that z-wave is a mesh network and failure of one node can cause massive delays or even transmission failures for neighboring nodes. However, how many people buying the product know all this?

The second issue is if you add a node, you must perform certain steps on the primary and secondary controller. Without actual study, a person is not going to know what these steps are. Further, most manufacturers do not tell you what the steps are really doing, making things seem even more confusing.
 
drvnbysound:

You'll find that a PC is much more versatile than a Vera. With an RS232 VRC0P, you can access the same z-wave classes the Vera is using (with the exception of some proprietary stuff that is not part of the zensys z-wave specification). Premise comes with an open-source VRC0P module, so if a device comes along, you can easily add it.

You can definitely do all kinds of neat things with Premise + VRC0P + z-wave door lock. You can track who unlocked the door and perform specific actions for that particular user (turn the TV to their favorite station, greet them, etc). Further, the kwisket z-wave locks are also smart enough to report how the door was locked or unlocked:
1. User code entered (to lock or unlock)
2. via key from outside or manual interior deadbolt knob

When they report this to Premise, different actions can be performed. This is nice because you only want to disarm the system if an actual code was entered. You want to arm stay mode if the deadbolt was locked form the inside, but arm away if a usercode was entered to lock from the outside.

Further, if you want to make modifications at the driver level, it's opensource and written in vbscript. Premise Builder makes use of vbscript or click and point. There's also an SDK that supports C++, .net or Java (while Vera uses only Lua and has nothing compared to Premise Builder's easy to use interface).

Understood. I guess my hesitation to doing a PC based solution now is I really don't know how I would 'easily' use a system based on PC controls - other than A) not actually interfacing with it, but relying solely on automation rules, or B) adding another app to my phone. Additionally, while I'm adding all of these "smart" devices to be automated, I still use them manually more often than not. What Elk+Zwave+eKeypad has really given me is the ability to check alarm status and control devices remotely (e.g. from my phone). I haven't yet found/seen the benefit of adding the PC-based solution [for me] yet. Heck, I've actually got a full subscription to CQC that I've paid for, and have yet to use... honestly, because I haven't found anything that I've wanted to do yet, that I haven't been able to do with Elk alone... :unsure:
 
No, Z-Wave is complicated!

Imagine: your kitchen dimmer's lamp burns out. Now when your PC triggers your living room light, it takes longer to come on?!?! This is confusing if you don't study things in advance and have an HA system smart enough to trigger events to tell you to change the bulb (the Premise module does).

Of course, we know from studying things that most z-wave dimmers only have control power when the lamp's filament is not burned out, and that z-wave is a mesh network and failure of one node can cause massive delays or even transmission failures for neighboring nodes. However, how many people buying the product know all this?

The second issue is if you add a node, you must perform certain steps on the primary and secondary controller. Without actual study, a person is not going to know what these steps are. Further, most manufacturers do not tell you what the steps are really doing, making things seem even more confusing.

Certainly. I wasn't trying to imply that it wasn't complicated, and I agree that it takes some time (or a failure) to learn about the nuances. Fortunately, I've been in my current home for almost 6 years and I think I've replaced 2 bulbs in the entire home.. but yeah, I suppose if I started to see awkward delays, I might need to go hunt down a blown bulb.
 
Back
Top