What is the best practice approach for this type of install?

[Amigo]

Good evening,

 

So I have a rectangular building, 90' wide and 180' long. About 22 rooms. Each room with get one Z-wave thermostat, powered.

I will use 6 repeaters, 1 at each end corner of the building and 2 in the middle.

ISY will sit somewhere in the IT room roughly in the middle of the building.

 

Would the best practice be to locate the repeaters in the final location first and add them into ISY then add the thermostats so they get the closest repeater as a neighbour?

 

Or should I just install one device at the time whether repeater or tstat and add it as I go?

 

I guess what my intended goal is to be sure that each thermostat tries to route through their closest repeater first.

 

cheers.

 

[mwester]

Start with the ones nearest the ISY, and add them as you go along, running a "heal" on the z-wave network after each device is added.

 

You have no control over the routing - the mesh network will figure out the best routing.  The only thing you can do is run a "heal" with makes the devices broadcast to each other to refresh their knowledge of who can see who -- and from that info, they will determine how to route.

[Teken]

Every time I read about how the Z-Wave routing and heal method operates I just have to shake my head. I've been known to bag on Insteon from time to time but in this case Insteons protocol method to communicate via simulcast is superior in every way.

 

The whole inclusion, exclusion, heal, routing seems like 1980 tech.

[mwester]

As we've discussed before, there are pros and cons to each.  The zwave approach cannot be dismissed as "1980's tech" - besides the fact that the mesh network principles are rather newer than that, the disparaging connotation is simply wrong.

 

As for the apparent superiority of the Insteon approach -- well, the incessant discussions about bridging and noise and rf-capable devices on extension cords is an indication that it too is not so perfect.  For instance, the limited number of hops is a major factor in its usefulness in an RF environment -- but the inverse is true for a power-line environment!  Quite a problem for Insteon.  And then there's the dreaded "All On" problem, which is claimed to happen because of the way Insteon handles collisions -- again, something that doesn't make the protocol seem too "superior" at all...

 

But all this is a tangent from the thread.  The z-wave devices do need a "heal" run to find their neighbors -- think of it, if you prefer, Teken, as a "software" equivalent of running all over the house pulling breakers, unplugging things, and swapping CFLs, LEDs, etc when you add a new Insteon device!  I know which of the two I prefer!

[Teken]

As we've discussed before, there are pros and cons to each.  The zwave approach cannot be dismissed as "1980's tech" - besides the fact that the mesh network principles are rather newer than that, the disparaging connotation is simply wrong.

 

As for the apparent superiority of the Insteon approach -- well, the incessant discussions about bridging and noise and rf-capable devices on extension cords is an indication that it too is not so perfect.  For instance, the limited number of hops is a major factor in its usefulness in an RF environment -- but the inverse is true for a power-line environment!  Quite a problem for Insteon.  And then there's the dreaded "All On" problem, which is claimed to happen because of the way Insteon handles collisions -- again, something that doesn't make the protocol seem too "superior" at all...

 

But all this is a tangent from the thread.  The z-wave devices do need a "heal" run to find their neighbors -- think of it, if you prefer, Teken, as a "software" equivalent of running all over the house pulling breakers, unplugging things, and swapping CFLs, LEDs, etc when you add a new Insteon device!  I know which of the two I prefer!

 

Maybe 1980 was over stated perhaps 1990 was a more accurate description? 

 

LOL . . .

 

Maybe spending a whole week end helping a friend set up his Z-Wave only home made me a little jaded. Watching him for hours trying to heal, include, exclude the various devices along with having no ability to see basic status of a light switch was pretty horrible to watch and endure.

 

Well besides the cheap feel and look of some of these devices . . .

 

Adding to his stress of trying to add a single lock set into his Z-Wave network only for me to tell him he actually had to bring it close to the controller before it could be done pretty much made his blood boil.

 

LOL . . .

 

I can fully understand why the Z-Wave camp has embraced and moved past the 1990 Z-Wave protocol and ushered in the Z-Wave Plus because the current RF range and reliability is just piss poor. Couple with the fact not one single vendor follows the spec as designed or called out.

 

I guess being pressed to explain why a $60.XX controller like the Wink / Smarthings which support multi channel support vs the 994 Series Controller which doesn't. Isn't something that is well known to people who purchase the Z-Wave module.

 

This is why he called me in because he knew I have been tracking the 994 for a long time and could offer some insight. I can't honestly say last week end left him with a good impression of HA in general never mind Z-Wave.

 

Having to explain to someone that a item needs to be told how to navigate and identify a route to reach a final destination? Maybe I have been pampered with Insteon over the last seven years and seeing other methods require so much work just shocks me.

 

With respect to power line issues I agree having to tell someone to unplug half their house or turn off all the breakers to find a noise maker / signal sucker is quite the hard sell. I for see Smartlabs increasing the RF output in all hardware as years go on. Because this will allow their devices to be operated in a RF only mode should it be required.

 

Thus far they have much work to do in this area but they have started the process so that is a good sign. I guess seeing random sirens thrown around a room just to couple a small house just doesn't make sense to me.

 

I don't believe a 1400 square foot bungalow could be considered big by any stretch of the imagination to be honest. If a lock can't be operated from 7 feet from where the controller is that doesn't lend to the whole RF is better.

 

In the end after schooling my friend on the subtle points of Z-Wave and how it works with the 994 controller after a grueling two day install. Its safe to say his experience with Z-Wave is typical of what I've seen over the last two years.

 

HA in general still needs lots of work and improvements in both Insteon and Z-Wave.

[stusviews]

For instance, the limited number of hops is a major factor in its usefulness in an RF environment -- but the inverse is true for a power-line environment!

 

That makes no sense at all. The Insteon protocol, whether RF or power line is identical, including the number of hops. Hops, BTW, is not related to how many devices repeat the message nor how many devices the message passes through. Hops it the number of times that the message is retransmitted until it's ACKed by the original sending device

 

And then there's the dreaded "All On" problem, which is claimed to happen because of the way Insteon handles collisions -- again, something that doesn't make the protocol seem too "superior" at all...

 

Only ISY users have reported the ALL ON phenomenon, and a small number of users at that, so it's not an Insteon trait, per se. Although both Insteon and Z-Wave require repeaters, most Insteon devices are repeaters as well as having another useful function.

 

There are areas where one excels when compared with the other, as well as devices that compete on equal footing. I prefer Insteon devices primarily because they were the original replacement devices for my X10 modules. Dual-band devices strengthened my preference. OTOH, I won't give up my 3rd generation (for me) Schlage locks--nor the Aeon Gen5 Siren required to get them to communicate reliably. An Aeotech Range Extender/Repeater was not adequate

[larryllix]

That makes no sense at all. The Insteon protocol, whether RF or power line is identical, including the number of hops. Hops, BTW, is not related to how many devices repeat the message nor how many devices the message passes through. Hops it the number of times that the message is retransmitted until it's ACKed by the original sending device

 

 

.....

The Insteon device responding with ACK has to wait until all hops are expended using the hop count otherwise a signal clash would result. The number of hops is determined by the originating Insteon device setting the max hops count in the command.

 

This is discussed in one of SH's whitepapers, if we can believe everything they say. I can't but they are just whitepapers and not technical manuals.

 

 

Also Insteon devices keep no routing information. They just repeat what they hear with a valid hops left count in the packet. There is no healing in Insteon. SH brags this as one of the Insteon advantages.

[Amigo]

Good morning,

Thanks for the input. Will start close and work my way out.

cheers.

[larryllix]

Good morning,

Thanks for the input. Will start close and work my way out.

cheers.

I have no Zwave experience but that technique sounds valid to me.

 

In addition I would NOT place repeaters in the far corners of the building. Having a device transmit to a repeater farther away from the ISY end target, and then transmitting, past the originator, to another repeater, to ISY, makes no sense.

 

I would try the ISY and see how far it reaches reliably with no repeaters first. Then cut the furthest successful distance to about 1/2 to 2/3 and place a repeater. Repeat the same process for the next repeater, at about 1/2 to 2/3 of the distance from the last placed repeater to the last successful device communicated with.

 

The last repeater will not be past the last end device (stat). It will be between the furthest thermostat and the previous repeater.

 

This may make the repeaters more accessible, eg: in a common room or hallway, rather than the far corner of a classroom, where disturbing the students is necessary, unless you like working after everybody else goes home.