WIth Iris shutting down and a lot of older "V1" devices becomming available dirt cheap, I've decided to explore how these work in hopes to find a way to get these to work with Hubitat. To get the project rolling @vjv was kind enough to donate a V1 SmartPlug to the cause.
I dug out my old Iris hub once again (I'm going to miss that) and sniffed the converation as I paired the device. There is some scattered documentation out there on others who've managed to get these devices to work on XBees. I compared my results to what others had done and confirmed a few things...
V1 devices use a different Zigbee profile. Hubitat should be able to handle that.
The SmartPlugs use completely different clusers for everything. Still not an issue.
AlertMe uses a custom Zigbee Key. That's a show stopper!
It means there's nothing that anyone outside of Hubitat devs can do to support these devices natively.
But there are other ways to tackle this. I was able to play around with some of the Python code samples that exist. They all were lacking complete information. One of the code samples I used even had a script to configure the XBee. Something must have been missing from it as the XBee took the configuration but never saw the plug.
I switched to another set of published code, using the XBee as it was left configured by the previous script. Still no joy. That code included in the documentation all of the settings that needed to be set. I checked, verified, and tested. No joy.
After about a half hour I decided to go through settings one by one, and low and behond, the coordinator enable (CE) setting was disabled. Flipped that on and immediately got the SmartPlug paired to my XBee which was now acting as a coordinator.
I was able to turn the plug on and off and get power reports to the XBee. That's a long ways from a usable integration, but the initial tests are highly encouraging.
The next step is to take the test code and build it into something more robust with queuing, error handling, and device management. The reference code is Python, one of my weakest languages. I also most decide if I want to keep using Python (which has some great 3rd party XBee libs) or re-write in something else.
Sure, I’ll take it. I do not have any of those around here. PM me and we can work out the details.[quote="trinaandandy, post:4, topic:10798, full:true"]
Why cant the Hubitat Hub use the Iris hub as a go between in a manner that Hubitat talks to Insteon Hub for their protocol?
[/quote]
That would be awesome if it were possible. The Iris hub is locked down and relies on a connection to the Iris cloud, which is shutting down March 31st. Lowe’s is supposedly releasing Iris as open source code, but I would be very, very, surprised if that includes anything that runs on the hub. I hope I am wrong.
Yes it is. There are still a lot of former Iris customers who have not bought into a system, or have picked one and been forced to start from scratch with new and unfamiliar devices.
Too bad you do not have any. I was going to invite you to the beta. Lol
I've been working with one of the admins on the Iris forum to line up a handful of Hubitat folks who have a lot of experience with the V1 Iris devices. I think we have got enough volunteers at this point.
I have many V1 Smart Plugs and Smart Switches and would love to help out on a beta too. I'm a C# programmer and have started looking into groovy code for some possible driver development of my own. I'm still trying to get my head around it, but before I shutdown my Iris hub I used the Iris portal to capture the event communication between the hub and the Orbit 12 station Iris sprinkler. That was my original Iris raison d'etre, and even thought Iris refunded the purchase price, I'd still like to get it working. With the ease of programming in Hubitat, I figure you wouldn't even need to control the schedules in the sprinkler, just be able to turn on or off a valve via a Hubitat schedule. I can't be that difficult, he says sarcastically!
Pretty sure this is a zigbee zha profile device, as opposed to the wonky v1 alertMe profile devices that we are testing. Point is writing a driver for this should be much easier.
