Zigbee 2-1 Switch (On/Off & Dimmer) | Project New Horizon (Blue Series)

This is my understanding, but there may be a way to, “associate” them similar to how Z-Wave works.

We’re definitely in a new territory for us with ZigBee, so there’s going to be a learning curve for us and a heavy leaning on our beta testers and the community to perfect it :slight_smile:

But totally see the need to have the speed of Z-Wave Association (but in ZigBee) + all the advanced automations the Hue bridge won’t allow.

I’m going to have to dust off my Hue bridge and start playing with it as well. Should be fun!

Hopefully someone else more familiar with ZigBee can chime in.

Zigbee has groups as a part of it’s functionality. I use them frequently when I want multiple cans to do the same thing.

Not sure how control WITHIN the group works, but one switch can control multiple cans simultaneously in my current setup (I have one switch controlling 8 cans in a room and they are indistinguishable when changing).

Here is one of the controllers explaining this function: Groups | zigbee2mqtt.io

2 Likes

Also learning more ins and outs of Zigbee myself lately, but…

This should work via Touchlink, probably the closest thing in the Zigbee world to Z-Wave Association, and theoretically that should work even if the devices are on a different network (say, Hue Bulbs on their own network, with the Inovelli paired to your Zigbee hub like Hubitat, SmartThings, Home Assistant, etc.). Or at least that’s what the Zigbee2MQTT docs say: Touchlink | zigbee2mqtt.io. (I was thinknig of playing around with this for a bit in Home Assistant, but I settled on the built-in ZHA integration for that usage. I also primarily use Hubitat so don’t dabble in this much at all.)

That being said, I’ve only used Touchlink when the devices are on the same network and I just wanted to leave the hub out of the picture in case it was down. The only oddity I’ve noticed is that the device state does not always (ever?) get reported back to the hub like it does if you change it from the hub, at least not without a refresh. Only noticed this with Hue Bridge bulbs on a Hue Bridge network (with a Touchlinked remote that I think I had on the same network) and Home Assistant with an Ikea bulb on ZHA and an Ikea 5-button “steering wheel” remote (which is definitely on the same network), but there still seems to be a pattern. :slight_smile:

1 Like

Zigbee has two ways of achieving this. One that is already mentioned and covered above is Touchlink, which is designed to let you communicate between devices that aren’t on the same network which are physically close to each other. Zigbee 3.0 also has ‘binding’ which is very similar but designed to let you use the zigbee network to connect two endpoints together (this supports groups and I also believe hopping between devices). For example, I’m currently using Home Assistant with Zigbe2MQTT for control of my zigbee lights, however, I also have a binding setup between my office light switch and the office lights group. This means that there is essentially no delay and the lights will continue to work even if Home Assistant or Zigbee2MQTT is offline.

Awesome, the UK? Great to have you here!

I’m about to show my inner ignorant American… but aren’t the switches a different design in the UK (ie: more of a box than rectangle)? I’m wondering how you’d be able to fit our switches in your walls.

Anyway, in regards to the beta testing, when the time comes, if we have room, we’ll for sure add you to the list! Thanks as well for volunteering!

@Eric_Inovelli I realised this just as I posted the question :frowning: I’m looking to see if there’s still a reasonable way for me to still use these switches, as so far they’re the only switches I can find that aren’t battery powered, will communicate with a smart bulb, don’t use ‘Zigbee Green Power’ (which isn’t directly compatible with Zigbee 3.0) and supports Zigbee Binding. I don’t suppose you’d consider designing a switch compatible with UK square wall boxes would you? :wink:

The beta list is probably already full, but I’m willing to beta test–I have a Hubitat with a Hue…and my topology approach thus far has been to link all non-bulb Zigbee devices to the Hubitat (e.g. gas leak sensors, RGBGenie remotes) and only Zigbee smart bulbs (both Hue and Linkind brands) to the Hue. All with, of course, a bridged connection setup from the Hubitat to the Hue.

UPDATE: I do have 1 RGBGenie remote touchlink associated directly to one unimportant Linkind bulb. That spot I could easily use and change around as my beta-test location

Project Update: Working samples are being sent tomorrow and should be here next week! They won’t have full firmware on them, but we can at least get a feel for how they look and feel and it will be the first indication of how they stack up against our Z-Wave switches.

Still targeting end of year for release (although personally I think that’s aggressive, but we’ll see). Once the switches arrive, and are approved, UL process can start, which is 6-8 weeks (this is why I think end of year is aggressive, but we can dream!)

4 Likes

One question,

If the intent is to be chip… Matter certified. Why are you not choosing to use thread out of the gate?

I am not passing judgement just thinking about it the other day.

Post 61 in this thread:

Thanks @kreene1987, @BertABCD1234 and @JeffersonBledsoe for the ZigBee explanation!

Hopefully one day! It’s definitely in our strategy – just need some capital to expand lol

Awesome, I haven’t put together a specific list yet, but when the time comes, we’ll definitely keep you (and everyone else) in mind! It does help to have the Hue bridge :slight_smile:

All good man! Honestly, the Thread thing has been brought up before to us and we just didn’t know much about it other than it seemed like that’s what Google used with their Nest products. Nothing against it, we just really didn’t know a lot about the protocol and were more familiar with ZigBee.

From a business case, we chose ZigBee for a few reasons:

  1. It’s the largest protocol (represents about 40% of the market)
  2. It’s a decent transition between what we’re already doing vs where we want to go in the sense that Z-Wave requires a hub and is incredibly advanced, whereas ZigBee can either be used with a hub like SmartThings/Hubitat/HA, but also could be directly paired to various Alexa devices
  3. The biggest push for us to go ZigBee was our smart bulb mode and seeing there’s still a huge gap in the market, especially with Hue. We want to be the company that solves the Hue smart switch to smart bulb issue as there are so many ppl out there that want this integration.

Hope that helps?

1 Like

Maybe my confusion is in the whole thread / matter relationship. Let me reiterare I will be purchasing from the first round of products.

I am just curious from a manufacturerers perspective, and it seems like inovelli is not alone, why come out with a Zigbee product this close to matter becoming a thing when thread is already out there. I have been reminded a number of times that thread and matter are not the same thing and they fit at different levels of the ISO model. Is it because the firmware was already so far along for zigbee? I understand why inovelli started with zigbee, matter wasn’t a thing, and I am sure a lot of work went to it before we first heard of chip. From a consumer point of view it puts a bit of hesitation in the excitement about matter and the mythical smart home.

I know the current voice assistant’s from Apple and Google have a thread radios built in (I would assume but can’t remember at the moment if the current echo devices have thread built in). I also know there are a few manufacturers such as Eve that are currently using thread.

Sorry this post is so long.

I posted another reply a few minutes ago not seeing this one. I know this has been in the works longer than matter has been a thing. So if I understand correctly it is more about getting the hardware to market so you can see some returns on the (much appreciated) investment s you have made? Also, personally I think I would hate to work on a piece of firmware for so long and have it scrapped at the finish line because the market switched direction.

Oh and I wouldn’t have waisted my time on thread before chip became a thing.

1 Like

You and me both my friend lol.

It’s a great question!

I guess the best way to answer this from our perspective is to go back to late 2019 / early 2020 (ahh the good ol’ days when lead times were 12 weeks, our manufacturer loved us, and we were naive to what was to come from a supply chain standpoint).

We had just started gaining traction from our Gen 2 switches and people were loving the, “Smart Bulb Mode” (known at the time as, “disable internal relay”).

We were also in early talks with Signify (owns Philips Hue) around making an in-wall smart switch for them that would work with their Hue bulbs. They loved our Z-Wave version and wanted us to create a ZigBee version.

Unfortunately that fell through due to the pandemic hitting and they pushed out that project bc they were more interested in the Aurora switch by Lutron bc there was no wiring needed, but we agreed to talk further when they wanted an in-wall switch.

However, we saw the huge potential of creating a ZigBee switch bc there are 1000x more ZigBee bulbs than Z-Wave and we could easily convert our Z-Wave firmware into ZigBee.

Around the same time, CHIP was announced (Connected Home over IP – now Matter) and we saw Philips was a part of that project.

CHIP/Matter aside, we felt ZigBee was still the right play bc while Google and Apple do have thread (I didn’t know about Apple, thanks!) we’d be stretching into an area that seemed a little off-brand and didn’t really have a, “hook” to convince people to buy our switch. If we went Thread, our major selling point, “Smart Bulb Mode” seemed hard to explain to people who are newer to the market and likely hadn’t experienced the pain of having a smart bulb with no smart switch. Whereas the Hue users know this pain all too well – if someone turns off the Hue bulb from a light switch, it doesn’t work.

Hopefully I’m explaining this well – I’m not sure the caffeine has hit yet lol.

The interesting thing is that we actually put ZigBee on hold for 1.5 years lol. We had the idea to come out with it back in late 2019 / early 2020 when we were in talks with Hue – and we even promoted it at CES, but ultimately had to scrap it due to the pandemic and all the issues that came with it (that still are there).

When CHIP was announced, I wanted to bring this back bc I thought it was important to be on the leading edge of something that has the potential to be huge. Imagine if we were one of the first switches that came to market and all these big dogs are now fighting over exclusivity? That’s super exciting for a small company and honestly has kept me up at night for a long time. To top it off, we were told that certain ZigBee chips would be easily converted to Matter (not so much true now, but it can still be done), so it felt like a no-brainer to start with ZigBee to work with Hue and then convert everything over to CHIP/Matter when the time comes.

We finally found a manufacturer that would take us on as our current Z-Wave one had no interest in working on innovation for us (rightly so, but that’s a story for another day) and we kicked off this project with them.

The thought process was this:

  • Best case scenario – we launch a Matter switch first and the market goes crazy bc all the other major companies put out their usual white-label stuff, whereas ours looks a lot different and we also know and understand our customer (bc we’re customers too)
  • Worst case scenario – Matter fails to come to fruition and we’re left with a ZigBee switch that still fills the need of Hue users and can still tap into mass market a bit (whereas we now play in a niche, Z-Wave/Power-User market) via Amazon Echo’s that also support ZigBee

So, to hopefully summarize, we’ve always wanted to have a Hue compatible switch bc of the sheer market-size that we can tap into and the problem that we’ve already solved (smart bulb mode) is easily translated from a marketing standpoint to our new target (mass market). We originally wanted to start with 100% CHIP/Matter when the project was announced, but since being delayed, we decided to stick with ZigBee and develop Matter firmware in parallel.

Haha, all good – I saw you typing while I was typing and figured as much :slight_smile:

Loving the dialogue btw, thanks for asking your questions!

2 Likes

It’s funny,
I put my dream smart switch search on hold when you announced yours at CES.

My first smart switch was a lutron switch and hated it. I would run into my kids room in the middle of the night and hit the on button instead of the fadu up button right below it. After that a decora style switch became very important. I was looking for a Zigbee switch with a rocker like yours. When I came across your switch I was ready to reluctantly switch to z wave because you had everything but zigbee. After you announced your zigbee switch at ces I started searching for a suitable switch to use in the interim. I ended up with tp link kasa switches because of their price, variety of products and the fact that they have been rock solid compared to others such as wemo.

I really look forward to buying my first switch from you.

1 Like

On a side note, have you heard of SuperHouseTV? If you are looking at marketing to the maker community I would love to see what the two of you could come up with.

Are there even any hubs that support thread? Even if the switch was thread how would someone even integrate that into their home?

The apple homepod mini has thread, the new nest hubs and nest wifi have it. The new nano leaf devices set up a thread network as well as the eve products. Espressif also has a new esp32 that supports it.

That being said, that’s about it for the mainstream market. Everyone else will be supporting it “in the future”

This is why I asked the question trying to understand what the hesitation towards Thread given how “close” we are to Matter. And I think Eric really answered it with if matter fails they still have a zigbee switch and in today’s market that is more desirable.

1 Like

Project Update: Prototypes have arrived for both the 2-1 and Aux switch. Heck yeah!

Here are some other call-outs:

Screwless faceplates!

Slimmer Design (left is prototype, right is current Gen 2)
image

Lower profile config button

No heat sink tabs

Edit: Since this is a prototype, the metal is plain and the plastic is a generic white. The final product will match our current switches (all black). We may possibly make the black blue, but I’ll leave that up to the community on what they think looks best.

Nice work @Darwyn_Inovelli!

10 Likes

I vote blue! I’ll be mixing and matching zwave and ZigBee so having the differentiator when the plate is off I think will be helpful.

5 Likes

I’d be curious on how it will look (the blue background) with a cover plate installed. If it just looks dark (shadow) between the switch and plate, I’d be down for a blue body.

But would prefer this:

Fun fact: This is @Eric_Inovelli hand modeling as a side job trying to fund the new switches.

4 Likes

Nailpolish. Pick any colour you like, put a dot on the Zigbee ones before you install them. It will never come off.

1 Like