Zigbee 3.0 Outdoor Battery Power Consumption and Line of Sight Experiments

The Zigbee antenna is clearly visible on the picture where @JDRoberts suspected it.

That zigzagging line there:

2 Likes

Thanks, Gabor, I was leaning toward that being the case, but was thinking I would eventually verify that from within JD’s links.

So, now I’m left wondering if I should cut the trace before affixing an external antenna. Maybe I’ll find that info, somewhere, but I’m thinking that maybe the redundancy would be detrimental. Another thing that I’m wondering is if there is typically a positive/negative relationship with the antenna or if there is any reason to be worried about it.

Thanks, again. I’m mostly just thinking out loud, and maybe I’ll find this info while digging through the posted resources.

My experience as a field tech (before ever getting SmartThings) is that Zigbee devices almost always work best with the manufacturer-supplied antenna. There’s just no reason to add an external one in most cases, and a lot of ways to get the resonance wrong. :thinking:

The main exception is devices which are designed to have swappable antennas to allow for directional antennas, but that’s not your typical home automation use case. It’s more for outdoor sensor nets using a different profile which might be as much as a mile apart. You can’t just add additional repeaters for those, which is what you do in a residential set up.

The following article is pretty typical, I think. This person spent a lot of time and effort testing range with different antennas on a typical low power home automation Zigbee device and ended up finding that there really wasn’t very much difference and they could’ve accomplished the same thing just by adding a few more repeaters.

Also remember that the more directional your antenna, the fewer routing choices your device will have, which can cause QOS issues of a different kind.

If you want to play around with it just because you think it’s an interesting project, go for it. But you’d probably get equal or better network results just by adding a few well-placed IKEA Tradfri lightbulbs. Just sayin’… :wink:

1 Like

I found myself thinking along those lines, regarding directional antenna. Being that if I were to eventually have a larger mesh, an omni-directional antenna is the only thing that makes sense.

The Zigbee components being as inexpensive as they are, I’ll not have very much concern experimenting a little with antenna modification.

My build cost per unit looks like I’m coming in at around $30, if I end up using the LED controller option with Chinese sourced 18v (12v) 9 watt panels, the previously posted charge controllers, and surplus lithium ion 12v modem battery packs (BMS onboard). That’s if I end up 3d printing the enclosures.

But as far as testing, I have what I need, minus a good way to determine power consumption. I’m figuring on doing capacity tests on batteries and working from there.

Thanks for the info, JD. Hopefully I’ll have something to share over the weekend, if other obligations don’t take priority.

*edit: so being a little OCD, now I’m wondeing what the DC conversion is on Zigbee bulbs. Doing my best not to be side-tracked. Eh… it’s too late now… already watching tear-downs.

** Man… I don’t know how they sell those bulbs for so little, can’t get an EM357 module for price of the whole bulb.

1 Like

OK, so… a quick test with the relay connected to a USB battery pack, gave me about 200 feet from whatever Zigbee repeater it was connected to. Whichever it was, it definitely didn’t have line of sight.

So, now I suppose I need to do what I have never really looked into, which is find out what the routes are… or however that is commonly known.

I did end up moving an Ecosmart Zigbee bulb to an outiside lamp, which I had line of sight with. That’s when I figured I needed some sort of mesh map, considering I had no way of knowing if it was contributing, or what Zigbee version the relay was connected to.

I’ll go looking around, but if anyone has any advice, I’d surely be grateful.

Smartthings doesn’t give us the tools to do a full zigbee map. :disappointed_relieved: You can see the most recently used route in the ID E, which physically is usually telling you the parent.

You can also add a separate device to map the routes, but most people don’t do that.

1 Like

Any easier way… heheh…

Nah, I guess that’s a little better than the worst case, that I was expecting. Like, a Zigbee Alliance membership, or something.

It’s definitely a good ability to have.

Thanks, JD.

1 Like

Rainy day today. I suppose I could make use of the atmospheric conditions to obtain some useful data…

But… I’m not going to be that thorough. At least I’m not planning on it.

I think, for now, rather than mapping the Zigbee repeaters, I will power off all but a select few. Or maybe use another hub for testing.

The initial 200 foot range rather surprised me, I have to say. At least I have a better idea of what distance I will start testing with multiple DC powered repeaters.

1 Like

Quick question for anyone who might shed some light on the subject:

After a little searching I’m still not real sure how LQI and RSSI might aid me in determining if an added repeater is contributing, or maybe they won’t prove to be useful metrics…

  • Last Hop LQI: 255
  • Last Hop RSSI: -52

I suppose my question is, say I added a repeater, and initially it were at a distance that I would expect to improve/change one or both of the values above?

Any idea if this could prove beneficial?

Edit: OK, so, I think I know why I wasn’t seeing any changes with the RSSI, anyway…

Apparently Zigbee needs a reboot, or at least with ST hub, before it will rebuild the mesh.

I think I got that right?

Fasten your seat belt…

RSSI is a measure of the strength of signal measured by the receiver. You’d think more signal is always better, but RSSI measures all the signal, whether it’s the one you want or not. If your receiver gets signals that aren’t the one you want, say because you are getting signal splash over from other transmitters on the same or nearby frequencies, Prof. Shannon will lower your effective bandwidth as the ratio of desired signal to all signals falls

This is where the Link Quality Indicator (LQI) comes in. It gives you a measure of the messages that successfully made it across all links, not just what is going on at your receiver. This is the number that will tell you whether the information is getting through. Ultimately, messages making it across all hops is what will make your Zigbee net work, but you should look at the two measurements together - a high RSSI and low LQI might indicate too many transmitters in the area or something interfering with the channel.

There’s a good overview in this article - it’s meant for drones, but still applies here.

2 Likes

Very informative. Thank you!

I’ll see how I can make practical use of that information.

The Professor looks like he knows what he’s talking about.