Sounds like a good idea but if there are plenty of zigbee repeaters will this still have the desired effect? Also some remotes with older firmware (pre-Zigbee 3.0 update) are working fine with months of battery life. Its only the newer remotes that are a problem for me.
Don’t know… I have tried everything also and nothing worked. There was no messages being exchanged between buttons and hub, so I has a light on my mind… Perhaps buttons are being waked by some wifi noise, so I toke the chance, saw wifi channels nearby and changed hub channel to as far as I could, and guess what, it seems to be working.
If I am understanding this right it seems like 26 (25 is the max available), 15, or 11 would be the best options. My hub has defaulted to 20 and I for sure have the problem.
This question is frequently asked, and frequently answered, in the forum. In fact, there are two separate community FAQs about it.
In order to figure out what is best at your own home, you have to look at two different things.
What Wi-Fi channel is your Wi-Fi router using? This will normally be 1, 6, or 11.
What Zigbee channel is your smartthings/aeotec hub using? This will be listed in the IDE.
The frequencies for these two sets of channels overlap each other, but the channel numbers are not identical. For example, Wi-Fi channel 11 is very far away from Zigbee channel 11, and that’s a good thing. you want them to be as far apart as possible so they don’t interfere with each other. (specifically so that Wi-Fi doesn’t drown out Zigbee, Wi-Fi is a much stronger signal.)
I can confirm that the battery life problem now is extending to the IKEA battery powered blinds too.
The first time I charged these the batteries easily last 6 months… The last charge barely lasted a month.
I have moved my Wifi band to Ch11 and my Zigbee to the opposite end of the spectrum… Let see if that makes any difference. These really cool and “low cost” IKEA blinds are costing me a fortune in coil cells.
I haven’t changed zigbee or wifi channels as it took me a while to get this right when I first set the hub up, and it’s been absolutely reliable ever since, so I haven’t tried this - however I fail to understand how this could be the cause of a battery drain issue specifically affecting ikea devices only. Are we suggesting that a weak or occluded zigbee signal to ikea endpoints means that they don’t go to sleep and constantly poll ST/their parent device/whatever?
I’m certainly not, but I am not ruling out the mesh topology contributing to the problem. So if you do anything that rebuilds your mesh you might get some benefit, however temporary.
If you have weak mesh, the device will have to try messages multiple times to get them through.
If the device is mistakenly using long polling rather than short polling, each individual message will use a lot of power, which is what tends to drain the batteries.
As far as I can tell from reading on other sites, the issue with the IKEA devices is that for whatever reason they have started using long polling all the time with some zigbee stacks. Thus burning through their batteries really quickly.
A weak mesh would cause the device to have to send a lot of extra messages just trying to get through, thus exacerbating the problem.
In fact, just a weak mesh on its own can cause excessive battery use for this reason, but then we’re usually talking about six months instead of 12, not just a few days.
The long polling issue can kill things really fast.
Reports from other platforms go all the way back to December, by the way:
Sonoff Zigbee Bridge users:
There is another theory, which is if they fall off the network and then use up all their battery trying to re-join, but for some technical reasons I think this is probably less likely than the long polling, mostly because of the fact that only some firmware versions are affected. But here’s a discussion of that theory:
By the way, I don’t know of any consistently reliable workarounds for this. If you have the problem and you are still in the warranty period I would contact IKEA. I have heard of some people being sent a replacement device which did not have the problem.
Thanks for the explanation! I’m dubious it will fix the issue as I (think I) have a strong mesh (small london flat, loads of aurora a-one spotlights, no neighbours either side so relatively little competing wifi), and nothing has changed in my config for ages. Only change is the recent firmware/ZigBee stack update on the hub.
I’m willing to fiddle with the the radio channels and toss in some new batteries though and see if that helps. What do people to use to test ZigBee mesh strength, is there a commonly used utility?
The easiest thing is to just look in the IDE for the device entry and see how many messages are being lost. If most of the messages are getting through, it’s just not likely to be a mesh issue unless there’s a specific time of day when things are bad.
BTW I got this from UK support today, so fingers crossed
We are aware of the issue regarding IKEA devices batteries draining very quickly, and we’re actively working on resolving the problem. I don’t have an exact ETA for when we could see this issue resolved, but we expect it will be in the very near future.
Personally, as an interested outsider with enterprise product support experience. My totally unscientific opinion says the problem is in Ikea’s Zigbee implementation (as evidenced its not just one Ikea product, the failure conditions are the same across product using a common feature and doesnt happen with only SmartThings) if so. This wont be fixed or worked around by ST because doing so brings the ST Zigbee stack out of compliance. (AFAIK The zigbee changes that trigger this in ST were intended to bring the stack into better compliance). Given those, if I were the PM in charge of accepting fixes theres exactly Zero way id take that change without escalation. (read: a very slow process)
All that to say IF ST can work around it. (and I dont believe they can) My bet is they won’t - because they shouldn’t.
Id be leaning on Ikea for ‘speedy’ resolution, not ST.
There is a little history of SmartThings modifying their firmware out of compliance to handle devices that don’t meet compliance. I believe it was certain Fibaro devices that broke when SmartThings updated to Z-wave S2 certification due to The Fibaro devices having something wrong with their firmware that should have never passed certification. So there might be a chance the same can happen for this Zigbee anomaly.
I remember that one Jimmy, and you are of course correct that it happened. I also believe strongly that as we commonly talk to folks on the forum that aren’t technicians, it needs to be VERY MUCH understood and pointed out (for any new folks who join the thread who don’t do IT every day) to be the EXCEPTION - not the rule. This isn’t Samsung behaving badly - they’re operating EXACTLY how I would expect any support org in this situation to act. (Seriously, If I were the PM I’d slap the person for just coming to my office to even ask…) It’s just not their problem to fix. With all the OTHER problems they have to fix in the platform, and do we know they have a lot… I’d generally prefer they don’t spin wheels on stuff that’s not theirs and make sure we as end users don’t need to hard reboot hubs after firmware updates or uninstall apps when you upgrade from the store.
If they decide to do it, it won’t be fast (the Fibaro thing took MONTHS) but again - I think we should absolutely expect that they won’t and bug Ikea about it instead.
