I am having the same issues with v2 - two sets of batteries since September.
considering you have configured it properly (disabled light sensor, fixed ip) it might happen.
Today new firmware is going to be released for DWs. There is a bug causing battery drain listed In changelog. worth a try. If not work I suggest to contact support and return as faulty device
Hi guys, same problem with all three of my Shelly D/W v2. I’ve spent something like 40€ on batteries right now since July (of course, July 2020!). I tried all that I could to make batteries last longer but nothing changed (fixed IP, the router at a maximum of 3 mt away, switched off led…). I’m very upset, today last 2 batteries drained. It’s enough. I will switch off those sensors and going to another supplier.
Did you contact Shelly support?
changelog of latest firmware (1.9.x) mentions battery draining bug fixed.
I wouldn’t bet on it, but I setup one of my DWs again to check it out.
Yes, they told me to set a fixed IP address… nothing changed. 2 to 3 weeks battery life.
Same experience with the new firmware? That was only released a week ago for DW2.
I think I’ll give it one more shot with one sensor, and if it is improved I’ll try the others
Expecting 2 Shelly Door Window 2 in the upcoming days.
I hope I’ll be able to put them inside my mailbox. My Zigbee sensor had two weak signal, I hope wifi will perform better, we’ll see. Or I will have have to get an external antenna out of the maibox.
I also hope batteries will last more than a few weeks, even with outdoor cold temperatures… 
Quick update, after a week the battery of my DW2 is still at 100%. During the last 8 days it’s going a bit up and down, not sure how come, but I’m happy with it till now. Seems that it will last a bit longer than just a couple of weeks.
Temperature related perhaps?
from what I observed those battery temporary drops might be related to establishing wifi connection. Eventually it may go up after a while.
But temperature was something I was thinking about too, since those sensors are exposed to temp changes even if mounted inside.
Hi all.
I have shelly door windows sensor 2, connected in my garage. My garage network has nothing in common with my home network where my HA is. Even two different ISPs. I have ShellyForHASS and it’s working fine-ish with my other shellys, but the new device is not appearing in my devices in HA. I can connect to it only via cloud. Do i need to add something else or is it even possible to connect with cloud only(like ewelink and sonoff) .
You have to have those devices in the same network to see each other. Because the only connection from garage is through Internet there is no easy way to connect them together.
Deam. I was hoping that it works like ewelink integration: either local or via cloud. Is it pulling only the names from the cloud?
IMO neither shellly integration in HA connects to Shelly cloud. All data are retrieved from devices available in local network
So… after more than month, 83 registered window state changes, battery is still on 100%.
Another DW mounted to not so often used window, only 10 state changes, also reports 100% of battery.
It is slowly going to seem that the most recent fw really has fixed the issue with DWs battery drain.
I have received 2 Shelly Door contacts but not tested them yet.
I have also received a button1 but I’m quite disappointed with the latency when the button is not plugged in. When i press the button, I have to wait about 10 secondes to get notified by mqtt. When the button is plugged in, it’s almost instant, like for all my other Shelly modules (not battery-powered).
Has the door contact about the same latency ?
I can understand some latency due the wifi protocol for a device which goes to sleep to save battery.
So far I can’t find any advantage to these Shelly battery-powered modules over Zigbee modules for whoever laready has a zigbee network in use.
the latency doesn’t come from protocol latency but from time needed to connect wifi.
Usually it is 2-8 secs with static address set in the device.
It has not been said where the difference comes from. Shelly says that 2 secs is optimal. My battery devices connect to wifi in 8 secs regardless the distance from ap
1 Like
Thanks for the confirmation.
Yeah sorry, that’s what I meant when I said “latency due to wifi” : I meant “time needed to establish wifi connection coming from a standby state”.
I have also observed that setting a static IP in the device configuration helped a little bit, no additionnal time to get an IP through DHCP.
Same observation for the Shelly Door and Window : up to 10 seconds to get notified by MQTT.
It works for some use cases, but I can’t find any advantage over Zigbee modules.