I purchased an Iotawatt back in late 2024, but only got it installed back in April of this year. It worked great! for about 24 hours. I fiddled with it a bit, pulled the SD card, looked at the logs, but couldn’t get it to consistently connect to the router. So I left it hanging out of the wall and forgot about it for a few months.
I got back to it this weekend and messed around a bit. One of the things I tried was a better power adapter with a shorter cable on the off chance there was interference or something from the long one tangled up right next to the wifi antenna. Suddenly it was communicating with the router again (same room, 10’ away, clear line of sight) It worked great! for about 24 hours.
I did see that it was collecting data the whole time. I have a good 3 months of tracking in there (though Home Assistant won’t import retroactively by itself 
). I just can’t get at it.
Throwing the router logs at Claude it diagnosed a client side DHCP issue:
DHCP client wedge (root-caused via router logs, 2026-08-03/04) — after ~2.5 hours of accelerating DHCP renewals (90min → 45min → 22min → down to every 1 minute), the device stopped sending DHCPREQUEST entirely and looped on DHCPDISCOVER broadcasts once a minute, indefinitely. Router logs confirm it kept responding with a valid DHCPOFFER every time — the device just never completed the handshake. WiFi association never dropped throughout (zero disassociations, hourly WPA rekeys all succeeded), so this is not a radio/AP-side issue — it’s a client-side DHCP state-machine hang in the device’s own firmware. Because it never disassociates from WiFi, the device’s documented self-recovery (“restart if not reconnected within an hour”) never triggers.
I tried power cycling this morning for about 15 seconds before work. It didn’t reconnect to the network. It don’t see any activity from it in the router log since this morning. I suspect if I power cycle it a few more times for more than a few seconds it might come back. I’ve kicked out the DHCP lease so that if/when I do get it talking again it might last a few days rather than one, but that’s a bandaid. I could us a Home Assistant automation to power cycle it repeatedly after it stops updating, but again, that feels like a bad solution.
Is this a firmware issue I can fix with an update? Is it a bad ESP8266 I can fix with a board swap (I have a handful of ESP8266MOD units around, but they’re half the size physically, so probably not a drop in + flash. I’d be fine getting a specific model off Amazon or Mouser)?
Thanks!
No
No
I think you may be lost in the weeds. It is a WiFi issue, but probably not as complicated as you describe. Let’s start with the basics:
What is the state of the LED when you believe it is not connected?
When you succeed in getting it communicating, however briefly, can you post the message log?
Hey, thanks for responding. Finally getting time to look at things again now that it’s the weekend. When it’s not connected the external green LED is on and lightly pulsing. Maybe twice a second or so? Here’s what the router sees with regards to its connection:
After a power cycle the LED blinks “red green green” on a loop, and the router displays this:
After a few minutes the LED switches to pulsing red. Sometimes it’ll broadcast its built in wifi access point, sometimes not (or it stops before I see it).
I wasn’t able to re-connect, but I pulled the SD card, and I’ve attached the iotamsgs.txt below.
You can see when I last worked on it (8/2) and it finally connected:
> 8/02/26 14:26:41 WiFi connected. SSID=Nec_Home_Iot, IP=192.168.53.48, channel=11, RSSI -50db
Almost exactly one day later it stopped updating:
8/03/26 14:29:25 Updater: Invalid response from server. HTTPcode: -4
8/04/26 13:26:51 timeSync: No time update in last 24 hours.
I tried a reboot, which had no externally visible effect. It does eventually reconnect to the wifi, but it never gets an IP address from DHCP, and it then gives up, assigning itself the local-link address:
8/05/26 09:24:40 WiFi connected. SSID=Nec_Home_Iot, IP=169.254.152.65, channel=11, RSSI -48db
You can see the multiple reboots today with similar eventual behavior.
I’m at a bit of a loss of where to go next. I don’t know why it would fail to get an ip address when the router clearly still has it assigned the same one. Presumably some communication breakdown.
I appreciate any direction or troubleshooting advice. Thanks!
iotamsgs.txt (42.7 KB)
It looks as if there is some incompatibility or mismatch between your router and the IoTaWatt. Apparently the settings on the router have changed during the period covered by the message log. The IP address assigned lately suggests that the DHCP server is not available. There are a lot of rabbit holes that we could go down, but first I’d suggest that you try to connect the IoTaWatt to another network - a different router (take the device, power supply and reference transformer to a friend’s house) or connect to a hotspot on your phone. See if it succeeds.
When you power up the IoTaWatt you should see an IoTaWatt AP while it is blinking RGG or RRG during the three minutes after power up.
