-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Lost connectivity on wired ethernet to powerwalls?
- Thread starter arnolddeleon
- Start date
-
- Tags
- powerwall 2
There's no arguing with that. I wonder whether the button is absent, or just the hole.
Mine was intalled in May.
You only have to remove the one screw at the bottom to open the panel covering the Gateway internal components. The reset button is probably there, but your panel just doesn't have the hole.
I don't have a reset button/hole like you do. I would need to remove the dead front cover to get at the gateway (which I will eventually get to unless I get Tesla to reset it remotely). I can ping the gateway and the TCP connection happens as @Sylvia describes. This makes me optimistic that I can fix my issue.
I just remembered I had a photo of the gateway before the cover went on and there is no obvious reset button.
And no LEDs that might be visible through the hole either. Seem like it's a quite different model.
I agree that it does seem like a network problem, and when it first started happening, that's what I was investigating. However, I had absolutely no problems communicating with the gateway from any device on my network from its installation in December last year, until the 1.20 update, then I progressively started losing access over a period of a few days, firstly web access then API access then ping responses. That's when I started checking threads here and on other forums and saw that I wasn't the only one. My WiFi network has around 33 devices connected to it (3 APs in a mesh), and the Powerwall gateway is the only device I'm having trouble with. The Ubiquiti manager shows that its WiFI connection is solid. On 1.20 I still can't get a web connection using the gateway's own AP, even after the restart.
My gateway is like the ones pictured without the reset button. It does have a red flashing LED on the left of the black contol box, partially visible when the front cover plate is unscrewed and removed.
BRAKING NEWS - while I've been typing this, I had a look at the web UI and noticed that the gateway has been updated to 1.21 firmware and I can now access the web UI via the gateway's AP again. Let's hope they're rolling out a fix for this mess!
There are three potential power sources that come into the gateway box. So unless the gateway computer has its own extra battery, disconnecting all of them should force the computer to shutdown, if you want to manually reboot the gateway.
The three power sources are: the grid (shut off the service disconnect), the Powerwall power wiring (shut off every 30A Powerwall breaker), and the communications line that connects to the Powerwalls (which I believe terminates on the green connector under the gateway computer, just unplug it). In fact maybe just try the communications line first, that alone might be enough
Cheers, Wayne
The three power sources are: the grid (shut off the service disconnect), the Powerwall power wiring (shut off every 30A Powerwall breaker), and the communications line that connects to the Powerwalls (which I believe terminates on the green connector under the gateway computer, just unplug it). In fact maybe just try the communications line first, that alone might be enough
Cheers, Wayne
Should be true, but unfortunately, not true always.
Some gateways do not have a shield around the HV components, so there are live bare metal parts inside the gateway. Be very careful opening the gateway, as you could easily accidently brush the panel against those parts while removing it.
Since the reboot initiated by Tesal reenabled my access to the gateway webserver, it's become apparent that the webserver often fails, and is restarted. The effect is for it to cease accepting the bearer token, necessitating a new login. Restarting seems to take about a minute, and during that time it refuses connections to the http and https ports. The failures seem quite random. They even occurred when I isolated the gateway from the Internet. Tesla Australia have confirmed that this shouldn't happen - the bearer token should remain valid until the next firmware upgrade (assuming no owner action to reset the gateway).
The system is also ignoring the backup reserve level, and depleting the battery overnight, only to recharge it again from medium priced energy the next morning.
The system is also ignoring the backup reserve level, and depleting the battery overnight, only to recharge it again from medium priced energy the next morning.
I would expect that to be a non-compliant installation. It should not be possible to reach any live parts without using tools.
I believe BJReplay is referring to what is behind the cover plate that has to be removed using a screwdriver. If you look at arnolddeleon's picture above, you can see exposed electric connections, for instance.
Yes you have to remove a dead front cover to get to it which secured by a screw. Here is a photo with the cover on: 2017-12-28 10.49.43
It is - but the shield cover to make it compliant is out of stock, and I haven't bothered poking the bear while I still have API control.
It may well have, but I had rebooted it the previous night which is when I got web and API access back, and it was showing firmware version 1.20 after the reboot that night.
Similar threads
