Monitor power post loss-of-power

I have a Tesla Powerwall2 solar system. Whenever the power goes off due to service interruption (thanks to PG&E here in Napa Valley, that is frequently) the Sense monitor takes a long time to reconnect to the Eero WiFi network (which is broadcasting more than enough signal beyond the Sense monitor location). Any idea what is going on here?

Not sure how you have everything hooked up but with the Powerwall2 configured correctly, unless you lose power for hours/days or have serious load during a power failure you should be able to keep the Sense monitor and your wifi network powered throughout. The nominal load of your network gear and Sense along with a fridge/freezer and other crucial gear should run for a long time on a Powerwall2.

As far as the Sense monitor reconnect:

Perhaps the Eero mesh is the culprit or are you hard-wired between hubs?

Thanks for your thoughtful response. Sense is still offline 5 hours later while Eero is working fine. This only happens when we lose Grid service and the PW2 kicks on. I’m thinking it’s likely a power issue and not a WiFi issue. When I get home tomorrow I’ll check out the load center.

Open a support ticket. Post the results as I am sure a lot of people with powerwalls would like to know (or those thinking of getting them).


I guess the place to start is to make sure the Sense monitor is actually powered when your Powerwall is active. Remember that the web and mobile are just loading cloud data so you’d need to confirm by looking at the Settings>Signals or the Mains Power Meter or with a realworld meter!

Got home. Sense still offline. Looked at breaker box…circuit energized. Flipped breaker and magic. Wondering out loud if Sense monitor power supply is sensitive to potential grid/PW2 spikes. Anyway…working…but it doesn’t feel like vicitry

If you look through the forum, you’ll see that brownouts and momentary power outages, rather than long “clean” power outages, are most likely to disrupt Sense’s network connection. They have put a lot of work into the reliability of their data collection plus robustness in the face of many possible complications:

But you can’t get away from the fact that the monitor only has finite bandwidth to do self-checking and we’re not dealing with a fully redundant high-reliability system here. Brownouts and momentary drops have been know to leave processors with unknown bad data inside of them - almost impossible to fix short of a reboot (but how does a slightly corrupted processor know to reboot).