Solar Panels

Starting March 15th, Tesla Forums will become read only. To continue the conversation with the Tesla community visit

Solar not working on new install/time to go back to grid power

My new solar stopped producing power 26 Dec at noon. I didn't notice until rather late in the day. When there was enough light today to produce power, it was still DOA. I shut everything down in the sequence on the troubleshooting page, reset the Neo gateway box as well as the FIOS boxes. The inverter is the Solar Edge without the display screen. The green LED says it is producing power.

Out of curiosity, I wanted to see if the Powerwalls were functional and flipped the breaker for grid power. The Powerwalls worked, but it took about 19 minutes for the grid power to resume being fed to the house. This was a bit disconcerting. Is this a normal amount of time? It also appears to mean that the gateway controls whether or not power can go from the grid to the house which makes me a bit nervous. I had assumed that there was a direct connection. In other words, a failure of the gateway could leave us with no power at all even if the grid is up. True?

I did just file a request for help from Tesla. The warning to wait three days was a bit daunting.

Thanks for any observations on this.


  • Hard to say what may have gone wrong. You seem to have taken all the steps I can think of. I expect Tesla can sort it out.

    I think 5 minutes is normal from grid power going on to switching off of Powerwalls. This is intentional to let the grid stabilize, as it can be a bit erratic in a real grid power off/on. 19 minutes does seem longer than I'd expect.

    Yes, the gateway controls the power going from the grid to the house. It's a legal requirement so that you never have power flowing from solar or the Powerwalls into a grid that is down. About 99.9% of the time it remains connected when the grid is up so excess solar can flow to the powered grid.

    As for failures, yes, there are a number of points that could cause your backup to fail - which is true of any technology. There are very expensive reducant systems ($100K or so) for ultra -high-reliability requirements such as some server environments. And even then, some of these fail.
  • Thanks! I was once in a Salomon Brothers building that was a backup for their NYC operation. It was actually two duplicate building connected together. Each was fed from different legs of the grid and were isolated from one another other. Each had their own BIG generator for backup and went off grid power when there was lightning within a radius I forget. Either building could take over from NY if needed. Some stuff would be put on hold, of course, but no trades would have been lost. I always wanted that building! The local government was dumb and didn't buy it. It was also hurricane rated and would have been a great EOC. This was in the late 80's so solar and batteries weren't that big yet for backup.
  • I saw something similar in Chicago, in the late '80s. I was in the Sears Tower, doing a com system test installation, in February! (I'm a "recovering" R&D electrical engineer.) The other end of the fiber-optic (a big deal, at the time) communications link was in a Chicago suburb. Nice town, but I don't remember the specific location. The remote telco central office was ten feet underground, behind a blast door and four feet of concrete. Fully stocked with food, water, plumbing and refrigerator-sized flooded lead-acid batteries, I suspect it was built to survive a nuclear war. I had to sign an NDA in order to visit and work there.
  • Reminds me of an underground Pacific ocean telephone junction point I visited about 40 years ago. Equipment racks were all mounted on huge springs to handle earthquakes or nuclear detonations. It also had lead-acid battery backup systems. Glad they never had to test it out!
Sign In or Register to comment.