Strange Behavior (Fresh Install)

[plucas777]

I want to give a disclaimer before I ask this question: My bidirectional meter has not been installed, therefore my system is not fully setup yet.

Having said that, I live in central Florida and just got an 11.360 kW solar system installed with 2 powerwalls. In full sun today the solar barely clocked higher than 2.3 kW and charged the powerwalls very slowly. After two days I am sitting around 70% charged. Even more strange to me is that most of the energy was flowing towards the "grid" but I can't technically send power to the grid yet without the new meter. What also doesn't make sense to me is at night the Tesla app shows higher solar generation (around 3.5 kW) than I get in the day time (all going towards the grid though). As expected, the powerwalls only charge during sunlight hours.

Can anyone tell me if this is normal behavior with a system that isn't finished yet ? Or should I be concerned. I was expecting much higher solar values during direct sunlight.

 

[Regnad]

Many issues are due to the CT (current transformers) being installed incorrectly. Check that first and good luck.

 

[morbidz]

I want to give a disclaimer before I ask this question: My bidirectional meter has not been installed, therefore my system is not fully setup yet.

Having said that, I live in central Florida and just got an 11.360 kW solar system installed with 2 powerwalls. In full sun today the solar barely clocked higher than 2.3 kW and charged the powerwalls very slowly. After two days I am sitting around 70% charged. Even more strange to me is that most of the energy was flowing towards the "grid" but I can't technically send power to the grid yet without the new meter. What also doesn't make sense to me is at night the Tesla app shows higher solar generation (around 3.5 kW) than I get in the day time (all going towards the grid though). As expected, the powerwalls only charge during sunlight hours.

Can anyone tell me if this is normal behavior with a system that isn't finished yet ? Or should I be concerned. I was expecting much higher solar values during direct sunlight.

there should not be any generation at night, duh! something is not hooked up correctly.

 

[charlesj]

It is strange indeed to say the least.
Hard to say if that uni directional meter has something to do with this, could be but certainly not solar generation at night.
When will the meter be changed?
May need to wait until then. May also want to call Tesla and ask if they know in the first place.

 

[JayClark]

I want to give a disclaimer before I ask this question: My bidirectional meter has not been installed, therefore my system is not fully setup yet.

Having said that, I live in central Florida and just got an 11.360 kW solar system installed with 2 powerwalls. In full sun today the solar barely clocked higher than 2.3 kW and charged the powerwalls very slowly. After two days I am sitting around 70% charged. Even more strange to me is that most of the energy was flowing towards the "grid" but I can't technically send power to the grid yet without the new meter. What also doesn't make sense to me is at night the Tesla app shows higher solar generation (around 3.5 kW) than I get in the day time (all going towards the grid though). As expected, the powerwalls only charge during sunlight hours.

Can anyone tell me if this is normal behavior with a system that isn't finished yet ? Or should I be concerned. I was expecting much higher solar values during direct sunlight.

One, or more, of the gateway CT clamps ( current measurements) is on the wrong line(s), or in backwards, or something similar. While I was waiting for my PTO I had experimented with moving the CT clamps around (not recommended, dangerous stuff to mess with if you’re not an electrician or ee) and saw similar behavior on my system until placed back correctly.

In fact the PWs can be made to trickle charge from the grid even at night with no solar, by doing something like this on purpose by moving the solar CT clamp to a non-solar circuit, but one that has current flowing all/most of the time.

Tesla (or your installer) will come out and fix that, call support and tell them it’s showing solar production at night, which will be a clear sign to them they have the CT clamps configured incorrectly. It’s an easy fix for them. But before calling Tesla, maybe make sure your installer (if different) didn’t do that on purpose to get some charge back into the PWs for some reason while waiting for the Solar meter to be installed.

 

[plucas777]

So the installer had to come back out for something, when I showed him the issue, he rebooted everything and one of the CT's was flagged red in his test utility. He fixed the issue he saw which seems to have made things better (I see higher kW's in the daytime), but I am still seeing strange behavior at night (2kW and above sometimes).

I guess I need to get him back out here.

 

[DDotJ]

I just had a system installed last week, and I'm also seeing very low generation.

I installed an 11.35 kW system but it's peaking at about 2.5 - 3.5 kW on a cloudless sunny day. I don't generate nearly enough power to charge up my two powerwalls. I called Tesla and they told me to check my inverter and see if all my panels are online. It shows 36/36 panels online and producing. They said it might be normal due to it being Winter here in SoCal and the angle of the sun but I don't believe there would that that much variance.

I escalated it and they're supposed to send out a technician but I haven't heard back about it yet. Hopefully it is underproducing, as it's only covering a small portion of my electricity usage.

 

[power.saver]

For reference, my 12kW system has produced around 18-20kWh per day for the past few days, excluding today since it's raining here.

Are you looking at the Tesla App or your actual inverter output?

 

[plucas777]

The Tesla App which I presume may be off due to some issue related to CT's, however I am not sure of this.
I also haven't gotten the credentials to my Solar Edge app yet either to try and confirm any of what I am seeing.

 

[plucas777]

UPDATE:

I continued to see discrepancies in my graphs and my local installer kept giving me bogus answers so I called up Tesla myself.
The most obvious crazy reading is that every day my home energy usage drops to zero around 4 PM and kicks back in around
8 AM. So I showed this to Tesla and they confirmed that it is a CT related issue. This was two days ago.

The tech on the line told me that the turn around time is "USUALLY" 48 hours, however I am certainly not optimistic from what I have
read here and other places. Anyway, she claims to have opened up a ticket on the matter so hopefully I hear back soon because I
begin payments on the system next month.

 

[McWill]

Home usage CT on the solar generation cable and the Solar CT measuring your home power usage?

Can you upload a pic containing both solar & home for reference?

 

[plucas777]

As requested.... I think you may be on to something McWill, however I am not very knowledgable in this.

 

[woferry]

The solar definitely looks more like house consumption, though it's odd that it rises with the sun (I guess it could be causing the Powerwall(s) to draw from the grid during this time). Sure seems like the CTs are not in the right place / configured properly. I assume you're not comfortable opening / playing with your electrical panels? If you are comfortable (and know how to identify your circuits, and aren't doing anything too complex [ideally whole-house backup]) you could probably fix it yourself fairly quickly, otherwise I'd say it's best to have your installer come out and fix things.

I'd suggest you keep an eye on your meter, especially if you're not getting any credit for production yet (not bi-directional yet), you might be better off just switching everything off as you may be drawing more from the grid right now due to the misconfiguration. Could you attach pictures of each 'layer' (home/grid/solar/powerwall) for the same day? The Gateway only measures grid and solar with the CTs, and uses this information to set the charge/discharge rate of the Powerwall. The home value is derived from what's left.

 

[plucas777]

Here is a sample with all activity displayed. One note is that I keep the Powerwalls in backup mode. They are for the most part fully charged other than when the tech came out and was attempting to troubleshoot. At some point I also simulated a grid outage for a short period of time as well.

I also through in a snapshot of the powerflow for right now which is about 1:00 AM in Florida. This was the initial red flag I began to see.

 

[McWill]

Certainly looks like polarity of grid is wrong, unfortunately my powerwall hasn't been installed but working on the info from @woferry about only 2 CTs. The images could be explained if grid CT was on the wrong way round and the solar CT was measuring house consumption.

At night, with no powerwall activity, your house consumption should be matched by the same grid consumption. You should however be able to identify what is happening with a few tests...

• Shut off the grid breaker to simulate an outage. Does the grid consumption graph flatline?
• Turn on a high power appliance, does this cause home consumption to spike (as it should) or is it seen as extra solar generation.

 

[plucas777]

Well I've got to give this a try. Appreciate the insight on this to all who have looked into it with me.

 

[plucas777]

I believe Tesla has fixed the problem remotely. I haven't seen a flow like this since I had it all installed. Now I just need the meter from the City and we should be good to go.

 

[woferry]

What does the graph look like now (with everything enabled)? Did the PWs fully charge?

 

[plucas777]

I keep them fully charged and in backup mode. The powerwalls always charged which is especially why I suspected something being off. Here are some shots with everything working - Solar didn’t kick in until around 8 AM

 

[woferry]

That definitely looks much better. I'm surprised they could have fixed that remotely, the grid measurement looked like it was just backwards, but I would have guessed that the solar CTs were on the wrong wires and would have needed o be physically moved. Good that you're finally getting sane readings.