8.16 kW DC Solar Panels but never generating more 6.5kW

[fatcatXK]

Hi,

I had Tesla 8.16 kW DC Solar Panels installed in June 2020 but they never generate more 6.5kW even after cleaning the panels...Across entire year, 6.5kW is the maximum generated power which is typically around noon to 1pm. I'm trying to understand whether getting 8.16 kW is realistic. I have a family member that installed a 4.8 and is getting 4.8.

The inverter installed is a solaredge SE7600H - USOOOBSU4 that maxes out at 7.6kW. Wanted to check here on whether this would be a cause as to why I'm not getting the maximum possible. My panels face south on the roof and has very clear and direct access to the sun across the whole day. Nothing obstructing.

I have a total of 24 Hanwa panels.

Thank you!!!

 

[EM CARS]

I want to follow this. I came here to post a similar question. I have a 6kw array that faces south. No obstruction and clear view of sky, but I'm only peaking out around 3.5kw.

 

[arnolddeleon]

I'm slightly skeptical of the report of 4.8 kW of DC panels regularly getting 4.8 kW of AC. The DC rating is under STC (Standard Test Conditions) and the main utility of that number is so you can compare panel to panel. The resemblance to real world conditions is only passing. In general, assuming that the inverter is big enough, the AC output matches or exceeds the DC rating only under edge cases.

In your case, the maximum will be 7.6kW because the size of your inverter.

PVWatts

Estimates the energy production and cost of energy of grid-connected photovoltaic (PV) energy systems throughout the world. It allows homeowners, small building owners, installers and manufacturers to easily develop estimates of the performance of potential PV installations

pvwatts.nrel.gov

might be helpful to you.

Of course, it is also possible that you have defective installation. You can post graph the output of your PV system to give us more information.

 

[fatcatXK]

Thank you. I will confirm my brother-in-law's 4.8kW and provide exact data.

I checked on PVWatts site. I should be peaking from May:7.24, June: 7.7, July: 8.07, Aug: 7.74...yet my peak all year is 6.5kW. Never higher than 6.5kW. The fact that the peak is always that across entire year also makes me think that there is something else artificially limiting the total output.

I have a 1 story home. The angle of the roof is not steep. The 20 degrees estimate by PVWatts is a bit conservative.

> You can post graph the output of your PV system to give us more information.

Could you help provide more details of how? Are you looking for a screenshot from the Tesla solar app? I was thinking of running pypowerwall but will that work if I don't have a powerwall? Thank you!

 

[arnolddeleon]

Thank you. I will confirm my brother-in-law's 4.8kW and provide exact data.

I checked on PVWatts site. I should be peaking from May:7.24, June: 7.7, July: 8.07, Aug: 7.74...yet my peak all year is 6.5kW. Never higher than 6.5kW. The fact that the peak is always that across entire year also makes me think that there is something else artificially limiting the total output.

I have a 1 story home. The angle of the roof is not steep. The 20 degrees estimate by PVWatts is a bit conservative.

> You can post graph the output of your PV system to give us more information.

Could you help provide more details of how? Are you looking for a screenshot from the Tesla solar app? I was thinking of running pypowerwall but will that work if I don't have a powerwall? Thank you!

Angle does matter.

A screenshot of Tesla solar app might give us more context. The shape of the graph can tell us if you output is being limited.

 

[fatcatXK]

 

[arnolddeleon]

Well, your output is not being clipped. It would look something like this if it were:

New item by Arnold de Leon

photos.app.goo.gl

The most likely explanation is the amount of light they receive is less than "ideal", most likely the tilt angle is less than optimal.

 

[fatcatXK]

I climbed onto my roof. The angle is 16 degrees, so more optimal than the 20 degrees as I have a one story home. Based on that, my peaks should be: May:7.3, June:7.81, July:8.15, Aug: 7.73 . Yet my peak is always 6.5 for whole year including May-Aug. That seems to indicate that there are issues with my installation. Greatly appreciate any feedback anyone has one this on what to investigate further.

There is absolutely no obstruction of any kind. No trees, chimney, etc...all clear for most of period from sunrise to sunset...in San Jose, CA

 

[jjrandorin]

I climbed onto my roof. The angle is 16 degrees, so more optimal than the 20 degrees as I have a one story home. Based on that, my peaks should be: May:7.3, June:7.81, July:8.15, Aug: 7.73 . Yet my peak is always 6.5 for whole year including May-Aug. That seems to indicate that there are issues with my installation. Greatly appreciate any feedback anyone has one this on what to investigate further.

There is absolutely no obstruction of any kind. No trees, chimney, etc...all clear for most of period from sunrise to sunset...in San Jose, CA

You are not going to see peak power greater than your inverter size of 7.6kW so the answer to "is expecting 8.16kW peak energy generation realistic" is "No, its not".

Getting peak out of your inverter assumes laboratory perfection. Even if your panels are at the exact right, laboratory angle, that doesnt mean your house is. PV production should be compared over months and years, as there are too many variables for peak production to be focused on that unless its drastically wrong.

There are more technical users than me here, for sure, but I havent seen anything that looks glaringly wrong with what has been posted so far in this thread. The graph doesnt show clipping. What does PVwatts say for your prospective Yearly production and how far are you from that? I suspect you are probably close.

 

[fatcatXK]

> You are not going to see peak power greater than your inverter size of 7.6kW so the answer to "is expecting 8.16kW peak energy generation realistic" is "No, its not".

Understood. That's why I'm thinking of asking Tesla to replace with an inverter capable of my max although the responses indicating that my max is not being clipped and hence might not be the inverter is confusing me.

I have a bell curve with a max of 6.5kW from April-Aug, it's always at 6.5. Shouldn't the peak vary based on month of year as PVWatt is indicating I should see? My peaks from Dec-Mar slowly grow and go up to 6.5. But then it never exceeds 6.5...so perhaps my panels just weren't made for 8.16kW?

> The graph doesnt show clipping. What does PVwatts say for your prospective Yearly production and how far are you from that? I suspect you are probably close.

Right. Could members please suggest on what could possibly be limiting my max if it's still a bell curve? I saw the there was a python script to be run to detect issues in installation but wasn't sure if that required a powerwall. Please advise.

> What does PVwatts say for your prospective Yearly production and how far are you from that? I suspect you are probably close.

I'm off by about 15% (lower). My peak in July is off by 25% though. That's why I'd like to tackle that first. This is based on full year data for 2021 and 2022. I only have partial for 2020 (and of course 2023).

Thank you!

 

[fatcatXK]

I saw this post in the forum: Disappointing Power Generation, and I need help with analysis of pypowerwall output please!

Would it make sense for me to try to gather similar data as that one? I'm unclear whether I need to run a powerall to run pypowerwall from GitHub - jasonacox/pypowerwall: Python API for Tesla Powerwall and Solar Power Data . Any advise on this would be greatly appreciated. Thank you!

 

[wwhitney]

What is the model number of the Solaredge Power Optimizers installed on each of your 24 panels?

Are all of the 24 panels facing the same orientation, and unshaded at solar noon? You indicated the elevation angle is 16 degrees (which would be about a 3.5:12 roof pitch, plausible), but what is the azimuth angle (e.g. due south, or off a bit)?

Cheers, Wayne

 

[arnolddeleon]

I recommend stopping obsessing about the peak production and instead apply the energy to looking at expected production over a large enough time period (e.g. a month). What did Tesla predict and how are you doing against that? You can use PVwatts to also check their numbers.

The nearly perfect shape of your generation doesn't suggest an inverter issue. If the system is under performing, then my guess is that you have a module (or two or three) that are not contributing.

Since you have a Solaredge inverter there are potentially other things you can check. Each module should have a Solaredge optimizer. If this inverter is a model that has a local display you can have it count out the number modules that you have. You may also be able to display the output of each module. They should all be roughly the same under non-changing light conditions. Tesla probably has the same information. Some users have reported being able to get access so they can use the Solaredge app to view the data, you can search the forums for that.

 

[Redhill_qik]

I climbed onto my roof. The angle is 16 degrees, so more optimal than the 20 degrees as I have a one story home. Based on that, my peaks should be: May:7.3, June:7.81, July:8.15, Aug: 7.73 . Yet my peak is always 6.5 for whole year including May-Aug. That seems to indicate that there are issues with my installation. Greatly appreciate any feedback anyone has one this on what to investigate further.

There is absolutely no obstruction of any kind. No trees, chimney, etc...all clear for most of period from sunrise to sunset...in San Jose, CA

You haven't mentioned the orientation of your panels. Can you post the layout picture from your approval docs? In the same approval docs there should also be an estimated annual production, how does that measure you against your production.

Personally, I have a 8.16kW of panels and max out at 6.2kW (except for short rare cloud edging events). This is because the majority of my large are northwest and northeast facing due to roof surfaces.

 

[fatcatXK]

What is the model number of the Solaredge Power Optimizers installed on each of your 24 panels?

Are all of the 24 panels facing the same orientation, and unshaded at solar noon? You indicated the elevation angle is 16 degrees (which would be about a 3.5:12 roof pitch, plausible), but what is the azimuth angle (e.g. due south, or off a bit)?

Cheers, Wayne

I will research the Solaredge Power Optimizers and revert back. Would that come from requesting that from Tesla? Or climbing onto the roof?

All 24 panels are facing the same orientation. They are all facing south (see picture below). When using google maps, here is what the panels look like. Straight down is south. So my panels face south with a tiny bit of South East. My roofs are not steep, 16 degrees when using the protractor app in Apple Appstore. Absolutely no obstruction from trees, chimney...none.

Thank you again!

 

[fatcatXK]

I recommend stopping obsessing about the peak production and instead apply the energy to looking at expected production over a large enough time period (e.g. a month). What did Tesla predict and how are you doing against that? You can use PVwatts to also check their numbers.

The nearly perfect shape of your generation doesn't suggest an inverter issue. If the system is under performing, then my guess is that you have a module (or two or three) that are not contributing.

Since you have a Solaredge inverter there are potentially other things you can check. Each module should have a Solaredge optimizer. If this inverter is a model that has a local display you can have it count out the number modules that you have. You may also be able to display the output of each module. They should all be roughly the same under non-changing light conditions. Tesla probably has the same information. Some users have reported being able to get access so they can use the Solaredge app to view the data, you can search the forums for that.

Sigh...I was not very thorough at the time of signing and the design doc says estimated output at 0. I was in a rush at the time to get the discounts (which have gone way up since then...small excuse for my extreme sloppiness).

Yes, I've seen a number of different user reported outputs. I'm a software engineer so I'm reading up on how best to get that data.

 

[fatcatXK]

Sigh...I was not very thorough at the time of signing and the design doc says estimated output at 0. I was in a rush at the time to get the discounts (which have gone way up since then...small excuse for my extreme sloppiness).

Yes, I've seen a number of different user reported outputs. I'm a software engineer so I'm reading up on how best to get that data.

Found the layout. Tesla estimate in that document was 13,500kwh. I'm 18% from that. PVWatt had estimated 13,100. Note that at the time of installation and design, there was a massive tree in my neighbor's house and so my a lot of my afternoon sun was blocked. That tree has since fallen. There are no trees, chimney, houses, walls, fences, poles,...nothing is blocking.

The panels are: Q.PEAK DUO BLK-G6+ / SC, 330-345 (from the layout doc).

I do not see the Solaredge optimizers listed. Will continue to research.

The actual installed panels face the same direction as the original layout design, but it's not exact. I assume this doesn't have a significant difference. I recall them saying that they had to move the panels due to vent outlets in the house.

vs

 

[wwhitney]

Gased on the elevation angle and the azimuth, then near solar noon near the solstice the sun's incidence angle to the panels should be near 90 degrees (optimal). That's because on the solstice the sun's maximum elevation is 90 - latitude + earth's tilt = 90 - 37 + 23.5 = 76 degrees. While your panel normal elevation angle is 90 - 16 = 74 degrees. So cosine error is not a plausible explanation for the discrepancy.

The main explanations I see that remain are that your panels aren't actually 340W each, or that a few of your panels aren't actually producing. Undersized optimizers would presumably give you signs of clipping in your daily power curve, which you don't have.

So I suggest figuring out how to access the optimizer level data for each of your 24 optimizers to see what's going on.

Cheers, Wayne

 

[fatcatXK]

Gased on the elevation angle and the azimuth, then near solar noon near the solstice the sun's incidence angle to the panels should be near 90 degrees (optimal). That's because on the solstice the sun's maximum elevation is 90 - latitude + earth's tilt = 90 - 37 + 23.5 = 76 degrees. While your panel normal elevation angle is 90 - 16 = 74 degrees. So cosine error is not a plausible explanation for the discrepancy.

The main explanations I see that remain are that your panels aren't actually 340W each, or that a few of your panels aren't actually producing. Undersized optimizers would presumably give you signs of clipping in your daily power curve, which you don't have.

So I suggest figuring out how to access the optimizer level data for each of your 24 optimizers to see what's going on.

Cheers, Wayne

Thank you very much!

If you have suggestions as to how to access the optimizer level data, that would be greatly appreciated. In the meantime, I will keep looking in the forums for more information.

Is there anything that you think might cause my peak from April to be the same as July while I have a normally looking bell curve in both? I imagine that if I didn't have the right panels that they still wouldn't have the same peaks? They would have relatively lower amount as I move away from June/July right?

It was harder to find an optimal day in Mar...but I found Mar 1, my peak that day was also 6.5. So my peak of Mar, April and May are all the same. Any thoughts on what could be causing that to help narrow down potential issues? Thank you again!!!

 

[fatcatXK]

Thank you very much!

If you have suggestions as to how to access the optimizer level data, that would be greatly appreciated. In the meantime, I will keep looking in the forums for more information.

Is there anything that you think might cause my peak from April to be the same as July while I have a normally looking bell curve in both? I imagine that if I didn't have the right panels that they still wouldn't have the same peaks? They would have relatively lower amount as I move away from June/July right?

It was harder to find an optimal day in Mar...but I found Mar 1, my peak that day was also 6.5. So my peak of Mar, April and May are all the same. Any thoughts on what could be causing that to help narrow down potential issues? Thank you again!!!

I read somewhere that if the inverter gets too hot that it will limit the power. My inverter is completely under the sun (no shade) for most of the day. Should I look into that? Forum users said that if it hits 115 degrees that it would start being left efficient. But I imagine that I wouldn't have a normal looking bell curve if that happened right? Thank you!