Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

undersize inverters?

Push for bigger inverters?

  • Yes

    Votes: 3 23.1%
  • No

    Votes: 10 76.9%

  • Total voters
    13
This site may earn commission on affiliate links.
I'm ordering a huge 35.02kW solar system with 4 powerwalls. Tesla is recommending 3 10,000w SolarEdge (my request) inverters. I know it's okay to undersize the inverters a bit, but that seems like way to much. Even if they went with 3 11,400w inverters it would still be undersized at 34,200w max.

Should I push for the bigger inverters?
 
30 kW AC and and 35 kW DC represents a ratio of 1.17, which is probably fine depending on how your system is laid out. If you've got all ~100 panels facing perfectly South, then it could be a problem because all of the strings would be producing at the same time and you could realistically generate 35 kW. But if you've got multiple arrays facing multiple directions, they'll all hit peaks at different times during the day.

Some more information on DC/AC ratio here: Understanding DC/AC Ratio - HelioScope Knowledge Base
 
Last edited:
You likely want more than 4 powerwalls, too. You probably want at least 5, as I would not be comfortable with more 7kW of solar per powerwall 2. You run the real risk of your system shutting down in a power outage if its actually outputting anywhere near its capacity during an outage.

(edit: I believe z'tesla and I were posting this at the same time).

On thread topic, it would depend on the direction your panels are, and there is a ratio to those things... but if it were me, and I could afford such a large system, I would likely ask for them to go with the larger inverter "just because". It likely wouldnt matter, but at that price point, whats a few more K to more future proof, even if it doesnt really help you?
 
  • Like
Reactions: preilly44
I have Microinverters on my setup and they are undersized, based on the LAB testing of my panels, but in the real world (most don't live in Arizona) the panels would never go near the top. To that end, I replaced two of my 240watt panels with 300watt units and there are days when I will get an error from those inverters, but it just caps the power at the max the inverter is rated for.
 
  • Like
Reactions: preilly44
I'm ordering a huge 35.02kW solar system with 4 powerwalls. Tesla is recommending 3 10,000w SolarEdge (my request) inverters. I know it's okay to undersize the inverters a bit, but that seems like way to much. Even if they went with 3 11,400w inverters it would still be undersized at 34,200w max.

Should I push for the bigger inverters?

With 35kW of solar on 30kW of inverters your oversize ratio is 1.17. Anything <1.2 almost never gets saturated.
 
Thanks for the quick responses. Just to give more information, there are several different roof directions (see layout below).

As far as the power walls go, it was my understanding that if I'm using enough power and the power walls are low enough, they will continue working and if I have one of my Tesla vehicles plugged in, any excess power would be used to charge the car. Should I be concerned? I'm trying to avoid adding more power walls because we probably won't be using them very often in the first place as we rarely lose power.
Capture.JPG
 
Thanks for the quick responses. Just to give more information, there are several different roof directions (see layout below).

As far as the power walls go, it was my understanding that if I'm using enough power and the power walls are low enough, they will continue working and if I have one of my Tesla vehicles plugged in, any excess power would be used to charge the car. Should I be concerned? I'm trying to avoid adding more power walls because we probably won't be using them very often in the first place as we rarely lose power.View attachment 624951

.... I seriously doubt 35kW of solar will ever saturate 30kW of inverters with that panel arrangement. IF it ever occurs it will be for minutes a year. Like... < 10 minutes a year. I predict the peak you see will be < 28kW.
 
Thanks for the quick responses. Just to give more information, there are several different roof directions (see layout below).

As far as the power walls go, it was my understanding that if I'm using enough power and the power walls are low enough, they will continue working and if I have one of my Tesla vehicles plugged in, any excess power would be used to charge the car. Should I be concerned? I'm trying to avoid adding more power walls because we probably won't be using them very often in the first place as we rarely lose power.View attachment 624951

zTesla can speak to this from personal experience, but the high level version is, if you outsize your solar over the number of powerwalls "too much" if you have an outage, and the solar is generating more than the powerwalls can take in at once, the entire system would shut down, even if the powerwalls were not full.

In normal operation it wouldnt matter much as any excess would be fed back to the grid (or not exported if you are under some sort of non export agreement). The one time it could matter is the time you would WANT to have power, after spending all that money, and that is during an outage.

In general, I believe tesla recommends 1 powerwall per no more than 7kW of PV, but thats just a guideline, it would depend on what your system could produce at peak. If peak is <28 then you are likely ok. If you actually need all that power, however, and it were me, I would add a 5th powerwall unless code / circumstances dictated I couldnt.
 
Last edited:
  • Like
Reactions: pilotSteve
This is 14.55kW on a 11.4kW inverter on a near perfect day.

View attachment 624954

I TOTALLY get what you are saying (believe me, I do), HOWEVER, if "I" had the money to put in 35kW of solar and 4 powerwalls, I would not want to see ANY plateau across the top of my chart, even if its "more economical" to have a little bit of clipping because the upsized inverter doesnt pay for itself.

At those costs, I dont know why anyone would. My response would be different if I was setting up a system <10k with 1-2 powerwalls, but at that scale? I wouldnt want any flat along the top (indicating clipping) even if it saved me a little money or was more economical.
 
  • Like
Reactions: preilly44
I TOTALLY get what you are saying (believe me, I do), HOWEVER, if "I" had the money to put in 35kW of solar and 4 powerwalls, I would not want to see ANY plateau across the top of my chart, even if its "more economical" to have a little bit of clipping because the upsized inverter doesnt pay for itself.

At those costs, I dont know why anyone would. My response would be different if I was setting up a system <10k with 1-2 powerwalls, but at that scale? I wouldnt want any flat along the top (indicating clipping) even if it saved me a little money or was more economical.

If you have N facing panels AND you're oversized by <1.2 there won't be any clipping aside from maybe during a 'cloud-edge' power surge which you'd need 40kW of inverters to avoid that...
 
Would you rather spend $200 on more inverter to capture another ~1kWh from cloud edge surge or $200 to produce ~300kWh more per year from another 300w of solar? It it really worth $200 to get another ~kWh/yr?
For what this system is costing, I am assuming money is not an issue. I am going with 2 11.4K inverters when I split my 14.5kw system. Totally stupid but I just want ability to upgrade if I wanted, which is probably never. So totally agree with you, I just did not like the clipping I got last summer. Now, if I was totally looking at costs and ROI, I would have done NONE of this stuff. But, only live once, and I love technology.
 
??? Why? Most rationally designed PV systems will pay for themselves in <10 years and last for >30. In terms of fiscal diversification that seems like a wise move. ~99% of the people I help with PV are doing it for cost savings and ROI.
Because I am 63 years old. No idea how long I will be in my large house. And I just do not believe a 10 year ROI is good payback. But, this is just me. Now, I have spent some of my kids inheritance on my tech fun. :)
 
I would agree with the idea that the proposed inverters should be sufficient as, among other factors, the mix of angles should mean there won't be a single peak. That said, the distribution of the panels among the inverters also matters - if you hook 1/3 of the panels to each inverter but put all the south-facing ones on a single inverter, it will be more likely to have clipping on that inverter with extra room on another.

Also, the amount of clipping, if any, would likely be not only for a small part of the day but for a small fraction of the year. PVWatts can give you the full numbers, but for us near you, even our south-facing solar is over a third down now versus our peak.
 
As others noted, I think a DC/AC Ratio of about 1.2 to 1.3 should be fine. But, what I'm not so sure about is how many panels are allowed on a single SE10000H to avoid clipping. Looking at your array, it's not like all your panels are perfectly lined up and can be evenly distributed across three SE10000H's. I feel like one thing to consider is to learn exactly which panels they're going to match up to each inverter.

Based on the level of investment here, I feel your own peace of mind will be boosted if you get the SE11400H instead. ROI and Pragmatism not withstanding, you'll be able to look back and know you got the best available.
 
  • Like
Reactions: preilly44
As others noted, I think a DC/AC Ratio of about 1.2 to 1.3 should be fine. But, what I'm not so sure about is how many panels are allowed on a single SE10000H to avoid clipping. Looking at your array, it's not like all your panels are perfectly lined up and can be evenly distributed across three SE10000H's. I feel like one thing to consider is to learn exactly which panels they're going to match up to each inverter.

Based on the level of investment here, I feel your own peace of mind will be boosted if you get the SE11400H instead. ROI and Pragmatism not withstanding, you'll be able to look back and know you got the best available.
Yep, 3 11.4 would be hard to clip, but as another posted, depends what panels are hooked up to what.
 
  • Like
Reactions: preilly44