Using 100% of the solar your panels produce [DC / AC ratio]

[dho112]

Hello All,

I have a new Tesla 9.6 Kw solar system and one Powerwall+ and I LOVE the system, and the install process was smooth and efficient!

I was initially confused because my energy production maxed at 7.7 Kw until I realized it’s because the solar inverter is rated for that. It’s still plenty for all my energy needs during the day including fully recharging the battery, even charging the car. I’ve been completely self powered for the last 32 days straight! It’s great!

But I’m capped by the inverter to 7.7 Kw, and I’m assuming, leaving about 2 Kwh during a few hours during the high sun.

I’d like to add another 3.8 Kw inverter to tap that extra peoduction and was wondering if its worth it, or not. I know the production is not entirely wasted since the extra collection area is used to maximize production during morning and evening sunlight.

Has anyone added inverters to their systems without adding another powerwall?

 

[wwhitney]

DC/AC ratio = 9.6/7.7 = 1.25, totally fine. Not financially or energetically worth adding more inverter power.

Right now sometimes your panels are underutilized because you don't have enough inverter power, but more often your inverter is underutilized because you don't have enough panel power. Since both inverters and panels cost money (or energy to produce), that is expected at the economic (or energetic) optimum ratio.

Cheers, Wayne

 

[RichardCullip]

You may want to model that using PvWatts ( PVWatts ) to see what, if any, additional energy production you can harvest. You will most likely see that it is nil, nada, nothing.

 

[travwill]

I’d also say in Winter you will produce less and as panels degrade over the years so your ratio of use will improve over time.

 

[ucmndd]

How many years will it take to recoup the cost of the second inverter if you start producing an extra 2-3kwh per day for half the year?

I’m pretty sure that ROI falls into the “don’t even bother with the math” category.

 

[holeydonut]

I agree with Wayne, the sweet spot around 1.25:1 DC/AC seems to be the ideal trade off between clipping and inverter cost.

Enphase sells micro inverters and you would assume they are actually motivated to get people to put in more AC than they need. And even they say ~1.25 is reasonable.

https://www4.enphase.com/en-us/support/technical-brief-why-my-pv-module-rating-larger-my-inverter-rating

 

[cwied]

The way I like to think about it is this: you can try to use 100% of the solar your panels produce or try to get as close to 100% of the AC your inverter can produce as possible. Thinking this way, it's easier for me to see that the optimal number is probably somewhere between these extremes, depending on the price of inverter capacity vs panel production capacity. If solar panels were free and didn't take up any space, you'd want your inverter to be clipping for as much of the day as possible.

 

[Unclepaul]

Goal should be to optimize your production vs investment.

 

[getakey]

Goal should be to optimize your production vs investment.

I'd put it slightly differently. Goal is to have NEM ~Zero, but not less than Zero.

 

[dho112]

Thank you all for the great replies! I’m learning a ton from this site! New solar concepts and practices I probably should have researched more fully before buying the system, but luckily, Whew! It looks like I made the proper choices to get close to maximizing my equipment potential despite my ignorance! Lol!

 

[tfan2018]

There’s just satisfaction seeing the full distribution curve on a sunny day

 

[DrSmile]

If you have a string inverter setup the cost difference for a larger inverter is pretty minimal. For a SE inverter the difference for example between a 10kw and 11.4kw inverter is $300. Install costs should be the same as nothing else changes. If you know you are going to significantly clip (not sure how you'd know from the get-go though) it would absolutely pay to get the larger inverter as 4 or 5 kwh per sunny day (say 50 days a year) would make the payback period similar to the solar system itself. My DC/AC ratio is 1.15:1 and I clip for 9 months out of the year. Because I have SRECs from NJ it would absolutely have been financially worth it to go with a 1:1 ratio. I did calculate that out by doing some calculus on my production graphs.

Obviously the system will degrade over time and the system will clip less, so one of the positives is your production losses due to degradation will be lower, but honestly my system is 3 years old and I haven't seen any significant changes. I think newer panels degrade much less than older generation panels.

 

[Bitslizer]

People also tend to forget, you GAIN production during the "shoulder" before and after clipping, usually that exceed or offset the clipping

All these clipping complaint is a side effect of Tesla's flat rate pricing structure, if folks were told, yeah you need to add $3000 for a second inverter most will rather not pay the extra

 

[wwhitney]

People also tend to forget, you GAIN production during the "shoulder" before and after clipping, usually that exceed or offset the clipping

I would like to see that quantified. For it to be significant, there would need to be significant difference in inverter efficiency between, say, 60% and 90% of rated power output.

Cheers, Wayne

 

[DrSmile]

People also tend to forget, you GAIN production during the "shoulder" before and after clipping, usually that exceed or offset the clipping

I see this quoted a lot. It is not really accurate. The panels are the panels. If you have 15 KW of panels you have the same production curve, period. The clipping is loss. The misinterpretation here is that people ADD more panels so they get clipping and compared to a system with LESS panels that wouldn't clip, the shoulder increase makes up for it. It isn't an apples to apples comparison.

 

[wwhitney]

I see this quoted a lot. It is not really accurate. The panels are the panels.

The claim is based on the inverter performance, and the idea that when your panels can put out, say, 5 kW DC, then a 6 kW AC inverter will give you a higher AC output (be more efficient) than an 8 kW AC inverter. Which might be true, but you'd really need something like PV Watts into which you could input the inverter efficiency curve to see what effect it would have over the course of a year.

Cheers, Wayne

 

[DrSmile]

The claim is based on the inverter performance, and the idea that when your panels can put out, say, 5 kW DC, then a 6 kW AC inverter will give you a higher AC output (be more efficient) than an 8 kW AC inverter. Which might be true, but you'd really need something like PV Watts into which you could input the inverter efficiency curve to see what effect it would have over the course of a year.

Cheers, Wayne

Yes I agree that the efficiency is better when the inverter is driven at higher wattage. But I'd argue that this is a very small difference compared to the clipping loss.

 

[cwied]

The claim makes more sense when you present the alternative as spending more $ on inverter capacity or more $ on panels. I think this is where it originated. If you add more panels, you will get more shoulder production. If you add more inverter capacity, you will regain the clipping losses. It's easy to see that some clipping is probably optimal from a cost perspective if you have no constraint on the number of panels you can install and you are bearing the full cost of the inverter upgrade.

 

[Vines]

The interesting part about this discussion is that traditional wisdom also came from the past and PV is moving quickly. It used to be that clipping was a sin, and in those days, an inverter might be 60% efficient at low loading. Then people started exploring 1.2 then 1.3, 1.5 and now 1.7 as acceptable stacking ratios.

IMO, nobody I work with is looking at the production curve and getting a rise out of those curves. The end goal is saving the customer money not making a 36-24-36 "perfect" curve for your production graph.

As a customer you input dollars, and you get some number of kWh per year out of those dollars. It absolutely makes sense to think about the best use of those dollars and consider all the factors.

Utility requirements are getting more stringent. Palo Alto recently introduced a soft cap of 20 kW of combined PV and ESS generation. Over this amount and a transformer upgrade is required. Larger inverters have rippling effects.

Panels and inverters both are getting less expensive, and more tolerant to high stacking ratios. The range of acceptable is much wider.

If a customer had a choice of putting $300 into another PV panel but with a higher inverter stacking ratio, or $300 into a slightly larger inverter, sometimes the extra panel produces more kWh per year for the same dollars spent.

 

[h2ofun]

The interesting part about this discussion is that traditional wisdom also came from the past and PV is moving quickly. It used to be that clipping was a sin, and in those days, an inverter might be 60% efficient at low loading. Then people started exploring 1.2 then 1.3, 1.5 and now 1.7 as acceptable stacking ratios.

IMO, nobody I work with is looking at the production curve and getting a rise out of those curves. The end goal is saving the customer money not making a 36-24-36 "perfect" curve for your production graph.

As a customer you input dollars, and you get some number of kWh per year out of those dollars. It absolutely makes sense to think about the best use of those dollars and consider all the factors.

Utility requirements are getting more stringent. Palo Alto recently introduced a soft cap of 20 kW of combined PV and ESS generation. Over this amount and a transformer upgrade is required. Larger inverters have rippling effects.

Panels and inverters both are getting less expensive, and more tolerant to high stacking ratios. The range of acceptable is much wider.

If a customer had a choice of putting $300 into another PV panel but with a higher inverter stacking ratio, or $300 into a slightly larger inverter, sometimes the extra panel produces more kWh per year for the same dollars spent.

wow, you and I would have been in trouble if we lived in Palo Alto.