New 20kw Solar Install - just interconnected - Did we get screwed?

[bcroe]

if OP has full access on monitoring then they can graph production for each inverter independently
as well as DC voltage ( a big indicator of inverter forced clipping when it jumps up).

This significant time in higher voltage clipping for the inverter will likely lead to a shorter life.

Shorter life could be a subject for discussion, it likely varies with the inverter. Have not been
monitoring voltage closely, but these Fronius have been running like this for 6 years on every
sunny day. I try to align panels to limit clipping, but it is happening. Bruce Roe
NScurJn17.jpg

[ButchDeal]

I don't doubt that the model's inverter eff. is no more than a general # I was sort of hoping for a better est. based on your longer/greater experience w/SolarEdge.

Understood about today's clipping. Given the array tilt angles and the location, and the solar altitudes/zenith angles this time of year, my guess is the smaller array w/ the 41 deg. tilt ~ = 41 deg. latitude is clipping a bit more as f(POA input /inverter output) per installed STC kW than the larger array with the 26 deg. tilt.

Then too, I'm not sure we've definitively determined that the 15 kW output limit is from the inverter capacity or is inverter limited.

But, with respect to inverter efficiency, if that's what is limiting output, since the P.O.A irradiance input beyond that needed for a 15 kW output is of no benefit (and in theory and practice - as in from an energy balance on the array - may well lower system eff. a bit by increasing array temp., or the inverter(s)' temp(s.), or all 4 of the devices' temps. for no increase in useful output).

I suppose the argument could be made that increased inverter eff. could mean that the greater output from that increased inverter eff. is also lost in the clipping thereby making the absolute difference between the clipped and unclipped output a slightly larger number. That is, a higher inverter eff. would deliver more energy to the load than would a lower eff. inverter if the output for either inverter scenario were not clipped. But I guess that would depend on the mechanisms of how the inverters clip and/or how the lost/dumped power is dissipated.

Overall, even though I'm mostly w/Peakbagger on inverter sizing, I'd still think/SWAG without more information that the overall annual losses to clipping for this application won't be all that much - at least not as much as the OP might have initially thought/feared.

The sizes are also of as the smaller is 7.1kw. They used larger pv modules than initially quoted.
they would have less clipping with the smaller pv modules as you stated but with the larger ones it is much more clipping.
Inverter efficiency should be 98.5%.

if OP has full access on monitoring then they can graph production for each inverter independently as well as DC voltage ( a big indicator of inverter forced clipping when it jumps up).

This significant time in higher voltage clipping for the inverter will likely lead to a shorter life.

[J.P.M.]

The efficiency is off in your model look at todays graph where OP was clipping from 11:00 to 2:00 pretty solid. As we don't have access to the individual inverter graphs likely the smaller was clipping longer as well.

I don't doubt that the model's inverter eff. is no more than a general # I was sort of hoping for a better est. based on your longer/greater experience w/SolarEdge.

Understood about today's clipping. Given the array tilt angles and the location, and the solar altitudes/zenith angles this time of year, my guess is the smaller array w/ the 41 deg. tilt ~ = 41 deg. latitude is clipping a bit more as f(POA input /inverter output) per installed STC kW than the larger array with the 26 deg. tilt.

Then too, I'm not sure we've definitively determined that the 15 kW output limit is from the inverter capacity or is inverter limited.

But, with respect to inverter efficiency, if that's what is limiting output, since the P.O.A irradiance input beyond that needed for a 15 kW output is of no benefit (and in theory and practice - as in from an energy balance on the array - may well lower system eff. a bit by increasing array temp., or the inverter(s)' temp(s.), or all 4 of the devices' temps. for no increase in useful output).

I suppose the argument could be made that increased inverter eff. could mean that the greater output from that increased inverter eff. is also lost in the clipping thereby making the absolute difference between the clipped and unclipped output a slightly larger number. That is, a higher inverter eff. would deliver more energy to the load than would a lower eff. inverter if the output for either inverter scenario were not clipped. But I guess that would depend on the mechanisms of how the inverters clip and/or how the lost/dumped power is dissipated.

Overall, even though I'm mostly w/Peakbagger on inverter sizing, I'd still think/SWAG without more information that the overall annual losses to clipping for this application won't be all that much - at least not as much as the OP might have initially thought/feared.

[ButchDeal]

I've been rooting around on your POCO's website and it looks to me like the size limitations on NEM systems is 10 MW.
Looking at your SolarEdge data, I'd say your output is inverter limited.

As a SWAG only, using PVWatts modeled output, using 10 % system losses and the information you provided, I got the following:

Your 12.350 kW array has a modeled DC output (to the inverter) of ~ 17,943 kWh/yr. or so. If the inverter took all of that and had the average PVWatts modeled eff. of 0.956 or so, the annual output after the inverter would be ~ 17,150 kWh/yr. If the inverter or some other force in the universe is indeed limiting your power output to 10 kW for that array, then the magic of the PVWatts model is suggesting that inverter limitation is causing ~ 3 kWh/yr. additional losses. According to the model, those losses occur on 17 hourly occasions over a TMY (Typical Meteorological Year).

Your 6.825 kW array has a PVWatts modeled DC output (to the inverter) of ~ 10,086 kWh/yr. or so. If the inverter took all of that and had an average PVWatts modeled eff. of 0.952 or so, the annual output after the inverter would be ~ 9,600 kWh/yr. If the inverter is limiting your power output to 5 kW for that array, PVWatts is suggesting that the inverter limitation is causing ~ 92 kWh/yr. additional losses. According to the model, those losses occur on 295 hourly occasions over a TMY.

So, in sum, the larger array appears to have an inverter that's likely to handle most of what gets crammed into it most of the time. Any losses will tend to decrease as the array ages and performance rolls off a bit.

Same goes for the smaller array. However, and as a perhaps educated guess, the greater apparent DC/AC ratio of the smaller array (1.30) vs the larger array's 1.10 ratio, plus the tilt difference from one array to the next are two factors that might account for the likely greater clipping losses for the smaller array.

IF it was my choice, I might have gone with a slightly larger inverter for the smaller array if the cost difference was small, but at ~ 100 kWh/yr. or so more that a larger array might get you for a few yrs. or less, I'm not sure the extra cost would be worth it.

Overall, and IMO only, I don't think you got screwed more than anyone else who buys PV without knowing what they're getting into, at least not as far as inverter sizing is concerned.

A caveat or two:

All this is based on modeled output. It's probably representative of some form of reality, but solar modeling, while it's pretty good when model inputs are reasonably close to reality, will always be an approximation, mostly due to weather variability. Reread the PVWatts help screens, particularly with respect to cautions for interpreting results.

The PVWatts output is a model for preliminary design. Most of what appears above is probably better called a bastardization of the model's intent and primary purpose. Even under the best of conditions, the model is not a predictor of actual output results, particularly over the short term - more like a probability distribution of possible system output over many years.

I used PVWatts inverter eff. of ~ 0.96 or so. Butch probably has much better and much more substantial experiential knowledge and info than I do with respect to Inverter SolarEdge inverter efficiencies.

Take what you want of the above. Scrap the rest.

The efficiency is off in your model look at todays graph where OP was clipping from 11:00 to 2:00 pretty solid. As we don't have access to the individual inverter graphs likely the smaller was clipping longer as well.

[J.P.M.]

Thanks peakbagger - that sounds consistent with what we've been reading about the interconnection tariff. The information we found did talk about a cap of 15 KW for the "simplified process" (based on the idea that at 15KW it was assumed the grid could absorb the impact, but anything higher than that it wasn't certain the grid could handle it). The link is available here: https://www9.nationalgridus.com/narr...on-process.asp

Based on our calculations though, a system with 47 panels (at 325 watts each) could have generated 15 KW. As I understand it, we didn't need 59 panels installed in order to produce the maximum amount of power we were going to be allowed to feed back into the grid. I understand the math might be a little fuzzy once you take into account location, tilt, panel degradation over time, etc., but since we're hitting that peak consistently and quickly every day it does seem like we have more panels installed than we can use (and, as such, we were oversold).

They neglected to tell us about these caps and told us we'd be getting a check for all this extra energy we'd be producing. I'm doubtful at this point that we'll even be able to offset all of our electricity given this cap.

While the technology may be new, consumer protection laws are not - if they did sell us an oversized system then it's going to be a problem for us and the installer.

I've been rooting around on your POCO's website and it looks to me like the size limitations on NEM systems is 10 MW.
Looking at your SolarEdge data, I'd say your output is inverter limited.

As a SWAG only, using PVWatts modeled output, using 10 % system losses and the information you provided, I got the following:

Your 12.350 kW array has a modeled DC output (to the inverter) of ~ 17,943 kWh/yr. or so. If the inverter took all of that and had the average PVWatts modeled eff. of 0.956 or so, the annual output after the inverter would be ~ 17,150 kWh/yr. If the inverter or some other force in the universe is indeed limiting your power output to 10 kW for that array, then the magic of the PVWatts model is suggesting that inverter limitation is causing ~ 3 kWh/yr. additional losses. According to the model, those losses occur on 17 hourly occasions over a TMY (Typical Meteorological Year).

Your 6.825 kW array has a PVWatts modeled DC output (to the inverter) of ~ 10,086 kWh/yr. or so. If the inverter took all of that and had an average PVWatts modeled eff. of 0.952 or so, the annual output after the inverter would be ~ 9,600 kWh/yr. If the inverter is limiting your power output to 5 kW for that array, PVWatts is suggesting that the inverter limitation is causing ~ 92 kWh/yr. additional losses. According to the model, those losses occur on 295 hourly occasions over a TMY.

So, in sum, the larger array appears to have an inverter that's likely to handle most of what gets crammed into it most of the time. Any losses will tend to decrease as the array ages and performance rolls off a bit.

Same goes for the smaller array. However, and as a perhaps educated guess, the greater apparent DC/AC ratio of the smaller array (1.30) vs the larger array's 1.10 ratio, plus the tilt difference from one array to the next are two factors that might account for the likely greater clipping losses for the smaller array.

IF it was my choice, I might have gone with a slightly larger inverter for the smaller array if the cost difference was small, but at ~ 100 kWh/yr. or so more that a larger array might get you for a few yrs. or less, I'm not sure the extra cost would be worth it.

Overall, and IMO only, I don't think you got screwed more than anyone else who buys PV without knowing what they're getting into, at least not as far as inverter sizing is concerned.

A caveat or two:

All this is based on modeled output. It's probably representative of some form of reality, but solar modeling, while it's pretty good when model inputs are reasonably close to reality, will always be an approximation, mostly due to weather variability. Reread the PVWatts help screens, particularly with respect to cautions for interpreting results.

The PVWatts output is a model for preliminary design. Most of what appears above is probably better called a bastardization of the model's intent and primary purpose. Even under the best of conditions, the model is not a predictor of actual output results, particularly over the short term - more like a probability distribution of possible system output over many years.

I used PVWatts inverter eff. of ~ 0.96 or so. Butch probably has much better and much more substantial experiential knowledge and info than I do with respect to Inverter SolarEdge inverter efficiencies.

Take what you want of the above. Scrap the rest.

[bcroe]

If your 3 phase is 3 X 120VAC neutral to line (208 line to line, like our lab) you should be able
to put 3 inverters line to line, adding 5KW to what you have. Bruce Roe

[Ampster]

That sounds like a good plan.

[LCF]

What i meant was allowing us to upgrade to 3 phase and then replacing the inverters. Not interested in removing panels. In order to add more capacity we need to be 3 phase which I believe has a higher limit of 25Kw. We had no way of knowing that the sales person was going to be tying our hands otherwise we would have insisted on appropriate inverters suitable for our long term goals which he was well aware of.

[Ampster]

Fix what? Reduce your DC capacity? I am not sure I understand why you cannot add more AC capacity in the future?
If I had the information you now have, I would have made sure i had specified inverters that were upgradable to 3 phase, especially when the inspector suggested the 3 phase upgrade path. The installer missed an opportunity for customer retention no doubt by not figuring that.

[LCF]

I had a similar situation, a 5.7kW DC to 3.8kW AC ratio. I had several simulations run and it turns out the the production of a smaller array to the same inverter would have been less kWhrs. I guess it all depends on whether you look at at the glass half full or half empty. The key issue for me with solar PV is the annual expected production. When I saw the numbers I was able to view the situation differently.

I still believe it would have been a more transparent process if they had explained why they were using such a high DC to AC ratio.

I certainly understand the numbers now that I came on this forum and have read some more information. I'm fully confident that the system will generate the amount of electricity that we needed to wipe out our current energy bill which is what we wanted. My problem is that it appears as though this is going to tie our hands from expanding to a larger system when we fire up our new business expansion which we were very clear with the sales person about being our goal. Putting in a new pole and new meter to one of the new buildings to allow us to add a second system won't be an option because we have the entire property tied to a massive brand new propane generator. We wish they explained this to us and allowed us to make this decision rather than making it for us and now our hands are tied. We are going to ask them to fix it.

[Ampster]

.........

While the technology may be new, consumer protection laws are not - if they did sell us an oversized system then it's going to be a problem for us and the installer.

I had a similar situation, a 5.7kW DC to 3.8kW AC ratio. I had several simulations run and it turns out the the production of a smaller array to the same inverter would have been less kWhrs. I guess it all depends on whether you look at at the glass half full or half empty. The key issue for me with solar PV is the annual expected production. When I saw the numbers I was able to view the situation differently.

I still believe it would have been a more transparent process if they had explained why they were using such a high DC to AC ratio.

[LCF]

Hi NewEclipse - I'm not sure we're missing that - we are hitting the cap daily, for hours every day, meaning there is excess capacity. In addition, we had extensive conversations with the installer about expanding the system as we brought additional buildings online - they assured us that wouldn't be a problem. It doesn't seem like that's possible now with a 15KW limitation - adding more panels won't actually produce more power, and fewer panels could have produced the same amount of power we're producing now. We just spent hundreds of thousands of dollars on a new commercial building on the farm (which will be filled with ranges, ovens and other electrical equipment), which the sales person toured. Our entire goal in going solar was for 100% of our energy to be generated by solar panels. It's something that's reflected on all our product packaging and something that now seems will be impossible given this cap.

The bigger picture seems to be that we had a salesperson sell us the biggest, most expensive system he could, even though it's producing more electricity than we could possibly sell back to the grid. To sell us the system he withheld material information that would have been fundamental to our decision making on whether we installed this system. If we knew there was a cap we would have upgraded to three-phase power and insisted on inverters with higher capacity.

[neweclipse]

I think you may be missing a bigger picture here...

Had you a 15kW system you would NOT hit 15kW early and for a longer part of the day or through periods or days of overcast skies...really, being over paneled is giving you much more production than a a perfect match at 15kW is capable of given same conditions.

Having 19k W will give you much more production and much more surplus to sell than having only 15kW rated size.

Send your installer a nice Thank You Card or Gift Card.

[LCF]

Thanks peakbagger - that sounds consistent with what we've been reading about the interconnection tariff. The information we found did talk about a cap of 15 KW for the "simplified process" (based on the idea that at 15KW it was assumed the grid could absorb the impact, but anything higher than that it wasn't certain the grid could handle it). The link is available here: https://www9.nationalgridus.com/narr...on-process.asp

Based on our calculations though, a system with 47 panels (at 325 watts each) could have generated 15 KW. As I understand it, we didn't need 59 panels installed in order to produce the maximum amount of power we were going to be allowed to feed back into the grid. I understand the math might be a little fuzzy once you take into account location, tilt, panel degradation over time, etc., but since we're hitting that peak consistently and quickly every day it does seem like we have more panels installed than we can use (and, as such, we were oversold).

They neglected to tell us about these caps and told us we'd be getting a check for all this extra energy we'd be producing. I'm doubtful at this point that we'll even be able to offset all of our electricity given this cap.

While the technology may be new, consumer protection laws are not - if they did sell us an oversized system then it's going to be a problem for us and the installer.

[peakbagger]

FWIW, 15 KW may be a regulatory cap. Many utilities have net metering caps which limited the system size. In NH the cap was a 10 KW array as determined by the installed inverter capacity. The array capacity could be larger but the inverter could not exceed 10 KW. In theory someone with storage could cheat by installing batteries and charge up the batteries during the day and export at night as long as the export inverter was 10 KW. I am using past tense as the cap may have been increased and expect the storage issue may have been addressed. Small systems are installed under an expedited interconnect where as long as the inverter is 1741 rated they are presumed to have no significant impact on the electric system. Anyone over the cap has to file and pay for a standard interconnect study and permit (which can be quite expensive). The permittee has to pay for any necessary upgrades to the electric system. Once they do get connected all they get is the standard commercial rate (which is far less attractive than net metering) plus SRECs that are not worth much in NH.

I expect this may not be the case with the OP but once three phase gets in the picture things can get complicated as the system may get out of the utilities expedited small solar permit exemptions.