Why do Dave's new 370W LG solar panels use lower rated 295W Enphase microinverters?
What gives?
Inverter clipping, solar panel overrating, DC/AC ratio, solar irradiance, and microinverter underrating explained.
And a look at data from Dave's old 3kW system to predict how many days a year the new 5kW system will clip.
I have a similar system in Melbourne. 12 panels, 335W Neon2, 270W peak Enphase S270s. I wondered what effect clipping would have, so I made display showing the generated power vs consumed, $, and predicted solar and predicted clipping. The attachment shows a good day, 8 Jan 2021. The solid red line is the predicted power; the dotted orange line is the predicted clipped value. My calculations don't take into account ambient light (that was too hard), which is why there is a large error in the morning and evening (5 panels get more morning sun, 7 get more afternoon sun).
Clipping starts mid-September and stops mid-March.
Have you made your PV data publicly available anywhere to download as a CSV? I’ve looked at what you’ve put up on ‘pvoutput.org’ and from what I understand I can’t download historic CSV data only the owner can? I ask because I have some MATLAB and R scripts I wrote awhile ago for Energy Queensland to visualise half hourly consumption and PV generation trends and I’m interested to see exactly how much those microinverters would have clipped your generation.
I have a similar system in Melbourne. 12 panels, 335W Neon2, 270W peak Enphase S270s. I wondered what effect clipping would have, so I made display showing the generated power vs consumed, $, and predicted solar and predicted clipping. The attachment shows a good day, 8 Jan 2021. The solid red line is the predicted power; the dotted orange line is the predicted clipped value. My calculations don't take into account ambient light (that was too hard), which is why there is a large error in the morning and evening (5 panels get more morning sun, 7 get more afternoon sun).
Clipping starts mid-September and stops mid-March.
Interesting, thanks.
Will need a full year to really analyse mine.
Have you made your PV data publicly available anywhere to download as a CSV? I’ve looked at what you’ve put up on ‘pvoutput.org’ and from what I understand I can’t download historic CSV data only the owner can? I ask because I have some MATLAB and R scripts I wrote awhile ago for Energy Queensland to visualise half hourly consumption and PV generation trends and I’m interested to see exactly how much those microinverters would have clipped your generation.
Apart from what's public on pvoutput, no.
I had to download in yearly chunks, and then I just got daily totals, haven't tried to download 5min data yet.
There's actually another reason why you would oversize your system: around here (weel, they're changing it atm) you pay a fee /kwh yearly and that is based on the inverter max power, not your panels. The fee is around 100$/kW, so with a 2kW inverter you pay 100$ (/year) less than a 3kW and as Dave says: you don't loose that much.
Truly affordable compact long term efficient home power storage needs to come about so you can truly say F-- to these stupid utility charges to send power back and go for the highest reasonable amount of power you can generate.
High efficiency electrolysis hydrogen fuel cells are hopefully coming with claims of 50% efficiency, but 20 years life span and huge kw storage.
https://youtu.be/0_bTjcjqN6c?t=269Maybe one day we will get even better efficiency or better batteries.
Already have a battery. And if more E-cars get LiFeO batteries, using those becomes more appealing as well...
It is not surprising to get the highest peaks on partially cloudy days: it is usually not that hot and when the clouds don't give shade, they can scatter additional light. This way one can get more than 1000 W/m², AFAIK up to some 1300 W/m² at peak times. However the peaks are often only short time and interrupted from times where the cloud blocks the sun.
It also makes sense for the gird to have the inverter smaller then the panels: this reduces the peak power and gives slightly more power in off peak times. Quite often the grid may be saturated at times with peak PV production. In some areas in Germany they strated to turn on street lampes at times with too much power from PV.
I was just watching this on full screen. I spotted all the numerous tabs open at the top of the screen and unplugged my network cable in case a dodgy virus was at work. Then I realised it was your computer screen being shown with 14 or so tabs open not mine
My friends 250kW wind turbine 'clips' if you want to call it that on windy days. The end 6 feet or so of the blades rotate to oppose the airflow to act as a brake when power gets upto 250kW. The three blades could maybe make twice the power from the strong winds but the gearbox and alternator are sized for 250kW and that is when the blade tip braking comes in to limit the power. The 'oversize' blades make a good amount of power for the majority of the time even in light winds, they just need to be air braked when the wind is strong.
He is now looking at using 100 acres of field for solar panels too, which makes my 4Kw system look miniscule.
For the last week or so we have had really good amounts of sun. We are not grid connected so inventing ways to use that power. I have a water distiller going so I make my own battery water which uses about 2kwh per day and a garden irrigation water pump going too just to make use of some of this power.
And it's a 10 year warranty on the microinverters, 5 years on the Envoy controller.
Their warranty is 25 years here on the inverters. Not sure why the difference.
And it's a 10 year warranty on the microinverters, 5 years on the Envoy controller.
Their warranty is 25 years here on the inverters. Not sure why the difference.
25 years seems crazy, are you sure?
It is not surprising to get the highest peaks on partially cloudy days: it is usually not that hot and when the clouds don't give shade, they can scatter additional light. This way one can get more than 1000 W/m², AFAIK up to some 1300 W/m² at peak times. However the peaks are often only short time and interrupted from times where the cloud blocks the sun.
You can get more than 1000W/sqm on a cloudless day, it's not some magic limit, the suns output varies.
Follow-up:
A very interesting comment I picked up on at the 9:38 mark was that your power dropped down by like 90% when you said the "Shading" hit them in the evening. I know you have the older SMA Sunny Boy Inverter, it would be a really interesting to see if the new model of your inverter with Patented "Shade Fix" technology works. Does it really fixes most of the problem?
A very interesting comment I picked up on at the 9:38 mark was that your power dropped down by like 90% when you said the "Shading" hit them in the evening. I know you have the older SMA Sunny Boy Inverter, it would be a really interesting to see if the new model of your inverter with Patented "Shade Fix" technology works. Does it really fixes most of the problem?
Fascinating, but apparently 'Shade Fix' is just a new version of OptiTrac Global Peak, which was introduced by SMA in 2009. No idea whether Dave's inverter has it
or if it is turned on.
https://www.sma-america.com/partners/knowledgebase/optitrac-global-peak.htmlOptiTrac Global PeakMany SMA inverters are equipped with OptiTrac Global Peak.
OptiTrac Global Peak is an additional function of the standard
MPP tracking OptiTrac and is deactivated by default. Please
visit our website for more information.
When a module is hit by the shaddow, one often looses the whole votlage or at leat a good fraction of that panel. The details depend on how they have wired diodes at the panel to cope with shade. They need some diodes for protection (to prevent excessive reverse voltage for some cells), but the details can vary.
With some inverters they need a significant minimum voltage to operate. With too much votlage drop due to shading, the MPP tracking may no longer work and give an extra hit on the performance.
The highest peaks on partially cloudy days are from a combination of direct sun-light and additional light scattered off clouds that are a bit off the direct path. In te extemeone gets direct light of a clear day plus nearly the intensity of a cloudy day.
Temperaute may give some 10% extra efficiency, clouds can give some extra 20%, maybe 30% under very favorable conditions.
The Panels likely get quite hot an a low wind sunny day. I remember a demonstration of litterally fried egg on a car-roof - they did not hit the 100 C, but it was close. This was in new mexico, so with really a lot of sun and thin air at some 2000 m.
I think the "shade fix" needs optimizers on the shaded panels to work. The difference is that only the shaded panels need the optimizer, not all of them. In theory, those optimizers can also be used to allow replacing a damaged panel with a new one of a different model, although I think the newer one should be of a lower current output (voltage can be higher as long as the total string voltage remains within inverter limits) since I'm pretty sure it can only buck, not boost.
