spring versus summer system results

Many system owners have probably seen changes in the performance of their systems between spring and summer. The days are longer and the sun is higher in the sky but it's also warmer and the sky is a bit hazier.

I have been able to collect various pieces of data on my system and have graphed them and thought I would share the results. First, the system is in south east Hants and faces roughly 170 degrees on a roof with a 40 degree pitch. It's 16 Sanyo 250w panels with an Aurora PVI-3.6-OUTD inverter. It's one of the later ones where the output is limited to 3.6 kWh.

Each of the 2 graphs shows the system output measured in watts as a purple line against the left axis. Against the right axis is the wind speed in blue (the zig-zag line along the bottom), the outside temp in pink, the temperature in the loft in yellow, the temperature on the back side of the panels in teal and the internal temp of the inverter in brown.

The first graph is from 2 May. Maximum temp during the day was about 18 degrees (ignore the rise to 20+ around 18:00 as the temp sender was exposed to the sun). The power output stayed at 3600 watts for quite some time due to a combination of cooler weather, a nice breeze, the angle of the panels and what was probably a clearer sky. The panel temp peaked at about 35 degrees. Note the loft temp temp peaked long after mid-day. I think the panels shaded the roof until late in the afternoon when the side of the house started to bear the brunt of the sun so the loft heated up. Internal inverter temp pretty much corresponds to power output. Total power for the day was 24.7 kWh.

The first graph is from 8 July. Maximum temp during the day was a summer-like 28 degrees (again ignore the rise to 30+ around 18:00). The power output peaked at about 3400 watts - probably due to the warmer ambient temps and higher panel temp (already 40 degrees at 11:30), less breeze and likely slightly hazier sky. The panel temp peaked at about 44 degrees. Again the loft temp temp peaked long after mid-day. Internal inverter temp pretty much corresponds to power output. Total power for the day was 22.3 kWh.

Questions and comments welcome - will try my best to answer them.

Interesting information, thank you. I have the same Aurora inverter and also find a difference between spring and summer, but more marked. my panels are maxing out at 3KW over the last few days, whereas in spring they go right up to 3.6kw (3.92kw panels). Temperature obviously has a key part. I live on a busy road and the panels get dusty, when I cleaned them earlier,power shot up to 3.3kw and then slowly dropped back down again to 3kw over the next 15 mins as they next warmed up again. I would be interested to know what products you are using to capture the temperature information as I would like to do some similar work for my system too.

I have an idea to overcome the 3.6kw limit on the Aurora inverter, if one or two of the panels are moved off the inverter to micro inverters, this will create capacity on the inverter and possibly reduce its temperature as well. From the manual, the aurora begins to limit output once the Temperature gets over 50 degrees, so could be a big win. I haven't done any cost calculations but that's not so important for me, If I have a 4kw system, I want to get 4Kw out of it!

All the best,

Whilst inverter de-rating over 50degreeC can have an impact, what you are more likely seeing is the impact of temperature on the module's voltage and power output - it's an inversely proportional relationship to the temperature and some panels will drop their voltage/power output more per degreeC increase than others....

Thanks for the comments so far. The inverter temperature is as reported by the inverter itself and represents its internal temperature. As far as de-rating goes here's an extract from the manual:
"For ambient temperatures below 20°C fans are off . No power derating.
• For ambient temperatures between 20°C and 25°C there is no power
derating. Fans are turned on and off according to a pre-set cooling
profile.
• For ambient temperatures between 25°C and 40°C there is no power
derating. Fans are kept on at a variable speed depending on
temperature.
• For ambient temperatures between 40°C and 55°C there could be a
power derating, even with the fans at max speed, with a minimum
output power of 2500W when temperature reaches 55°C."

As the ambient temp is only in the mid 30's there shouldn't be any de-rating occurring. And if there was I would guess a step-change in the output would be visible.

The other three temperatures are measured with an old Oregon Scientific weather station. I think the senders are accurate to within a degree or two. One sender is held in place against the back of a panel; when I checked the temp of the front of the panel with an IR thermometer they were within a degree of one another.

According to Sanyo's data the panels should only produce about 6% less power at 44 degrees than at 25 degrees so 4 kWp of panels should still produce roughly 3.76 kWp. I suspect the hazier sky is having an effect. And I should calculate the best roof pitch for those 2 dates and see how far off my roof pitch is.

I have the Aurora PVI-3.0-OUTD with 13 250w suntech panels, today the inverter maxed out at 2440w, in cooler temps the inverter has reached 3303w.....do you think the 2440w is too much of a drop??

My system peaked at around 3500 watts today; the panels are capable of 4000 watts under standard conditions. Your panels are capable of 3250 watts under standard conditions and today they produced 2440. So roughly speaking my system was at 1/8 less than peak and yours 1/4 less than peak. But it's not that simple. What is your roof pitch? Where is your inverter located and how warm does it get in that space?

A drop to 2440w is quite big, but I have seen similar figures. Have you been monitoring this over a few days?

SolarCity said:

A drop to 2440w is quite big, but I have seen similar figures. Have you been monitoring this over a few days?

Click to expand...

Yes.....during "full sun" days peak output is consistently around 2500w, today is was 2513

I've checked a few systems and figures are showing a comparable drop to be honest. The norm seems to be around 20% less than maximum yield. It is very hot and very hazy - not particularly great for good PV yields. When did you last notice your system peaking?

We've seen this effect consistently across loads of site that we monitor were the peak output and power days are actually the spring and autumn equinox and they beat midsummer.

Measured the temperature on our panels yesterday as equally saw a drop in output at around 1:30pm, panel temperatures 52.8°C.

Just measured them 2 minutes ago today.. Outside Air Temperature is 23.3°C Black on Black Mono: 46.8°C Silver on White Mono :43.2°C Surface our of decking area outside the office (great for iced tea on days like this ) 38.6°C power output is down by about 20% from cooler days.

Inverters are in ideal location, outside not in direct sun, plenty of ventilaltion.

Worcester said:

We've seen this effect consistently across loads of site that we monitor were the peak output and power days are actually the spring and autumn equinox and they beat midsummer.

Measured the temperature on our panels yesterday as equally saw a drop in output at around 1:30pm, panel temperatures 52.8°C.

Just measured them 2 minutes ago today.. Outside Air Temperature is 23.3°C Black on Black Mono: 46.8°C Silver on White Mono :43.2°C Surface our of decking area outside the office (great for iced tea on days like this ) 38.6°C power output is down by about 20% from cooler days.

Inverters are in ideal location, outside not in direct sun, plenty of ventilaltion.

Click to expand...

Thanks for this, can you monitor panel temp with power one inverters, if so how?

No these are thermometers speicifically designed for data capture and ultimately home automation - part of our comprehesive logging system (17 sensors and growing - openenergymonitor.org)

The changing angle of incidence between the suns rays and the solar panel is important.
If the angle of incidence between the sun's rays and the solar panel isnt optimum, it wont generate well regardless of light intensity, temperature, length of day etc.
Direct light contributes more to generation than diffuse light.
Modal peak solar generation tends to peak within an hour of midday in N West England,
when the sun is highest in the sky. In spring the UK sun tracks lower with a shorter orbital track. 30 Degree south facing roofs generate well at this time of year because the angle of incidence between the suns rays and the solar panel is closer to its optimum 90 degree angle. Seasonal wind and rain also help cool the panels.
In summer the sun tracks higher in the sky on a longer orbital track.
The angle of incidence between the suns direct light and the solar panel reduces at this time of year, average temperatures increase, weather tends to be more stable, less windy.
Fixed inclined 30 degree south facing solar systems see a reduction in midday peak generation, but generate more KW per day because they are exposed to light longer - as days lengthen.

Vegelen said:

The sun's orbit changes throughout the year.

Click to expand...

Galileo would argue otherwise.

SolarCity said:

Galileo would argue otherwise.

Click to expand...

yeah, but then he also got burned at the stake for his troubles didn't he?