Winter generation

We’ve had our installation since 25/11/2022. It’s a 4.3kwh, hybrid inverter system, with a 5.2kwh battery. Panels are 12x360w. 2strings (x5 south facing, x7 south east facing).

Over that period, a simple calculation yielded the following (figures from the GE app):

Total generated: 51.84kwh
Number of days: 39
=1.33kwh per day. Divide that by 4.3

This works out to 0.31kw of power generated per kWh of PV on the roof per day on average. Needless to say the battery hasn’t really taken much charge from generation.

Now I know it’s winter, cold, gloomy etc. but how does that compare with other people? I appreciate that conditions differ person to person, install, location in country, climatic impact etc. For what it’s worth, I’m in south Leicester. I think it might still be valid to get rough comparisons….

I've just run a report from 25/11 to yesterday - 38 days I make it? - we generated 95.7kWh - so an average of 3.6kWh - which is more than I expected TBH.
Our system is 50:50 split south:west and about 4.6kwp.
We are South East of Birmingham so not that far from you.

South Oxfordshire, 12x390W panels (7 ESE, 5 SSW), 146.9kWh for Nov and 103.3kWh for Dec. Bright sunny day will give ₇.5kWh, heavy cloud and it's down to 0.5kWh.

anglefire thanks for doing that. The difference between the 2 is, I think, quite large. At 38 days ours is 0.31, yours 0.54. That’s like a 40% difference. I’m going to be speaking to Eon tomorrow as there is defo something not quite right.

Jumbo99 sounds about right. Expect 8 - 10x that amount in June/July as long as it doesn't get to hot.

December (31 days) - generated 166 kWh. 14 panels x 400W Perlight. North Wales.

We have a thread running on OCUK forums where generation numbers is a common topic, and a small pvoutput team that we submit to for real world comparisons. Thread link here if you feel like signing up and joining in!

I have 13 x 370W panels in a South (6 panels)/ West (7 panels) 2 string orientation based in Hampshire, October was around 300kwh, November 150kwh, December was 110kwh or so. I can see from the inverter data that my South facing panels are doing most of the heavy lifting at this time of year, but I get some shading as the sun is so low and I have some taller houses near me.

Happy enough with these figures and more or less in-line with expectations. interesting to see where 2023 takes things.

LordHippos that’s interesting. How are you getting data from the inverter for each string?

Jumbo99

If you go into the GE site, then into My Inverter => PV Data, you can see the string performance and even go back and look at previous dates.

The image below perfectly shows what I mean really, it's for today but you can see my String 2 is performing much better than String 1, it's nice and sunny out but String 1 is my west facing set, and they won't get any sun until the afternoon.

Right click the image and open in new tab for full size! PPV1 is string 1 in the table, PPV2 is string 2 in the table.

LordHippos what's the name of your PV Output team?

stevelewis

This one - https://pvoutput.org/listteam.jsp?tid=1661

People are welcome to join I think but it's not my team, I'm just a member. UK focused numbers and I guess it's preferable for people who are in the team to also be part of the forum discussion in the thread I linked above.

Mid-Wales 900ft altitude - misty, foggy, grotty
Panels 6.84 kW split East/West

November: 97 kWh generated = 20% of our needs
December: 57 kWh generated = 11% of our needs

Gatwick, Surrey.
Weather round here has been abismal for December!!

14 panels - max genteration approx 5.4 KW. All panels south facing with no obstructions.

Sept - 449 kWh generated
Oct - 428 kWh generated
Nov - 213 kWh generated
Dec - 160 kWh generated

Rugby based system made 120 kw from a 3.55kw system in Dec
Compared to Dec 2021 59kw 2022 was a good year…

LordHippos

Here's mine for comparison: https://pvoutput.org/intraday.jsp?id=18903&sid=16752

175 kWhr for December in Kent. Best day was just over 16 kWhr. 16 off 400w panels facing ESE and SW.

Preston, Lancashire

11 x 370w panels facing south max gen 4.1kw system 3.6 Gen 1 Hybrid.

Sept 383 kw generated
Oct 262 kw generated
Nov 121kw generated
Dec 105kw generated

With your setup i would say in summer, i would say you should be getting around 30kwh + a day

It's just unlucky you was installed in winter, but you could say that's a good thing as your fully set for this summer

Just a note, winter is crazy bad, some days you could get 1kwh solar or less on your size of system, all depends on the weather

Soon it will be march, that's when solar starts to pick up, you will start to see things pick up nice

lifco thank you, actually all of you!! Great community here!!

4.5kW array, SSW facing in Suffolk, Gen1 3.6HY

October - 344kWh
November - 130kWh
December - 137kWh

There's far to many variables to make comparisons worthwhile. Much better to set a (free) account up on Solcast, input your details and then compare the forecasts to your actual outputs. I find that Solcast is usually very close to the actual numbers.

I think considering everything, we y it st seem to have it a bad month for day light where we are. It’s been very gloomy and dark here since the install and that’s probably the cause. Looking at PPV data for both arrays, they are working so I suspect it is just the light.

New 6,4kW system with 9.5kWh battery. SSW facing in Wiltshire. Generated 145kWh in the first 25 days, up to the end of December.

Jumbo99

I'm in South Yorkshire, have almost the same set up as you, but facing a few degrees West from South and get a bit of shading this time of year from trees over the road and I got 84kWh over the same time period if that helps.

In July and August I got 540kWh and 590kWh respectively, roll on summer 😀

Jumbo99 to reassure you a little... I have a 8.64kwh system split 50/50 over my east and west roofs on the house.

Today I have generated 0.3 kwh. And I'm not expecting to generate anything for the test of today. Lots of dark clouds out there.

There seems to be a floor level of light for my system (a Solis). If you're below that then the generation is literally zero. Once you're above that is scales nicely.

Just the way it goes sometimes.

Jumbo99 We have a 10.08 kwp system split 50:50 East/West. It's been running for 12 days and we've done just over 30kwh, so around 90kwh extrapolated for the month. This is about what we expected given low sun, East/West and some trees on the south side. However, come March/April we expect that to start increasing considerably. The batteries and little bit of solar are easily covering the house load and we expect quite considerable excess in summer. The key at the moment is to avoid the punitive peak energy rates.

One thing that occurred to me Jumbo, are you sure that both strings are connected up? My parents had 2 strings fitted on their bungalow back in 2010/11 and the installer did not connect one of them up. It's only because I checked that it got sorted out.

Jumbo99 4.4kW panels, SSW facing, little bit of an obstruction from a tree during winter months and for the same period we generated 113.7kW in North Lincolnshire. Surprised yours is that low, but then having a the split of 5/7 might be a larger issue in winter as you won't get the peak solar generation hitting all panels at the same time? Just a thought.

sponev Well I'm clearly still learning. Just logged on to the main desktop portal and see that we've generated just over 50kwh in 12 days. The 30 kwh I quoted earlier is the generation to the house shown in the app and did not include generation excess to the battery. Pleased I checked!

Yes the default chart under Energy is home, need to switch to generation instead to see that!

LordHippos Thanks for that. I just followed your advice to view the string performance in the inverter settings of the portal - fascinating. Really handy to check that all looks well, particularly as we opted to go without optimsers as we have no real shading as such.

It’s really weird that ours appears to have generated less than pretty much everyone commenting.

@Jumbo99
I have a 5kW array West (slightly north) facing so not the best, which was installed end July Near manchester = often cloudy!
Aug = 500 kwh
Sept = 307
Oct = 182
Nov = 67
Dec = 43 ....

My numbers are much lower than others (i'm very jealous!) , but it reflects the angle and position of my panels.
A colleague has a similar array, but south facing and easily gets 2-3x my generation on good days.

I think your system is probably fine, but I would check that both your PV1 and PV2 strings are showing generation in the portal. One of my strings wasn't properly connected by the installer...

TRBob that was my main concern. Having looked at “My Inverter” on the GivEnergy cloud portal, I can see that Ppv1 and Ppv2 are both generating.

Since my system was installed on 10th November it has generated 345kW, 18 x 370 panels, 5kW Inverter and 9.5kW battery.

7.02 kWp in 10:8 split S. West/ N. East. the NE aren't doing much at present but do match the SW on bad days!
2022-04 697.18
2022-05 815.53
2022-06 899.34
2022-07 914.08
2022-08 780.88
2022-09 515.24
2022-10 376.94
2022-11 189.9
2022-12 143.87

Jumbo99 , have you used the PVGIS5 web portal to enter your panel and orientation details to estimate the expected output for each string, then compared with what you actually got?

https://re.jrc.ec.europa.eu/pvg_tools/en/

You can select your house location, and input panel orientation and roof slope angle.
I measured a window surround, then counted bricks horizontally and vertically, then counted again on the triangle of the gable end of the house. Using schoolday Maths, tan(angle) = gable height/(0.5 x house width).

Run the pdf report for each of the 2 strings, then add the 2 figures together for each month of the year.

With 10x410W West and 6x410W East at 30 degrees slope, my Dec22 yield was 74.8 kWh, so 81% of the PVGIS5 estimate of 91.8 kWh.

We are in Sheffield, and have 18 x Jinko 420 kW panels, so 7.56 kW in total, installed at the beginning of November. G1 5kW inverter and 9.5kWh battery.

String 1 is 9 x ENE
String 2 is 6 x WSW and 3 x SSW

November: 90 kWh
December: 64 kWh

@Jumbo99 Compared to your 0.31 kWh per kWp, I therefore have a figure of 0.27 in December, and 0.40 in November, which suggests we're in the same ballpark as you.

@sponev provided 50kWh over 12 days with 10.08 kWp, which equates to 0.41 kWh per kWp in December, which is much better - however they may be further South and/or seeing more sun.

I've got a similar concern, and have started wondering if either the 6 or the 3 panels on string 2 are possibly disconnected. Looking at the numbers people have provided above reinforces that concern. You can't tell of course as both sets of panels are on the same string and the inverter is reporting output for that string, however the actual output of string 2 is a little less than the forecast for either the 6 panels or the 3 panels. It is considerably less than the forecast for both together, whereas the 9 panels on string 1 are only a little less than forecast. In addition the peaks and troughs for the actual output most closely match the forecast for the group of 6 panels.

My forecasting is based on Forecast.Solar, but I'll try Solcast to see if there is much difference.

That said, the installer has mentioned that string 2 is more shaded, so perhaps that accounts for it (we also have optimisers involved too). My gut is telling me otherwise, something wrong, but my head trusts his expertise!

andyn ........................are possibly disconnected......
If you look at the string Voltages it should give an initial indication

Just West of Norwich

11.4kw of panels
4.6kw West facing 280 degree
6.8kw East 100 degree (4.4kw was added 26th October)

Nov 213.78kw generated
Dec 179.1kw generated

Oh how I wish my roof faced south :-(

Simon_C thanks, I’ll try this!!

anglefire My system is most similar to this, but 5.6kWp of panels on 2 strings SE/NEW in Devon, total generation 133kWh over the same date range 25/11 to 2/1. I think it's been very overcast apart from 2 or 3 days. We had ours installed roughly 1st Sept when the weather was lovely and generated about 30kWh on the best day. I'm glad we have the 8.2kwh battery as it just about covers our daily usage (with some care!). The only time we struggle to stay within the battery capacity is at weekends when we are both at home, so we have learned to spread out the washing during the week and use the oven less!

10x405W East + 7x405W west installed mid November. Just South of Bristol.

November (Partial month) 77.3kWh
December (first full month) 100.6kWh
January as of time of post 16kWh

West Sussex 15 panel 5.85kw system split 50/50 ESE/SSW on two sides of roof installed end of September

October 404kwh
November 180kwh
December 113kwh

SilverArt

Ah, that's a good idea, thanks!

I'm not sure how to interpret the voltages, but my guess would be that because they are similar though a little higher (bearing in mind the same number of panels, and 3 x SSW + 6 x WSW on string 2 versus 9 x SSW on string 1), that suggests both sets of string 2 panels are connected correctly?

By the way, part of the reason I'd like to understand this better is that we have the opportunity to add an additional SSW panel, and a second battery. At the moment it feels as though SSW output doesn't justify this, which seems very odd!

Last 5 sets of data this morning (sunny, and at this time of day string 2 is mostly shaded):

Time VPV1 (V) VPV2 (V) IPV1 (A) IPV2 (A) PPV1 (W) PPV2 (W) PPV (W)
10:35:58 254.7 285.4 2.8 0.3 734 94 828
10:30:56 251.9 301.6 4.1 0.3 1056 104 1160
10:25:54 253.7 291.9 3.7 0.4 962 113 1075
10:20:52 244.7 295.7 5 0.4 1226 119 1345
10:15:50 250.3 295.3 5 0.4 1263 120 1383

Here is Wednesday from 1pm (sunny-ish day round here):

13:24:53 306.3 283 0.8 0.8 260 246 506
13:19:51 305.5 304.5 0.6 1 198 308 506
13:14:49 307.3 267.3 1.3 1.7 429 475 904
13:09:47 309.6 267.4 0.9 1.1 286 303 589
13:04:45 307.1 222.9 1 1.2 304 281 585

Edit: Well it's well-formatted when I post it 😂

So it seems the higher the generation, the lower the voltage on that string?

What concerns me is that we just never seem to see the same highs on string 2 as string 1 (roughly a half as you can see from these two examples). But maybe this is a Winter thing - afternoons more likely to be cloudy than mornings, or morning/Eastern sun more intense than afternoon in the West.

I can't think why the latter should be, but my geography teacher wife tells me that cloud is more likely in the afternoon (generally) as the air warms during the day, and warm air rises, which aids the creation of cloud. That's way beyond my pay-grade though! 🤷‍♂️

If that's the case, I'd expect Winter (when other factors also encourage cloud formation) output for South and West facing panels to be proportionally lower than output for East-facing panels than in Summer. In Summer when cloud is less likely, South and West should contribute more (proportionally) I guess.

andyn I would suggest that the cloud in the afternoon thing also depends on where in the country you are. Quite often weather fronts hit the west country during the day and in the evening in East Anglia?

andyn

Unless your panels have optimisers on, I am not sure if having a string with two different orientations of panels makes a lot of sense. Shading on any part of the string can impact the generation of the whole string as it has to work in concert as one entity so far as the inverter sees it.

I have 6 panels on a S facing string and 7 on a W facing string, the S facing ones are doing significantly more than the W facing ones at this time of year, but if I had those on 1 string I think the W facing ones would be dragging down the performance of the whole setup quite a lot.

Do your panels have optimisers on them to handle individual panel performance?

Not an expert on this by any means, but I wonder if that's part of your issue.

andyn The numbers appear to be sensible compared to what I see on my system.
It looks as if String 2 is in the shade in your 10:15 - 10:35 slot but by 13:04 is is holding its own, so no shade(?) but there was a bit more cloud cover..

Cloud cover is more complicated, as I understand it the panels can benefit from reflected rays off of clouds and from rays focused by the clouds as well as the obscuring effect, plus, add to that the effect the solar panel efficiency dependant on panel temperature

LordHippos Yep, there are optimisers on string 2, in fact a couple of panels on string 1 too, as they are in the lee of part of the roof. However, it may be worth checking if they are working/connected properly - that's a good thought, thanks!

I'm doing a calculation in Home Assistant to work out real-time efficiency of PV Power against "theoretical maximum" at each point in the day.

So regardless of the time of year or Sun position, I can know if the PV is behaving correctly. Basically if the sky's clear* then the figure is always 100%. It's a bunch of Sin & Cos - I can post the calculation if anyone's interested.

Graph from yesterday. You can see I've got a lot of shading through either side of midday (which I expect) and then I hit my 100% around 13:30 - 15:00 when the sky was clear.

*and it's daytime...

Riscy Yes please, thats kinda interesting stats 🙂

East facing, 5.8kw system and thus far in January only 57.6kw😩

My system was finally finished on the 9th of December. 4.5kw south facing. I'm getting terrible numbers of between 1 to 3kw per day (3kw being this week which has been sunny and fairly cloud free.

However, as the sun is so low at this time of year I get lots of shading moving across the panels from nearby trees. More or less as one trees shadow goes another one takes over. I also have the disadvantage of a van der valk (valkpro+) ground/falt roof mounting system where by the first row of panels partial shades the second row and the second row partially shades the third row for part of the day again because the sun is so low in the sky. I wish the ramp system had been available but sadly it wasn't and I didn't know about this design issue with the valkpro+ until after it was installed.

As the sun gets higher in the sky the shading from trees and self shading from the mount system will stop being an issue but still a bit disappointing.

I'm wondering about getting tigo optimisers installed, but I've seen mixed reports as to their effectiveness. I think I'll make a post about them.

The battery system which I debated its worth in getting has been amazing though, I think the battery will probably save me more per year than what the panels do.

As you have shading optimisers would help. But obviously that is at extra cost.

THALL I doubt there will be any real benefit. The system is only a few months old. Unless the installers were very naughty and installed old (even very old) panels that are not split cell and have no bypass diodes and the inverter is not MPPT I think optimisers would be a waste of money. Optimisers and even micro inverters are a solution to a problem that has largely been solved by newer panel and inverter technology.

Its a HY 5.0 (GE) inverter. With Eurener 375w Diodes by-pass built-in (3/6) I'm pretty certain they're split cell.

TimI Sure! Taken from my Node-RED flow, this gives you a Scalar (i.e. a number from zero to one) saying what the potential output of your array should be, with the sun in that position.

Obviously I'm already grabbing the sun's position (elevation and azimuth) from somewhere else. Exercise left to the reader! Multiply this by your array's installed kWp figure, and then divide your current PV output into that. And voila, tells you your efficiency percentage.

  function rads(deg) {
      return deg * (Math.PI / 180.0);
  };

  const inc = 45;  // Inclination of the roof
  const azi = 180; // Azimuth of the panels (e.g. 180 degrees is South)
  const alt = msg.sunpos.attributes.elevation;
  const omega = msg.sunpos.attributes.azimuth;

  const scalar = Math.abs(Math.cos(rads(inc)) 
              * Math.cos(rads(azi))
              * Math.sin(rads(alt))

              + Math.sin(rads(inc))
              * Math.sin(rads(azi))
              * Math.cos(omega)
              * Math.sin(alt));
`

Riscy nice job!

Or, for those who prefer the easy option, upload the generation data to pvoutput.org and get the efficiency info straight out of the box.

I am lucky enough to have an almost perfect location for my 10kw ground-mounted array, facing south on an elevated site with no obstruction for most of the day, this is my record for 7 years. at the time of installation, there was no really viable storage solution so I just took the FIT payments and the free electricity during the day, it took the invasion of Ukraine and the resulting energy price hike to motivate me to try and use all this power i was almost giving away to the grid. In April 22 i got an IBoost fitted to stop using the LPG boiler for hot water and ordered an AC coupled inverter and 9.5kw battery. that was on a long lead so didn't actually arrived until November and i also got a second battery, so by 2nd December i had commissioned my system with 19kw of battery. just in time for the worst time of the year. however i am already impressed by the saving, December was good in that we are now using all the surplus, so far in January due to some nice clear frosty days i have been amazed to manage 10 days without taking anything from the grid & very little on the other days. by April Octopus will be paying me 😄

Feb production was 176kw. South facing 3.55kw system in Rugby. This year was 1kw more than in 2022