Solar PV 'rule of thumb'

[SoreTween]

I think I'm about to pull the trigger on a small PV system, an 800W DIY 'plug in'¹ system.  As a guesstimate of time to return the investment I searched for a rule of thumb as to system output vs boiler plate capacity in the UK and came up with this link. This states suggests:

A rule of thumb in the UK is that the average output over a year will be around 750kWh per kilowatt-peak of panel installed.

How does that sound for those with experience?  There's no date on the page but it talks of Feed In Tariffs in the present tense so it could be fairly dated and thus low.  OTOH it's from an organisation that despite being 'independent' has a vested interest in punters Doing Shit Like This so it could be talking the tech up.

Thanks.

¹Lie², it doesn't.
²not mine.

[Diver300]

I don't think that the technology has much to do with it. If the solar panels are less efficient, you need more area per kW, but the question is whether 1 kW peak can produce 750 kWh per year, not whether so many m2 can produce 750 kWh.

There are 8760 hours per year, half of which are daylight. You are only looking to average 17% during daylight hours, which seems reasonable.

[Brucey]

it varies with locality depending on the average sunlight.   



the above chart is showing above your figure but in eastern England, rainfall is about 1/3 what it is in the west country (for example) and that -assuming prevailing SW winds- presumably means less cloud, and/or thinner cloud than some other parts of the country too. So more solar output.  Even quite small hills can affect weather locally too, and with it PV output.

The above chart is typical (for the UK) in that the six months in the middle of the year contribute about 2/3rds of the annual total output.

Another thing that makes a big difference is how clean the panels are kept.  Some shed dirt better than others do, but in every 'solar farm' it is economic (or essential) to clean the panels on a regular basis. 



shows how the angle of the panel is expected to alter its output



shows that sunshine makes about 2/3rds of the power, even though it is sunny only a fraction of the time.

cheers

[Kim]

There must be websites where solar power nerds (or PV kit molishers?) collect real-time stats of their output, where you could see what someone nearby has generated over the last couple of years...

[morbihan]

Following this.
just installed solar here in Bermuda. As well as a  battery system. It actually went on line today.
Good points. lots of sun.
 Batteries offer redundancy when Hurricanes knock the grid out.

Bad points. Everything is imported to remote Island so super expensive...but then so is electricity.
Time lag of planning permissions (painful)
Haven't got to grips with the figures yet but it was really nice on day one watching the solar bank run the house, fill the battery charge an electrical car, and send power back into the grid.
Can update when I have my head around the stats. The app is pretty detailed and Im no physicist.

[SoreTween]

There must be websites where solar power nerds (or PV kit molishers?) collect real-time stats of their output, where you could see what someone nearby has generated over the last couple of years...

Nerds - not found that site yet.  The PV fora I have dipped into have been chock full of installers touting for business.
Kit molishers - Enphase have the micro inverter market pretty much sewn up (world wide) and their kit reports back to the mothership.  They have a handy map showing you locations of all their installations, there's a few nearby but no data is shown.
Domoticz (the route I'm tentatively treading to home automation & monitoring) can talk to Enphase kit, that's a route to experience data I've not yet investigated.

There is an installation on a roof top I can see from my office window, I might pop round and ask.  It's old so technologically not comparable but if they know their numbers it'll be indicative.  They may not, it's sufficiently old it may not be locally monitored (rent-a-roof FIT era).  It will be string for sure, probably two as they have 6 panels facing east-ish and 8 south-ish.

[pcolbeck]

Try this:

https://www.withouthotair.com/

It wont tell you how much energy a particular PV solution will generate but it will tell you what the maximum you could possibly generate per square metre is and give you formulas for working out what a reasonable expectation would be based on the efficiency of a solar panel. The main thing to remember is that no matter if a solar panel ws 100% efficient (which they aren't) you can't possibly generate more energy than the sun delivers per square metre, if you want more energy you need more area.

[SoreTween]

Thanks.  23% ish seems to be the current bleeding edge, I'm looking at ones that claim 18.56%

[ETA]

Figure 6.5. Solar photovoltaics: data from a 25-m2 array in Cambridgeshire in 2006. The peak power delivered by this array is about 4 kW. The average, year-round, is 12 kWh per day. That’s 20 W per square metre of panel.

1095 kWh P/A per kW    That system produces 20W/m² which is below the theoretical figures on pages 38-39.

At 750 my pay back time should be 10 years simply from the electricity I don't buy at 16.6p.

[Quisling]

I have previously been a solar PV owner, having installed (in Oxfordshire) a WSW facing (sloping) rooftop array in 2010.  Over the next 5 years it had a specific output around 930 kWh/kWp/annum.

I work as an energy consultant and have recently reviewed outputs from 9 arrays for a public sector client in the Bristol/Bath area.  With the exception of fault issues, all arrays have produced approximately 800 - 1050 kWh/kWp/year over the last two years.  The lower of these probably had periods of fault.  All are fairly well sited rooftop arrays.  Review of other client's projects has shown similar performance.

I have mates who are solar installers/investors/PV system managers.  The larger PV arrays tend to have pyranometers fitted to check irradiance vs system output and to quickly trigger fault alerts.  Pyranometers are pricey so aren't fitted as standard but generally speaking, the PV panels perform very well.  As a rough guide, I'd assume that output will never be greater than 90-95% of rated output - e.g. a 4 kWpeak array won't produce more than 3.8kWp

Hope that helps.

[SoreTween]

Thank you Quisling, it very much does.  In fact I could hardly ask for better

Everything so far pointing to 750kWh being a safe conservative figure to use.

[mrcharly-YHT]

It should be obvious that location and positioning matters.

When I ran a boat of PV, the output would drop to 10% of rated if the panel was flat.

[Quisling]

All the above considers no, or very limited shading.  Small amounts of shading can have a significant impact on generation.  My old array caught shade from a tree but only at the end of the day.  Regular partial shading may effectively knock out a whole string of panels.  Some of the decent inverters (Solar Edge etc.) allow configuration of strings to try to minimise impact of shading.  Hopefully you'll have  a clear view of the sun.

On smaller commercial jobs I'm seeing estimated paybacks from 6 years upwards at the moment, including some with co-located battery storage coming in under 10 years.

Check warranties on any quotes you get.  You should be looking for 10 years on the inverter nowadays.  The panels should have longer warranty on performance.  Make sure any quotes include operation and maintenance documentation include details of how to safely isolate the system etc.  It should all be well labelled up in line with the electrical regs anyway.

[aidan.f]

Can anyone explain to me why inverters seem to have a 10 year design life? I work with power electronics so the 'oh they are highly stressed it's so demanding' does not ring true to me.

[simonp]

We have a 9.9kWp system (feeding two 4kW inverters). So we have some clipping on sunny days. The panels are split east/west rather than facing south.

Total output for the first 12 full months was 8716kWh. This is higher than the estimated output.

[Brucey]

Can anyone explain to me why inverters seem to have a 10 year design life? I work with power electronics so the 'oh they are highly stressed it's so demanding' does not ring true to me.

I am told that one of the more common point of failure is the capacitors in the inverter stage. 

cheers

[SoreTween]

Trigger pulled

[Kim]

Can anyone explain to me why inverters seem to have a 10 year design life? I work with power electronics so the 'oh they are highly stressed it's so demanding' does not ring true to me.

It does if you read it as "we cut corners on the thermal design".

[Diver300]

Can anyone explain to me why inverters seem to have a 10 year design life? I work with power electronics so the 'oh they are highly stressed it's so demanding' does not ring true to me.

I am told that one of the more common point of failure is the capacitors in the inverter stage. 

cheers

Electrolytic capacitors have a finite life. Their life is very dependent on temperature. Increasing life by over-specifying capacitors or heat sinks will add cost.

[grams]

Is there any kind of consumer(-ish) electronics that's warrantied for more than 10 years?

[SoreTween]

Well, the Enphase microinverters I've just ordered are warranted for 20 years.  Is that consumer(-ish) though?

[morbihan]

Early results of the system we installed last week point to being off grid about 18 hours a day.
We have been over producing during the day (enough to run the property in a 24 hour period) and and will get to sell that back into the grid at a reduced rate once the meters have been switched.
The batteries crap out somewhere around midnight on at the moment. One battery is not taking full charge and is having some cells replaced which will help. Thats a month away.
The other smaller battery can take a 2.5 KW bolt-on that would also help with storage capacity (2 feeds)

There are some simple measures that could mitigate consumption like timers on water heaters etc that we need to get down in the weeds with.

Hopefully we will end up with the system producing more than we need throughout the year and use the electric company here as a sometime power storage company buying back what we can't store in the small hours.

[SoreTween]

Sounds like a much more interesting system than mine will be.

What's the buy/sell ratio like there?  Here 5.5p sell is the best I've found so far.  Buy rates vary hugely from 12p (Daily tracker + standing charge) to 21p with no SC.  So roughly 2.5 times your sell rate is the best you can hope to buy at.

[morbihan]

Sounds like a much more interesting system than mine will be.

What's the buy/sell ratio like there?  Here 5.5p sell is the best I've found so far.  Buy rates vary hugely from 12p (Daily tracker + standing charge) to 21p with no SC.  So roughly 2.5 times your sell rate is the best you can hope to buy at.

Yup we have a pretty complex system, but we both work from home so it involves workshops, studios etc. There is also a vacation cottage that we run on site hence the double electrical feed.
The one and only local electricity company was originally buying back at 100% on the dollar up until a couple of years ago but now I think its closer to 50%. A few early adopters who were getting full return are miffed but we don't see an issue with it. The electricity company finance and run all the hardware that connects everything (lines, sub stations) and have to repair it all after hurricanes etc. They are storing the power off site for everyone and need to be recompensed for that.
The more they become a power storage company and the less oil we have to ship in the better as far as Im concerned.