Inverter Sizing

Ed050458

Hi
I haven't seen anything on the sizing (capacity) considerations of an inverter.
I've recently had a 3.5kW system installed but noticed that the Eversolar inverter is rated at only 2.8kw (3.08kW peak). Apparantly this is better for overall efficiency but I cant help resent the automated wasted on a sunny day. I am facing West with a 45 deg roof.
Any advice appreciated. Deep down I want a bigger inverter fitted but need some technical backing for this.

Martyn1981

Ed050458 wrote: »

Hi
I haven't seen anything on the sizing (capacity) considerations of an inverter.
I've recently had a 3.5kW system installed but noticed that the Eversolar inverter is rated at only 2.8kw (3.08kW peak). Apparantly this is better for overall efficiency but I cant help resent the automated wasted on a sunny day. I am facing West with a 45 deg roof.
Any advice appreciated. Deep down I want a bigger inverter fitted but need some technical backing for this.

Hiya Ed, here are some discussion points from previous mentions, don't really answer your question, but supply some info and background [edit: just click on the white chevron in the blue box next to the poster's name]:

A post of mine, relaying info from another poster on another site:

Martyn1981 wrote: »

Holy Exploding Undersized Inverters Batman!

And a more recent post, from an installer:

Energetic wrote: »

Hi Guys, thanks for the questions and responses, pretty much already answered but I’ll give you my take on it.

Depending on how I read your figures you have an undersizing of 10 to 20%. So appears near the max recommended. Doubt you want to wait too long, but at least in a years time you'll be able to compare estimated and actual generation, and get a feel for whether you are up, down or neutral.

Might be worth you asking Energetic for his thoughts and options on his thread (the second link), but I'm not sure how often he looks in.

Mart.

jimjames

Ed050458 wrote: »

Hi
I haven't seen anything on the sizing (capacity) considerations of an inverter.
I've recently had a 3.5kW system installed but noticed that the Eversolar inverter is rated at only 2.8kw (3.08kW peak). Apparantly this is better for overall efficiency but I cant help resent the automated wasted on a sunny day. I am facing West with a 45 deg roof.
Any advice appreciated. Deep down I want a bigger inverter fitted but need some technical backing for this.

You aren't wasting much if anything on a sunny day. As panels are less efficient as they warm up the maximum generation can be significantly below their rated output.

For example on our 3kWp panels on a cooler sunny day we have peaked at 2.7kW but on a warm sunny day it is only around 2.5kW peak generation. Your panels will not produce much over 3kW peak output so the inverter will not be capping much output at all.

http://uk-solarpanels.blogspot.co.uk/2012/05/temperature-solar-panel-electricity.html

Kernel_Sanders

Martyn1981 wrote: »

(the second link),

Is anyone else not seeing these links? Must be bewildering for the OP!

Martyn1981

Kernel_Sanders wrote: »

Is anyone else not seeing these links? Must be bewildering for the OP!

Clicking the white chevron in the blue box next to the name Energetic takes you to the post which is contained within 'his' thread 'Ask a renewables installer'.

Was 'link' the wrong term? Perhaps I should have said second post link, or something like that.

Mart.

Kernel_Sanders

Martyn1981 wrote: »

Clicking the white chevron in the blue box next to the name Energetic takes you to the post which is contained within 'his' thread 'Ask a renewables installer'.

Was 'link' the wrong term? Perhaps I should have said second post link, or something like that.

Sorry Mart, didn't realize that those chevrons were links.
[IMG]http://forums.moneysavingexpert.com/<a href=http://imageshack.us/photo/my-images/829/screenshotat20130613234.png/ target=_blank rel=nofollow>[/img] [IMG]http://forums.moneysavingexpert.com/<a href=http://imageshack.us/photo/my-images/829/screenshotat20130613234.png/ target=_blank rel=nofollow>[/img]

Martyn1981

Kernel_Sanders wrote: »

Sorry Mart, didn't realize that those chevrons were links.
[IMG]http://forums.moneysavingexpert.com/<a href=http://imageshack.us/photo/my-images/829/screenshotat20130613234.png/ target=_blank rel=nofollow>[/img] [IMG]http://forums.moneysavingexpert.com/<a href=http://imageshack.us/photo/my-images/829/screenshotat20130613234.png/ target=_blank rel=nofollow>[/img]

No probs Kernel, in fact, if you didn't know then there is little chance the OP did, so I couldn't have made it less obvious. I'll amend the post.

Cheers for letting me know.

Mart.

Edit - PS, still doesn't really answer the question. What do you think? All three of my inverters are very slightly over-sized, but that's just down to availability and panel sizes. I'm torn on the subject, 10% undersizing doesn't sound unreasonable, but 20% is nearing max, though to further complicate, the OP has an inverter with ratings that cover both percentages. If I was a betting man, I wouldn't know what to go for - less, more or the same generation as a 3.5kW inverter! M.

rogerblack

Martyn1981 wrote: »

No probs Kernel, in fact, if you didn't know then there is little chance the OP did, so I couldn't have made it less obvious. I'll amend the post.

Cheers for letting me know.

Mart.

Edit - PS, still doesn't really answer the question. What do you think? All three of my inverters are very slightly over-sized, but that's just down to availability and panel sizes. I'm torn on the subject, 10% undersizing doesn't sound unreasonable, but 20% is nearing max,.

The actual degredation can be surprisingly small.

To repost:

I did some calculations based on a localish sites reported output.
Doubling the panel size is not unreasonable (at least for that site, assuming the reporting was accurate.
https://forums.moneysavingexpert.com/discussion/comment/49694945#Comment_49694945


I came to wondering, after noticing on a local PV installations instantaneous generation graph how little time it's close to its maximum.

 *    orig  new  ideal  %
  1 1911 1911 1911 100
1.2 1911 2285 2293 99
1.4 1911 2628 2675 98
1.6 1911 2932 3057 95
1.8 1911 3200 3440 93
  2 1911 3441 3822 90
  3 1911 4399 5733 76
  4 1911 5102 7644 66

This was a simple calculation based on computing the 10 minute interval instantaneous outputs of a local 4kW installation with larger panels (for about 5 months)

Columns are:
Excess installed capacity - so the '2' row is for 8kW solar with a 4kW inverter.
Original output in kWh over the period.
New output in kWh over the period.
Output if there was a properly sized inverter installed.
Efficiency (how much you get relative to a 'properly' sized system.

If it was legal to install more panels, then for this period, for this system, quite large increments in panel size seem to make sense.
For example, a 7kW set of panels with a 4kW inverter generate 90% the output of a 7kW inverter.

This assumes that the inverter only draws 4kW from the panels, and does not switch off - if the panels voltage is too high, it will switch off - the install would need to be properly designed.

The above calculation could be inaccurate if there are spikes in the data that are not reported - if a lot of the time a 50% reported output is 5 min of sun, and 5 miniutes of deep cloud, then a double sized panel will have half the above expected output.

As an aside - I finally got round to asking OFGEM - and have sent them an email asking them to clarify my understanding that according to their rules you can (for example) place a 3kW panel on east and west roofs, with a 3.5kW capped inverter, and declare a capacity of 3.5kW.

Martyn1981

rogerblack wrote: »

The actual degredation can be surprisingly small.

Columns are:
Excess installed capacity - so the '2' row is for 8kW solar with a 4kW inverter.

1.2 1911 2285 2293 99%

Hiya Roger, being a bit thick today, can I just check I'm understanding that correctly, so the 1.2 row, represents a kWp that is 120% of the inverter kW. Have I got that right?

If so, then that seems to support Energetic's approx rule of 20% undersizing as it seems to show negligible difference - though he does go on to say location, within the UK needs to be considered.

Also fits with the bits and bobs I found 1 to 2 years ago suggesting 10% 'forced' undersizing via a 3.68kW capped inverter on a 4kWp install could actually be beneficial, whilst simultaneously meeting DNO and FIT bandings.

Moving on to the down side, are there concerns over inverter life expectancies? I appreciate that an undersized inverter will be working harder (as a percentage of it's rating) for more time than an equal/over-sized inverter, but are they built to last for x years regardless or for x years @ y% with a drop off as the percentage goes up?

Or is this a 'how long is a piece of string' question, that can only be answered in 10 years or so? I suppose panel degradation also needs considering after 5+ years too!

Sorry for all the questions. As nosey as ever I'm afraid.

Mart.

rogerblack

Martyn1981 wrote: »

Hiya Roger, being a bit thick today, can I just check I'm understanding that correctly, so the 1.2 row, represents a kWp that is 120% of the inverter kW. Have I got that right?

Yes.

If so, then that seems to support Energetic's approx rule of 20% undersizing as it seems to show negligible difference - though he does go on to say location, within the UK needs to be considered.
Moving on to the down side, are there concerns over inverter life expectancies? I appreciate that an undersized inverter will be working harder (as a percentage of it's rating) for more time than an equal/over-sized inverter, but are they built to last for x years regardless or for x years @ y% with a drop off as the percentage goes up?.

This was from a site in Scotland.
I would expect this to have the least effect of undersizing - long summer days, a large fraction of which the panels are not straight-on to the sun, and relatively fewer periods with no cloud cover.

Will be a close approximation of how badly you'd be affected by undersizing in summer 2012.
The more of your generation is outside the period 10AM-2PM (or so), and the less from direct sunlight, the less you'd be affected.

zeupater

Hi

Looks like we've got two issues here, marginal over/under-sizing and massive. Whilst marginal sizing differences would have marginal effect, massive is simply illogical as there is either a loss of generation, & therefore revenue, or unnecessary capital expenditure combined with a likely small loss of generation efficiency.

I tend to agree with M that a substantially undersized inverter would likely have a shorter lifespan, this being due to higher average operating temperatures having an effect on component degradation, however, considering the time that an inverter in the UK spends generating at anywhere near it's full potential, any sizing differences of 10% to 20% would likely have negligible effect.

Regarding Roger's "10 minute interval instantaneous outputs" , there is no such thing available on our/most inverters, there is simply an average over a period - eg, on a 4kWp system the total generation was 'x' 10 minutes ago, now it's x+600W, therefore the average power was 3600W (600x(60/10)). Our inverter regularly displays instantaneous power at the full capacity of the inverter, but rarely records a 5 minute averaged power at over 95%, can hold 90% for considerable periods if conditions allow, and rarely generates below 80% when the sun is positioned optimally on a clear sunny day (mid Spring to mid Autumn), the system being close to WSW, so not that much different to Ed050458's West facing system.

Depending on the panel pack which was received and the actual capability of the inverter (is it 3.08kW instantaneous, clipping to 2.8kW after a set period ?), then, in our experience, I would have thought that 80% was slightly too small and personally would have preferred a unit which was capable of sustaining at least 3kW (even better parity), even if the overall effect of this would only be around £10/year at the currently applicable FiT rate.

HTH
Z