I'd be patient with it until you get more data from spring/summer. I guess that's how they're physically laid out? South facing roof with no shading?

For reference, my 3kWp East + 3kWp West system (Sheffield area) produced 180kWh from Decemeber - End of Jan. It was installed in September and was generating 20kWh+ some days, so like you I'm waiting to see how it performs once we have more daylight hours..

 

Yep - the layout reflects real life. The panels face south east, so I’m open to slight variations in angle leading to different production per panel when the sun is as low as it is at the moment. Fingers crossed!

 

I know that the sun is low in the sky. I know that my panels don’t face due south. But should I be a bit concerned about 20% variation between the power generated by the best panel and the worst panel? All the panels are aligned. There is a little variation in shading, but I don’t think this can be the explanation due the pattern of variation (see the layout diagram).

My installation was in early December, so all the sunlight has been at quite a severe angle so far and I’m trying to be patient before jumping to conclusions.

View attachment 111376

December generates 1-2% of your annual yield so a variation of 20% is a 0.2% ish yield variation for the month but that’s only on the panels that are varying so it’s actually less than that.

Typically year to year overall generation will vary by upto 10% just based on weather so generally nothing to worry about.

Also if you look at measured variations between panels as they come out of the factory they could vary by upto 5% and be within tolerance.

That said is there any shade on the panels as would have thought they should be fairly consistent. What panels are they?


Sent from my iPhone using Tapatalk

 

What panels are they?

JA Solar 280Wp.

 

Any bird droppings? In the middle of summer, cooling differences will cause power variations, but would not be a factor in the winter

How are you measuring the individual panel outputs?

 

I don’t think there are droppings, but I’ll take a quick look. They were only fitted at the beginning of December, so I don’t think the geese have done any bombing runs on them. I just think I need to wait a couple of months for stronger, more direct sunlight to see. Peak output is currently under 50% of the panel rating.

I’m measuring the individual panels with the Solaredge website.

Edit: no droppings.

 

That looks to be very well within normal expected tolerance to me. We've had 25 panels working in our roof for nearly five years now, and with no shading at all there is a big variation when the sun is low in the sky. It helps a bit if you have panels with an anti-reflective front face (ours have this as a planning consent requirement) but the downside with the anti-reflective glass is that it gets dirty more quickly, so the panels need to be cleaned once a year or so.

As mentioned above, winter generation is a very small part of your overall generation. We tend to get slightly better winter performance on good days, because our roof pitch is 45 deg and faces 20 deg west of south, but even so there are plenty of days in December and January when there is next to sod all generation (I suspect today will be one of them).

We generate around 6,000 kWh per year, so an average of about 500 kWh per month. Looking at the December figures we generated a total of 88 kWh, and in January we generated 110 kWh. Looked at in percentage terms we generated around 1.46% of the annual total in December and 1.83% of the annual total in January.

In general, PV generation tends to just fall off a cliff in October and stays at a low level until about March, when it picks up again. We don't bank on getting any useful excess energy generation through these months, but we do get the occasional bright, clear, cold day that can be very good. These days tend to be rare, though, we might get a handful of them through the winter.

 

I’ve just checked the February variation and it’s down to 11% following two out of three days being relatively high generation. I assume the variation will be less noticeable as generation increases.

 

There will always be variation, it's normal and within installation/manufacturing tolerances. IIRC the allowable manufacturing tolerance alone can be at least 10% or so between one panel and the next. Manufacturers tend to supply panels that are slightly better than the stated specification to get around the normal tolerance issue, and ensure that delivered panels start off always exceeding the minimum output by a small margin.

 

That looks to be very well within normal expected tolerance to me. We've had 25 panels working in our roof for nearly five years now, and with no shading at all there is a big variation when the sun is low in the sky. It helps a bit if you have panels with an anti-reflective front face (ours have this as a planning consent requirement) but the downside with the anti-reflective glass is that it gets dirty more quickly, so the panels need to be cleaned once a year or so.

As mentioned above, winter generation is a very small part of your overall generation. We tend to get slightly better winter performance on good days, because our roof pitch is 45 deg and faces 20 deg west of south, but even so there are plenty of days in December and January when there is next to sod all generation (I suspect today will be one of them).

We generate around 6,000 kWh per year, so an average of about 500 kWh per month. Looking at the December figures we generated a total of 88 kWh, and in January we generated 110 kWh. Looked at in percentage terms we generated around 1.46% of the annual total in December and 1.83% of the annual total in January.

In general, PV generation tends to just fall off a cliff in October and stays at a low level until about March, when it picks up again. We don't bank on getting any useful excess energy generation through these months, but we do get the occasional bright, clear, cold day that can be very good. These days tend to be rare, though, we might get a handful of them through the winter.

Wow that’s a lot of hours, take it your a 6kW system then, yes main generation is 2nd week in Mar through 2nd week in Oct. It’s the point that the sun reaches 17 degrees (roughly) above the horizon at midday. You’ll also notice that’s the point the weather seems to feel warmer / cooler as well. It’s to do with the angle of sun through the atmosphere and incidence angle to panels.

If you’ve ever got any questions fire away I’ve installed over 20MW and monitor over 25MW so there’s rarely a question I haven’t seen or a problem we haven’t already come across.


Sent from my iPhone using Tapatalk

 

It's a 6.25 kWp system built in to the roof, using the GSE Intégration system. The system performs pretty much exactly as PVGIS predicted, when averaged out over a few years. From around March onwards we generate more energy than we use; since I built the house it has used around 10 MWh (it's all-electric) and the PV has generated around 30 MWh.

We run our hot water from a phase change thermal battery (heated by a mix of E7 off peak in winter and excess PV generation via a diverter in summer). Heating is from an ASHP, also run from off-peak in winter, to "charge" our passive slab, that then acts like a storage heater to keep the house warm through the day and evening.

The car can either fast charge overnight at E7 off-peak rate, or can charge at a variable rate if the EVSE is set to use variable excess PV generation (2nd priority after charging the phase change battery). 9.6 kWh of battery storage that will be operational in the next few weeks should mean that we can run the house entirely from either off-peak electricity in winter or PV generation in summer.

 

That’s a cool system, pretty much what I’m doing at mine, have 4kW PV at moment just about to add another 4kW but limit the export still to 3.68kW as DNO won’t allow me more, add a Leaf as a battery with bidirectional should mean will be off grid most of time with 8kE of PV.

I’m in the process of buying a solar thermal company and was thinking about adding thermal with a heat store hadn’t decided between big tank vs Sunamp storage. Also have a screed in kitchen but not sure the volume will be enough there’s only about 4m3 in screed

May also add In a Tesla PW2 with backup gateway as we are an Tesla partner but want to see what the balance on use / consumption looks like first


Sent from my iPhone using Tapatalk

 

I looked at solar thermal briefly, but I've experience of fixing broken systems (replacing pumps, controllers, sensors etc) and frankly didn't want the hassle. I'm using a Sunamp UniQ eHW 9 phase change thermal battery, and before that we had a Sunamp PV. The Sunamp PV worked very well indeed, the Sunamp Uniq not so great, they need to do some work on the controller to make it work efficiently when charged from excess PV generation. I've designed a solution to the controller problem, but am reluctant to implement it whilst the unit is still under warranty, and I'm hoping that Sunamp come up with a fix before too long (they are aware and are looking into it).

We had a thermal store initially, but the heat losses from that, even with an additional 50mm layer of insulation foam around it, were just too high. Our services room was reaching 40 deg C in summer, and the adjacent bedroom was over 30 deg C, so the thermal store had to go. The losses from the Sunamp are tiny in comparison, which both alleviates the over-heating problem and increases efficiency a lot.

We were lucky in that we found an "unknown" 95mm² 3 phase Wavecon cable under our plot, which didn't have any wayleave or easement to be there. The DNO moved it, to a location where it wasn't causing us a problem, and where it ran underneath the external meter box kiosk I installed. That meant our incomer is barely 2m long. When I asked for G59/3-1 consent the DNO just gave me "up to 12 kWp on any single phase" permission. It turns out that the 95mm² Wavecon only supplies one other house...

I can't get the sums to add up for the PW2; it's just too costly at the moment to recover the capital outlay from through life savings. Even the much cheaper battery system I'm putting together will struggle to recover the capital investment, but as it has a UPS function, and as we get half a dozen power cuts every year, I'm prepared to pay a modest premium to keep essential circuits working.

 

That’s an interesting point about the heat loss, not really considered that although yes it’s an obvious issue as we see same thing from commercial inverters when we’ve 200kW in one room essentially it’s a 5kW heater!

PW2 should give about a 10-12 yr payback but know what you mean, less if you don’t have FiTs, that’s the reason for wanting an old leaf 24 with bidirectional as it’s a cheap battery that you happen to get a spare car with. The PW2’s do backup from next month you just need the new gateway 2, essentially it’s a seamless backup unit (but not advertised as that). How you doing the grid compliance on your unit, presume a G59 changeover unit, did the DNO want much info on fault current etc?

The waveform was a lucky find!


Sent from my iPhone using Tapatalk

 

It's any easy calculation to make, if you can accurately estimate the value of the stored usable energy (which is around 85% to 90% of the charge energy). Taking the PW2 as an example, it has a usable energy of 13.5 kWh, and takes a bit over 15 kWh of charge to get that usable capacity. Unless you have a large PV array, then the PW2 won't charge from PV for a fair bit of the year. For example, our 6.25 kWp array is averaging about 3 kWh per day of excess generation right now, perhaps 20% of the excess generation capacity needed to charge a PW2. That means we'd only charge a PW2 up to capacity once every five days on average at this time of the year.

After 5 days of charging from excess PV generation we'd get about £2.20 worth of usable energy back out of the PW2. This would improve markedly as the summer approaches, but our electricity demand also drops a great deal at the same time. By April we'd not be able to use more than about 1/3rd of a PW2 battery capacity, despite it being charged pretty much every day, simply because our electricity consumption during the night would be lower, and it makes more sense to utilise excess PV generation during the day for loads that we'd otherwise put on at night (saves the battery round-trip losses).

To just recover the capital investment of £6.5k in a PW2 installation, you would need to be able to use about 40 MWh of stored excess PV generation electricity from it at current prices before the PW2 dies of old age. We've used a total of 10 MWh in the last 5 years, so even if we could use all of the capacity in a PW2 then it would still take 20 years just to recover the capital investment. I doubt that the calendar life of the cells would extend more than about 15 years at the very most, and my experience with my 9 year old electric motorcycle battery pack is that calendar life ageing has been significantly greater than cycle life ageing, as that pack is now down to about 50% of it's original capacity, after around 600 cycles.

If you are able to use close to the 13.5 kWh capacity of the PW2 every day, and more importantly, are able to charge it from excess PV generation to full capacity every day (which probably needs a PV array of 20 kWp or more at a guess), then yes, it will probably recover the original capital investment within about 12 years or so, around the same as the payback for a PV installation, perhaps.

 

Interesting post. It’ll take more brain power than I have right now to work my way through it, but one thing is easy to point out - my Powerwall has 100% cycled nearly every day, because apart from a couple of nights before cloudless days, I’ve filled it with economy 7 electricity. It powers some electric heating and the car pre-conditioning in the morning, then charges from the solar array during the day. I’m drawing 14kWh per day on average since the beginning of December.

Who knows how long the Powerwall will last? Significantly beyond the warranty, I hope. Might do a bit of research on that, although they’ve only been around for a couple of years...

 

The return on capital investment time if charging from E7 is roughly double that from charging from excess PV generation, as generally E7 off-peak rates are around half the peak rate.

My intention is to charge a battery system from E7 during the winter, too, and the 18 year return on capital investment period that I worked out for a PW2 reflects that mix of excess PV generation charging and E7 off-peak charging. That 18 years includes around 10,000 charge/discharge cycles (allowing for the twice a day partial cycling that will occur during the shoulder months), which in EV terms would be roughly equivalent to around 400,000 miles (again, allowing for partial cycling). I believe that the warranty covers up to 5,000 cycles with 20% allowable capacity loss (happy to be corrected if this has changed), so it would seem to match the life warranty reasonably well.

I'm looking at a capital investment for a battery system that is a fair bit less than that for a PW2, so can get the return on investment period down to about 12 to 15 years, but, as I mentioned before, I'm not convinced that what I'm doing will be truly cost effective, part of the reason for the investment is to have a back up supply for the inevitable power cuts we get. The reason the house has two DBs, split into essential and non-essential circuits is because I was originally going to install a back up generator and change over relay to the essential circuit supply. It's a happy coincidence that I can just slightly re-jig the wiring to be able to easily install a battery powered back up supply.

 

I’ve been looking for the warranty this afternoon. All I’ve found is a headline “10 years” on the Tesla UK website. I’m sure it’s more complicated than that.

I would love a backup configuration, but I don’t believe Tesla offer it yet in the UK. We have less than one power cut a year, so it’s not a big deal.

 

Here's what I got from Tesla, still differs to the version they publish on the website.

 

Interesting to see what is presumably a heat effect causing 50% worse battery degradation.

I’ll be keeping a close eye on degradation. I’ve heard the Powerwall chemistry is different to the car chemistry, so I’m not making any assumptions. Also, the battery is tiny compared to a Tesla traction battery as is the max power output. Lots of unknowns...

 

Still 11% variation between my best and worst panels over the last two days...