Adding batteries to 12 year old solar system (and considering heat source pump)

[GlynG]

Um. So, I'm not in the UK, I'm on the other side of the pond. But my system was installed back in 2008, which also makes it 12 years old. And there's Stuff.
So, back in the day, one had a choice between amorphous silicon (has this mottled blue color), which was cheaper than the alternative, crystalline silicon (solid blue, no mottled color, more efficient so one needed fewer panels for the same power. Further, the power variation was more severe (over temp, daylight angle, and all that) with the amorphous silicon than with the crystalline.
Next: The inverters. At the time in the U.S., batteries were pretty much of the lead-acid type so, unless one really needed them, one didn't install them. Given that that was the case, strings of solar panels in series and/or parallel (say, 9 panels in a string, put in parallel with another 9 panels in a string) were wired up and applied directly to an inverter. When the sun was down, 0 volts. When the sun was just over the horizon, maybe 100V; when it was noon, 240V, and so on. So, wildly varying voltage, the inverter would wildly vary the current drawn from the panels, always hunting for the point where the product of V*I was maximized. So the power output would very definitely vary with time of day, cloud cover, and so on. And throw into that mix the temperature, because (natch) the panels tend to be a bit more efficient when they're cold, and so on.
Now, lets talk about losses. Say one has that example of 9 panels in series, paralleled with 9 other panels in series. Say that the sun shineth at some angle and temperature. Those panels have manufacturing variations: Some will have higher voltages, some will have lower voltages, and the variation will be at random because, well, manufacturing. This means that the nominal voltage on one batch of 9 panels won't be the same on the other batch of 9 panels - so the higher guy will kind of back-drive, but not completely, the lower batch. Result: One will not get the max power out of the panels on the lower-voltage string. Which leads to losses. It's not unusual to get 5% to 10% less power out of the panels on one's roof than one might think by summing up the power available by multiplying the manufacturer's spec'd power times the number of panels.
It gets worse. Remember that inverter? If one designs an inverter that has a wide-ranging input voltage/current, it's not going to be quite as efficient as an inverter that has, say, a fixed input voltage.
Finally: Just because some panel with some illumination has a particular voltage and current where it maximizes its output power, the panel next in line might very well have different voltages and currents that maximize its output power - which might be very close, or different, than the first panel.
So, in my case, I happen to have a nominal maximum power of all the panels on the roof of some 9.02 kW. The inverters (I happen to have two) have a maximum output power of 7.8 kW - and they very rarely hit that power level. It's all about the losses/manufacturing variations. That's a 13.5% loss at max power, and we're not even talking about what happens at 1/2 or 1/4 illumination, temperature, and all that jazz.
However, Technology Has Changed, And For the Better.
First, let's talk about how string/panel technology has changed.

1. Typically, each individual panel has upon it a DC-DC converter that takes the voltage and current from the panel and converts it to a different voltage and current on the output. These converters are typically highly efficient, in the 95% to 98% range. Further, the panels in series can communicate with each other, for the purpose of setting the output voltage. You'll see why in a minute.
2. Next: Wire up a bunch of these panels in series. The outputs are in series. Because they're in series, they all have the same current. But! They talk to each other. The total voltage across the string is set to a fixed value, typically 300V.
3. Let's say we have 10 panels in series. So, each panel's DC-DC converter output would be, nominally, 30V. But, say that one panel is doing well today: It increases its voltage compared to the rest a bit, so it's V*I = P = Pmax for that panel, at that time. The others, who aren't doing as well as our well-doing one, reduce their voltage a bit. But, given Variations, we end up with an enforced 300V across all of them, with different voltages, but the same current across each panel, with the result that each panel is independently operating at its maximum power given the conditions.

Breather time. So, say that these DC-DC converters have, say, 3% inefficiency. But that more than makes up for the lack of losses because of manufacturing-variation-mis-matched panels!

Further, since all these strings of, say, five strings of 10 panels each are in parallel, the overall voltage is still 300V, with each panel doing its personal level best.
Finally: We have a nifty, fixed, 300V on the Entire Array. Feed that to an inverter that:

1. Is optimized for 300V operation. Yea! Less losses.
2. And here's the tricky bit: Has a 300V battery built in.
3. And talks to all the DC-DC converters out there.

Charging/discharging the battery works well, it's on the solar panel side of things, pretty much, so one doesn't have to go crazy (like with my 12V lead-acid battery example) upconverting or downconverting the voltage into the batteries. The inverter side of things converts from 300 VDC to city power..

The electronics is cheaper.
The panels run at the full efficiency that they're capable of, illumination, temperature, and all.
There's losses in the DC-DC converters, but they're much less than the 13% to 15% I was citing for my system.
With greater efficiency in the panels (crystalline vs. amorphous) one needs fewer panels to get the same amount of energy.
With greater efficiency in the electronics, one needs fewer panels to get the same amount of energy.

OK, fine. But how does this apply to you?
1. You don't have any of those nifty DC-DC converters. Um. Oops.
2. Don't know how this will affect installing a different inverter.
3. Don't know how this will affect installing a battery.

The World, Somewhere, probably has an inverter/battery system that can work with your panel system, but it might take some searching and comparing this-with-that to find one that meets what you want.

Not saying it can't be done, mind you, but, yes, The Technology Has Changed.

Your clearly a lot more knowledgeable than I am, most of your post went straight over my head (Are you an installer or electrician?) What i did understand (I think) is that systems have completely changed since inception and perhaps your terminology is a little different where you are but the DC to DC converters you refer to would be called "Optimisers" in the UK (fitted to every panel) and is particular effective if any panel gets shade from trees or perhaps other buildings as the sun traverses it course, without optimisers, when one panel is shaded it reduces output from every other panel to the same level as the shaded panel, With optimisers each panel generates according to the illumination it receives independently and makes the system a lot more efficient.

Lead Acid batteries enjoyed a brief spell in the UK but most deploy the lithium Iron technology now

 

[Dilly]

My system battery never exports to the grid, its always been that way and don't know if it could be changed - not that i would do that because its just adding cycles to the battery and that will shorten their life. My batteries are rated for 6000 cycles and a ten year guarantee. I have the PV inverter in the loft and my battery system is in the garage (integral garage) - that too has an inverter, the batteries run around 400volts. My export is not recorded just the 50% deemed of my total generation - so I wouldn't ever want to be paid by what i actually export.


I'm pretty sure i don't fully understand the formulas for fit payment levels either, its just so complicated, My understanding was confirmed by the installer as my system is rated at 4.380Kws, I was told that over 4Kws then in order to get approval they had to show the roof orientation wasn't true south and the output would be limited irrespective of what the panels could achieve in optimum sun conditions by the 3.68Kw inverter. If I had the 5Kw Inverter then my panels could theoretically generate the full 4.380Kws and therefore i would only qualify for the lower rate of FIT.

If my understanding is correct, your 7Kwhr system would be paid at the lower FIT rate of the day, but you say you are paid the higher rate, Interestingly you have two 3.68kw inverters, Is it possible the installers only have one Inverter set to export? and therefore the higher FIT rate to be paid - but the house could access the full 7Kws the panels could generate? The more i get to know the less I actually know.

I should have said my roof is 1° off true south. the whole lot is permitted to export the full 7KW
there are four strings of 7 panels; 2 strings to each inverter. I think the inverters join at the DC output
FIT rates …pass…

 

[Tronguy]

Your clearly a lot more knowledgeable than I am, most of your post went straight over my head (Are you an installer or electrician?) What i did understand (I think) is that systems have completely changed since inception and perhaps your terminology is a little different where you are but the DC to DC converters you refer to would be called "Optimisers" in the UK (fitted to every panel) and is particular effective if any panel gets shade from trees or perhaps other buildings as the sun traverses it course, without optimisers, when one panel is shaded it reduces output from every other panel to the same level as the shaded panel, With optimisers each panel generates according to the illumination it receives independently and makes the system a lot more efficient.

Lead Acid batteries enjoyed a brief spell in the UK but most deploy the lithium Iron technology now

Um. I'm an electronics engineer, mostly involved in telecom. And, in terms of "engineer", I mean, "Design, Build, Troubleshoot, code for, and Generally Beat Upon Electronics until the blame stuff works." Down to the transistor level. Telecom power supplies, fiber optic amplifiers, LED indicators, phase-locked loops, touchy GHz GAAS parts, and picking up the sometimes smoking bits from the field (most of my work these days) to figure out why they're smoking and to Change Things so it never does That again. So, yeah, DC to Daylight work.
I'm not a solar installer or an electrician but, what with solar on the roof and an electric car or two, I have, shall we say, an interest. And have the maths and technical background to be able to read the data sheets and make sense of it all.
Agree about the lithium iron tech; of course, that wasn't around 12 years ago, or at least not much, and not at the current price point.
The part I'm truly not sure about has to do with retrofitting batteries into an older grid-tied system. It's pretty clear that if one is running the kind of optimisers I'm vaguely familiar with, then going from a 300V solar panel string to a 300V battery looks pretty simple, as would also be powering the house from said battery when the sun goes down. Not so sure what happens if the voltage from the solar on the roof varies all over the map. Multiple inverters? Standard practice? different equipment? Don't know, but would have to check and see what the various installers would say.

 

[Tony Hoyle]

@GlynG I think PV/inverter and deemed export is based on a generation meter at the Inverter - so even if you have a battery downstream it shouldn’t affect the PV-Inverter- generation meter part. You just won’t get paid for anything the battery exports to the grid (which is fine, I’d set mine to not export)

My generation meter is at the end of the solar feed just before it enters the CU. If I were to install a battery before that it would massively reduce anything recorded by that meter and hence the deemed FIT (or make it go higher when the battery was in use.. in short.. the number would become essentially meaningless).

Of course no sensible installer would do that, they'd patch in beyond the meter somehow.. Only about an inch of cable 'downstream' but I guess that's fixable easily enough.

 

[Mrklaw]

My generation meter is at the end of the solar feed just before it enters the CU. If I were to install a battery before that it would massively reduce anything recorded by that meter and hence the deemed FIT (or make it go higher when the battery was in use.. in short.. the number would become essentially meaningless).

Of course no sensible installer would do that, they'd patch in beyond the meter somehow.. Only about an inch of cable 'downstream' but I guess that's fixable easily enough.

I think (again all my guesswork from what I’ve read up about) that its outside of the CU - the battery sees excess export from a CT clamp, so it knows what is being exported, and creates a demand for that number of watts - effectively diverting the export into the battery. The generator meter will still see what the PV is generating as the battery will pull from after the CU

 

[Mrklaw]

wife seems semi ok with the idea of £6k on a battery. Its about what we’d pay on a big family holiday to Japan and we haven’t been for two years so that was my angle. Along with stability of pricing over time even with a slow payback period.

still curious if I can wangle 0% with some solar on the N roof but probably not worth it

 

[Tony Hoyle]

To be honest if electricity keeps going up 6k is looking better all the time... A year ago I'd have laughed at the concept of a battery being worth that.

 

[Mrklaw]

To be honest if electricity keeps going up 6k is looking better all the time... A year ago I'd have laughed at the concept of a battery being worth that.

There was an interesting video I watched this morning trying to calculate return on investment and savings and they included a comparison with just putting the cost of solar/battery/both into a stocks and shares ISA over the 10-15 year period. That was quite interesting. I already have tiny solar so the savings are a bit better, but honestly I just want to get some stability against future volatility too if possible - blunt off the spikes in prices as much as possible

 

[Dilly]

There is no doubt in my Mind that solar and battery do go hand in hand and make a big difference in both smoothing out peaks and troughs and providing free/Low cost energy during darkness.
That said, the payback for batteries is quite long winded but I’d certainly still have one.
I have a pretty large system and these days my actual grid usage is a fraction of what it used be so my savings going forward, are on the face of it, quite small.
But the overriding fact is, the more I need to use the more my system will save. I just hope I have enough slack for ASHP when the time comes!

 

[GlynG]

My reasons for having Solar and storage batteries in November 2018 was driven by the fact I'm a bit of a gadget freak - and Solar just happened to lodge in my head and become my latest project. I'm no eco warrior, saving the planet didn't and doesn't enter my head though the list of energy saving ideas I've implemented would suggest otherwise.
I changed every single light bulb in the house to LED many years ago, the walls are cavity insulated, I did the loft insulation too - well over the required thickness, windows are all double glazed, I had a new boiler fitted and Hive controls and thermostatic valves to radiators - I sort of did it all to see just how well i could improve the energy rating of the house (A 1970s detached and took it from an "E" rating to a "B") The "B" rating is very close to being an "A" and the EPC was done prior to and as part of the stipulation of having the Solar and Battery - so I would assume a new survey would lower the score a little more but to get an "A" rating I would need to dig up the concrete floors and insulate which clearly just isn't cost effective.

As usual, when I throw myself into something I start at the middle level of tech available and talk myself into having the best using man maths to justify changing the spec and future prof the systems (I must be a sales persons dream) I'm also very thorough in my research before committing.

I had the one battery to start with (4.3Kw) and was pleased with getting the house supplied free of the grid for maybe 8 hours before it was depleted and back to the grid supply - but me being me I bought and installed myself the additional two batteries to take the system to its maximum capacity and enjoy about 24hrs of free energy when fully charged. My Solax triple power batteries were only available in 4.3Kw at the time and the system was marketed as up to a maximum of 4 batteries - it was new tech at the time but the company found issues with the voltages with 4 batts and revised there system down to 3 batts max - which p*****d me off a bit. a little later they introduced 6.4Kw batts then a new system they called "Gen 2" - Gen 1 and Gen 2 are not compatible so my max storage capacity of 13.5Kw cant be expanded unless I can find 3 x 6.4kw batts in Gen 1 standard and of course find a buyer for my 3 x 4.3kw batts. I could swap the BMS to a gen 2 and have three new gen 2 6.4kw batts but the cost is rather stupid - so I will wait until these batteries come to the end of their life before considering what's new to replace them with.

From day one I have kept a spreadsheet - meter readings every month religiously, my daily average consumption is between 17 and 18 KWhrs so a column has been created to show me what the daily cost would be if I didn't have solar or batts, also columns for generation so i can establish what the FIT payments will be - the spreadsheet is to the penny accurate so i can see each month what my savings have been and what the system has paid back as ROI over its life, the latest energy price increase will build those savings significantly as time goes by.

The money paid out is in my mind is "dead" money - its already been spent and forgotten about - the savings will go on for many years well beyond what I have left, I see the installation as being my pet project of the time, I've enjoyed it, I think its been good value for money, I anticipated energy price increases to be around 8% per year at the start of this journey and predicted a ROI in 12 years, its looking very likely to be in 7 or 8 years now.

As for heat pumps for the future - Not sure where i will go with this, my "new" boiler is now 7 years old so plenty of time left before this needs to be considered but current heat pumps are big, ugly, expensive and a bit crap and cant see myself getting one, however I'm sure the tech will improve in both performance, cost and aesthetics.

 

[Dilly]

@GlynG Man after my own heart. I too have a spreadsheet going back in part to 2008.
I cream all the available data from my kit and calculate just about all the savings that I can’. Some will be always unseen or incalculable.
my oil boiler is 19 years old. It was top spec at the time and the service guy was a real pro. That said it’s a 180,000bt h running 16 radiators. 2/3 of the system is less than 20 years old and the rest, twice that.
I don’t really have an external space that I want to give over to ASHP. They are a big old lumps! I guess things will change as tech improves.

 

[GlynG]

@GlynG Man after my own heart. I too have a spreadsheet going back in part to 2008.
I cream all the available data from my kit and calculate just about all the savings that I can’. Some will be always unseen or incalculable.
my oil boiler is 19 years old. It was top spec at the time and the service guy was a real pro. That said it’s a 180,000bt h running 16 radiators. 2/3 of the system is less than 20 years old and the rest, twice that.
I don’t really have an external space that I want to give over to ASHP. They are a big old lumps! I guess things will change as tech improves.

Having read your replies on this and other subjects i think we are from the same mould, I always thought my views and actions were a bit creative or OCD, nice to know I'm not totally weird.

 

[Dilly]

Having read your replies on this and other subjects i think we are from the same mould, I always thought my views and actions were a bit creative or OCD, nice to know I'm not totally weird.

There are plenty like us about. The others just don’t admit it

 

[Mrklaw]

There is no doubt in my Mind that solar and battery do go hand in hand and make a big difference in both smoothing out peaks and troughs and providing free/Low cost energy during darkness.
That said, the payback for batteries is quite long winded but I’d certainly still have one.
I have a pretty large system and these days my actual grid usage is a fraction of what it used be so my savings going forward, are on the face of it, quite small.
But the overriding fact is, the more I need to use the more my system will save. I just hope I have enough slack for ASHP when the time comes!

worth considering ASHP may double your daily electric usage. If your PV can't handle that (And you'll want some heating in winter/nighttime) then you may want to at least ensure going with a modular/expandable battery and possibly faster charging (4 hours octopus go will only fill 12kwh with a 3kw inverter/battery capacity)