Solar panels and battery - any real world reccomendations?

Most systems can be timed to charge at night during eco7 tariffs.

If you're with octopus their API can be plugged directly into your system's app which can then automate it for you. But if you're a home assistant pro then of course that's just as easy.
 
Last edited:
Bug One said:
Not sure if there's any systems that will read the weather forecast, and charge a bit overnight automatically in response.

Can you trigger an event based on atmospheric conditions using home assistant?
Click to expand...

That is coming to givenergy apparently. But solar forecasting is quite flakey other than real time stuff.

But I think @katie279 is doing that with HA
 
Last edited:
I was kind of interested in weather integration but assuming your getting a medium/large install vs your consumption and similar with battery the benefits of trying to vary charge in battery seems pretty limited.
Eg assuming a 5-6 or so KW array and 10kw or so storage with an daily usage of mid/high teens to 20kw a day

Roughly 6 months of the year from April-Sept the majority of days should see you basically grid neutral.
Roughly 4 months of the year from Nov-Feb you almost certainly going to want 100% battery charge.
Which leaves a couple of months March and Oct where there is likely to be quite spiky days and low days. On those I would full charge, assuming you have SEG setup.

Logic is, say you need 10kwh to charge you battery at Go rate of 12p (new rate) and you have a SEG of 4p
You spend £1.20 charging the batt. But its a OMG great solar day so you could have got away with zero, you end up exporting 10 units and get 40p back, net cost of 80p ie your net cost of taking an offpeak unit is only 8 pence if you get it wrong and draw too many
Alternatively you dont charge, its not a great day and you end up sucking in 10 units at 34p, you spent £3.40. So you need to be right to not charge 34/8 or roughly 4 days vs 1 day you get it wrong.
Ie your kind of gambling at this point and you need to be right a lot more frequently than you get it wrong.

The forecast predictors seem ok as in getting low/ok/good/great level info right, but not great when the predict you will generate say 18kw and it ends up as 13kw (@HungryHippos was showing me data to this effect but he may have tracked more since and have a better comment in that regard)
Lets face it how often do you look at tomorrows weather before you go to bed and its different when you get up!?
 
Journey said:
You don't really need a kit, you can just buy what fits your needs best.

Are you saying you have 5.5m both high and wide, on two faces? So a total of 60.5m2? If so that is a lot of panels/generation potential, and you could use some of the larger panels that work out slightly less £/pW, e.g. the 600w Canadian solar that are 2.17m x 1.3m, that would allow 8 panels in portrait (per side) assuming your measurements are correct, so 9.6kWp if you have two sides, the panels in question are about £260 inc VAT. :)
Click to expand...
Sorry no 5.5x5.5 flat roof so 27sqm or whatever it is, and would prbably need a plug in inverter type thing I'm assuming.
 
Last edited:
Mercenary Keyboard Warrior said:
I was kind of interested in weather integration but assuming your getting a medium/large install vs your consumption and similar with battery the benefits of trying to vary charge in battery seems pretty limited.
Eg assuming a 5-6 or so KW array and 10kw or so storage with an daily usage of mid/high teens to 20kw a day

Roughly 6 months of the year from April-Sept the majority of days should see you basically grid neutral.
Roughly 4 months of the year from Nov-Feb you almost certainly going to want 100% battery charge.
Which leaves a couple of months March and Oct where there is likely to be quite spiky days and low days. On those I would full charge, assuming you have SEG setup.

Logic is, say you need 10kwh to charge you battery at Go rate of 12p (new rate) and you have a SEG of 4p
You spend £1.20 charging the batt. But its a OMG great solar day so you could have got away with zero, you end up exporting 10 units and get 40p back, net cost of 80p ie your net cost of taking an offpeak unit is only 8 pence if you get it wrong and draw too many
Alternatively you dont charge, its not a great day and you end up sucking in 10 units at 34p, you spent £3.40. So you need to be right to not charge 34/8 or roughly 4 days vs 1 day you get it wrong.
Ie your kind of gambling at this point and you need to be right a lot more frequently than you get it wrong.

The forecast predictors seem ok as in getting low/ok/good/great level info right, but not great when the predict you will generate say 18kw and it ends up as 13kw (@HungryHippos was showing me data to this effect but he may have tracked more since and have a better comment in that regard)
Lets face it how often do you look at tomorrows weather before you go to bed and its different when you get up!?
Click to expand...

Thanks for that, that’s well thought out and I hasn’t considered that actually it’s quite marginal once SEG if factored in. Particularly if they fix the SEG to be a more reasonable rate.

Let’s just say I don’t like spending money if I don’t have to :D. Being able to automatically control it sounded ideal for maximising the benefits and we get loads of sunny days here through winter.

I guess actually the smart play is to just fill the batteries daily Oct to March and any excess will go into the car initially anyway.

Beautiful blue sky sun set out their tonight.
 
Last edited:
SoliD said:
Sorry no 5.5x5.5 flat roof so 27sqm or whatever it is, and would prbably need a plug in inverter type thing I'm assuming.
Click to expand...

Oh flat roof is totally different then, and you'll need a more discrete mount method to stay within planning laws. By plugin inverter, you mean a direct to 3-pin device? If so then I don't rate them for larger arrays, you'll be limited to 13A realistically, and that isn't something I'd trust over a very long sunny day.
 
b0rn2sk8 said:
Thanks for that, that’s well thought out and I hasn’t considered that actually it’s quite marginal once SEG if factored in. Particularly if they fix the SEG to be a more reasonable rate.

Let’s just say I don’t like spending money if I don’t have to :D. Being able to automatically control it sounded ideal for maximising the benefits and we get loads of sunny days here through winter.

I guess actually the smart play is to just fill the batteries daily Oct to March and any excess will go into the car initially anyway.

Beautiful blue sky sun set out their tonight.
Click to expand...

Right so avoided mentioning the solar diverter I am having as well, so any excess would go there first anyway
Since not everyone has a hot water tank to replicate this but for me that would happen before exporting and probably actually suck in any excess I could generate unless we were away
 
Mercenary Keyboard Warrior said:
I was kind of interested in weather integration but assuming your getting a medium/large install vs your consumption and similar with battery the benefits of trying to vary charge in battery seems pretty limited.
Eg assuming a 5-6 or so KW array and 10kw or so storage with an daily usage of mid/high teens to 20kw a day

Roughly 6 months of the year from April-Sept the majority of days should see you basically grid neutral.
Roughly 4 months of the year from Nov-Feb you almost certainly going to want 100% battery charge.
Which leaves a couple of months March and Oct where there is likely to be quite spiky days and low days. On those I would full charge, assuming you have SEG setup.

Logic is, say you need 10kwh to charge you battery at Go rate of 12p (new rate) and you have a SEG of 4p
You spend £1.20 charging the batt. But its a OMG great solar day so you could have got away with zero, you end up exporting 10 units and get 40p back, net cost of 80p ie your net cost of taking an offpeak unit is only 8 pence if you get it wrong and draw too many
Alternatively you dont charge, its not a great day and you end up sucking in Lets face it how often do you look at tomorrows weather before you go to bed and its different when you get up?
Click to expand...
 
Last edited:
Journey said:
Oh flat roof is totally different then, and you'll need a more discrete mount method to stay within planning laws. By plugin inverter, you mean a direct to 3-pin device? If so then I don't rate them for larger arrays, you'll be limited to 13A realistically, and that isn't something I'd trust over a very long sunny day.
Click to expand...
Agreed on that last point. I’ve seen loads of melted 13A sockets from people using 10A EVSEs on them for as little as a couple of hours.

There is very little in your house that actually takes a lot of current for long periods. You could have a dodgy socket in the house and not even know it because it never really gets heaven use.

Things like 2kw heaters are pulsing the heating element so they won’t pulling the current all the time, governing the socket the opportunity to cool.

Mercenary Keyboard Warrior said:
Right so avoided mentioning the solar diverter I am having as well, so any excess would go there first anyway
Since not everyone has a hot water tank to replicate this but for me that would happen before exporting and probably actually suck in any excess I could generate unless we were away
Click to expand...

That’s next on the list, we currently have a 20 year old Range power max which is a gas boiler and heat store in one unit.

The gas heats the water based heat store which directly runs the CH and it has a heat exchanger for mains pressure hot water. Interesting device for its time (ignoring its fairly dangerous flaws).

Will be getting a cylinder with either a ASHP or gas boiler to replace it.
 
Freefaller said:
That is coming to givenergy apparently. But solar forecasting is quite flakey other than real time stuff.

But I think @katie279 is doing that with HA
Click to expand...
Yup! Pretty easy actually Solcast integration gives you the solar forecast, Givenergy Integration (or whatever you need for your setup) gives you all your battery/solar stuff, then simple automation to suit your needs - eg if forecast greater than X kWh, then charge y

I go pretty conservative with my estimates given how accurate weather forecasting is in the uk
 
Last edited:
General muse guys - I have a second inverter arriving in Jan, other half is getting a Tesla Y in a few weeks and I already have a Leaf.

I know re DNO they like to check I'm not over exporting, however is there any issues/risk of over importing? Given we're on Octopus Go, I'd aim to use the overnight window to charge bother cars @ 7kw and the batteries @ 7.2kw - any issues with ~21kw of draw at peak time?
 
katie279 said:
Cool! Gonna be some intensive activity 00:30-04:30 then!
Click to expand...

You could always request a change to go faster 5 hour window, 21:30 - 02:30 and keep the equivalent dated tariff, if you are on 7.5p then it would be 8.25p instead. They have made this change for customers at their discretion, they may offer you a later window, or you could move to intelligent Octopus with your new Model Y.
 
Journey said:
You could always request a change to go faster 5 hour window, 21:30 - 02:30 and keep the equivalent dated tariff, if you are on 7.5p then it would be 8.25p instead. They have made this change for customers at their discretion, they may offer you a later window, or you could move to intelligent Octopus with your new Model Y.
Click to expand...
Thanks journey - yeah, I'm gonna need to work this out once the Tesla arrives to figure out the maths a bit on all those options.

I think most seem to think intelligent octopus would be better when things are a bit more predictable economy wise - maybe in April I might check this again, but go faster is definitely an option!

In fact the more I think, the more sensible this sounds - even in short term as Octopus let you change contract after 30 days I believe, so even whilst I wait for my second inverter, this would let me get another 2.6kw per night during the winter months....
 
Last edited:
Hmm, sorry getting confused - thought you meant Octopus Agile when you said Intelligent Octopus! That's the six hour one.

That's a really good idea, the only challenge is the rate has been whacked up to 12p rather than the 7.5p it was.

Are you suggesting if I request the change, they might honour my current rate or do I have to accept it'll be much more expensive?
 
Yeah, just checking Intelligent Octopus - it's actually at 10p and slightly cheaper day rate, so for six hours, that's pretty good.

Only thing that puts me off is all this faq stuff about how it tries to intelligently charge my car - can I still just schedule to charge through the night my batteries and car or am I tied to it 'smart' charging when it chooses?
 
You must log in or register to reply here.
Back
Top Bottom