Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Solar Panels UK - is it worth it?

This site may earn commission on affiliate links.
To buck the trend of GivEnergy / MyEnergy installs in this thread I thought I would share my experience with a SunSynk install.

We had started looking last summer but very few local people were even returning calls, we eventually got 3 different quotes from £16K - £20K which I felt were all too high. We then started with First4Solar based on some of the positive experiences earlier in this thread. Originally we were going to have a 5KW GivEnergy Hybrid with 9.5KWh battery install from First Solar for about £14K, they advised it would take 6 months, but then went bust after about 5 months. Their customer list was purchased by Contact Solar who offered to complete an equivalent install for the remaining price after deposit, which is therefore 80% of what we were going to pay (I seem to have got my deposit back from credit card provider successfully).

Sunsynk
Contact Solar have an interesting approach, more on them later, but they said for that price we would have to move to a SunSynk Ecco 5KW Inverter with 2 x 5.32KWH batteries which have a 90% dod, giving us the same as before. I've not seen Sunsynk talked about often in this forum like this but eFixx had a positive opinion of them and they seem to have a fairly full product range. They do seem to have a rather quirky CEO Keith making interesting videos so they seem pretty actively developing their range. I quite liked that the inverter has a touchscreen to program the device on the front as well as an app, so even if Sunsynk disappeared I feel the equipment will be functional, I don't like being dependent on clouds for vendors I don't know will last the warranty period. Talking of which, 10 years warranty included, 6000 battery cycles.

The Sunsynk app is entirely functional, they've got integration with Octopus Agile and Flux already and apparently there's going to be a big app update coming soon.

1690651549638.png


1690651503716.png

I've done an integration with Home Assistant already which pulls data from a web api, seems to work pretty well. There are also people that integrate locally to the inverters RS434 port, either replacing or in addition to the Sunsynk monitoring dongle, I might go down this route in the future for more realtime monitoring.

Contact Solar
Slightly unusual sales process for a fairly large cost, there was no contract or deposit prior to the installation, just a verbal agreement of the price. They claimed to be doing us special deals as former First4Solar customers, the price was certainly very competitive.

16 x 405w JA Solar All Black split panels
5KW Sunsynk Ecco
2 x 5.32KWH Sunsynk batteries
Full install across 2 roofs for £11,400

They did recommend 9 panels have Tigo optimisers due to a chimney, I decided to not go for them having watched some videos comparing optimisers to the built in bypass diodes. I feel that given I have a fairly steep 40 degree roof and a 50:50 split East/West I'm really not going to be effected by shadows when the suns not high enough to crest the roofline anyway. Installed price would be £600 so that would be an awful lot of optimisation. We'll see who is right over time.

Everything through the process seemed to run like clockwork, survey, booking scaffolding, installation date within 6 weeks. The install was contracted out to another party who arrived as expected with all the materials from Contact Solar. After they finished I wasn't totally happy with how they had left the batteries freestanding on my loft floor, Contact sent them back 2 days later to secure them to the wall. They were willing to follow a slightly complex cable route through the house rather than on the outside front aspect, and also extending the EPS sockets down into the house rather than being in the loft with the inverter (I'm not sure what I'll do with them, haven't had a powercut here in years, but it would have seemed annoying if we did and were unable to use the battery).

It's not really the neatest install compared to the Youtuber/Instagrammer electricians, but I dare say you are paying for it. Given it's well out of view in the loft I'm happy.


20230729_143126922_iOS.jpg


Post install they ask for payment immediately, and don't give you the inverter access until you pay, fair enough. Within a week I've had the scaffolding removed and all the MCS paperwork stuff organised. They feel very efficient.

Conclusion
I feel I've had an excellent deal, and a low hassle process, I would recommend Contact Solar and Sunsynk to anyone like me who got a bit fed up of the process for getting solar, swift and cheap. I'm surprised Sunsynk don't get more coverage, their products are very flexible. There's going to be some kind of HW diverter coming in a few months, and there is some kind of EV charger in the works. It would be nice to have a native Home Assistant connection like GivTCP, but I don't rule it out being the sort of thing that may come in the future, and there are solutions for that today through connecting other devices to the Inverter.

The last week has been great, negligible grid use as you would expect with 16 panels in July. We are desperately trying to use up excess when it's sunny while we wait for the export tarrif stuff to be set up.
 
was sunny this morning so put the zappi on eco mode. Forgot to turn it off and noticed it was raining. By the time I stopped it my battery was down at 70%. Its now at 53% as I’m using the oven for dinner. Should still scrape it to off peak hopefully. Don’t mind if the battery needs to top up on off peak - effectively thats just transposing what went into the car, but peak kwh would be painful
 
It's not just in my head... July has been really bad for solar here (South Oxfordshire). Today we're forecast 10kWh and it's rain and heavy cloud, so entirely likely. And that's it, putting it almost level with April!
 

Attachments

  • SolarOutput.PNG
    SolarOutput.PNG
    56 KB · Views: 33
It's not just in my head... July has been really bad for solar here (South Oxfordshire). Today we're forecast 10kWh and it's rain and heavy cloud, so entirely likely. And that's it, putting it almost level with April!
It's been nasty here in Edinburgh too. We were away for most of it fortunately, but limping through just now as we are mid transition to IO so don't have any cheap time overnight ☹️.
 
  • Like
Reactions: Pete UK and phil4
yeah, it is horrendously tragic :D

but still managed to live more or less solar/batteries only whole time since installation (apart that epic fail we had).

now still waiting for the audit report...

also, as the result, system installation (at least for now) is modified from the Smart meter to CT Clamp, which has a major drawback - at low currents (when house usage is more or less below 300w) it sees more or less no house usage.
 
July has been pretty disappointing for solar here too. We should manage around 760 kWh generated by the end of today. PVWatts estimate for July is 856 kWh for our two arrays, so well under theoretical. June made up for it though, with just over 1 MWh generated versus 837 kWh estimated.

I have to say Solcast's forecasts have been abysmal this month too. More often than not over estimating daily production by a fair margin.

Still, net cost for July is around £13 :)
 
  • Like
Reactions: WannabeOwner
We went with PAS24 Security (Secure by Design) patio doors which included heavier frames, deadbolts, secure locks & hooks, plus 6mm laminate glass.

If you want to know the difference between standard glass, toughened glass, and laminate glass... watch the test below

Have burglars git bright enough to carry s blowtorch to melt the laminate interlayers? Can that be done?
 
I have to say Solcast's forecasts have been abysmal this month too. More often than not over estimating daily production by a fair margin.

Me too ... although I haven't reviewed the raw data I get form API - I have that data, but what I usually look at is the graph on their website (which they told me was using their old algorithm, unless it has been updated since, so the raw API data should be "better")
 
Me too ... although I haven't reviewed the raw data I get form API - I have that data, but what I usually look at is the graph on their website (which they told me was using their old algorithm, unless it has been updated since, so the raw API data should be "better")
It's not just me then? I thought it was lauded as the best thing since sliced bread, but have found it really no better than HA's default.

I've tweaked the system efficiency on their site down to the 70's, as was always over estimating.

Since doing that, it's been vaguely right... today started off as ~13kWh estimate and is now down at 9kWh. Sure that's 30% out, but both are low numbers so it's had little impact. I've given up using it's forecast any further than tomorrow, and I take today's estimate with a pinch of salt.
 
Me too ... although I haven't reviewed the raw data I get form API - I have that data, but what I usually look at is the graph on their website (which they told me was using their old algorithm, unless it has been updated since, so the raw API data should be "better")
The raw data's accuracy does improve as the day progresses; it's the day-ahead forecast which is always too optimistic. Here's today:

Screenshot 2023-07-31 at 10.56.23.png


I have a program which polls the Solcast API throughout the day, updating the values in my InfluxDB, graphed above next to actual solar generation. Today's estimate started off at around 17 kWh first thing this morning. I expect the current forecast of 14 kWh is still wide of the mark.
 
anyone managed to automate the HA to have Inverter settings changed if IO time slot is provided?

lets say - if IO on from 1:00 - 5:30 then same time frame on inverter to be set as Battery first?
Sort of. I don't touch the inverter as I don't care what that does. I bump the charge level up on the batteries when we're in an IO smart slot... and drop it back down when out of it. The IO api will tell you if you're in a smart slot, and or upcoming ones. The thing to watch in my experience is the 5:30am end of the normal smart slots... for some reason Octopus keep indicating it's a smart slot for a couple of minutes after, which if you're using that will mean a couple of minutes every day that your battery could be charging, but it shouldn't.

Here's my YAML, note it also sets the Zappi charge mode to Fast, so the car charges. The calcs are to decide what level to charge the battery to/keep it at:
alias: Intelligent Octopus, Charge PW
description: ""
trigger:
- platform: time_pattern
seconds: "15"
condition:
- condition: time
after: "05:35:00"
before: "23:25:00"
weekday:
- mon
- tue
- wed
- thu
- fri
- sat
- sun
- condition: state
entity_id: binary_sensor.octopus_intelligent_slot
state: "on"
action:
- device_id: 60e4b5f3e9834f5eac44bdab9b14eace
domain: select
entity_id: bbf1498fbacb81d3e3307a40598300cc
type: select_option
option: Fast
- service: number.set_value
data:
value: "{{100-((states('sensor.forecast_today') | float)*1.5)|int }}"
target:
device_id: d8411b6adba0fec69752749ff56cec72
mode: single
 
if you're ok doing it manually (my suggestion earlier hopefully keeps it less likely which I'd prefer) then you can also drop the battery discharge right down to 0 to pause or to 300w or whatever works to cover baseload. 0 would be ok as it'd pull from the grid but it'd be cheap rate.

I'd quite like a battery tweaked version of IO to stop all this fuss

option 1 - Octopus nice but dim
- give me the 6 hours still
- keep the app control
- do *not* give me extra slots. As a battery owner its annoying
- you *can* adjust which slots you pick in that 6 hour window to bias towards cleaner/cheaper slots and still have that dynamic control
- I may be willing to negotiate on max kwh that can be allocated if you have to use the 6hr window. Eg assume 4 is usable so 28kwh


option 2 - octopus genius
- basically intelligent, but also integrates with your battery API to prevent discharge. This is something they can apprantly do with givenergy already with the intelligent flux offering, so how about bringing that to intelligent?
 
I thought it was lauded as the best thing since sliced bread

IME it [very significantly] better than anything else I have found ... but don't try putting your sliced bread next to it!

The raw data's accuracy does improve as the day progresses; it's the day-ahead forecast which is always too optimistic.

I am using prediction data during my Off Peak period, in order to try to end Off Peak with "just the right battery %age" ... so if it can be right by towards-end-of-Off-Peak I'm happy ... and I'm on E7 so that gives me a lot more scope than a 00:00 - 04:00 tariff

I mostly don't care too much during the day, as my battery is in whatever SoC it is ... I only use the prediction during the day for whether I should charge one/both cars - e.g. if the battery will easily fill by end of day then charge car(s) concurrently, to reduce the charge rate to battery - for longevity, or simply because PV will generate more kW than I can dump into a car! so need car charging to be concurrent with battery, rather than just "when exporting", because once battery is 100% the export will be more than 7kW x 2 cars!
 
I am using prediction data during my Off Peak period, in order to try to end Off Peak with "just the right battery %age" ... so if it can be right by towards-end-of-Off-Peak I'm happy ... and I'm on E7 so that gives me a lot more scope than a 00:00 - 04:00 tariff

I mostly don't care too much during the day, as my battery is in whatever SoC it is ...
That's exactly what I use it for... but taking today, during the off peak this morning (all the way to 05:30am) it was predicting today to be 12.81kWh, so I charge the batteries to that, since then the prediction for today had steadily dropped to 7.63kWh. That's a pretty poor prediction for just 12 hours ahead. Like you, I'll not be doing anything about it, out of the cheap window, but equally I suspect I'll be absolutely fine, as batteries have enough charge even for this.
 
  • Like
Reactions: WannabeOwner
I suspect I'll be absolutely fine, as batteries have enough charge even for this.

I'm increasing my PowerWalls from 2 to 3.

On best days, this time of the year, I get to 100% (from PV) and maintain that until around 21:30 and then overnight falls to about 25% ... so i have some slack for a following cloudy day.

But on the cloudy day I may fall such that I only just reach Off Peak ... and then the [API :) ] question is how much to charge during Off Peak

Once I have 3 PowerWalls I'm going to arrive on the cloudy day at much more than 25% (across 3x PWs), so likely to have enough for "tomorrow night" but i suppose I will still like days where "yesterday was cloudy", and I manage to run all three PowerWalls down to 0% just as the sun comes up ... and then they are 100% full at SunDown ... and also needing to handle multiple-cloudy-days-in-a-row.

So (I think) the upgrade is just going to provide more leeway - in particular on the days when SolCast is "way off"

(Of course this upgrade purchase is more to do with Winter and being able to get through the whole day having charged on Off Peak, but also recognising that it allows "Looser" prediction accuracy in Summer PV)
 
  • Like
Reactions: phil4