Solar Panels UK - is it worth it?

[ACarneiro]

Not a bungalow but the passageway runs the full length of a single storey extension. I wouldn't want it above the extension on the side of the main house as it would be really exposed (to weather and people). It will be nicely hidden away and well sheltered on the side of the extension.

I'm not sure what price the PW3 will be but I've seen estimates of between £11k and £15k. I'm paying a lot less than that for the PW2. Over a 10 year period I'm not sure I'll lose enough round trip energy to warrant the extra cost of a PW3. And I'm too impatient to wait even longer

Well, yes, but the PW3 includes the solar inverter. I suppose it's designed to be fitted at the same time as fitting your solar panels.
I wonder how much the extra PWs to expand the system (presumably with no inverter included) will cost.

 

[notmeguv]

Well, yes, but the PW3 includes the solar inverter. I suppose it's designed to be fitted at the same time as fitting your solar panels.
I wonder how much the extra PWs to expand the system (presumably with no inverter included) will cost.

That's true, I forgot you didn't need a separate inverter.

 

[notmeguv]

Wow, installer has just come back to say DNO (SSE) has approved the G99 application with no restrictions, yay! That was just under 3 weeks for approval. Scaffolding booked for Nov 28th and fitting scheduled for the 30th.

 

[thewanted]

I had the same conundrum for my battery only install. Went GivEnergy in end due to:

• Pricing: This has changed the 2 devices should be at parity cost wise but some installers are not passing on the price drops.
• Battery chemistry: Not a fan of NMC (PW2) in the household storage scenario feel more comfortable with LFP (GivEnergy)

If I had the patience I'd have waited until 2024 and putting in PW3.

• Battery chemistry changed to LFP
• Max output of 11.5kW will power just about any household
• Built in Solar Invertor. So although behaves like a normal AC coupled battery, DC solar can be wired straight in and either charge battery or be DC-AC inverted and feed house or grid.

Tidy looking setup, which should reduce install costs (I believe that was the aim with PW3)

Have you had any issues with the GivEnergy AIO? I'm seeing quite a few moans in various Facebook groups about the SOC randomly dropping to almost nothing which causes the house to draw from the grid. Apparently it'll be fixed in a future software update, but who knows when that will be.

I'm very tempted by the AIO, but I am nervous about being a beta tester at this level of expense.

 

[thewishy]

Have you had any issues with the GivEnergy AIO? I'm seeing quite a few moans in various Facebook groups about the SOC randomly dropping to almost nothing which causes the house to draw from the grid. Apparently it'll be fixed in a future software update, but who knows when that will be.

I'm very tempted by the AIO, but I am nervous about being a beta tester at this level of expense.

It's a fairly standard issue for Giv, but the problem is more to do with LFP BMS as a whole than Giv.
LFP has a fairly small voltage difference across the charge range. So for an NMC battery just grab the voltage, shove it in a lookup table, here's your percentage.
BMS for LFP is more like an accounting exercise, you count Wh in and Wh out. The process isn't perfect - we saw some of the original LFP model 3's dying at lowish-but-not-empty charge because of this - hence the recommendation to charge to 100% weekly.

The account error will continue to get worse until the battery reaches a fixed point, at which things can be corrected. For Giv that fixed point is "when empty"
My 9.5kWh Giv Hybrid goes about 0.5kWh every day it's not empied. So after 7 days without emptying I can expect the battery to disconnect and recalibrate at approximately 37% reported SOC.

If you empty the battery semi-regularly, it's a non-issue. But given the AIOs are quite large, it seems likely that some people don't empty their battery for weeks on end. (For me, I have a home assistant automation which will delay the charge until later at night, combined with a dishwasher and washing machine, that's normally enough to empty it)
There probably isn't a huge amount lot Giv can do to "fix" the "issue" - I'd describe this as a characteristic of LFP rather than a fault as such. It'll be interested to see if Tesla get similar issues with a PW3.

 

[thewanted]

It's a fairly standard issue for Giv, but the problem is more to do with LFP BMS as a whole than Giv.
LFP has a fairly small voltage difference across the charge range. So for an NMC battery just grab the voltage, shove it in a lookup table, here's your percentage.
BMS for LFP is more like an accounting exercise, you count Wh in and Wh out. The process isn't perfect - we saw some of the original LFP model 3's dying at lowish-but-not-empty charge because of this - hence the recommendation to charge to 100% weekly.

The account error will continue to get worse until the battery reaches a fixed point, at which things can be corrected. For Giv that fixed point is "when empty"
My 9.5kWh Giv Hybrid goes about 0.5kWh every day it's not empied. So after 7 days without emptying I can expect the battery to disconnect and recalibrate at approximately 37% reported SOC.

If you empty the battery semi-regularly, it's a non-issue. But given the AIOs are quite large, it seems likely that some people don't empty their battery for weeks on end. (For me, I have a home assistant automation which will delay the charge until later at night, combined with a dishwasher and washing machine, that's normally enough to empty it)
There probably isn't a huge amount lot Giv can do to "fix" the "issue" - I'd describe this as a characteristic of LFP rather than a fault as such. It'll be interested to see if Tesla get similar issues with a PW3.

Thanks It has been in the back of my mind that the very flat voltage curve for LFP batteries could be the issue here.

I had a company over to quote earlier so I'm still very interested in this. As easy as a Powerwall 2 would be, I don't really want to buy into NMC technology at this late stage.

 

[ringi]

It a shame that giv can't empty half the connected batteries every other day rather then spreading usage over all the connected batteries.

 

[thewishy]

It a shame that giv can't empty half the connected batteries every other day rather then spreading usage over all the connected batteries.

Seems like a decent fix for multibattery hybrids (Not sure how many of them are there since the larger 8.2/9.5kWh came out?). Not sure if the AIO is wired such that individual banks can be independently charged / discharged (I know they'll have balancing options, but those don't tend to get used for high current charge or any discharge?)

 

[spon88]

There are still quite a few issues with the GE AIO if the posts on the GE forum are to be believed. It's not easy to determine the ratio of good:bad experience as we all know that the disgruntled folk justifiably shout loudest. That said, there are enough concerns being expressed at the moment to steer clear if it was me. I know there is also a lot of work going on to fix it, insofar as it can be fixed. My guess is that GE will get there or thereabouts with it as they (finally) did with the SoC issues on the earlier batteries, at least for the vast majority of folks.

We have two 9.5s running the latest stable firmware (3.12) and other than a few quirks, the SoC issues have been very manageble. BUT the key is cycling the batteries full to near empty fairly regularly as noted above. In summer this was trickier although Flux export facilitated forced discharge very nicely. In winter, our ASHPs handle the cycling pretty well, but if we don't get sub 20% for any reason, I still force discharge once a week to let the bottom end self-calibration do its thing. It's not perfect but it's good enough to prevent any unexpected SoC dumps, at least with our kit. Hopefully GE will bring the AIO to a similar level of consistency reasonably soon.

 

[GRiLLA]

It's a fairly standard issue for Giv, but the problem is more to do with LFP BMS as a whole than Giv.
LFP has a fairly small voltage difference across the charge range. So for an NMC battery just grab the voltage, shove it in a lookup table, here's your percentage.
BMS for LFP is more like an accounting exercise, you count Wh in and Wh out. The process isn't perfect - we saw some of the original LFP model 3's dying at lowish-but-not-empty charge because of this - hence the recommendation to charge to 100% weekly.

The account error will continue to get worse until the battery reaches a fixed point, at which things can be corrected. For Giv that fixed point is "when empty"
My 9.5kWh Giv Hybrid goes about 0.5kWh every day it's not empied. So after 7 days without emptying I can expect the battery to disconnect and recalibrate at approximately 37% reported SOC.

If you empty the battery semi-regularly, it's a non-issue. But given the AIOs are quite large, it seems likely that some people don't empty their battery for weeks on end. (For me, I have a home assistant automation which will delay the charge until later at night, combined with a dishwasher and washing machine, that's normally enough to empty it)
There probably isn't a huge amount lot Giv can do to "fix" the "issue" - I'd describe this as a characteristic of LFP rather than a fault as such. It'll be interested to see if Tesla get similar issues with a PW3.

Other Inverter/battery solution don't regularly have the same problem, my Sunsynk LFP batteries don't do this at all and it's not an issue in their forums.

Also, while charging to 100% on LFP Tesla's is to help the BMS remain calibrated, I've never heard on here of an LFP Model 3 suddenly dropping large amounts of charge.

100% a Giv Energy issue, maybe not one they can fix in software alone.

 

[thewishy]

Other Inverter/battery solution don't regularly have the same problem, my Sunsynk LFP batteries don't do this at all and it's not an issue in their forums.

Also, while charging to 100% on LFP Tesla's is to help the BMS remain calibrated, I've never heard on here of an LFP Model 3 suddenly dropping large amounts of charge.

100% a Giv Energy issue, maybe not one they can fix in software alone.

Limited experience of others, but I think my key point is that this is manageable if you discharge regularly.

There were definitely some initial problems with Telsa M3 LFPs. I suspect they updated the BMS to be more conservative as to where the bottom is when the top hasn't been seen for a few charge cycles.

Ultimately the Giv battery having a SOC error means you run on mains for a bit, an M3 with a SOC error is a roadside recovery. From my point of view it's all economics after that - Once every couple of weeks you might have an evening on grid, it doesn't add up to a huge cost vs cost of system.

 

[notmeguv]

Scaffolding is up. If the weather holds off fitting should be commencing tomorrow! Knowing my luck it will snow overnight.

 

[notmeguv]

Finally! It’s all in and working. They couldn’t finish it last week as we had some troublesome cable runs from the loft to the porch where the Tesla gateway and inverter are located. They couldn’t come back until today but have completed it all and hidden most of the wiring really well. Just got to apply for an export MPAN with Octopus.

So now I have to learn a load new stuff!

First thing I need to work out is how to charge the PW2 and the car both in the Intelligent Octopus off peak window. I’ve set the Powerwall in the Tesla app to Time Based Control and set the off peak window in there (can’t set fractions of a pence so does it matter if it’s set to 7p or 8p?)

Will the car charge as normal through IO as well, in parallel with the battery charging, or do I need to set anything else?

Any recommended forums or groups to discuss the set up or this as good as any?

 

[Avendit]

I think the car should do it's own thing. Here as good as anywhere for overall system questions, PW etc. There is an octo IO thread that may have people that know more about that, but let's face it, a big overlap!

 

[KennethS]

Finally! It’s all in and working. They couldn’t finish it last week as we had some troublesome cable runs from the loft to the porch where the Tesla gateway and inverter are located. They couldn’t come back until today but have completed it all and hidden most of the wiring really well. Just got to apply for an export MPAN with Octopus.

So now I have to learn a load new stuff!

First thing I need to work out is how to charge the PW2 and the car both in the Intelligent Octopus off peak window. I’ve set the Powerwall in the Tesla app to Time Based Control and set the off peak window in there (can’t set fractions of a pence so does it matter if it’s set to 7p or 8p?)

Will the car charge as normal through IO as well, in parallel with the battery charging, or do I need to set anything else?

Any recommended forums or groups to discuss the set up or this as good as any?

We have a TWC (the dumb version 2) and set one of our Tesla's to charge via the IO app. The other Tesla we use the schedule in the car or app and have it start charging after 11.30pm and set the charge limit to a value that I am sure it won't run over 5.30am.

For the Powerwalls, £0.07 or £0.08 for the off-peak import tariff is up to you, it doesn't affect how the PW chooses to charge. It might make a small difference in the car app's calculation of at home charge cost but will be pretty close either way. I have our settings for the PWs on Time-Based Control, Export Everything, and Grid Charging Yes. Then just input the rate plan info. Prior to the recent Octoplus sessions I just had two rate periods: Super off-peak (£0.07 in and £0.15 out) and Peak (£0.31 in and £0.15 out). When Octoplus was introduced I added a Mid-Peak rate of whatever they are paying per kWh for export and the same £0.31 as the import rate. Since these don't seem to be one-offs I found it easier to leave the Mid-Peak setting and just use the usual Peak rates for the days when there is no Octoplus. Then I just have to change the time of the Mid-Peak and the export rate on the days when an Octoplus is taking place. It means some manual intervention but it works.

Under the picture is worth 1,000 words theory, here is how I have the rate plan setting:

 

[Fat Wolfie]

We are on Octopus Intelligent and our PW will charge up at the same time as the car. The only thing we have found is that when the car is using IO cheap rate outside the core 23.30 to 05.30 timeframe the PW doesn’t recognise this and so contributes to the car charging. The car tends to draw 3.6kwh from the grid and 3.6kwh from the Powerwall under these circumstances. I’m sure there’s some clever automation that can be done but it’s beyond me. 90%+ of the charging takes place in the core cheap hours and I just live with it for the odd 30 minute periods the PW discharges to the car.

out of interest I have set the rates up to show a lower export rate than import rate (so I’ve got off peak import set to 8p and all export time value of 4p). I found that the battery was tending to charge up from the grid cheap rate overnight and exporting loads of solar in the day as the battery didn’t need it. Mathematically it’s the most advantageous thing to do but my personal wish was to be less reliant on the grid than maximise margins/reduce payback periods so by doing what I have done the in the summer the PW would charge from solar mainly, and once full would export. I still get 15p per kWh from Octopus, it’s just that the PW doesn’t know this.

enjoy the new system

 

[notmeguv]

We have a TWC (the dumb version 2) and set one of our Tesla's to charge via the IO app. The other Tesla we use the schedule in the car or app and have it start charging after 11.30pm and set the charge limit to a value that I am sure it won't run over 5.30am.

For the Powerwalls, £0.07 or £0.08 for the off-peak import tariff is up to you, it doesn't affect how the PW chooses to charge. It might make a small difference in the car app's calculation of at home charge cost but will be pretty close either way. I have our settings for the PWs on Time-Based Control, Export Everything, and Grid Charging Yes. Then just input the rate plan info. Prior to the recent Octoplus sessions I just had two rate periods: Super off-peak (£0.07 in and £0.15 out) and Peak (£0.31 in and £0.15 out). When Octoplus was introduced I added a Mid-Peak rate of whatever they are paying per kWh for export and the same £0.31 as the import rate. Since these don't seem to be one-offs I found it easier to leave the Mid-Peak setting and just use the usual Peak rates for the days when there is no Octoplus. Then I just have to change the time of the Mid-Peak and the export rate on the days when an Octoplus is taking place. It means some manual intervention but it works.

Under the picture is worth 1,000 words theory, here is how I have the rate plan setting:

View attachment 997649

Thanks. I don’t have the Energy Exports setting in the Tesla app so I don’t think they’ve configured the battery correctly. It won’t matter at the moment as I don’t have an export MPAN yet so better to use whatever little solar production for now. I’ll phone them on Monday.

Good shout on the extra schedule for the Saving Sessions, I’ll add that, thank you.

*edit* how did get the super off peak sell price set to 15p? When I try that it says the export price cannot be higher than the import price.

 

[ACarneiro]

Thanks. I don’t have the Energy Exports setting in the Tesla app so I don’t think they’ve configured the battery correctly. It won’t matter at the moment as I don’t have an export MPAN yet so better to use whatever little solar production for now. I’ll phone them on Monday.

Good shout on the extra schedule for the Saving Sessions, I’ll add that, thank you.

*edit* how did get the super off peak sell price set to 15p? When I try that it says the export price cannot be higher than the import price.

I have the same problem. It doesn’t really matter, just set both the import and export prices to 8p/kWh and the system will still behave as desired.

Regarding your export, that’s not for the installers to sort out. You should have an export limit set by your DNO.
Just email [email protected] with your Gateway serial number and tell them what the export limit is and they will enable it remotely within a day or so

 

[notmeguv]

We are on Octopus Intelligent and our PW will charge up at the same time as the car. The only thing we have found is that when the car is using IO cheap rate outside the core 23.30 to 05.30 timeframe the PW doesn’t recognise this and so contributes to the car charging. The car tends to draw 3.6kwh from the grid and 3.6kwh from the Powerwall under these circumstances. I’m sure there’s some clever automation that can be done but it’s beyond me. 90%+ of the charging takes place in the core cheap hours and I just live with it for the odd 30 minute periods the PW discharges to the car.

out of interest I have set the rates up to show a lower export rate than import rate (so I’ve got off peak import set to 8p and all export time value of 4p). I found that the battery was tending to charge up from the grid cheap rate overnight and exporting loads of solar in the day as the battery didn’t need it. Mathematically it’s the most advantageous thing to do but my personal wish was to be less reliant on the grid than maximise margins/reduce payback periods so by doing what I have done the in the summer the PW would charge from solar mainly, and once full would export. I still get 15p per kWh from Octopus, it’s just that the PW doesn’t know this.

enjoy the new system

Thank you. I have a Zappi charger and the electrician changed a setting on that to ensure it wouldn’t charge the car from the battery.

 

[notmeguv]

Regarding your export, that’s not for the installers to sort out. You should have an export limit set by your DNO.
Just email [email protected] with your Gateway serial number and tell them what the export limit is and they will enable it remotely within a day or so

Thanks, I’ll have to ask the installers what the import limit is. All they said was the DNO had given a full green light.