Considering getting a Powerwall

[NastyNick83]

tl;dr - for us, battery capacity made very little difference in cost savings.


It was based upon the capacity that we knew about the original Powerwall spec at the time and a few simplifications such as not taking into account power limits so was simplified but the answer that I was after was ballpark not exact - even if 100% out, it would be a useful number and it would have been far better than that - fact checked again my actual annual bill - its impossible to save more than my actual bill for the period. I was on a fixed tarrif so a simplified approach would have given an almost best case scenario.

The simplified approach:
Maximum stored PV excess for the day = deemed days solar export capped at capacity of the battery so 100% usage of excess solar - assumed 100% round trip efficiency and no charge limit (latter likely to have no effect as days when peak power limits breached would likely to exceed battery capacity limit)
Maximum stored energy reuse = measured days energy usage less maximum stored PV excess for the day - assumed no limit on import.

The above had very small difference in saving between a 4kWh battery and a 6.1kWh battery. Above the daily energy usage, size of battery had no effect. This however assumed that residual charge in the battery was lost at end of the day but this would be offset by the assumption that energy could be used before it was generated - averaging these out, there would not be much in it, but the larger the battery capacity, the larger the difference, but pretty minor.

I've just gone back and revisited the spreadsheet and done some what-if's on battery capacity. These values are based upon our energy usage - others will have different usage patterns. But i've put in our energy usage for that year, 2015. Figures were based upon 15p/kWh electricity price (we are less than that so current savings would be even less). Values also based upon CT clamp meaturing, so not that accurate at low power levels.

4kWp PV System, ~3360kWh generation, ~2060kWh calculated export (no export meter), 1830kWh imported (current ~2100 metered), at 2015 price/usage 0.15p/kWh = £308 (current price and usage 0.138p/kWh = £290). IF WE IMPORTED ZERO ELECTRICITY THESE VALUES REPRESENT THE MAXIMUM ANNUAL SAVING POSSIBLE THAT WE CAN MAKE. The figures predate EV charging so savings now potentially better.

What ifs...
Usable battery capacity (kWh) / £ per year export stored / £ per year saving at 15p/kWh
1 / £42 / £42
2.4 / £87 / £79
4.8 / £155 / £115 - probably best bang for the buck - a system may cost around £2300
5.8 (7.2kWh @80%)/ £178 / £120 - just added so not reflected in text below. System price around £3050.
7.2 / £210 / £121
10 / £257 / £121
14 / £292 / £121 - approx Powerwall spec at I believe around 9k installed.
20 / £308 / £121

System prices do not include installation

So based upon the above and our 2015 usage patterns, somewhere between 4.8 and 7.2kWh storage, delta savings converges to 0. So no cost saving to be had exceeding 7.2kWh storage - within caveats mentioned above. Sweet spot seems to be somewhere between 4.8kWh and 7.2kWh. We have gone 7.2kWh, 5.

Assuming cost saving of £121/year, 10 year payback of £1210 - the battery system that I have ordered is £3k give or take a few ££. So that on paper is not going to pay back even in 20 years which is not good as a 10 year life span is probably to be expected.

As a sanity check, this gives self consumption at around 80%, which is inline with many published values. So I am fairly confident that the simplified approach is within reasonable bounds.

However, at time of calculating, I took other factors into account, such as energy inflation (its actually deflated) and assumption that my figures were 100% out being generous to wanting to justify a battery system, the real saving for us is somewhere between these approx figures and £308 (currently £290). Now with EV charging, and ability to time shift in winter and possibly move to something like Agile (this will flatten the evening peak) these figures above will be lower than current usage patterns, but I suspect cost saving wise, our system will come in close to 10 year or less break even.

All above assumes 100% round trip efficiency.

Love this post!

Once you've got your batteries, I would get switching to an economy 7 tariff or better still, an agile tariff - if you can use the batteries to avoid drain between 4-7pm you'll be considerably better off than the 13.8p/kWh

Nick

 

[ACarneiro]

Love this post!

Once you've got your batteries, I would get switching to an economy 7 tariff or better still, an agile tariff - if you can use the batteries to avoid drain between 4-7pm you'll be considerably better off than the 13.8p/kWh

Nick

E7 is about 8-9 p/kWh so the difference to Agile may not be quite so significant

 

[Battpower]

Love this post!

Once you've got your batteries, I would get switching to an economy 7 tariff or better still, an agile tariff - if you can use the batteries to avoid drain between 4-7pm you'll be considerably better off than the 13.8p/kWh

Nick

If you take the cost of the battery into account and the fact that lithium has a pretty well defined maximum energy throughput, energy costs need to be very low (3-4p pkwh) for it to make sense time shifting energy. If you have solar PV then you have the added dilemma of if you are paying to charge over night compared with free solar charging the following day. Even using weather forecasts it often doesn't work out how you expect.
Of course recently there has been no need to charge over night, but still worth paying for night rate E7 rather than drain batteries:

Compared to winter months.... This from Feb before lock down so charging cars, heating thermal stores and charging home batteries too. Really have to be careful not to get too close to 100 amp!

What this graph doesn't show is the night time stored energy being used during the day to run the heat pump, but notice that even in Feb you can often take a decent chunk out of daytime electricity use.

 

[MrBadger]

@Battpower what is your solar installed kWp and how much battery kWh store and kW store/discharge?

If I understand, direct use = from Solar PV?
From your comment and from the graphs, you are not time shifting electricity - other than to run the heat pump possibly - see below?
The winter 3/6am hump - is theat heat pump running to get heat into the thermal stores (water?) and thermal mass of the house, to discharge the rest of the day or the battery charging which then runs the heat pump during the day but is not showing that discharge on the graph?

 

[Battpower]

@Battpower what is your solar installed kWp and how much battery kWh store and kW store/discharge?

If I understand, direct use = from Solar PV?
From your comment and from the graphs, you are not time shifting electricity - other than to run the heat pump possibly - see below?
The winter 3/6am hump - is theat heat pump running to get heat into the thermal stores (water?) and thermal mass of the house, to discharge the rest of the day or the battery charging which then runs the heat pump during the day but is not showing that discharge on the graph?

It's complicated!

But relating to this graph, 6kwp panels (tiles) ssw aspect near Derby.

There are potentially 'humps' due to:

Heatpump (2.5kw in / 8kw out) heating a thermal store as you say to bank energy for day use. Also heats some underfloor.

Charging a couple of Pylontech batteries for general use during the day.

Charging 9kwh LG Chem batteries specifically for backup power if grid fails, running Heatpump during the day and emergency charging EV during day as well as additional power to allow 7.5kw charging during day when solar PV output alone isn't sufficient.

Charging EV.

 

[Battpower]

There are so many 'wheels within wheels' it's hard to convey simply what's going on, but this annual graph is interesting.

Jan to Mar 2019, red grid energy is 75% E7 night use, 15% unavoidable day use and 10% from a separate solar PV system that's included with Grid Supply on this graph. In September I finished installing the Heatpump so over winter Grid consumption obviously went up significantly, but approx 85% night use and time shifted night rate and dropped my oil heating bill. Still waiting for oil cost calcs but looks like approx £500 pa saving switching to Heatpump.

The summer months show virtually all the Jan to Mar grid use including car charging being replaced by solar PV.

I should add that all the blue on this graph is day solar stored for later use and no time shifting. The time shifting is on the other two systems. But the batteries make a steady contribution throughout the year and this has remained constant since 2013 when installed.

I have one more interesting graph from the Immersun immersion heater diverters which can heat the store hotter than the Heatpump so although they don't give the extra gain of the heat pump, it allows me to keep heating the thermal store up to 80 Dec C before finally having to start exporting! I'll try and get that graph off another device to post.

 

[Battpower]

The CONSUMED figure here is bogus because energy use gets double counted, but red import is correct as is exported.

 

[MrBadger]

3 x PylonTech 2000 batteries, 1 x Sofar ME3000 inverter, box of cables, mounts, other installation bits delivered today. I'm sure a PowerWall would be better packed though!

Next job, installation... I'll probably do the DC and comms side of that myself. But getting a sparky to install an additional consumer unit (to double up for dedicated EV charging too) may not be so easy these days. If anyone can recommend one North Surrey/Hants, please PM me.

A nice thing about being modular, is that it is easy to expand, upgrade, or fix. The battery side looks to be very much DIY with supplied plug in cabling and a easy to fit earth strap.

 

[NorfolkMustard]

Keep us updated on installation progress

 

[gangzoom]

Half way through the month, and so far our total household grid electricity usage is a crazy 6.4kWh including car charging, or 80p!!

 

[SirRob]

I haven't read the entire thread, but my belief had been that my position on the globe meant that I simply didn't get enough sun to make it worthwhile. Is that not the case?

 

[MrBadger]

Yes. That is typically the case, although there are exceptions. Financially, the sums do not stack up for most users and even cheaper solutions, its hit and miss whether financial breakeven in 10 years is achievable.

 

[Battpower]

I haven't read the entire thread, but my belief had been that my position on the globe meant that I simply didn't get enough sun to make it worthwhile. Is that not the case?

Where exactly on the globe is 'Not-Required'?!

 

[gangzoom]

I haven't read the entire thread, but my belief had been that my position on the globe meant that I simply didn't get enough sun to make it worthwhile. Is that not the case?

We live in Leicester, 4KW panels split between 2KW SE facing, 2KW SW facing (so not ideal). In November - March solar hardly made an ident into our energy use. Combination of inefficient cold car been used for short trips, no sun, meant out grid use was pretty high.

However since the lock-down started (actually a bit pre), with the extra sun, our gird electricity usage has fallen dramatically. Its really quite impressive to see that we are pulling less than 1kWh a day average from the grid. That's for a family of 3 with electric cooker, EV been charged to do about 400 miles per month at present, and all the other normal electrical white goods.

We only had our PowerWall install in November, so it will be interesting to see how it changes our grid energy usage over the whole year. But certainly when the sun is out, even if a small number of homes shifted to battery storage it would solve any concerns the grid has about meeting EV charging demands.

If its worthwhile financially is a different matter.........Our solar PV panels are part of the old FIT scheme - so we get 14.4p per kWh generated, and assumed 50% export rate of 5.5p. The panels generated 400kWh last months, so £67.7 in FIT payments. Add in energy saved by not needing grid power - lets assume 14p per kWh rate, thats a potential saving of £123 last month versus using just grid power. Not huge amounts, but not bad.

 

[Battpower]

£123 last month versus using just grid power. Not huge amounts, but not bad.

Try making that on any similar investment.... And for many, the electricity saving is worth more because you would pay for grid consumprtion with post-tax earnings.

 

[MrBadger]

Keep us updated on installation progress

Batteries installed and linked. Inverter will be in a separate space. So far, very much DIY.

Each 48V battery is 2.4kWh nominal, total stack 7.2kWh nominal, 5.8kWh usable at 80% discharge, charge/discharge 60A/48V = 2880W into 3kW inverter. Warranty 7+3 years, life ~ 6000 cycles. Cost of each battery as installed (vat/delivery/stacking frame) ~£800. Inverter extra.

 

[arg]

life ~ 6000 cycles.

That of course is the flip side of the "how big do I need?" question - the larger the capacity (of given tech), the longer it will last at a given usage. So when we were saying that above a certain size was uneconomic because it doesn't often get used, that is partly offset by by the fact it will last longer. But actually putting a figure on it is very tricky.

 

[MrBadger]

That of course is the flip side of the "how big do I need?" question - the larger the capacity (of given tech), the longer it will last at a given usage.

I agree. Putting that into numbers though, as cycles normally means full cycles, not top ups, for this stack, its likely to be around 36MWh cycled lifetime capacity conservative give or take - it will still work, just not at 'within spec' capacity. If it was used solely for my solar export, which if I captured 100% is around 1.8MWh/year, it exceeds the 10 years that I am expecting out of the system, even if its at reduced capacity. By which time, battery technology will have moved on, and prices possibly reduced.

Adding my real financials to the 36MWh mix, thats £1800@5p, £[email protected], £[email protected] time shift delta on an battery unit value of ~£2300 (total cost ~£3100 ex fitting). And thats before taking into account round trip losses. So breakeven numbers before warranty expires only make sense if solar was offsetting full unit prices. The sums are more complex for Agile - if only used to offset peak rate, I suspect numbers look better (but its a huge amount of normal [not fake ie consuming energy in this period that would not normally be consumed during that period, ie car charging] energy needed to be consumed 4-7pm to fulfil this assumption), but worse if looking at daily average.

So if wanting to look solely at a financial return, what/when you use the stored energy for make a big difference to the cost effectiveness of the system. For Powerwall, I would imagine (I don't have the exact numbers immediately to hand but, if terms unchanged since article written, warranty seems to be limited at 37.8MWh Graph Of The Day - Powerwall 2 warranty is under 9 years for most households if cycled once per day | RenewEconomy so not necessarily hugely different warranty wise, but you would expect battery still to perform out of warranty) that you could better the numbers above assuming (a big if, you would need a pretty big PV export that you could self consume) that you could use the full capacity of the battery before something else failed, but the battery is just shy of £10k installed, so financials even tighter - imho, not achievable for average Joe.

So its easy to see why financially, battery storage does not stack up for most. But its not just the financials.

imho.

 

[Battpower]

So breakeven numbers before warranty expires only make sense if solar was offsetting full unit prices.

That was (is) my approach. I looked at what I was 'throwing away' with solar PV export and looked how I could make best use of it. Simplest was a diverter to turn on immersion heater, but with my original conventional hot water tank, that had only limited possibilities / storage potential. When you add together the green (stored in battery) and blue (exported) energy, that looked like by far the largest proportion of my solar PV 'crop'. In terms of my ability to make use of what I generated, solar PV shifted for use before E7 rate cut in had the same value as direct PV during day time.

This is with no EV charging in 2015. All the dark blue 'export' looked like free energy if I could find a cost effective way of using it. Even inefficient use was better than 'throwing it away' as long as the setup costs / payback period made sense.

And here is the matching consumption graph.

Yellow is direct use, light blue shifted through battery and red consumed from grid. This was one factor that pushed me to get my first EV because avoiding buying fuel for a car represented the highest value use of PV energy.

At that time Zappi didn't exist, but that would probably have saved me a bunch of kit that was needed to charge the car at 7.5kw.

7 years on and the original lead acid batteries are making a bigger absolute contribution to my energy production today than when they were new.

 

[Battpower]

the battery is just shy of £10k installed, so financials even tighter

That would have been a killer for me. You need to be at half that figure or less for it to make sense - and even then as you say, primarily using PV because time shifting night rate is generally not worth it. Of course, if you get paid to consume or 'free', then why not, but I don't see that as a solid expectation to design a system around.