Smart Meter #3: My Smart Meter Journey Part 2

Introduction

As promised an update on my smart meter; how did Octopus respond to my latest e-mail and follow up phone calls on 2nd and 18th Sept...

...Engineer booked to call on Friday 20th Sept with a flying lead for the comms hub.

 

Engineer Visits with a plan: 2 Things to Try

 

1. Single Band Comms Hub:

He changed the comms hub to a single band unit because he hard heard on the Octopus smart meter installation engineers' grapevine that the single band HAN network hubs were better with weak WAN signals (he actually installed the same model as was installed on my 1st smart meter in June 2021, the EDMI CS010A. This was mounted directly on the electricity meter, as per the June 2021 installation, but this time the KAIFA MA120 meter.

 

Result: commissioning did not work 1st time (his office based colleague could not connect to the comms hub), so he tried the next thing...

2. Flying Lead:

Using the same single band comms hub and electric meter he added the KAIFA FL100 Flying Lead. The location of the hub was selected as the furthest convenient location from the meter as the flying lead length would allow (further left or up was restricted by central heating and gas pipes). This gave about 53cm extra height and was about 13cm to the left (see images below).

 

The same model of comms hub as was installed back in June 2021 (above photo of June 2021 install)...	...but this time with the FL 100 Fly Lead to get more height and a little to the left

 

The two ends of the FL100 mount on the electricity meter and the comms hub. It appears that connections are common between brands as here the KAIFA meter connects to the EDMI comms hub directly an via the FL 100 flying lead (see close up images below).

 

The engineer did not believe that this would work and I must admit that the short lead length giving only a 50cm gain in height did not seem likely to have a significant impact.

 

KAIFA FL 100 flying lead mounted on top of the MA120 electricity meter	Other end of the FL100 Flying Lead mounted under the comms hub

Result: his office based colleague connected to the comms hub straight away, allowing the electricity meter, gas meter and In Home Display (IHD) all to commission 1st time.

 

Discussion

The various hardware configurations and commissioning attempts are summarised below:

 

Event	Comms hub	Electric meter	WAN worked	Notes
June 2021 Engineer Visit	EDMI CS010A Direct mount	EDMI ES-10B	No	No WAN communications - smart meters working in dumb mode
May 2023 Engineer Visit	EDMI CS010A Direct mount	EDMI ES-10B	Yes	WAN communications established a few weeks after the engineer's visit. Electric meter working correctly, gas meter and IHD not communicating.
March 2024 solar system installed	EDMI CS010A Direct mount	EDMI ES-10B	intermittent	After my solar system was installed, Octopus starts to miss readings on around 5%  of days. Manual retry from Octopus would capture missing reading.
June 2024 Engineer Visit	EDMI CS020A-06-32 dual band hub Direct mount	KAIFA MA120	No	No WAN communications - smart meters working in dumb mode
Sept 2024 Engineer Visit	EDMI CS010A Direct mount	KAIFA MA120	No	No WAN communications  after single attempt to establish communications - no extended test.
Sept 2024 Engineer Visit	EDMI CS010A FL 100 Flying Lead mount	KAIFA MA120	Yes	WAN communications established on 1st attempt. Currently showing just over 30days of data with no missing days.

 

There is some danger here of making a generalisation from a limited data set, but here goes:

 

• Signal is weak where my electric meter is located, but smart meters are designed to keep on trying to commission indefinitely (eg KAIFA MA 120 will try every 4 hours) and so the original installation sprung into life after the May 2023 engineer's visit.
• It is likely that the installation of my solar system (inverter and battery) did impact the WAN signal so that after this about 5% of daily readings were missing (but could still be read when Octopus performed a manual retry - see diagram below).
• After installing the comms hub around 50cm higher the best WAN signal was seen:
• Commissioned on 1st attempt
• No missing daily readings (so far)

The comms hub is now about 5cm below the top of the battery and located opposite the gap between the battery and inverter (see photo below)

 

DCC provide communications hubs and also supporting information - the diagram below, from DCC, would indicate that my solar power inverter and battery would not be classified as 'significant metallic obstructions' by DCC. Neither the solar installation engineer nor the smart meter installation engineer highlighted this as an issue. However, the battery &/or inverter do seem to have some impact, causing some missing daily readings. I am assuming that the battery and inverter meet electrical interference standards and so that leaves their earthed metal boxes as a likely cause of the problem.

 

Diagram taken from DCC document Communications Hub Supporting Information v2.0 In the 'North' region (Arqiva network) the comms hub should not be installed with a 'Significant Metallic Obstruction on 3 or more sides' where dimensions of the object are greater than: X: 32cm Y: 18cm D must be more than 18cm   In this case dimensions are: X (battery): 80cm Y (battery + gap + Inverter): 128 cm D: 285cm Ie a significant object on 1 side only and more than 18cm from the comms hub.

Top of battery (left) is about 5cm above the comms hub. Both the Inverter (right) and battery  are earthed metal boxes, mounted on the opposite side of the garage (approx 2.8m wide).

 

My take is that the DCC specification of a 'significant metallic obstruction' is too 'black and white':

• Objects that are more than 18cm away and/or not on 3 or more sides will have a small impact
• In weak signal areas then a small impact may cause failures.

 

There are just under 54 million electricity meters in the UK and DCC report that their network covers all but 0.7% of these, or 380,000 properties. I live in a village within spitting distance of the Greater Manchester border; I struggle to believe that my house is within the 0.7% area of non-coverage (it would be interesting to have DCC comment). So should that mean that a smart meter installation, that meets the DCC installation specification, should always work?

 

Perhaps surprisingly, the additional 50cm height of the comms hub makes a big difference, and this also helped the comms hub 'see past' the inverter and battery to the (unknown to me) location of the Arqiva/DCC antenna...

 

...of course all based on this one install (it would be interesting to hear from others in the 'North' on their experience of the flying lead).

 

Conclusion

The FL 100 Flying lead enables a slightly higher comms hub location that, in this case, makes a significant difference to the signal strength and also somewhat helps the comms hub 'see past' metallic objects.

 

Suggested Future Work:

• Octopus &/or DCC/other installers trial the FL 100 unit at other premises that are thought to have a weak signal - what proportion of the Northern region non-working smart meters start working with this additional, approx 50cm, of height?

 

• DCC specify some longer flying lead lengths. What is the maximum cable length possible? Perhaps a 10m and 15m option could be produced to allow the comms hub to be located in an attic and maybe a weather proof box to mount the comms hub on an outside wall.

 

• DCC publish a map of signal strength. It would e useful to know if a target smart meter location is in the 0.7% of properties that DCC believe are not covered.

 

I believe that the above will save money, by increasing the proportion of smart meters installed first time AND improve customer satisfaction...

 

...I have loads more ideas for stream lining smart meter installation, I feel another post coming on.

 

And Finally:

Many thanks to all the folks at Octopus who helped to get this resolved: Matt, Becky, Pardip, Joe, Lexis, Jordon, Tia, Amie,  Mercy, Arran, Lou, Georgia, Harrison, Jack, Leigh, Daisy, Katie - the journey took over 3 years, but, thanks to your hard work and persistence we made it in the end.

 

Feb 2025 Update:

A neighbour sent me this photo of his smart meter. He had a number of elecricity suppliers and their subcontractors try to get his smart meter comms working. He used this post to convince his supplier (this time Ovo)  to try the FL100 flying lead to get the comms hub working; it did the trick and he is now getting cheap rate electricity to charge his EV.

 

 

Smart Meter #2: "4 Million Smart Meters Don't Work"

Introduction

4 million smart meters are not working in smart mode (BBC).

 

Why this matters

For net zero to work it requires every household to invest in alternatives to fossil fuel consumption. Every year millions of people buy new cars and new heating systems for their home (or a new build home). All these people will make the trade off between investing in something that will use fossil fuels for the next several years or something that will consume electricity (increasingly from renewable sources). Roof top solar is a small contributor to the UK electricity production (solar is around 4%, a third from roof top solar); domestic roof top installs have potential to increase this significantly.

 

Which is the Best Use of Land: Solar Panels and Houses or Solar Panels?

With growing use of EVs and heat pumps electricity demand will rise; shifting demand to off peak periods is one of the paths to minimising new power grid infrastructure, as well as reducing the need to fire up gas power stations to meet peak demand.

 

How do Smart Meters Help?

A working smart meter allows automated 30min energy usage readings and billing; The user can then benefit from the range of flexible tariffs that pass savings on to the user for off peak electricity consumption:

• EV: Flexible tariffs might save £400 per year (based on 10,000 miles from home charging using Octopus Go night rate Vs standard day rate and 4miles/kWh (see Carwow for some example EV energy efficiencies)

 

Octopus Go tariff: low cost electricity at night to charge  an electric car

• Heat Pumps: Flexible Tariffs make the case for heat pumps - for example The Guardian shows the change from operating a heat pump with standard tariffs Vs for example, the Octopus Cosy Tariff is typically around £200 per year and can make the difference between a heat pump being less expensive to run than a gas boiler and more expensive on the standard tariff.  https://www.theguardian.com/business/article/2024/may/13/are-heat-pumps-more-expensive-to-run-than-gas-boilers).
• Solar with Battery: Flexible tariffs help the domestic battery (and solar) investment case, for example the Octopus Flux rate means that you can buy electricity at cheap rate and sell at peak rate with a difference of around 9p/kWh. So for a 10kWh battery this means around £300 per year.

 

On top of this you can run other high consumption appliances at the cheap rate and add to your savings, eg my dishwasher consumes around 2kWh per wash, a convenient and easy way to save £100/year by programming an overnight run.

 

These investments are filled with uncertainty (how long will an EV last and how much will it cost to insure and repair?) and all are big ticket items for most people. Of course, there is the emotional side to this too, about feeling good on saving the environment and your view of the latest tweet from Elon Musk Vs the very nice cars that his Tesla company builds. The point here is that a working smart meter pushes the calculation significantly in favour of an investment that contributes to net zero.

 

So How Many People do not Have a Smart Meter that can tap in to Flexible Tariffs?

The UK Government states that working smart meter uptake is now at around 55%. So 45% of users are missing out on the chance of the return on investment boost of flexible tariffs. A good chunk of these have the 4 million smart meters that do not work and await a solution from their energy provider. The rest need to be convinced to embark on the smart meter journey.

 

Each person in this set of 45% potential users will need to be prepared to wait in all day for the engineer to visit, perhaps more than once. Perhaps e-mail and telephone their provider, maybe many times...

 

We all know that 'Time is money'; many of the people with money to invest in 'green technology' are time poor and even at minimum wage (£11.44/hour) a saving of £300 is only worth investing 26hours...

...the smart meter install needs to be achieved with minimal effort for the user or the above savings are meaningless. Martin Lewis points out that the smart meter installation record risks brand damage. It is easy to see why: as word of mouth from the users of these 4 million meters spreads no amount of advertising and PR will make an impact...

...the only way to fix the reputation is to fix the 4 million smart meters.

 

The Smart Meter Dream:

If all smart meters worked then the switch from fossil fuels will accelerate as the investment case for millions of people to buy EV's and heat pumps will be that much more attractive. Home owners will see a bigger return on a solar panel and battery investment adding to renewable energy supply and domestic batteries that shift demand from peak periods...

...but we get more:

• Flexible tariffs work by encouraging users to shift demand away from peak periods reducing the need for suppliers to turn to gas generation and flattening demand on the supply grid
• Roof top solar panels do not require fields to switch from food production to solar farms.
• As growing electricity demand is shifted to off peak periods the need for new grid infrastructure and the associated environmental impact, is minimised.

 

My Experience of the Smart Meter Dream:

• Without exception, all the Octopus people whom I spoke with were polite and as helpful as they could be.
• The Octopus team do have some good processes in place - for example each of the people whom I spoke with had access to our history of calls and e-mails.
• The processes and information available to the Octopus team, relating to smart meters, was very patchy and sometimes contradictory:
• Am I on the "North" ("Arqiva") or "South" ("O2") network? (correct answer - "North", I think)
• How long is data held in the meter if the data connection is not working? I still don't have an answer for this that I trust.
• Is there sufficient signal strength in my area?
• Do my neighbours have working smart meters? (knocking on doors was the engineer's solution here)
• Which direction is the radio mast? What might be obstructing the radio signal path?
• What are the options if my meter does not work (external aerials, signal boosters and having DCC increase the signal strength were proposed by various enthusiastic members of the Octopus team, but as far as I am aware none of these options exist in the "North" network).
• While weak signal strength was pointed to as the cause of the issue by all the engineers that visited, none of the engineers had an instrument to measure signal strength nor a specification as to what is required.
• There was no escalation path for difficult installations - after 5 engineer visits the proposal by the Octopus Customer service representative for a 6th visit with no new plan seemed like groundhog day.
• "North" of England (ie Arquiva provided) communications network does not provide good coverage of my home using the equipment and processes available to an Octopus smart meter installation engineer.
• The  network provider, Data Communications Company (DCC), Arquiva, are often pointed to by the Octopus team as the source of problems (eg weak signal), but, while I heard Octopus people say that the issue will be raised with DCC I have never heard anyone at Octopus say that DCC has responded.
• My take, based on this experience: Octopus could operate more efficiently (eg 1 or 2 engineer visits in place of 5), but in the end they do not have the tools to do the job, not the backup from the network providers that they really need.

 

So How Do You Fix Smart Meters?

 

The subject of another post (spoiler alert, I don't have all the answers, yet!...)

 

I will also provide an update on my smart meter journey

Smart Meter #1: My Journey

Introduction

A decent Return on Investment (ROI) for a solar system relies on variable tariffs and variable tariffs rely on a working smart meter to report back half hourly meter readings...

...it turns out that this is not straight forward (at least for my installation)

 

A 'brief' history:

• My smart meter journey started in 2021 when, while I was aware of smart meters, the only benefit that I knew of was that meter readings could be automated.
• At the request of Octopus (ie several phone calls from Octopus it just seemed easier to give in and say “yes”) the smart meter was installed on 30th June 2021. This was year 2 of the pandemic and I  was working from home, so it was not too much effort wait in for the meters to be installed. The meters (gas and electric) both gave readings that seemed in line with the old meters, but
• The Home Area Network (HAN, ie connection to the gas meter and in house display) and Wide Area Network (WAN, ie connection back to base) wireless connections did not function.
• The engineer did not have any equipment to measure the WAN signal strength but told me that it was weak - it is not clear how he came to this conclusion.
• He said that the providers of the communications network would be notified and in a few weeks and, when they had done the work to increase signal strength, the meter should spring into life.
• At the time I was being billed on an Economy 7 tariff but the electric meter was set up to provide just one electric meter reading for 24hours a day (as opposed to night-time and day-time use)

 

EDMI ES 10B Electric Smart Meter with EDMI CS010A-08-92 Zigbee Comms Hub mounted on top	EDMI GS-60B Gas meter - this should communicate to the comms Hub via the Zigbee HAN.

 
 

• Subsequent engineer visits:
• I just got on with life, every so often submitting meter readings on the Octopus web site, until I realised that I was being over charged because Octopus were billing for Economy 7 and charging the single electricity meter reading (ie all my electricity consumption) at the (higher) daytime rate. Octopus did resolve this after several months and they were always friendly and helpful (eg I was never asked to pay the £2,500 over charge, it just sat as a debt on my account). However, as part of the resolution Octopus suggest that my smart meter was fixed....
• 19 May 2023: a new engineer visits - he did not have any equipment to measure the signal strength but told me that it was weak - it is not clear how he came to this conclusion.
• After this visit I was moved from the Economy 7 tariff to a single rate.
• Soon after the visit the WAN connection to my electricity meter sprung into life. I was able to see half-hourly readings via the Octopus app and website. The gas meter and In Home Display (IHD) wireless connections did not function and I was told that this was the HAN not functioning.
• The electric smart meter continued to communicate via the WAN over the next few months; this turned out to be a requirement for a solar system installation (or at least variable rate tariffs), so good news. Neither Octopus nor me put any further effort into getting the gas smart meter and IHD working.
• In March 2024 my solar power system was up and running and I applied for an export/flux tariff - this was up and running by the 23rd March (see post on application process).
• In subsequent weeks I was able to see import and export data in the Octopus app and this matched the values in my solar power system app (GivEnergy), but on odd days data was missing on the Octopus app, eg:
• Export: Wed 3rd April
• Import: Tues 2nd April
• Import: Sat 6th April
• My observations:
• It always seemed to be a whole day that was missing (ie either no data for that day or all 48 half hourly readings were showing in the app, or sometimes just a couple of readings for the day would show, perhaps a British Summer Time display and underlying data is held as Greenwich Mean Time?)
• Import and export data missing events were independent (ie most days had data for both, a few just import and a few just export)
• Roughly 3 or 4 days per month had missing data
• If the data did not show up in a few days then it would not show up without me contacting your customer service team - they had the capability to request the data to be uploaded. They could also do a quick search for missing data (as opposed to my manual scanning from the bar charts in the app). Typically this request resulted in the missing data showing up within a few days.
• 14th May 2024: I spoke with an Octopus customer service representative who suggested an engineer visit to resolve the missing data issue (and, as a 2nd priority get the gas meter and IHD connected).
• 24th June 2024: another engineer called - his information for the job was:
• Recommission the gas meter
• Replace the electric meter
• Nothing about the issue symptoms or diagnosis just the prescription.
• We discussed the symptoms and his diagnosis was weak signal (most likely) or faulty communications hub/hardware (least likely). His suggestion was that he replace the whole system (communications hub, gas meter, electric meter). Given the low rates of hardware failure that he sees then we agreed that if the system still did not work then faulty hardware was very unlikely to be the cause. He did not have time available that day to replace the whole system, so we rebooked a new slot...
• 27th June 2024: the engineer returns to replace the electric meter, gas meter and communications hub. This involved lots of safety checks etc, and the result was:
• He could not commission the WAN (ie comms hub to Data Communications Company (DCC)/Octopus link) and so could not commission the gas meter to comms hub link, nor the IHD.
• The engineer’s view at the time was that the root cause was a weak signal (but again, he had not instrument to measure this).
  
• The engineer phoned me later that day to say that the internal Octopus Engineers' messaging system chat was that DCC were having some problems on that day and he thought that another go at commissioning might be worthwhile.
• 18th July: The engineer is back again to have another go at commissioning the WAN. This did not work. He replaced the hub with a dual band hub (apparently it is the HAN/Zigbee network that is dual band/extended range, the WAN is the same as previous hubs). This gave the same fault. The engineer discussed the operation of the green flashing WAN LED:
• 1 flash every 5 secs indicates that it is working

My system did this:

• 1 flash every 5 seconds repeating for around 30secs to 1minute then
• 1 or 2 quick flashes then
• 1 flash every 5 seconds repeating for around 30secs to 1minute etc

I could not find an explanation for this WAN LED behaviour online.

• 20th Aug: I spoke with an octopus customer services representative who attempted to remotely commission the communications hub with me pressing buttons on the electricity meter:
• After I pressed a few buttons on the meter, under his guidance, we got stuck with my meter ‘searching…’ when another response was expected
• It turned out that this was his 1st ever commissioning job; however, he was in communication with a colleague to confirm that he was doing this correctly and they tried doing something on their end to get the commissioning process to work but to no avail.
• He then suggested that another engineer visit me, but could not say what this engineer would do that was different to the previous 5 visits and we also discussed the only other alternative, that I would have to return to a fixed rate...

…we left it that he was going to schedule another engineer to visit, but with no ideas on what would be different this time.

 

A few days later I wrote the following e-mail to Octopus…

…I will let you know the outcome in a follow up post.

 

My e-mail to Octopus. I am intrigued to know if my proposed recompense for a non-working meter makes any difference to Octopus; the Ofgem website hints that they might be sympathetic to my case if I raised it with them: https://www.energyombudsman.org/case-studies/smart-meters-not-sending-readings

 

 

Solar #5: Tariff Choice and Sign Up

Introduction

The wait for solar panels to be installed gave me time to choose an electricity supplier and a tariff.

 

My Existing Supplier is Octopus - Should I Change?

I want a supplier that I can trust long term and that imposes minimum effort:

• I am not interested in switching supplier every few months so to get the best offer, so I ignore special introductory offers and focus on the regular rates available to all users
• I do want low rates, but the rate does not need to be the lowest, just close to lowest
• Good, friendly service for any issues is important; issues will arise and I wanted to be looked after.
• I had been with the Coop who, a few years ago, gave up doing energy and passed on all their customers to Octopus. This 'relationship' had been fine over a number of years; rates reasonable, minimal effort on my part and friendly communications.

Good News: When I did a little Googling of electricity suppliers it turned out that Octopus is amongst the leaders in special rates for Solar. Reading a bit more it looks like Octopus has green energy as a strategic direction; while this could just be green washing it does seem that they will keep innovating (eg new solar tariffs) and that should be good for customers with solar/battery systems.

 

And Which Tariff?

Octopus Flux is one of the options that the Octopus website recommended for solar with battery and looks like it might fit what I am after:

Octopus Flux is the Tariff that I signed up for and the one that I used for my return on investment calculations

 

• Good Return on Investment (ROI) for battery storage (eg buy cheap rate electricity at night and sell back at peak in the evening) - roughly 10p per kWh difference between buying at cheap rate and selling at peak rate (when I did the ROI calculation in summer/autumn 2023).
• Set up and forget - no need to tweak scheduling each day (eg Vs options with ever changing rates that track the wholesale energy market)
• We are free to use power at home for our convenience, when we want without significant penalty (eg we can decide when to cook based on when we want to eat, not on the changing electricity price).

 

Another Octopus tariff option is Intelligent Flux where Octopus take control of the battery charge and discharge. This sounds quite clever, but I worry about the complexity and also the website did not give any return on investment figures. My decision was to keep it simple and go with Octopus Flux. With this up and running and acting as a baseline over, say, 12 months I could then consider other tariff options.

 

How Was the Sign Up Process?

It is important to note that we had a working smart meter, at this point, sending electricity consumption meter readings back to Octopus every 30mins. The sign up timeline was:

• Solar panels commissioning completed and left on battery charge/discharge cycle in Friday 8th March
• In full operation on Sat 9th March
• Installer e-mailed me the MCS certificate/number and Distribution Network Operator (DNO) notification letter on Monday 11th March
• Tried to sign up for Octopus flux on Mon 11th March - website crashed - 3 attempts
• Tues 12th: 1 more attempt and called Octopus - sounds like a website issue and my request has actually been submitted.
• 23rd March: export and Octopus flux now set up and working:

 

Getting Paid by Octopus for my Solar Generation, Much of it Exported at Peak Rate via the Battery

 

How Do I Schedule the GivEnergy Inverter for Octopus Flux?

For this setup I used the GivEnergy Cloud interface for the inverter (https://givenergy.cloud/login). There is an app for both iOS and Android, but (at the time of writing) these do not give access to all the scheduling features that are available on the cloud interface.

 

My Setup:

• Fill the battery at cheap rate 2am to 5am
• In the winter this needs to be a 100% battery fill
• In the summer this can be a lower %age fill as the solar power will usually fill the rest.
• The cost of cheap rate import is slightly less than day rate export, so, in theory, you actually gain just over 1p/kWh by charging from the grid at cheap rate as opposed to letting the solar complete the battery charge. However, I think, that this comes with additional losses and heating/wear of the inverter/battery (just my unproven theory).
• My take is that it makes little difference to the final ROI, but I would be interested to hear from anyone who knows more about losses and inverter/battery wear.
• The GivEnergy scheduled charge does the above.
• After 5am (day rate) home use power source in priority order of:
• Directly from solar
• If not enough solar, from battery as well
• If not enough battery, from grid as well
• An use excess solar above home use in priority order of:
• Top up battery
• If battery full or charging at max power then export to grid
• Good news: the GivEnergy ECO mode does all of the above.

• Peak rate, 4pm to 7pm. This is the tricky bit, objectives are:
1. Do not import at peak rate unless there is no other possibility
2. Export everything that you can without breaking rule #1
3. Freedom to use power when we need it, eg what if we want to have the electric oven on between 6pm and 7pm? I don't want to delay tea so that we can cook at a lower electricity unit cost.
• The last and first points make it tricky - you cannot just dump all the stored battery energy to the grid at 4pm. It is important to have a staged discharge and keep some back right up until 7pm.
• After 7pm it makes little difference if you have managed to export the last kWh from the battery at 22.885p / kWh and then import at the day rate (24.66p / kWh) or if you keep back the odd kWh to supply the house in the evening.
• You do want a (close to) empty battery at 2am to start filling it at cheap rate, so you have a margin of error of 7hours minimum domestic use so that, even if you are on holiday, any residual battery charge that is left at 7pm is used up by 2am. For us we can run as low as 200W when we are on holiday, so a margin of error of 7hours x 200Watts = 1.4kWh.
• Givenergy provides 10 discharge periods, so by dividing up the 180minute peak rate period into 10 x 18mins (or 9 x 20mins for simplicity) mean that you can discharge a step at a time and leave, say 1kWh at 7pm. However, there is one problem, a peculiar behaviour on charge and discharge that means that this (currently) does not work in all situations...

...more on this in another post. In the end I used just 3 discharge periods (as can be seen from the graph below) - again more on this in the future post.

 

• So what is the ROI? - see 3 examples of recent Flux tariffs below (I will use the July '24 tariff in the following text):
• "Summer" ROI: here my assumption is that there has been enough daylight up until 4pm (peak rate start) that the battery is 100% full. On bright days, during the day, after the battery is full, we export at the day rate (at 14.85p/kWh). After 4pm get to sell all the cheap rate electricity that we bought in the night before (2am to 5am) at peak rate - a difference of 9.3p/kWh. We also programme our dishwasher to run between 2am and 5am (roughly 2kWh at about the same rate saving). Roughly speaking 10kWh (9.5kWh battery + 2kWh dishwasher - losses - baseline battery 'reserve' charge (set to 4%, but seems to vary between 3% and 6%)) at 9.3p/kWh rate difference is approximately a 10kWh or £9.30 return on investment each day.

System Power Graph showing Typical Summer Day. Charge up to 50% at cheap rate, using solar power to get to 100% change and then discharging to the grid in 3 steps during the peak rate.

Solar Generation Energy Graph for typical Summer Day.

• "Winter" ROI: The battery capacity (9.5kWh) is around the same as our daily consumption (~10kWh). In winter, my assumption is that solar production is minimal and we only have enough battery charge from cheap rate to run the house. Discharge at the peak rate can only be minimal if any. When we are at home we use an average of around 11kWh (including inverter & battery loses - see upcoming post), but the typical range is 8kWh to 16kWh. Solar power production can be as low as 3kWh on gloomy days (based on my experience so far from March to September - I assume we will go even lower in the next few months). So in these situations the return on investment is the difference between cheap rate import and the day rate import that we are avoiding, ie 9.03p/kWh difference giving around a £9.03/day return on investment.
• Note: this is the return for using the Flux rate over a regular fixed import and export tariff (eg at the day rates below). While this depends on having a battery, the battery has an additional return not covered here, when the solar system is producing more than is being consumed at home, the battery can charge. When there are peaks in demand (eg kettle on), or dips in solar production (eg cloud cover) then the battery is used as opposed to importing ie comparing a solar system with and without battery the battery removes both import and export at day rate. This return on investment for the battery is true of all import/export tariffs and is not calculated here.

 

Octopus Flux Tariffs Have Changed in the Last Few Months:

From 1st July 2024

	Import	Export
Day rate	22.586p / kWh	14.85p / kWh
Flux rate (02:00 - 05:00)	13.552p / kWh	4.816p / kWh
Peak rate (16:00 - 19:00)	31.621p / kWh	22.885p / kWh
Standing charge	49.119p / day

 

Battery profit (summer)= 22.885-13.552=9.333 p/kWh

Battery profit (winter) = 22.586-13.552=9.0340 p/kWh

These figures are used for the above post.

 

May 2024

	Import	Export
Day rate	24.66p / kWh	14.85p / kWh
Flux rate (02:00 - 05:00)	14.80p / kWh	5.26p / kWh
Peak rate (16:00 - 19:00)	34.53p / kWh	24.99p / kWh
Standing charge	49.12p / day

 

Battery profit (summer)= 24.99-14.80=10.19 p/kWh

Battery profit (winter) = 24.66-14.80=9.86 p/kWh

 

To April 2024

	Import	Export
Day rate	28.44p / kWh	17.44p / kWh
Flux rate (02:00 - 05:00)	17.06p / kWh	6.06p / kWh
Peak rate (16:00 - 19:00)	39.82p / kWh	28.82p / kWh
Standing charge	49.77p / day

 

Battery profit (summer)= 28.82-17.06=11.76 p/kWh

Battery profit (winter) = 28.44-17.06=11.38 p/kWh

 

Over the last year the battery ROI has been going down from over 11p per kWh to only just over 9p per kWh...

...this was identified in the risks for this investment (See Solar Panels Investment Case)