Earth Notes: Saving Electricity (2025)
Updated 2026-01-04 19:59 GMT.By Damon Hart-Davis.
Meter readings at home #defossilisation #fossilFree
Meter readings and analysis of our electricity carbon footprint at home in 2025. The gas meter has gone: we are all-electric with a heat pump now!
See the daily PV generation logs for 2025 as See the heat-pump activity fade out (heating off on the ) and diversion take over for DHW by the end of the month.
With the additional grid battery
and the energy stored in the DHW cylinder and Thermino heat battery, self-consumption was good and grid dependence low (~4%).
Weekday 15-minute Some efficiency and load-shaping was done before July.
Effort was put into reducing standing losses from the DHW tank, thus leaving more energy to spill to the grid and trim carbon emissions.
Also, some Thermino top-up (~2kWh) is now postponed until solar noon, to help counter the huge flood of PV to the grid then. Diversion to the Thermino is now prevented between and , and when grid carbon intensity is high, and when local grid voltage is low, along with when grid frequency is low as before.
Full-week 15-minute March and September shoulder months are when the
Enphase AC-coupled batteries work the hardest, when PV generation is patchy but is similar to consumption overall.
.csv and .png chart.
: Trendline
hph4 heat pump electricity consumption (kWh) vs heating degree days (HDD12) with a linear fit. (Source CSV data, and as ODS spreadsheet.)
hph4 heat pump electricity consumption (kWh) vs heating degree days (HDD12) with a degree-2 polynomial fit. Electricity demand is expected to rise faster than linearly with HDD because higher HDD implies higher heat demand and lower CoP from the higher 'lift' required. (Source CSV data, and as ODS spreadsheet.): Heat-pump Fadeout
: Self-consumption
con consumption and net imports for generated with sh script/storesim/load_profile.sh 202504.
: Less Loss
con consumption and net imports for generated with sh script/storesim/load_profile.sh 202507.
: Batteries and Shoulder Months
sh script/storesim/load_profile.sh -batt 202509. Times UTC. Data and other views are available.
: Heating Half On
4o outside. Second: radiator temperatures seen at the radiators and at Heathrow over a month. Space heating was officially on from as we had a guest, but had been automatically intermittently on low for pipe-freeze protection from keeping inside temperatures typically at a little over 16°C.
| 2025 | Electricity | |||
|---|---|---|---|---|
| Month | Consumed kWh (/d) | Generated -kWh (/d) | Period | Notes |
| 2025-01 | 425 (13.7/d) | -52 (-1.7/d) | 20241231T18:02Z to 20250131T18:00Z (31d) | Meter readings: electricity generation 51.7kWh, export 0.5kWh, import 374.0kWh (thus net consumption (I+G-E) 425.2kWh, self consumption ((N-I)/N) 12%). See load profiles.
The Enphase emailed monthly report for January (triggered ) says: produced (generated) 58.5kWh (max 4.8kWh on the 30th), consumed 425.9kWh (max 18.0kWh on the 4th), grid energy imported 378.7kWh and exported 0.1kWh, grid dependence 90% (min 68% on the 30th). The Ecotricity energy bill for this month was lower than last January's even though this month was colder. The new bill was ~£130 for just electricity (about 100kWh more than last January) now that gas has gone. The previous January's bill was ~£160 of which ~£50 was gas. In effect ~100kWh of electricity more than substituted for ~500kWh of gas: an effective CoP of ~5, maybe, though ~100kWh of Thermino CoP 1 storage was being avoided too this month. Allowing for that suggests the equivalent of a ~200kWh rise in electricity against a ~600kWh fall in gas, ie an overall CoP of ~3. (Note that the months' billing days were slightly different, and the electricity unit rate fell but the standing rate rose.) For the month reported hph4 was 204kWh (heat 207kWh) at ~89% coverage, thus likely ~230kWh. Daikin stats report 247kWh for hph4 (possibly not including OpenTRV unit consumption). Linear regression of hph4 against HDD12 yields n=28, slope=0.96kWh/HDD12, intercept=0.07kWh, R^2=0.87. Due to an Eddi firmware fault between ~23rd and 28th, the underlying reported consumption data is partly wrong in that interval or entirely absent. Avoiding partial dubious data for those days yields n=25, slope=0.87kWh/HDD12, intercept=1.07kWh, R^2=0.88.
HDD12=235 vs 199 last January. |
| 2025-02 | 353 (12.6/d) | -110 (-3.9/d) | 20250131T18:00Z to 20250228T18:20Z (28d) | Meter readings: electricity generation 110.3kWh, export 6.0kWh, import 248.2kWh (thus net consumption (I+G-E) 352.5kWh, self consumption ((N-I)/N) 30%). See load profiles.
The Enphase emailed monthly report for February (triggered ) says: produced (generated) 111.2kWh (max 10.1kWh on the 28th), consumed 333.7kWh (max 22.2kWh on the 16th), grid energy imported 244.8kWh and exported 4.8kWh, grid dependence 73% (min 32% on the 18th). The Ecotricity energy bill for this month was ~£20 lower than last February's even though this month was colder. For the month hph4 was 182kWh (heat 190kWh). Daikin stats report 182kWh for hph4. Linear regression of hph4 against HDD12 yields n=28, slope=0.58kWh/HDD12, intercept=2.76kWh, R^2=0.64.
HDD12=180 vs 91 last February. |
| 2025-03 | 356 (11.5/d) | -351 (-11.3/d) | 20250228T18:20Z to 20250331T17:08Z (31d) | Meter readings: electricity generation 351.4kWh, export 60.3kWh, import 64.7kWh (thus net consumption (I+G-E) 355.8kWh, self consumption ((N-I)/N) 82%). See load profiles.
The Enphase emailed monthly report for March (triggered ) says: produced (generated) 353.0kWh (max 18.7kWh on the 30th), consumed 330.3kWh (max 15.2kWh on the 30th), grid energy imported 62.6kWh and exported 57.5kWh, grid dependence 19% (min 2% on the 29th). The heat battery (indeed all storage including off-grid) filled up for the first time this year on the 31st. Sunny month, but colder than last March. The Ecotricity energy bill for this month was ~£50 lower (more than halved) compared to last March, even though this month was colder. For the month hph4 was 98kWh (heat 179kWh). Daikin stats report 95kWh for hph4. Linear regression of hph4 against HDD12 for the first 18 days of the month until we turned off space heating (though cold nights means that pipe-freeze protection runs compressor and pump) yields n=18, slope=0.50kWh/HDD12, intercept=1.61kWh, R^2=0.69.
HDD12=124 vs 100 last March. |
| 2025-04 | 360 (12.0/d) | -533 (-17.8/d) | 20250331T17:08Z to 20250430T17:23Z (30d) | Meter readings: electricity generation 532.8kWh, export 188.6kWh, import 15.3kWh (thus net consumption (I+G-E) 359.5kWh, self consumption ((N-I)/N) 96%). See load profiles.
The Enphase emailed monthly report for April (triggered ) says: produced (generated) 534.0kWh (max 25.3kWh on the 29th), consumed 340.5kWh (max 17.5kWh on the 25th), grid energy imported 12.8kWh and exported 185.0kWh, grid dependence 4% (min 2% on the 29th). This month we passed 42MWh exports and 47MWh generation on the first array (just under 62MWh generation total). A fifth Enphase battery was installed mid-month. The Ecotricity energy bill for this month was ~£20 lower (about halved) compared to last April. This month was marginally warmer. For the month hph4 was 17kWh (heat 187kWh). Daikin stats report 9kWh for hph4.
HDD12=50 vs 67 last April. |
| 2025-05 | 319 (10.2/d) | -600 (-19.3/d) | 20250430T17:23Z to 20250531T17:13Z (31d) | Meter readings: electricity generation 599.6kWh, export 293.4kWh, import 12.5kWh (thus net consumption (I+G-E) 318.7kWh, self consumption ((N-I)/N) 96%). See load profiles.
The Enphase emailed monthly report for May (triggered ) says: produced (generated) 602.6kWh (max 27.9kWh on the 16th), consumed 298.8kWh (max 16.9kWh on the 8th), grid energy imported 9.9kWh and exported 290.7kWh, grid dependence 3% (min 2% on the 16th). Generation this month was quite high (not beating though), maybe partly driven by Solar records in UK and Germany as cold snap hits Eastern Europe. The Ecotricity energy bill for this month was ~£20 lower (about halved) compared to last May. Very similar to April, being mostly standing charge. For the month hph4 was 25kWh (heat 159kWh); this reflects the change to reduce standing losses and overall consumption, by ~1kWh/d in months where diversion is available.
HDD12=20 vs 9 last May. |
| 2025-06 | 281 (9.4/d) | -634 (-21.1/d) | 20250531T17:13Z to 20250630T17:08Z (30d) | Meter readings: electricity generation 633.6kWh, export 364.9kWh, import 12.2kWh (thus net consumption (I+G-E) 208.9kWh, self consumption ((N-I)/N) 95%). See load profiles.
The Enphase emailed monthly report for June (triggered ) says: produced (generated) 637.2kWh (max 28.1kWh on the 18th), consumed 263.4kWh (max 15.2kWh on the 8th), grid energy imported 9.7kWh and exported 362.9kWh, grid dependence 4% (min 2% on the 13th). Generation from the first PV set passed 48MWh and from the second set 15MWh, so 63MWh total! The efforts to reduce standing DHW losses seem to have contributed to ~1.5kWh/d reduction in total consumption compared to the previous two months (10.2kWh/d in May, 12.0kWh/d in April) as hoped for. The Ecotricity energy bill for this month was more than halved compared to last June. Very similar to May and April, just over £20. For the month hph4 was 24kWh (heat 120kWh).
HDD12=1 vs 8 last June. |
| 2025-07 | 246 (7.9/d) | -558 (-18.0/d) | 20250630T17:08Z to 20250731T17:28Z (31d) | Meter readings: electricity generation 558.0kWh, export 321.9kWh, import 10.0kWh (thus net consumption (I+G-E) 246.9kWh, self consumption ((N-I)/N) 96%). See load profiles.
The Enphase emailed monthly report for July (triggered ) says: produced (generated) 559.6kWh (max 29.0kWh on the 8th), consumed 226.7kWh (max 11.4kWh on the 3rd), grid energy imported 8.1kWh and exported 319.0kWh, grid dependence 4% (min 2% on the 11th). The Ecotricity energy bill for this month was well under £20 for 9kWh imported and lots of standing charge. The house was almost never fully occupied in July. For the month hph4 was 22kWh (heat 98kWh).
HDD12=0 vs 0 last July. |
| 2025-08 | 202 (6.5/d) | -492 (-15.9/d) | 20250731T17:28Z to 20250831T17:20Z (31d) | Meter readings: electricity generation 491.1kWh, export 298.8kWh, import 9.2kWh (thus net consumption (I+G-E) 202.5kWh, self consumption ((N-I)/N) 95%). See load profiles.
The Enphase emailed monthly report for August (triggered ) says: produced (generated) 495.0kWh (max 23.3kWh on the 1st), consumed 185.5kWh (max 12.3kWh on the 1st), grid energy imported 8.7kWh and exported 298.1kWh, grid dependence 5% (min 2% on the 2nd). Holidays away for most or all of the family this month reduced consumption, eg: At 16WW over 16 days from to : electricity import 0.1kWh/d (net -12.7kWh/d).That 0.1kWh/d represents 'leakage' of ~10W in from the grid at night with minimal demand in spite of the battery. The Ecotricity energy bill for this month was again well under £20 for 10kWh imported. For the month hph4 was 17kWh (heat 94kWh).
HDD12=0 vs 0 last August. |
| 2025-09 | 279 (9.3/d) | -341 (-11.4/d) | 20250831T17:20Z to 20250930T16:42Z (30d) | Meter readings: electricity generation 341.0kWh, export 84.4kWh, import 22.2kWh (thus net consumption (I+G-E) 278.8kWh, self consumption ((N-I)/N) 92%). See load profiles.
The Enphase emailed monthly report for September (triggered ) says: produced (generated) 341.9kWh (max 17.5kWh on the 5th), consumed 255.0kWh (max 13.0kWh on the 25th), grid energy imported 19.3kWh and exported 80.9kWh, grid dependence 8% (min 3% on the 12th). Generation from the first PV array/meter passed 49MWh. It seems as if the extra Enphase battery (plus this relatively sunny month) has extended the run of 'summer' months importing <1kWh/d from 5 to 6. The Ecotricity energy bill for this month was just over £20 for 23kWh imported, mainly standing charge. Ecotricity's annual (import) estimate is ~1.4MWh/£600 including VAT. For the month hph4 was 35kWh (heat 104kWh).
HDD12=11 vs 16 last September. |
| 2025-10 | 219 (7.1/d) | -149 (-4.8/d) | 20250930T16:42Z to 20251031T16:42Z (31d) | Meter readings: electricity generation 149.2kWh, export 10.5kWh, import 80.0kWh (thus net consumption (I+G-E) 218.7kWh, self consumption ((N-I)/N) 63%). See load profiles.
The Enphase emailed monthly report for October (triggered ) says: produced (generated) 150.5kWh (max 10.5kWh on the 8th), consumed 200.5kWh (max 9.5kWh on the 30th), grid energy imported 80.6kWh and exported 8.8kWh, grid dependence 40% (min 5% on the 5th). Total grid flow ( imp+exp) was the lowest for any month since at 90kWh. So careful use of expanded storage, heat and electrical, is working.
The Ecotricity energy bill for this month was just under £40 for 82kWh imported, cf over £70 for last October. For the month hph4 was 49kWh (heat 58kWh). Space heating was still off for this month.
HDD12=23 vs 27 last October. |
| 2025-11 | 246 (8.2/d) | -76 (-2.5/d) | 20251031T16:42Z to 20251130T15:14Z (30d) | Meter readings: electricity generation 76.2kWh, export 1.1kWh, import 170.8kWh (thus net consumption (I+G-E) 245.9kWh, self consumption ((N-I)/N) 31%). See load profiles.
The Enphase emailed monthly report for November (triggered ) says: produced (generated) 76.4kWh (max 4.9kWh on the 2nd), consumed 229.6kWh (max 13.7kWh on the 21st), grid energy imported 172.6kWh and exported 0.2kWh, grid dependence 75% (min 17% on the 3rd). The Ecotricity energy bill for this month was just under £70 for 176kWh imported, cf ~£80 for electricity for last November. Ecotricity's annual (import) estimate has fallen a little to ~1.3MWh and under £600 including VAT. For the month hph4 was 94kWh (heat 100kWh). DHW was being provided by the heat pump throughout the month, with the main daily slot's temperature raised to 48°C (comfort) from 45°C (eco) . Space heating was officially on from as we had a guest, but had been automatically intermittently on low for pipe-freeze protection from .
For the year ending (last reading at) from (first reading) , Pilio reports heating performance for electricity imports vs HDD12: slope 1.36kWh/HDD, baseload 2kWh (per week), R2 0.76. HDD12=95 vs 113 last November. |
| 2025-12 | 329 (10.6/d) | -45 (-1.4/d) | 20251130T15:14Z to 20251231T17:53Z (31d) | Meter readings: electricity generation 44.7kWh, export 0.5kWh, import 284.2kWh (thus net consumption (I+G-E) 328.4kWh, self consumption ((N-I)/N) 13%). See load profiles.
The Enphase emailed monthly report for December (triggered ) says: produced (generated) 45.2kWh (max 2.7kWh on the 24th), consumed 315.6kWh (max 13.8kWh on the 31st), grid energy imported 286.0kWh and exported 0.0kWh, grid dependence 91% (min 79% on the 10th). The Ecotricity energy bill for this month was just over £100 for 288kWh imported (~15% is standing charge at 45.32p/d + 5% VAT), cf ~£120 for energy for last December. Probably the flow/return fault fix, and subsequent tuning, helped, given that HDD barely changed. For the month hph4 was 146kWh (heat 148kWh). For the year ending (last reading at) from (first reading) , Pilio reports heating performance for electricity imports vs HDD12: slope 1.26kWh/HDD, baseload 3kWh (per week), R2 0.72.
HDD12=132 vs 133 last December. |
| 2025 total (Jan–Dec) | 3619 (9.92/d) | -3947 (-10.81/d) | 20241231T18:02Z to 20251231T17:53Z (365d) |
HDD of 871 (vs 764 for 2024), below the – mean of 984.
PV generation of 3947kWh was good, close to maximum. Self-consumption calculated as (N-Imp)/N was 64%, up from 57%. Total grid flows (import+export) about the same at 2934kWh (previously 2909kWh). |
| kgCO2 | 1556kg (4.3/d) | -1683kg (-4.6/d) | ~0.127tCO2 [though see below] (~0.35kgCO2/d) gas/electricity emissions total for 3.x people for most of the year, based on fixed (outdated but used for consistency with previous calculations) 0.43kgCO2/kWh for electricity (netted), ie slightly carbon positive. Half of the family did fly short-haul this year. We still do not own a car. With data from the Enphase metering and the FUELINST log I compute a potentially more accurate CO2 electricity balance, noting the relatively low carbon grid intensity when exporting summer PV and high intensity when importing in winter. (There is still the argument that any balancing at the moment may be done by a gas turbine in practice.) See below for the computations — the estimated figure is now a very slightly positive 0.019tCO2 overall for electricity. | |
: ToD Carbon Calculation
% sh script/analytic/16WW-elect-CO2-Enphase-FUELINST.sh 2025 INFO: created/updated out/yearly/Enphase-1h-2025.csv.gz INFO: script/analytic/FUELINST-bucket-1h.sh: no FUELINST full-year data archive found at data/FUELINST/log/2025.log.xz INFO: script/analytic/FUELINST-bucket-1h.sh: collated FUELINST part-year data at out/yearly/FUELINST-1h-2025_part.log.xz INFO: created/updated out/yearly/FUELINST-1h-2025.csv.gz INFO: input data collected. INFO: created/updated out/yearly/Enphase-join-FUELINST-1h-2025.csv.gz INFO: unique hours in joined data: 8703 (363 days) INFO: scale factor to allow for missing data: 1.00655 INFO: raw to-date 16WW electricity CO2 kg for 2025: 19 INFO: adjusted 16WW electricity CO2 kg for 2025: 19 INFO: created out/yearly/16WW-elect-kgCO2-2025.txt
So small residual carbon emissions of ~19kgCo2 for the year.
The were net electricity exports of 328kWh this year, out of (high-ish) PV generation of 3940kWh (cf mean ~3815kWh/y).
If electricity grid had been a fixed intensity at ~CCGT/OCGT levels, that would imply ~-100kgCO2, ie negative, emissions. The difference comes from importing most in winter when intensity is higher.
Increasing exports without reducing services to the house (eg comfort levels) could send that effective emissions value negative in .