WEBVTT NOTE Transcript 2024-03-20 from FLAC by https://elvery.net/prototypes/transcribe/ on my MacBook Air M1 on maximum quality setting. NOTE Body lightly edited. 00:00.000 --> 00:03.360 Statscast for April 2020 00:03.360 --> 00:09.120 [Intro Music] 00:09.120 --> 00:18.840 Hi, I'm Damon Hart-Davis, and welcome to Earth Notes podcast on all things eco and green and efficient at home! 00:18.840 --> 00:21.360 3rd May 2020 00:21.360 --> 00:29.280 Turn-of-the-month update of home energy and related stats at 16WW (ie EOU Towers) for April: 00:29.280 --> 00:35.560 'leccy consumption, generation and import, gas for hot water, mains water temperature 00:35.560 --> 00:40.200 and other significant GB, Europe world numbers over the monthish. 00:40.200 --> 00:42.480 A spot of ordification too. 00:42.480 --> 00:46.760 Maybe not all the numbers have to be directly eco-related to hold interest. 00:46.760 --> 00:49.080 Here is this episode's mystery number. 00:49.080 --> 00:51.760 33821 00:51.760 --> 00:52.920 Don't sue me. 00:52.920 --> 01:00.480 If you have been especially thorough in absorbing the minutiae of the host EOU earth.org.uk website, 01:00.480 --> 01:06.680 especially the foot of the full fat desktop pages, you may already know what this ordinal is. 01:06.680 --> 01:11.680 Knowing might be evidence of your needing to get out more, but under lockdown that's forgivable. 01:11.680 --> 01:15.000 Beans will be spilt later in this episode. 01:15.000 --> 01:18.280 Here's a mystery noise to ponder too. 01:18.280 --> 01:22.080 [mystery noise] 01:22.080 --> 01:23.080 Data 01:23.080 --> 01:28.960 There are many interesting anecdote to pick from for April, home grown and beyond. 01:53.360 --> 01:57.920 Look at my saving electricity and gas 2020 page for most of the detailed energy numbers 01:57.920 --> 02:00.080 that I'll talk about next. 02:00.080 --> 02:06.440 Gross electricity consumption at 16WW, that is everything used by appliances and gadgets 02:06.440 --> 02:10.880 ignoring all generation and battery activity, is higher than I'd like. 02:10.880 --> 02:16.440 A typical value is 5-6kWh per day at this time of year, but for the last 4 months it's 02:16.440 --> 02:19.020 being around 7kWh per day. 02:19.020 --> 02:24.520 The first two months were likely largely bumped up by dehumidifier use and March and April 02:24.520 --> 02:29.560 were higher as a result of extra Xbox and cooking workouts from us all being at home 02:29.560 --> 02:32.520 all the time for some of March and all of April. 02:32.520 --> 02:37.320 However the lockdown has resulted in less air pollution and thus apparently more blue 02:37.320 --> 02:38.320 skies. 02:38.320 --> 02:43.240 Our mains water has been warm and April's PV generation a record on our grid tide system 02:43.240 --> 02:47.200 - 18kWh a day against a more typical 15. 02:47.200 --> 02:51.960 The Met Office says it has probably been the sunniest April on record and the solar power 02:51.960 --> 03:00.240 industry reported its highest ever production of electricity - 9.68gWh - in the UK at 1230 03:00.240 --> 03:02.400 on Monday 20th April. 03:02.400 --> 03:07.280 Being home to be able to choose the optimum time to run loads such as the dishwasher and 03:07.280 --> 03:13.200 be able to run my laptop off grid more easily has improved self consumption from 69% last 03:13.200 --> 03:18.880 April to 77% this with imports staying under 2kWh per day. 03:18.880 --> 03:22.760 On occasion imports are well under 1kWh per day. 03:22.760 --> 03:26.400 Both the grid tide and off grid systems are almost making it through the night without 03:26.400 --> 03:29.880 emptying, each has done it about once in the last month. 03:29.880 --> 03:36.920 Dinner time is still tricky and forces imports in the usual 4pm to 7pm weekday peak slot. 03:42.120 --> 03:47.240 case typically importing ahead of the old 6pm peak time. 03:47.240 --> 03:51.480 We can sneak the roughly 2kWh peak demands of the washing machine and dishwasher under 03:51.480 --> 03:58.400 the PV generation curve on a sunny day, avoiding imports, but the kettle at 3kWh and cooking 03:58.400 --> 04:03.600 in the morning and evening particularly if the 2kWh oven is used are beyond the end phase 04:03.600 --> 04:05.260 to cover. 04:05.260 --> 04:10.400 From lockdown right now but with rising clean energy supply and continuing likely lower 04:10.400 --> 04:14.720 demand there is an interesting though anticipated issue. 04:14.720 --> 04:18.920 National Grid warns that Britain could be at risk of blackouts on Friday from renewables 04:26.960 --> 04:28.960 energy that can't be used. 04:28.960 --> 04:31.000 This isn't a bug, it's a feature. 04:31.000 --> 04:36.159 I hope that we will overbuild renewables massively so that we can usually turn some off and then 04:38.320 --> 04:43.320 This minimises the need for storage, likely to stay relatively expensive or even fossil 04:43.320 --> 04:44.960 fuel backup. 04:44.960 --> 04:49.200 Pollution has dropped significantly because of reduced travel, quite likely cars being 04:49.200 --> 04:51.240 a large, large part. 04:51.240 --> 04:55.600 I have relatives in Rome and Paris and the city mapper mobility index, percentage of 04:55.600 --> 05:02.840 city moving compared to usual for 1st May, is showing Rome at 3%, Paris at 4% and London 05:02.840 --> 05:04.520 at 9%. 05:04.520 --> 05:10.240 At the other end of the scale are Seoul at 45% and Hong Kong at 51%. 05:10.240 --> 05:15.840 An email I received from TFL, Transport for London, on the 30th as a regular user of its 05:15.840 --> 05:21.400 services in normal times said in part "The number of people using tube and rail services 05:21.400 --> 05:27.240 has fallen dramatically by around 95% and the number of people using buses by around 05:27.240 --> 05:32.520 85% - this is important to stop the spread of coronavirus." 05:32.520 --> 05:38.040 In a piece also on the 30th by Heidi Alexander, the London Mayor's Deputy for Transport and 05:38.040 --> 05:43.240 Deputy Chair of Transport for London, she said "On Easter Sunday we recorded the lowest 05:43.240 --> 05:46.960 level of tube journeys in a single day since the 19th century. 05:46.960 --> 05:50.040 How will social distancing be observed on the tube? 05:50.040 --> 05:54.720 Each carriage could carry just a sixth of normal rush hour loading, normal planning 05:54.720 --> 05:59.560 assumptions for the tube that four people occupy each square metre." 05:59.560 --> 06:04.800 We also got an interesting glimpse into the massively inflated demand for some, not all, 06:04.800 --> 06:07.120 products from retailers during panic buying. 06:07.120 --> 06:14.520 Sainsbury CEO Mike Coop told BBC that sales growth peaked at 48% in the week to 21st March 06:14.520 --> 06:19.880 and said "We sold more for five days running than we would normally sell in our busiest 06:19.880 --> 06:21.440 day at Christmas. 06:21.440 --> 06:23.520 That's why you saw the gaps on the shelf. 06:23.520 --> 06:27.440 We have seen a very marked change in the way our customers are shopping. 06:27.440 --> 06:32.920 They're coming less frequently and are buying about twice as much when they do come. 06:32.920 --> 06:35.600 Some facts and figures that caught my attention." 06:35.600 --> 06:40.840 Audification Here's what the mystery noise from earlier 06:42.120 --> 06:46.920 Audification is the process of turning numbers into sounds to help understand them in the 06:46.920 --> 06:51.560 same way that visualisation turns numbers into images such as graphs. 06:51.560 --> 06:57.240 Starting from the episode "Battery Sounds, How Does a Battery Sound in Time Lapse?" 06:57.240 --> 07:02.800 My target is to find weekly, that is seven day, signals in my energy data. 07:02.800 --> 07:03.800 Why? 07:03.800 --> 07:07.920 Because I want to see if our consumption patterns are mainly daily with relatively little change 07:07.920 --> 07:09.460 for weekends. 07:09.460 --> 07:14.660 Under lockdown such a barely weekly finding might be less astonishing, but my thesis is 07:14.660 --> 07:19.080 that our life is driven most strongly by a constant daily cycle anyway. 07:19.080 --> 07:21.480 Other households may well be very different. 07:21.480 --> 07:25.880 One slight issue with hearing a weekly cycle if I find one is that at the typical one year 07:25.880 --> 07:30.720 to one second time lapse that I've been using until now the cycle will be a touch over 52 07:30.720 --> 07:31.720 hertz. 07:31.720 --> 07:36.360 That's only a little higher than, and might be confused with, UK mains hum. 07:36.360 --> 07:41.600 I can hear 52 hertz in my headphones, but not on my MacBook Air speakers, so I might 07:41.600 --> 07:46.480 have to crank up speed a bit, maybe to three years per second, to ensure that all listeners 07:46.480 --> 07:47.520 can hear it. 07:47.520 --> 07:51.920 But then I feel a bit short of data, at not even two full years captured with the in-phase 07:51.920 --> 07:53.920 battery system so far. 07:53.920 --> 07:57.800 Listening again to some previously captured PV generation time-lapse from just over a 07:57.800 --> 08:03.600 year in duration, the seasonal fluctuation is quite audible. 08:03.600 --> 08:11.640 Again and in a plotted spectrogram a strong daily signal is also clear at about 365 hertz 08:11.640 --> 08:13.880 and is much of the buzz. 08:13.880 --> 08:16.680 Again. 08:16.680 --> 08:20.440 Sunshine doesn't know about human weeks, though there may be more rain at weekends due to 08:20.440 --> 08:24.880 changing pollution levels, so a strong weekly signal would be unexpected. 08:24.880 --> 08:29.360 I failed to find a weekly signal in battery discharge but rediscovered the high-pitched 08:29.360 --> 08:33.360 whistle of the fridge freezer consumption. 08:33.360 --> 08:37.919 I saw a hint of something like weekly in the spectrogram of grid imports and there's an audibly quieter summer when demand is covered by PV. 08:43.120 --> 08:49.320 But it really is only a hint, a relatively low peak, and at about 104 hertz, so twice 08:49.320 --> 08:51.660 per week, which is puzzling. 08:51.660 --> 08:53.760 This is the mystery noise from earlier. 08:53.760 --> 08:58.120 So then I had the brainwave to look at gross electricity consumption as well. 08:58.120 --> 09:02.320 That's backing out the effect of the battery and PV, what we actually use. 09:02.320 --> 09:07.120 There's a couple of small quiet bits for August holidays but otherwise as fairly constant. 09:07.120 --> 09:12.960 And a fainter twice weekly 104 hertz part. 09:12.960 --> 09:15.160 This quest is not finished yet, therefore. 09:15.160 --> 09:18.840 Finding a good weekly cycle and making it audible eludes me. 09:18.840 --> 09:21.880 No cut to the chase for this episode. 09:21.880 --> 09:25.520 See more audification details in the show notes. 09:25.520 --> 09:26.880 Bonus sound. 09:26.880 --> 09:33.640 The intro and outro sound is a new audification of 5 years worth of solar PV generation data 09:33.640 --> 09:37.360 from EOU towers via the SunnyBeam display unit. 09:37.360 --> 09:43.360 Again, more in the show notes. 09:43.360 --> 09:44.360 Mystery number. 09:44.360 --> 09:46.840 Now for the mystery number reveal. 09:46.840 --> 09:53.400 As I come towards the end of putting together this episode it now stands at 34034. 09:53.400 --> 09:59.760 What is this mysterious value that has risen by 213 since I started this piece on the 25th? 09:59.760 --> 10:05.680 It's the revision number for the SVN repository, in which I keep all the source files, and 10:05.680 --> 10:10.000 many of the data files, for this and some related sites. 10:10.000 --> 10:14.440 Each increment represents the capture of one or more changed files. 10:14.440 --> 10:19.560 Most of those changes are manual, but some are, for example, the new daily consumption 10:19.560 --> 10:23.280 values for the grid tide and off-grid PV systems. 10:23.280 --> 10:29.160 Quite a few of those 213 commits are edits and contributions to this piece, including 10:29.160 --> 10:34.320 the words and the sounds, and even the code to make those sounds. 10:34.320 --> 10:41.920 I said not to sue. 10:41.920 --> 10:47.880 There's more on my Earth Notes Web site at Earth.Org.UK.