Control of domestic night-storage heaters

#1
I have read some of rmeldo's threads here from 2008 and wondered if the project was successful. This is a topic dear to me, as I have three old but sound night-storage heaters that are hard to control, and have often thought about automating them. Now that I have a better idea of what PICAXE controllers can do (and can't do), they seem ideal for this.

Electrical heating seems to be coming back into vogue, at least in the UK (although with our power costs it's hard to understand why). There are now at least two commercial ranges that use 'intelligent' controllers, Elonur's Ecombi (Spanish) and Dimplex Quantum (UK?). They are costly, especially the Dimplex ones, so retrofitting a controller to old heaters is appealing. Has anyone tried this?
 
#2
Greetings Baker Steve,

For more than 5 years I have been using Picaxe 08Ms to control a 2KW convector heater and a 150W electric over-blanket.

Both use separate mains to 5V plugtop power supplies, zero crossover opto-isolated solid state relays, 40A for the convector and 5A for the blanket.

Burst firing controlled from a potentiometer which is preset for both heat and program sequence.

Dave
 
#3
Control is not the issue. It's the interface between PICAXE and mains power that is of concern.
My advice would be to get some RF plug sockets and use the PICAXE to 'press' the button on the remote.

If however, you are happy and qualified to connect to mains, then there are plenty of solid-state relays available that can be controlled by PICAXE. Just be careful about connecting direct to a PICAXE pin. Some SSRs require more drive than a PICAXE can safely provide and will require transistor drive to keep them within spec. The data-sheet will of course tell you that.
 

hippy

Technical Support
Staff member
#4
I must admit I am intrigued as to what PICAXEing or automating a night storage heater would amount to.

I have never had storage heaters myself but thought they were powered on over night ( or when electricity tariffs were lower ) then released their heat over the course of the day.
 
#6
If I had some (which I don't), I would charge them during the day from my solar panels.
I do this with my hot water immersion.
PICAXE measure inverter output and house load via current clamps and a precision rectifier. (no direct connection to mains)
It then determines how much excess power is being generated (it's actually a PID loop) and applies an appropriate PWM value to a burst fire SSR.
Hence, immersion is run at the highest value possible without ever drawing from the grid.
One day, I'll get around to sending all the data to an ESP8266 and posting it to "thingspeak" or similar.
So far, I bought the ESP8266, downloaded new firmware and bricked it! (it's still waiting in the corner to be forgiven for wasting 2 days of my life).
 
#7
Some interesting replies here: thanks.

To hippy, yes, that's how they work: basically a pile of bricks that are heated with cheap electricity at night and then cool down during the day. The problem with night storage heaters is that you have to guess how much heat to put into them based on how cold you think it's going to be 24 hours ahead. Down here in the pointy bit at the bottom-left of the UK the temperature can change very quickly in the winter due to oceanic winds, as it is doing as I write this. The traditional controls for old-style night-storage heaters consist of a 'charge' control (as far as I can tell just a thermostat on the core), and a 'boost' or 'release' control, basically a mechanical thermostatically-controlled damper. The latter tends to get jammed with age. The Elonur Ecombi, and as far as I know, the Quantum series, try to improve on this by determining how much heat was lost from the core during the last discharge (i.e. daytime) cycle, and increasing or decreasing this proportionally during the next night-time charge. Their control systems are proprietary.

To achieve this with a microcontroller would seem relatively simple, only slightly complicated by the fact that a night-storage power circuit is only powered, er, at night. Warnings about working with mains voltages are appropriate, but using an opto-isolated Triac pack, a radio link, or even a relay, should keep things safe for those competent to work with such things. My electronics background, although it was quite a long time ago, was in the design of medical diagnostic systems. As you might imagine, we had to pay a fair bit attention to not electrocuting the patient!

Beanibots: yes, true, but the issue with solar PV (which I also have) is that you want the heat in the winter, when there tends not to be enough PV power to do much with. And especially at night ;-) For hot water it's a great idea, true, and I must get around to doing that.
 

hippy

Technical Support
Staff member
#8
Thanks for the clarification. It seems there is quite a bit to it which I hadn't appreciated and makes it quite an interesting challenge.
 
#9
Yes, I think it's worthwhile. I have two clock projects on the go, one purely electronic and the other electro-mechanical, both using PICAXE controllers, that should keep me busy for a while, but as soon as they are either completed or abandoned I will turn my attention to the storage heaters.
 

tmfkam

Senior Member
#10
To throw something into the mix...

I've recently (been forced to) changed over to an Economy10 tariff. Our heating is purely electrical, with storage heaters as our only space heating. Economy10 comes on from (Approximately) 13.00 - 16.00, 20.00 - 22.00 and 00.00 - 05.00. This gives a reasonable spread of heat across the daytime, without it being like the Bahamas in the morning, yet like Siberia in the evening as we were when on our original (night time only) Economy7. We were previously on a slightly cheaper for heating, HeatWise tariff but this is no longer available or so Eon told us.

If you are already on this type of tariff, forget I mentioned it.
 
#11
Surprising, as Economy 10 is notoriously hard to get hold of. I have not pushed to get it, as my Economy 7 hours suit one of my other occupations, baking. We get a slightly odd mix: off-peak from 11-12 pm, then from 1 pm to 7 pm. So I fire the ovens (up to 9 kW) from 11 pm onwards, as the bulk of the power is taken while they heat up.
 

premelec

Senior Member
#12
I'm curious as to just what the rate differences are between the tariffs - I live in a large thermal mass [stone house...] and appreciate thermal storage but am on methane combustion heat mostly and 100% wind power for electric [utility has the nerve to charge more per wind KWHr... ]. I have modified my boiler controls some to use residual boiler heat and therefore run boiler at lower average temperature hydronic 4 zone system...
 

JimPerry

Senior Member
#13
In Berwick-upon-Tweed - with a bedroom wall that is 30-foot thick ... off-peak ( 11:30pm to 8:30am ) is half the cost of normal. So my wife and I only use dishwasher/dryer and washing m/c on delayed start :eek:
 
#14
The tariff I'm currently on (with eON) is day: 13.556p, night: 5.208p, both per kWh, and a daily standing charge of 16.42p. This is a good rate for these days, achieved through the Cornwall Collective bargaining scheme. So my night rate is slightly over one-third of the day rate. Previously (with Scottish Power until 2014), the corresponding figures were 16.036p day, 7.393p night, and standing charge 27.39p/day.

These are Economy 7. I don't have any data for Economy 10, as I've never had it.
 
#15
The traditional controls for old-style night-storage heaters consist of a 'charge' control (as far as I can tell just a thermostat on the core), and a 'boost' or 'release' control, basically a mechanical thermostatically-controlled damper. The Elonur Ecombi, and as far as I know, the Quantum series, try to improve on this by determining how much heat was lost from the core during the last discharge (i.e. daytime) cycle, and increasing or decreasing this proportionally during the next night-time charge.
As the proud owner of a single Cornish Storage Heater - which I visit as often as possible :D - this is something I have given some thought to. The Dimplex XLS24N (a current product) uses a mechanical whirly-gig: https://store.gdcgroup.co.uk/gifs/product_images/large/dimplex/006001_0_1.jpg on its input control. It's in free air on the underneath of the unit - so I'm not entirely clear what it's supposed to be measuring or responding to. It's described as: "Hydraulic close differential room temperature sensing thermostat with external sensor" which doesn't really help:confused:

I was thinking along the lines of delaying charging (ie power-on), based on tomorrow's weather forecast - pulled off Yahoo! weather or similar service.
 
Last edited:
#16
The 'intelligence' of the Dimplex Quantum series is described here:

http://www.dimplex.co.uk/products/d...nstalled_heating/quantum/contractor/index.htm

The XL range uses 'old style' controls, of the type I described in an earlier post. Viz. 'The heating level of XLN range is adjusted by means of a variable input control, which the occupant sets in line with the changing weather conditions.' I.e. look out the window and guess. The 'S' variants have some sort of weather compensation:

http://www.dimplex.co.uk/products/d...ne_storage_heater/technical_specification.htm

As for your idea of using a weather forecast, go for it! If you make it work you'll have something you can sell ;-)
 

papaof2

Senior Member
#17
Doing estimation for tomorrow's heating needs sounds like a good place for an outside temperature sensor and a wind speed sensor.

The outside temperature sensor would be checked regularly (hourly?) and the values stored for perhaps a rolling 24 or 48 hours. A check of the temperature trend, up or down, and the rate of change plus some recorded history should provide reasonable direction on needed the heat needed for tomorrow. The recorded history could be converted into a lookup table where the temperature plus the rate of change was the index to the entries in the table.

The wind speed sensor would provide some indication of the rate of heat loss, over and above the temperature changes. Both the wind speed and its rate of change could be combined with the temperature rate of change to give a "heat needed" forecast. That forecast would control when and how long to storage heater(s) had power.

I've tried none of this, being in the US and having relatively inexpensive natural gas fired central heat.

I'm aware of the value of structural mass in maintaining a building's temperature. Most of this house is brick and part of the lower level is below ground (daylight basement). We also have a large masonry fireplace (now with gas logs) and I know how long the brick continues to radiate heat into the room after it has been warmed by a fire.
 
#18
As I started this thread, I'll reply. Your idea sounds like a good one. In the UK none of our heating is inexpensive, relatively or otherwise :-( I have an 'optimizing' (Vassmann) gas-fired boiler that provides the main heating, the storage heaters being only for the upper floor. The boiler controller uses an outside temperature sensor, although I'm not sure how; the good thing is that it just all works.

Certainly if you are able to profile a building in terms of its heat loss vs external temperature and wind direction and strength, it should give a good basis for optimizing heater control, bearing in mind that you would always be guessing the forthcoming weather conditions. A wise person once commented that the UK is like a 'Clapham Junction' for weather (Clapham Junction is a very busy rail junction south of London) - it's not like the more predictable weather that you get on a large continental land mass like the US or Central Europe.
 

premelec

Senior Member
#19
The outside temperature sense units usually have a thermistor somewhere on the outer wall of building... effectively allow running the boiler at lower temperature when outdoors is warmer... As I mentioned earlier similar effect can be had by delaying start of flame to see if you can make up the heat from boiler - turn on circ pumps and wait to add fire... i suppose now there are Internet weather report connected devices... :)
 

tmfkam

Senior Member
#20
The costs for Economy10 are (per kWHr): Off peak (made up of three charging periods) 8.11p, Peak rate 16.21 . For Heatwise, the rates are split into three values, across four charging periods (per kWHr): Off peak (rates 1 & 2) 8.28p, Off peak (3) 6.83p, Peak 16.21p. Heatwise has the advantage that the longer early morning rate is at the lowest cost. On Economy10, while in an 'Off peak' period, all electricity consumption is charged at the lower price so the TV, PicAxe programmer etc. used during this time costs half as much to run as it would during the peak period. On Heatwise, only loads connected to the 'Off peak' circuit of the electric meter is charged at the lower rate. The cost of Economy7 here next to the cold wet North Sea through Eon is Off peak: 6.83p, Peak 16.21p so I would say that Baker Steve has managed to negotiate a very good price. Standing charge for all tariffs for us costs 24.78p per day.

Our heaters are the old Dimplex type with an input thermostat and a Bi-Metallic vent flap type device to control the output. We leave the vent flap side turned so it has zero effect and vary the input stat only if the weather gets very cold. As both my Wife and I work, we find that the Economy10 or Heatwise suit us well. When we were on Economy7 (only for a few weeks during the first winter we moved in) we needed to have the heaters so hot in the early morning 'charge up' portion of the day to have enough heat remaining when we returned home. We couldn't bear the heat at breakfast, and shivered late evening.
 
#21
True, our rate is good, but I can take no personal credit for it. Community Energy Plus, which I believe is a charity, instigated the collective bargaining scheme. This really is the way to go with the big power companies. I switched last year, getting a better rate even than the one Scottish Power offered me when I told them I was leaving. This year the eON deal is even better. Because my part-time business is bread, I use a lot of electricity: before I switched I was paying Scottish Power in the region of £1800 a year for electricity alone (there is no mains gas here). I also have solar PV, which offsets the cost a bit.

I will return to the storage heater control idea, hopefully before next winter, when my current projects are further forward, after which I will have more experience with PICAXE, but I've appreciated the discussion here.
 

PhilHornby

Senior Member
#22
Community Energy Plus, which I believe is a charity, instigated the collective bargaining scheme.
Definitely worth knowing about - I've signed-up for the next one, which is sometime in June.

(there is no mains gas here).
My cottage has no mains gas, no mains sewerage and electricity delivered by overhead power line. However, it does have fibre-optic broadband :).

I also have solar PV, which offsets the cost a bit.
How would storing that Solar power into the storage heater directly, compare with selling it to the grid and then buying it back as economy 7, cost-wise? I got embroiled in an argument with a PV salesman at the Ideal Homes show last year; he couldn't understand why I would want to 'store' my electricity, rather than sell it.
 
#23
I've looked briefly into the PV *battery* storage idea. Although it has been 'productized' by the likes of SMA, Samsung and Tesla, it's not yet cost-effective due to the massive cost of the batteries. It's not very 'green', either, as AFAIK there is still no commercially viable method of recycling Lithium batteries - although this might have changed since I last looked.

There is no way with a domestic PV of specifying where the power generated goes - it goes into the grid and/or house wiring. If you meant attempting to capture some of it in night storage heaters, the problem with this is the seasonal one: you need the power in the winter, but PVs generate the bulk of their output in the summer months.
 

tmfkam

Senior Member
#24
You can get devices that automatically 'steer' PV generated energy into immersion heaters, I suppose these could be connected to storage heaters for space heating if required. They work by measuring the incoming supply and working out whether electricity is being exported (sold to the grid) or imported (bought by you). If the load within the property is low enough that electricity generated by the PV exceeds the load, the immersion (or storage) heater is activated. I applied for a job with a place locally that designed and manufactured them.

The company is: https://www.immersun.co.uk . I didn't get the job sadly although I'd got a better offer at the time so never followed up the interview. I have no connection with them, no doubt other companies make similar devices. They did say that the sense technique they used was patent pending, and superior to other competing products. I was given an outline of how it worked at the time, but I can't spill the beans as I simply can't remember!
 
#25
Yes, true, there are a load of systems that are 'almost there' with a view to intelligent dumping of PV power: Geo Solar comes to mind. There are some actual products too, but my impression is that the market forces driving these have withered due to the UK cutbacks on subsidies.

Smart dumping makes sense if you *do* have PV but *do not* have solar HW. However the same rule applies for storage heaters: you need the power in the winter, and at night: not much sun ;-)
 

Jarubell

Senior Member
#26
I really do not know much about PV systems, how big are these systems mentioned in this thread. I have two storage heaters (4.5kW each) at my cottage with baseboards in the bedrooms, now I forgot to turn off two baseboards on over Christmas and my daily average use was 86kWh! I would think that this would be too large for an average PV system, am I wrong?
 

premelec

Senior Member
#27
You're right... Do your heaters have thermostatic controls to shut them off? 86/9KWHr = 9+ hours on time - would seem yout might be a candidate for more sophisticated controls... :) Solar residential PVs around here seldom go up to 10KW peak and mostly 5KW peak - not a lot of sun hours in winter...
 

Jarubell

Senior Member
#28
They do have a controller that from what I read, charges the ETS units during the night, taking into consideration the outside temperature and thermostat setpoint (11pm-7pm). I was guessing that I would require a PV farm to power it! I figure I'll figure something out to supplement my heat cost, other than wearing more clothing! (Evac Tubes? but that's for another day and forum)
 
#29
You can get devices that automatically 'steer' PV generated energy into immersion heaters, I suppose these could be connected to storage heaters for space heating if required. They work by measuring the incoming supply and working out whether electricity is being exported (sold to the grid) or imported (bought by you). If the load within the property is low enough that electricity generated by the PV exceeds the load, the immersion (or storage) heater is activated. I applied for a job with a place locally that designed and manufactured them.

The company is: https://www.immersun.co.uk . I didn't get the job sadly although I'd got a better offer at the time so never followed up the interview. I have no connection with them, no doubt other companies make similar devices. They did say that the sense technique they used was patent pending, and superior to other competing products. I was given an outline of how it worked at the time, but I can't spill the beans as I simply can't remember!
You can also do it with a Picaxe: https://openenergymonitor.org/emon/node/2429 If you scroll down that page you'll find an excess PV generation diverter that I built using a Picaxe 08M2. Works a treat and has been on continuously for a couple of years with no hitches.

A fraction of the cost of an Immersun, too, and since building that original I have upgraded the design so that I do the sensing in the meter cabinet and then transmit data using ERF modules to another Picaxe controlling the water heater switch and another in a small plug in energy display so I can have a real-time readout of import/export anywhere in the house.
 

tmfkam

Senior Member
#30
You can also do it with a Picaxe: https://openenergymonitor.org/emon/node/2429 If you scroll down that page you'll find an excess PV generation diverter that I built using a Picaxe 08M2. Works a treat and has been on continuously for a couple of years with no hitches.

A fraction of the cost of an Immersun, too, and since building that original I have upgraded the design so that I do the sensing in the meter cabinet and then transmit data using ERF modules to another Picaxe controlling the water heater switch and another in a small plug in energy display so I can have a real-time readout of import/export anywhere in the house.
Thanks for that interesting read. While I have no PV, I am going to look into the AD7755 as I have some minor interest in monitoring the power and think it could be a useful device.
 

PhilHornby

Senior Member
#31
You can also do it with a Picaxe: https://openenergymonitor.org/emon/node/2429 If you scroll down that page you'll find an excess PV generation diverter that I built using a Picaxe 08M2. .
Fascinating stuff (I particularly enjoyed the PDP-11 reminiscences!).

me said:
How would storing that Solar power into the storage heater directly, compare with selling it to the grid and then buying it back as economy 7, cost-wise?
Having done a bit more research, am I right in saying that you could store all the (otherwise surplus) energy you generate and still get paid for generating it? (and also still be deemed to have exported it to the grid!). (I was reading this: http://www.fitariffs.co.uk/FITs/principles/generation/ )
 
#32
I'll reply to Phil: yes. The FIT payments are divided into two parts - the payment for what you generate and the payment for what you export. The former is much greater than the latter. In theory at least - and if smart meters had been implemented - the exported quotient could be measured. In reality it usually cannot, so a system called 'deeming' was set up (by Ofgem?). This makes the assumption that you export half of what you generate, and rewards you accordingly. So under this scheme no-one would ever know if you'd exported nothing.

Typically, the payment for generation is at about ten times the rate for generation, although the various waterings-down of the FIT schemes over the years may have shifted this balance a bit. So *in theory* it's well worth your while to use as much of the electricity you generate instead of exporting it. As electricity costs are only going to rise over the years, this is going to become more and more true.

The problems arise when you start to look at how you might store electricity/energy. Batteries are costly, heavy, generally not very 'green' (especially Lithium) and have limited service lives. When I got interested in this sort of thing round about the mid 1990s, the folks who were setting up off-grid supplies could benefit from a large scrapment of MOD lead-acid batteries from the submarine fleet. These were built to last and got snapped up quite quickly. Coming back to today, you can find a lot of information on the current commercial solutions from SMA, Tesla, Samsung and the like on the respective companies' websites. A major capital investment is involved, so a careful cost-benefit analysis is necessary.

The other method of storing PV power is as heat. For this you need a large dump load - the larger the better. There have been experiments with phase-change systems that store heat based on the latent heat of liquefaction. To do this you fill a large tank with a suitable salt that melts at some convenient temperature. You use surplus electricity to melt the salt; the salt then gives this heat out again as it cools. Phase-change systems like this can store vastly more heat than mere bulk heating of water, but you need a cellar or a large outbuilding to hold them. I do not know whether any such systems have reached commercial production.

Coming back to domestic reality, it's a good idea to use PV power to heat your hot water. There is little point though if you already have solar HW, as both systems generate at the same time. If you have a swimming pool, that might make a good thermal reserve, although experiments with a large underground HW storage system at the Centre for Alternative Technology in the 1980-90s was generally deemed to have been a failure due to the cost of the engineering works and maintenance. An electric Aga could also act as a thermal dump load: I used to know a designer of wind generators that used his wind turbine to heat his Aga. But if you don't already own an electric Aga don't even think about one, as they cost as much as your electricity bills for the rest of your life, and eat the stuff too ;-)

What we need therefore is a cost-effective long-term power storage system that scales down to the domestic level. I don't know of one.
 
Last edited:
#33
I have a commercially available phase change heat storage system, charged by excess PV generation. It is small, very efficient (it has a fraction of the heat losses of a hot water tank) and works extremely well. It's made by a company called Sunamp who have been very good to deal with. The system is modular, and I have their smallest domestic range version, the Sunamp PV, which stores 5 kWh of heat as instantly heated (like a combi) hot water at 75 deg C, mixed down at the outlet to a safe hot water temperature with a standard thermostatic mixer valve. The unit is around the size of a small wall mounted gas boiler, and needs nothing other than an electricity supply and a cold water in and hot water out pipes, plus a vent drain pipe in the event of an over-pressure problem.
 
#35
I know of another chap who's installed two Sunamp PVs in parallel, to get a max instantaneous heating capacity of 70 kW, with 10 kWh of stored heat energy. The Sunamp Stack will be available soon, allowing larger scale, modular, units to be built to suit larger heat storage requirements.

We replaced a 260 litre double spray foam insulated thermal store, that had an additional 50mm of PIR foam added in the form of a taped up and foamed in place box, with the Sunamp PV. The performance of both in terms of real hot water delivery is pretty much the same, but the measured losses on the double foam insulated thermal store were originally around 3.4 kWh per 24 hours, which reduced to 2.5 kWh per 24 hours after I added the extra 50mm box of PIR foam. The Sunamp PV losses are around 0.6 kWh per 24 hours, and the compact case of the thing never even feels warm on the outside.

It is the massive reduction is standing heat loss which accounts for the ability of a 5 kWh heat store to do the job of one that could store around 7 kWh in theory. We bought the 260 litre thermal store in the belief that the manufacturers test spec of 1.6 kWh per 24 hours heat loss was real, in fact none of the manufacturers give the true loss for a tank kept hot (as it is when fed from excess PV), they all use a "boiler cycle tapping test" which gives a very optimistic standing loss figure.
 
#36
This is encouraging, but they're all quite small in terms of energy storage capacity - although comparable with Samsung's domestic electrical storage products:

http://www.samsungsdi.com/ess/energy-storage-system-application.html

What we really need is access to massive - i.e. inter-seasonal - energy storage. One can only dream... as I in fact once did, considering filling an entire barn with a gigantic heatstore consisting of coupled tanks. In the end all I filled the barn with was a small farm of freezers, but that's another story.
 
#38
The market for this is the (I guess) relatively small number of households that *do* have solar PV but *do not* have solar HW.

A big part of the cost of such systems would appear to be the energy monitoring, as you need to measure differential AC current. A couple of years ago I looked at PV monitoring systems, but I only found cheapish toys that didn't do the job properly (I can provide details if anyone is interested), or costly systems like the Elios. Once you have effective energy monitoring in place, intelligent power diversion is easy.
 
#39
The energy monitoring is dead cheap, one chip for a couple of pounds, plus a current transformer at around the same price. The chips I used are just one of many designed for energy metering, and the market for them is huge, hence the very low unit price. They are very accurate (to electricity meter standards, with 16 bit fast A/D converters to measure the voltage and current waveforms) and are bidirectional, so they can measure whether you're importing or exporting and store totals in internal registers or just export instantaneous energy (or power) data.

When you think that a bidirectional (import/export) electricity meter, complete with housing, display, IRDA port etc retails at one-off prices of around £30, then you can appreciate just how cheap the internals of these things is.

If someone doesn't want to mess about with current transformers etc, then the easy thing to do is just buy an off-the-shelf electricity meter and take the data out of the IRDA port. I'm doing this for household monitoring, I bought a £30 Elster A100C, which has an IR port that constantly transmits data from the internal registers. It's dead easy to just read this with an IR receiver and translate the serial data to do what you want with it. This chap shows how to do that pretty easily, and I'm sure a Picaxe would be up to the task: http://www.rotwang.co.uk/projects/meter.html
 
#40
I am posting a follow-up to this thread, as I've found a way of improving control of old night-storage heaters (as long as they're not rated over about 2 kW), in case anyone else is struggling with old but sound heaters with their awful controls.

There is one - and I've only found one - wireless thermostat that can switch a 2 kW heating load: the Honeywell Y6630D1007.

<https://heatingcontrols.honeywellho...ontrols/wireless-temperature-controls/Y6630D/>

The UK price for the sender/receiver pair is about £80 - a lot cheaper than replacing an old storage heater with a new 'optimizing' one.

I had assumed that it would work in the same way as a conventional mechanical thermostat, but in fact it implements a PID controller. As the sensor unit is battery powered, it continues to monitor zone temperature while the storage heater (assuming it's on an off-peak circuit) is powered down. Although an unconventional application, this is showing vastly improved control over guessing the weather and fiddling with the storage heater's mechanical stat. I have graphs.

Hope this works for someone.
 
Top