White Paper 1 showed that existing systems don’t work very well, don’t collect a lot of hot water and certainly nowhere near what is claimed in the marketing hype.  They are also extremely cost ineffective.  But in that first Paper I did say the Tranquility (T) project had significantly developed the basic concept to solve the problems and make it hugely more effective.  The figures (just to remind) are that T’s hot water cost for the year 2009/10 was just over £7.10 and for 2008/9 – £8.24 so it can be done.

In reality SHW systems as sold have advanced very little since the black radiator or hose pipe which had cold water passed in one end with warmer or hotter water coming out of the other when the sun shone.

An Existing SHW System:

All SHW systems currently consist of:

  • A collector (Plates or Tubes) usually on the roof facing the sun
  • A water circuit between the collector and hot cylinder including antifreeze
  • The Cylinder which has an additional ‘solar heat exchanger’ to transfer heat from the roof to the water while the other heat exchanger transfers heat from the alternative energy source – of choice a gas boiler
  • A pump to transfer the available heat
  • A controller which turns the pump on when heat is available and off when it isn’t
  • The replacement water  is from the Mains

That is it.

What technical improvements have been made over time?

The only real improvement is in the collector so we will explain the difference between flat plates and vacuum tubes; my preference and what the technical improvements have been.

  1. Flat Plate collectors:  These are simple effective devices made from two pieces of sheet steel welded together round the edges and variously across the surface. There is sufficient space between them for water to flow, but not too much so the volume is small which heats quickly.  The entire surface area is the collector.

Product development has been in the surfaces.  Black absorbs more heat than white, but black painted surfaces are as good at emitting heat as absorbing it – a lot gained but a lot lost. The trick is to have a surface that absorbs a lot but emits little.

The ratio between what is absorbed and emitted is the ‘absorptivity/emissivity’ index – for black paint it is about 1.  Not special.  But for Nickel oxide the figure is 11 so it absorbs 11 times as much heat as it emits.  Terrific.  It is only the surface molecules that are the stars, so the art is to make just the surface nickel oxide.  This is not difficult if the steel is a high nickel content stainless.

  1. B.      Vacuum Tubes: The disadvantage of flat plates is they operate in ambient air, so in winter they are in cold air.  The heat lost is therefore high as the air can be well below freezing – and in winter when the sun is shining – it normally is. 

In vacuum tubes on the other hand the collector, a very small water pipe, is in the middle of a big glass tube which is evacuated so doesn’t lose heat, making winter collection more efficient but the collecting surface is extremely small.  The tube therefore has a reflective back so the sun falling on most of it is focused onto the small collecting pipe.

What was the collector choice for T and Why?

These two options were examined carefully; the flat plates won easily and I have every reason to maintain that conclusion.   The primary reason: the vacuum tubes are considerably more expensive per sq meter of collector – generally about half as much again – but if you spend the same amount on each you therefore get 50% more area of flat plates than tubes, and that extra 50% will give you the same heat collection in winter  but 50% more in summer.  However, the vacuum tubes do sometimes fail and the cost of replacing one is usually very high as scaffolding is often required.  Interestingly when I took the decision for T, the majority of installations were the tubes, but that has completely changed around for the reasons given. 

So how do we improve SHW performance?

The Tranquility System

This is of course the nub of the issue and as always I didn’t try to modify conventional wisdom or existing products.  I started with a blank sheet headed with the task:

Collect heat from the sun and transfer it into water.  A lot of heat – to supply as much of the hot water as practical. That is all there is to it.

The golden Rules:

               i.      Catch everything we can as effectively as we can

             ii.      When we have caught it – keep it and

           iii.      Use it wisely

The way I work 

To reach an optimal solution – never stop at the first idea.  If the first ‘idea’ solves 60% of the problem cost effectively – that leaves 40%.  Either the first idea was not good enough or the 40% must now be tackled.

If the next ‘idea’ saves 60% of the 40%, 84% of the problem is cracked – but there is still 16% to go.  When there is no further cost effective solution we are there.

Designing for Existing Houses

My objective with SHW is to crunch the amount of fossil energy used nationally to heat water, but with only 150,000 new houses being built annually and 23,000,000 existing houses – solutions for existing houses are crucial.

The full new T system has been developed on from what was developed here and is designed to be retrofitted.  It is ‘Patent Applied for’ so I apologise that the detail cannot be given but we can explain more or less how T currently does it.  You will see that a number of the T solutions explained here would be between difficult and impossible for most existing houses but there is a way to achieve probably all of this with retrofitted systems though they may not look remotely like the systems described here or work like them!

Working through the hierarchy of ‘solutions’ as I think they were thought of:

Solution 1:

Here is a case where many if not all problems must be considered simultaneously even at the earliest design stage as any one might impact on any other.  Water is a huge global problem and as technically the UK is officially ‘Water Short’, it was always a requirement that T should be almost if not totally water independent. Could the water solution be part of the Hot Water solution? [The problem isn’t that it doesn’t rain in the UK or even that it doesn’t rain for long enough periods but that we are hugely over-populated.  Dividing quite a lot of annual water by a huge number of people means we only have about 1,260 cubic metres/ person/ year.  1,750 is the UN figure defining water short].

Drilling horizontally into the hill behind T would actually find water, but not too many of us can do that so using that as the water solution here would develop nothing and prove nothing.  Except there is water in the hill.  So the water had to come from the rain, but to be water independent means providing ALL the water for the house for ALL its uses including drinking, bathing and cooking.  ‘Rainwater Harvesting’ systems cannot do that and are not allowed to for safety reasons as the water could easily be contaminated with legionella, salmonella, leptospiridium, cryptosporidium or any of the other very serious nasties.  They would also be hugely expensive and not financially cost effective.  T is approved to do what it does but please don’t try to copy it without proper equipment AND control systems as you could contaminate the mains which could affect neighbors as well as you.

Once the requirement is to provide all the water for a substantial building there will need to be a substantial storage facility – so where to put it?  The bacteria and other problems with water don’t occur below 20°C, so it must be below ground, but could it be stored where the location will partially warm the water?  What about inside the house?

After the raid on the MET office archive I had a chart of local soil temperature versus depth through the year. The main tank, which is 15,600 litres, is therefore installed inside the perimeter of the house and underground but at a specific depth, the resulting temperatures being as calculated to within a degree.   Water is then drawn into the house from just below the surface where the water is warmest and cleanest.  Taking the warmest also helps stop it overheating.

Rainwater Tank temperatures through 2009


Notes: The red is the surface temperature – green is on the bottom. The mains are taken to be at 8.5°.

It was a poor summer; the temperature in this tank peaks in August at or just below 20°, and is only at 8° for a very short period through the winter.

This has 3 benefits when trying to collect sufficient heat to have enough hot water for the summer:

  1. Hot water should be delivered from a tap at about 42°, so normally needs heating through 33.5° from the 8.5°.   If the average supply temperature from this tank is about 18° through the summer period, 9.5° of this or 28% is already achieved. Just from where the tank is.
  2. The cold taps are now delivering at the 18° which means it isn’t COLD.  In fact it is easily warm enough to use for hand and face washing as well as much else.
  3. If the water in the cylinder is quite hot it must be blended somewhere even if at the tap itself to make it safe.  But blending with water at 18° makes the hot go further than if it is 8.5°.

Don’t knock it.  We have saved about ? of the cost of heating the water through the summer already and a fair bit through the winter as well.

Solution 2:- again solving 2 problems at the same time.

The water must be delivered to the taps at some pressure, but as it is underground a pump is needed.  [There are many water issues being dealt with simultaneously which are covered in a further White Paper on ‘Water Supply’, and these benefit hugely from a 450 liter Break Tank being installed in the top loft].  But if the water is batch processed into this tank it sits there for some time and if this a ‘semi warm’ space, this water is warming gently hence a second passive heat gain.  Even in mid winter this space is above the temperature of the water arriving so it always warms.

Break Tank against R/W tank Temperatures – a year

Notes: The Break tank cools when filled and warms slowly as the water is used hence the saw tooth, but the mean temperature gain is significant and almost constant through the year.  The number of oscillations shows how many times the tank was refilled.

The gain is generally over 2.5°, so now we have passively gained 12° or 36% in total.  Already.   And as this is the cold for the house, even the cold water used in the kettle or pans needs 12° less heating.

Solution 3: surely we can do better

The water now has to travel from the Break Tank to the hot taps, but the Solar Room is almost in the way and always relatively hot so what if we ran the pipes through this space so they could gain a few more degrees?  Can we do better even than that?  Of course we can.  We add another cylinder.

The water that will become the hot passes into and through a first stage 300 litre cylinder (C2) stored below the Break Tank but above the glass roof of the Solar Room.  This is connected to a passive vertical homemade solar panel installed inside the south facing glass.  As it cannot freeze, the water is heated directly saving the cost and inefficiency of heat exchangers.

The beauty is that no power is required.  The flow is thermo dynamic so when the temperature at the top of this panel is higher than the water at the top of C2 – circulation starts – and if the panel is colder no flow will occur so no heat can be lost.  Terrific.

The surface of this collector needs to have a high absorptivity/emissivity ratio (as covered above) but as there was so much going on at the time I ran out of time to solve this one – putting it aside for another day.  I can achieve a figure of between 6.7 and 9.7, but as there was no plant in the UK that could apply what was required on a collector of this size, it is black painted – so has a figure of about 1.  However it would be very easy to reach 3.4!

But the output is superb, though the temperature in C2 doesn’t go much above about 43° it spends all summer at about 33° more and in winter is generally above 18° – even with a poor surface. Taking the summer average, the cumulative gain is now 24.5° or we are 73% of the way to hot water.  And the solar panel hasn’t seen a ray of sunshine yet.  Oh and today on 17/6 it is at 37.5°.

Passive C2 against Break Tank – June

Notes: Another rather good chart showing the temperature in C2 compared with the Break Tank. Each stage of the passive gain is demonstrated in the charts above, and as these are dynamic they already allow for the hot water being drawn by the house.

The water arriving in the hot cylinder therefore only needs a further 9° added through summer.  Even on a poor day the SHW will often reach the 42° needed whereas a conventional cylinder stands no chance at all.

Solution 4:

The hot cylinder (C1) can be at up to 92° here but with the passive pre-heat (C2) sitting behind it at maybe 33 degrees or more.  And we want water to the taps at 42°!  Even C2 is nearly hot enough and sometimes is.

It is crucial that a blender valve be fitted to make certain the water is delivered at a safe 42° so the water from C1 must be mixed with cold water – or must it be cold?  It must be colder, but what if the ‘colder’ side is supplied from C2 – already at 33°.  Suddenly the hot water in the main cylinder (C1) is going to go a very long way indeed as most of the hot water in the ‘blend’ is going to be supplied from passive C2 with just enough coming from C1 to make it up to the 42°.  Now we are beginning to talk. In fact these two cylinders between them can hold an equivalent of about 1,600 litres at the 42° – which is massive.

But when showering one day the shower stopped.  Damned technology I thought. What’s it sulking about now?  I realised the blender valve had an impossible task.  Water was sitting on one side at over 90° and on the other at about 40° – so as it couldn’t make water at 42° it locked down!

Solution 5:- and a modification to the original.

The ‘cold’ side of this blender valve can now be supplied optionally from C2 (37.5° today) or from the domestic cold supply which is from the Break Tank (20.4° today).  With a motorized valve on the optional pipes, the supply can be switched optimally between the two so when C2 gets too hot, the blending water is from the Break Tank, and when it isn’t, it comes from C2.  Getting better.

Solution 6:

Given the hot cylinder is frequently getting extremely hot from mid spring onwards; the excess heat must be removed or stopped from arriving.  In T there are two options – one being a bit of a luxury, but the original calculations showed there was a lot of heat to remove.  T has a sauna that can be heated from the roof because panel temperatures can reach about 130° or be dangerously hot.  This water is switched by another motorized valve to a ‘radiator’ in the sauna BUT BE CAREFUL and please don’t copy unless you know exactly what you are doing.  Not many radiators can handle water at 130° nor can almost any Motorised valve, and the heat coming off the roof is many kilowatts and not too many radiators could emit that amount of heat especially into a very hot space.  But it can be done.

Alternatively the excess heat can be dumped straight back into the bottom rainwater tank by bleeding the excess hot water from C1 via another auto motorised valve.  In this way the heat is retained in the system and no water is lost.  On such a bleed the temperature of the water in the Rainwater Tank can be raised about 2 degrees, but that pushes the entire supply chain up that 2° so the gain is proportionately substantial.

Impact of bleeding surplus hot water from the Hot cylinder to the water Tank

Notes: This chart shows the impact on the temperature in the rainwater tank of bleeding excess heat from the Hot Cylinder – C1. It covers the last week to 22/6 and shows 1 full and 3 part bleed offs which resulted in a temperature rise of 2.9° in the R/W tank.  But this rise will affect each passive stage and of course C1 heated by the Solar Panels.

Solution 7 Could there be a 7?

Well, yes there could and is.  So far we have only dealt with ‘Catching the heat’ but we have to keep what we caught.

When we think of insulating a house in 2010 we quite casually accept 300mm or more in the loft – yet the loft maybe around freezing with the rooms below are at say 20° so 300mm for a 20  difference.  Yet a Solar Cylinder can go to 90°; the space around it may be about 25° and there is 50mm for the 65°difference!  Enough said?  Normal cylinder insulation is hopeless even on a Solar Cylinder.

So the insulation on the cylinder needed to be radically rethought and improved which it has been.  In White Paper 1 it was shown that a normal solar cylinder actually loses about 21.55kw.hrs each week, but T’s now loses only about 7.9kw.hrs. This is an extremely cheap and cost effective solution for all existing cylinders so an instruction sheet will be downloadable on HYPERLINK which explains what to do, why and how much it should cost.  

Solution 8:

So we have caught it and kept it: now we just need to use it wisely.  Here we will just consider washing dishes and clothes – both of which can be significant hot water users (or not as the case may be) – and using the sink.

Most modern machines have a single fill pipe plumbed into the cold so the water is always heated by electricity even with oceans of solar hot water available.  The manufacturers worry because clothes can be damaged if the water is too hot, so as they don’t know the temperature of your hot water – the safe thing is to cold plumb.

But water that is too hot for clothes is too hot for people, so there must be a blending valve in Solar systems to control the hot water. You will know if you have one if:

  1. The hot pipe leaving the very top of the hot cylinder travels horizontally to this blender valve, into which a further pipe comes usually from the opposite side, and below which it goes to all the outlets. If you run a hot tap you will feel the water leaving the cylinder is very hot; that on the other side of the valve is cold, and that leaving the bottom will be hot but not as hot as that leaving the cylinder – if the cylinder is say at 60° or above.
  2. When the cylinder is very hot, the hot water at the taps or showers is still quite safe to put your hands under.

When you know the water is safe, check the supply pipes into the Washing and Dishwashing machines, and if they are on the cold – switch them to the hot.  The hot water in T does supply these machines all through the year so the hot water cost figures here include all machine use.

Using the sink is interesting.  Normally the hot tap is run until the water is hot and the sink then filled (or horrendously the hot tap is left running).  In T about 5 litres has to be run before the hot arrives, so 5 litres of hot water is ‘used’ every time the hot is run which is left in the pipes to go cold afterwards.  The solution is to boil a kettle on the gas hob. It takes 0.34kw.hrs here to boil 2 litres in a kettle, and mixed with 3.3 litres of cold from the tap gives 5.3 litres of hot water to wash up in.  Much the lowest cost and carbon solution to the problem, but before anybody wonders how much it has cost to boil all the kettles here – T’s total gas consumption on the hobs over exactly 2 years has been 61.3 m³ which has cost about £22. And this covers everything done on the hobs including lots of homemade soups and jams.

How much fuel is used to heat water with the Tranquility System?

The conventional SHW system – as shown in SHW White Paper 1 – needed 54.12kw.hrs of gas to be burned on average each week during the period analysed, but as is shown here Tranquility actually burned 0.20kw.hrs.

The energy difference between a ‘normal’ system and that in T is really huge, with almost no external heating required here at all.  But when so many passive energy gains are made the outcome has to be between exceptional and extremely good.

What of the retrofit developments from the T system for existing houses? 

Using the same thought processes a number of products have been invented/designed which will dramatically improve SHW performance for ‘normal’ houses and may be able to achieve all, or almost all, that T does.  Indeed in some ways they will do more as the White Paper ‘The Route to a Zero Carbon Britain’ argues that compound solutions are again required, and one is to match energy demand with the available supply.  The T retrofit solution does try to do this as well.

A retrofit solution could provide all the hot water for a house through the summer period without the use of a boiler.  This does assume 48.28kw.hrs/week of hot water is drawn (no one hour showers); that an even better insulated cylinder is developed, and that the sun’s energy is collected in more than one way.  The energy supplied by the boiler will then be ? 0.  It is of course theoretical but all earlier calculations proved extremely accurate and my expectations on this are no less.

There is a great hurry to bring these products to market so a building specification will be defined that allows T products to be retrofitted if that specification is followed.   But while for building energy this is practical, for SHW I am afraid no item currently available would be useful, not the cylinder, the collectors, the control system and probably not even the pump.  In order to achieve these high outputs at low cost each item has to be optimized and none is currently available on the market.


It has been shown that while existing SHW products don’t actually achieve much; certainly ‘don’t do what is claimed on the tin’,  and are not cost effective – superior and cost effective performance can be achieved. To significantly improve the performance it is necessary to pre-heat the water before it arrives at the hot cylinder.

The winter period has not been analysed here because the original objective was simply to discover what the existing systems produce but was extended into this Part 2 to show Tranquility has solved these problems.