The House at the Bottom of the Garden

Nice description Terry, I'm hoping to get ours installed in the next week or so. We actually moved them to the site yesterday, prior to that they have been "interesting" kitchen ornaments. As you know we have gone with a pair of Sunamps but in a master/expansion unit configuration. For our typical usage model I didn't see need for a parallel pair.

Yup the main difference is the max flow rates.  I don't understand why they plumb up the cells in series rather than parallel.

Maybe there's an advantage to charging or discharging sequentially - something like the series PCM equivalent of maintaining stratification in a thermal store?  

God I can't even grasp all that water phase malarkey even when dumbed down. 

 

They do sound good thought. 

 

What happens when you run out of the 10 kw and want a shower? Can the sunamp heat on demand?

 

i know you pointed out gas as being dear but my last connection was below 200. It's subsidised

Simple question regarding an unvented cylinder, what forms or documentation is required and what enforcement exists regarding this?

4 minutes ago, Oz07 said:

God I can't even grasp all that water phase malarkey even when dumbed down. 

 

All you need to understand is that when you heat something like ice, it takes a certain amount of energy to change its temperature by 1 degree (say from -10 deg C to -9 deg C).  The amount of energy required for each degree temperature change is fairly consistent.  However, once you reach 0 deg C - the temperature where the ice melts - something interesting happens: the ice stays at the same temperature as you keep putting energy in.  You keep pouring lots of energy in, and slowly the ice turns into water, but without actually changing temperature.  As Terry said above, it takes the same amount of energy to change the temperature of ice from -58C to 0C as it does to convert all the ice to water without changing its temperature.

 

The Sunamp phase change materials use the same principle.  The main difference is that they are selected to have a phase change temperature that is adapted to the desired application.  In the Sunamp case, the temperature is around 60 deg C (I believe - it may actually be a temperature range more than a specific temperature).  

 

That means that when you first heat the material, it'll get to 60 degrees, then stay at the same temp while you pile more energy in.  In this way, it can store a lot more energy than water over the same temperature range (incoming cold water temperature up to a safe maximum of, say 80C), because the phase change system has its phase change temperature in the working range of the fluid.  The phase change temperature is also high enough that it can usefully heat water above the desired temperature for supply to the DHW side of the plumbing.  Having a phase change material with a phase change at, say, 35 deg wouldn't be much use.

 

It's really interesting stuff.  I looked at PCM panels when we first started planning the house.  They're supposed to help buffer heat to keep the house temperature more constant, but I think they're overkill (and expensive) in a passiv-ish house.

Probably a good time to repost this September 2015 entry by Andrew Bessell (SunAmp CEO) from the other place.

(As usual if the original poster objects to this being reposted please contact us and it will be removed)

 

=======================================================================================================================

 

Phase Change Materials

Phase change materials work by storing heat when they melt, and releasing it when they freeze again. Their mode of operation is simple in principle.

Melting/freezing stores/releases a lot of energy. As an example, think of water as a PCM that melts at 0^oC.

Water ice has a specific heat capacity of 2.000 kJ/kg/K at -10^oC (and density 918.9 g/L)
Latent heat of fusion of water (ice to liquid) is 334 kJ/kg
Liquid water has a specific heat capacity of 4.182 kJ/kg/K at 20^oC

Over the temperature range -10 to 20^oC there is (about) 10*2.000 + 334 + 20*4.182 = 20 + 334 + 84 = 438 kJ/kg of total heat capacity.
NOTE that over 76% of the stored energy in that 30 degree band is latent heat and less than 24% is specific heat.

For comparison, ice would store ~60 kJ/kg over 30 degrees from -31 to -1^oC; water would store ~125 kJ over 30 degrees from 1 to 31^oC.

(See:
http://www.engineeri...ties-d_162.html
http://www.engineeri...ties-d_576.html)

So the effect of having a solid/liquid phase change in the temperature range of interest is to dramatically increase the heat capacity over the range. That's why we use PCMs.

Of course not all materials are suitable to be PCMs. To be a PCM, a material has to have the right melting point for the application, a suitably high latent heat capacity (and ideally a high specific heat capacity too), and be safe and reliable over the long term.

Choice of PCMs

There are two main classes of Phase Change Materials for room temperature class applications (i.e. 10 - 90^oC):

• Organic - paraffins, waxes, fats, oils, fatty acids, etc
• Inorganic - salts, salt hydrates, metals (e.g. Woods metal)

It was easy to choose between these, and then hard to execute the choice.

Organic materials are common and easy to get, easy to work with, have a wide range of tuneable melting points, but suffer from defects that, in my view (and Sunamp's), make them fundamentally unsuitable for use in household heat stores.

The outright show-stopper for me is the high flammability of organic materials. I would not put tens or hundreds of litres of hot oil in a store in my home, and I wouldn't ask anyone else to do the same.

The other key defects of organic materials are lower energy density (typically half that of salt hydrates) and poor sustainability ("bio-derived PCMs" often derive from palm oil, which is strongly linked to tropical deforestation - sustainable sourcing may be available, but how can we be sure?; paraffins are directly derived from oil refining).

I don't absolutely rule out use of organic materials in future, in some capacity (perhaps in industrial heat stores, or ones buried underground), but I can't recommend their use inside homes.

Of the inorganic PCMs, the metals are costly, physically very dense (I mean g/L not kWh/m³) and toxic. Pure salts generally don't melt until much higher temperatures.

Salt hydrate PCMs

Sunamp uses salt hydrate PCMs.

We do so because they are non-flammable and have high energy density (very similar to water and water ice), and offer a range of melting points.

As you know we work with families of materials (where we have applied for patents) based on:

• Sodium Acetate Trihydrate (primary melting point: 58^oC; tuneability 46 - 58^oC)
• Strontium Bromide Hexahydrate (primary melting point: 88^oC; tuneability 75 - 88^oC)

Only SU58 (our 58^oC PCM using Sodium Acetate Trihydrate aka SAT) is in full-scale production as yet. It is our mainstay material, used in the heat batteries (HB58) used in SunampPV and other forthcoming products.

Our PCMs were developed with University of Edinburgh, and there are forthcoming academic papers due. For reasons of academic priority, I can't say everything I'd like to about the materials yet.

What I can say is that conventional SAT PCM is used (as some here have observed) for hand warmers. These can be stored at room temperature and can be activated by clicking a metal disk inside. They are recharged by fully melting the SAT inside by heating the hand warmers in a pan of hot water.

There is something to like in this approach - storability at room temperature - BUT there are some serious drawbacks:

• Conventional SAT materials used in hand warmers degrade after hundreds of cycles due to "phase segregation" - an anhydrous fraction precipitates out like a sediment; what is left behind is more dilute and less effective
• They need to be activated - when do you choose to do that?
• When the materials cool back to room temperature for storage,the specific heat is lost, so total energy storage is lower than it might be
• They don't reach 58^oC when re-activated and the energy released is less

Sunamp's SU58 is formulated differently. It contains additives that achieve two things:

• A polymer that eliminates the degradation effect. All the science points to the degradation being completely eliminated. Details to follow in the scientific papers, so please be patient. However what we have learnt does allow us to offer a 10 year limited warranty similar to the best electric batteries.
• It includes a nucleator, which means that the material starts to freeze cleanly when cooled to 58^oC. This means that there is no control issue about when to activate the material, no lost energy and we achieve the full 58^oC.

We have evaluated many other materials, and continue to do so. The 88^oC family is very exciting and essentially new. Many of the best ones at other temperatures are already in the public domain, though not always with their best formulations.

Sunamp Heat Batteries

A heat storage appliance like SunampPV needs to contain a heat store. In Sunamp's case we use Heat Batteries. After several generations of refinement (the first Sunamp heat batteries were deployed in field trials in 2013), we have arrived at a red plastic-cased cell, the shape of a large cereal box (115 mm wide x 463 mm long x 499 mm tall, externally occupying 26.5 dm³), weighing a little over 30 kg.

A heat battery cell offers:

• 45 Litres of instant hot water or
• 2.2 kWh of thermal energy

when discharged through a TMV set at 55°C, from 73°C to 45°C. Higher or lower performance may be delivered for other applications and operating conditions.


For example jsharris wants hot water at 42^oC. In that mode, two cells in series would deliver about 4.4 kWh before dropping below 42^oC. Three cells about 6.6 kWh, provided those cells were charged to an average 73^oC using, say, off-peak electricity or diverted solar PV electricity.

SunampPV includes two such cells (sitting side-by-side, connected in series) and hydraulics to allow the electric charging of the cells.

When discharged, cold mains water flows into the appliance and through heat exchangers in the cells. The cold water picks up heat from the PCM. Initially it transfers heat from the hot liquid PCM (which cools from 73 to 58^oC progressively), and then from the PCM as it freezes (at 58^oC). Finally once the PCM is all solid, the solid cools further (from 58^oC downwards), continuing to transfer heat to the mains water.

Is the end of useful energy when output drops below 45 or 42 or 40^oC (e.g. in a hot water application replacing a hot water tank with immersion heater or an instant hot water heater), or when the output approaches cold mains temperature (e.g. when SunampPV is connected as a pre-heat to a combo boiler**)? That's for you to judge based on your application. It affects useful heat output: there is about 20% of total heat to be had from 45^oC to ~11.5^oC. We quote heat capacity without this at cell level, and with this heat at SunampPV level (as SunampPV is designed for use with combis, where pre-heat is useful).

(For SunampPV spec sheet please see:
http://sunamp.co.uk/...ure-for-Web.pdf
Apologies that it needs a little updating. For example it overstates the mass by 10 kg (it has come in below estimates at 80kg +/- 1 kg)

Power and Flow Rate

One very critical parameter that almost all other PCM system developers tend to fudge is power. This is because the thermal conductivity of PCMs tends to be very low (<< 1 Wm^-1K^-1), so its hard to get the heat in and out fast. Its all very well that PCMs offer high energy densities, but if you can't get at the energy, is it useful?

Sunamp Heat Batteries are designed to overcome this. A pair of heat batteries in series can sustain over 12 litres per minute of discharge when fed with cold mains water at 10^oC and deliver most of the energy in a plateau at just over 50^oC. Power is over 40 kW peak, 35 kW sustained, with over 30 kW still being delivered when falling to 45^oC.

A couple of points to note:

• The plateau temperature doesn't change much whether the mains cold inlet is 5, 10, 15 or 35^oC. This means power scales UP in cold conditions, when mains temperatures would be say 5^oC. This is desirable.
• There is very little water stored in the appliance - for SunampPV about 5 litres is present. This water comes out at high temperature (whatever temperature the PCM is at, whether 70+^oC at the end of charging or 58^oC during the freezing plateau) so there is IMMEDIATE hot water - no waiting for a heat exchanger to heat up; no initial slug of cold (except what is in pipework).

(Little water in the appliance, a different slug between each discharge and high temperature storage and, at least after every charging, over 65^oC, absolutely minimises legionella risk.)

Frankly I wanted to stomp the power problem into the ground, and I am proud that we have. One consequence is that SunampPV outperforms most combis it could be paired with (which is what we want). Another is that SunampPV can be used as a high flow rate electric hot water heater at >30kW when connected to a 3kW spur circuit (albeit not continuously, but if the average duty cycle is 10% it works).

We haven't yet tested maximum flow rate at maximum rated mains pressure. We'll update when we know.

Heat Loss

When building a small appliance, smaller than a water tank of equivalent heat capacity by 3 to 4 times, you risk being a mouse (i.e. having the surface area:volume ratio work against you). You also risk having total insulation volume end up close to or greater than your heat storage volume. For this reason SunampPV uses Vacuum Insulation Panel. We can do that because we have flat surfaces to insulate, and because the total surface is quite small (so we could afford to do the right thing - VIP is expensive per m²).

So far preliminary heat loss testing has been performed in-house using the approach used to test an electric storage water heater, wherein there is thermal draw-off and recharge using a tapping cycle during the hours of 07:00 to 22:00. This suggests 0.7 kWh / 24 hour heat loss from SunampPV. We are about to send a unit to a test lab for confirmation and other tests, so should have better data soon.

I am sorry that the 5W figure clouded this. That is the upper bound of the electrical standby load.

I hope we can further significantly reduce the thermal losses later.

Regulatory

If you read this after 26 September 2015, either we will have ErP certification from the above-mentioned test lab OR SunampPV will temporarily be off the market until we get ErP certification. Equally SunampPV doesn't yet have CE mark (although it has passed all the related lab tests for EMC and LVD). We believe SunampPV meets the UK water regulations, but WRAS certification is still pending.

SunampStack

Other Sunamp products will be composed of different numbers of cells in series, parallel or series-parallel arrangements. We offer cells to OEMs and for our own products. While we are not offering bare cells direct to the general public,we expect to offer SunampStack, a configurable product line, soon.

SunampStack is composed by integrating multiple layers of heat batteries

• Each layer is composed of 2, 3, 4, 5 or 6 cells
• Layers may be stacked up to 4 high (2,350 mm)
• High degree of flexibility and modularity to meet different storage requirements from 4 to >50kWh (90 to >1000 litres)

Example SunampStack:

• 5 cells wide by 3 layers high = 15 cells
• Occupies 655 mm wide x 540 mm deep x 1800 mm high
• Delivering 675 litres above 45^oC via a TMV set at 55^oC or 33 kWh

That's all for today folks.

Next I'll try to tackle some of your specific requests. Please be a little patient.

Andrew

2 minutes ago, jack said:

 It takes the same amount of energy to change the temperature of ice from -58C to 0C as it does to convert all the ice to water without changing its temperature.

Typo.  It's amazingly -158°C.  The entropy of a crystalline structure is a lot lower than the liquid phase which is why converting from one to t'other soaks so much heat.  And this is also why the specific heat of ice is a pretty much half that of water.  This effect is also why steam is so dangerous.  If it condenses into water on your skin it chucks a lot of heat and causes serious scalding.

 

28 minutes ago, Oz07 said:

What happens when you run out of the 10 kw and want a shower? Can the sunamp heat on demand?

If you run out of heat, then you run out of heat. It's store not a generator.  But there's nothing to stop you topping it up during the day if you need to and one of the standard uses is to dump your excess PV into it during the day.

 

31 minutes ago, Oz07 said:

I know you pointed out gas as being dear but my last connection was below 200. It's subsidised

My quote wasn't cheap

 

22 minutes ago, joe90 said:

Simple question regarding an unvented cylinder, what forms or documentation is required and what enforcement exists regarding this?

 

John, this one is a bit off-topic but l will answer this.  In essence a UVC is classified as a potentially dangerous appliance, which is not surprising really since if one burst at 3 bar and blasted a few hundred litres of scalding water around your house, then people could get seriously hurt.  This is covered in Part G paras 3.17 to 3.21.  Installation of a UVC is notifiable and must be installed and certified by a qualified installer.  Your B Insp won't sign off your completion certificate without this if you have a UVC.  

Sorry it was off topic, yes I have a neighbour/plumber who will sign this off and I understand why , I just wondered how it's enforcable to make sure it's done annually?

I don't know. I suspect that it's not, but it one of those things that will bite you badly if you don't and something goes wrong. Will you be insured? Will you be criminally liable?

Hi @TerryE, great blog post, very interesting reading. 

Can I ask why you decided not to fee your UFH from the SAPV? 

1 hour ago, Barney12 said:

Can I ask why you decided not to fee your UFH from the SAPV? 

 

My logic is this: the SAPV is a thermal store that enable you to charge it with E7  or PV electricity at times when electricity is cheap, and because f its low heat loses, you can then discharge it into heating your DHW when you need it.  It is optimised for heat delivery in the 40-60°C range.

 

In an MBC house with a passive slab (or other equivalent vendor offerings) you already have equivalent thermal store but one that is optimised for storing heat at 20-25°C and that is the slab.  I had to get my head around using it this way, which was the reason behind my topic Modelling the "Chunk" Heating of a Passive Slab, but my current plain is to use a small electric water heater to pump a chunk of heat into the slab each night when I am in net heating days.

18 hours ago, jack said:

 It takes the same amount of energy to change the temperature of ice from -58C to 0C as it does to convert all the ice to water without changing its temperature.

Typo.  It's amazingly -158°C.  The entropy of a crystalline structure is a lot lower than the liquid phase which is why converting from one to t'other soaks so much heat.  And this is also why the specific heat of ice is a pretty much half that of water.  This effect is also why steam is so dangerous.  If it condenses into water on your skin it chucks a lot of heat and causes serious scalding.

 

still don't get it completely!

 

anyway so once your out of power your out of hot water? Do you have any form of backup? How long do these take to charge and can you use them at any amount of charge?

1 hour ago, Oz07 said:

Anyway so once your out of power your out of hot water? Do you have any form of backup? How long do these take to charge and can you use them at any amount of charge?

 

AFAIK the discharge mode is essentially passive, so you can get H/W even if the power is out.  But as to charge mode, have you got a standby generator for your current gas central heating system?  If not, then you are in exactly the same situation today. 

 

I have two SunAmps in parallel so both have to fail to lose H/W.  And then I have a Propane 2-ring gas hob in my kitchen so tea and coffee are still available.  And the power would have to be off for more than 3 days before the house starts to feel a little cool.

 

5hWh / 2.8kW = just under 2 hrs if you've exhausted the store totally.

 

In practice squeezing the last 0.1kWh of heat out of the device probably isn't worth it so the trick is to size your system so you rarely run out, but all of this is exactly the same as if you had a conventional UVC. 

I can confirm that with the electricity to the house off the Sunamp PV still delivers hot water as usual, until it runs out of stored heat.  In our case the amount of hot water delivered is less, because we lose the preheat facility with the power out (the preheat uses a pump to circulate water from an ASHP heated low temperature buffer tank).  Still enough for a shower plus a fair bit left over, though, if you start with the Sunamp PV fully charged at the start of the power cut.

 

At our old house we have a gas combi, and one of the slight annoyances with that is that it stops working during a power cut, so we have no hot water.  That's not the case with the Sunamp PV, but in most other respects it works just like a combi, in that it's an instant water heater with a high power output that only heats water when there is a demand (i.e. a hot tap or similar is opened).

I have just been unpacking my SunAmps.  What a neat and well laid out piece of kit! I looks like it belongs in a computer server room, not in a cupboard. 

1 hour ago, TerryE said:

I have just been unpacking my SunAmps.  What a neat and well laid out piece of kit! I looks like it belongs in a computer server room, not in a cupboard. 

 

Mine are placed on the opposite wall to the 19" networking racks. I'm sure if I put a Cisco or HP sticker on them there won't be many who would spot them as part DHW system,

Terry, thanks for the post. I’m leaning towards using a Sunamp unit for DWH.

 

The one thing I’m not clear about is heating and/or cooling in a near passive house. How do you achieve this?

@Triassic, not sure why I missed your post.  Sorry.  8 months into occupation and the SunAmps work fantastically.  OK, they are Gen 1 product and the Gen 2 kit is more cost-effective, but that's the price of being an early adopter.

 

I need to do another post on our space conditioning.  Heating hasn't proved to be very much of an issue.  The 3 kW Wills is easily man-enough to handle sub-zero winter demands; the 1st and 2nd floors are 1-2°C colder than the target GFL temp (21.5°C), but that isn't an issue.  Our house has no S facing windows and the roughly E/W aspect works well in the summer.  (At least until this week.)  We just leave windows open on the shaded side (yes with the MVHR running) and a roof light in the apex, and this gives us enough heat dump to keep the house cool and airy. 

 

This current hot spell has been breaking this operating model.  The GFL is now climbing to ~24-25°C and we have a strong thermal gradient up the house with the 1st floor maybe 2°C warmer and the 2nd another 1-2°C again.  This seems to be bearing out my hypothesis that an ASHP might have a marginal payback for savings in heating costs, but might prove essential for height of summer cooling.  Though I need to think more about the heat gradient issue.

 

16 hours ago, TerryE said:

 

This current hot spell has been breaking this operating model.  The GFL is now climbing to ~24-25°C and we have a strong thermal gradient up the house with the 1st floor maybe 2°C warmer and the 2nd another 1-2°C again.  This seems to be bearing out my hypothesis that an ASHP might have a marginal payback for savings in heating costs, but might prove essential for height of summer cooling.  Though I need to think more about the heat gradient issue.

Hi terry. 

Do you think a 'geothermal' brine loop to drop the incoming MVHR air temp would have been enough to put a dent in the temp rise? 

@Nickfromwales, I considered a geothermal loop in the early days, but the simple issue is that we didn't have a big enough plot to fit one in.  An ASHP in ground floor UFH chill mode would help, but the main issue is my son's room in the loft.  The bugger has to many electronics chuffing out maybe 300-400 W, and with him, the thermal gradient and the high insulation, his room gets too hot in this weather.