Hot water

It works!

Yes, indeed. The wonky water heater, cobbled together from old gas bottles, made the trip from Scotland to the tiny house in France. I used it every day for a period of about 3 weeks and I’m pleased to say that it performs as expected. Perhaps even better than I had hoped for.

Installation

The photo above shows the water heater installed a few metres away from the tiny house.

One addition to the heater was the home made ‘spark arrestor’ fitted to the top of the flue. It is actually a steel mesh pen/pencil holder (cost 0.69 Eur!). I don’t know how well it really works as I did see the occasional spark escape. Maybe it would have been worse without it.

You can just about see the copper ‘tails’ from the heat exchanger coil connected to the red plastic pipe. It wasn’t too difficult to find the fittings to join the UK 10mm copper microbore pipe to the 12mm PER plastic pipe (PEX in the UK).

The red plastic pipe came covered in protective conduit, so the pipe was simply laid on the surface of the ground for the duration of my summer stay.

At the other end of the plastic pipe I connected short sections of 14mm diameter copper pipe (I used this size of pipe throughout the house). These sections went up through two holes drilled in the tiny house floor. These were then connected to my hot water mixer valve system, via flexible pipes, as shown below.

Admittedly the plumbing ain’t pretty, but it does work. Surprisingly, none of my solder joints leaked but I did have some trouble with the isolator valves which had compression fittings. I tried using PTFE tape and using replacement olives, without finding a reliable fix. After lots of trial and error, I eventually got the leaks down to one small seep, which actually stopped after a few days.

The mixer valve controls the temperature of the hot water that goes to the kitchen sink, bathroom washbasin and shower. The valve itself cost about 34 Euros from Amazon France. It was simple to install and worked very well.

Although the sink and washbasin have mixer taps, the shower does not. Instead I installed a simple ball valve. This is the red valve next to the soap dish in the photo below. In this way, I was able to run the shower pipe neatly around the edge of the corrugated steel shower walls. This was important as I didn’t want the complexity of a conventional shower mixer which would have required holes to be drilled in the shower walls and pipes installed behind them. Another advantage of my system is that it takes up less space inside the shower, which is only 70cm square.

Operation

Operating the water heater was simplicity itself. For firewood, I have lots of small pieces of timber offcuts. These were stacked in a cross-cross fashion in the heater firebox and a gas blowtorch used for a few seconds to light it. I could also have used paper, or firelighters, were I less lazy/impatient.

Once lit, the heater smoked a bit for 5 or 10 mins then burned extremely clean. At dusk it was easy to see the flames leaving the top of the flue. Perhaps there was secondary combustion happening in the flue? In any case, it was something of a shame to see this wasted heat. In retrospect a taller thermal store (from a larger gas bottle) would have enabled me to make use of this energy.

Anyway, by the time I had cleared my tools away at the end of the day, the water was up to temperature (70-90 deg C) and I could have a well earned shower!

I found that an indicated setting of 35 deg C on the mixer valve gave a good temperature for showering and washing up (I’ve no idea what the actual water temperature was). Once set, it was not necessary to change it.

It takes perhaps 20 or 30 seconds for the hot water to arrive at the shower head. The little ball valve allows the flow rate to be easily adjusted. With a ‘medium’ flow rate, the shower temperature is maintained without a problem for the length of the shower. This was a nice surprise given the small size of of the copper coil heat exchanger.

I was often in the tiny house alone, so I could afford to be quite indulgent with the length of time spent in the shower and the amount of hot water used. With more care, it should be possible to have 2 or 3 showers without needing to add more firewood. Not that adding a few more sticks and waiting 15mins would be any great hardship.

One final point: in order to hide the messy plumbing, I boxed in the mixer valve system and added a door. This was painted a similar colour to the kitchen units and a small black knob added. Now it looks quite neat and unobtrusive.

Final thoughts

I am really pleased with the water heater – it was such a thrill to get hot water from the tap and see that my crazy idea actually worked!

Although I have no complaints about the performance of the system, there are some improvements that could be made. Fitting some insulation to the thermal store is the most pressing. At the moment it is an effective radiator and loses heat quite quickly.

Another thing I may do in future is fit a flue damper. This might reduce some of the heat being lost up the flue. I suspect that when the fire gets to the glowing ember stage, it would be good to be able to close off the flue so that more of the residual heat is used for water heating.

I’m not sure if it is worth insulating the firebox. Yes, it does lose some heat but it is also quite nice to sit outside next to this fire with a cup of coffee. I imagine in the colder seasons this could become a daily ritual :-). When the firebox door is opened the radiant heat can be felt some metres away.

Unsurprisingly, the water in the thermal store became quite orange and rusty. The gas cylinder and flue are, after all, just made from mild steel. Sooner or later a rust hole will appear. I guess, at that point, I’ll just have to patch it up. I’ve no idea if adding corrosion inhibitor to the water would really help. As it’s not a sealed system it does need to be regularly topped up to replace the water that has evaporated, so I’d be continually adding corrosion inhibitor too. At about £14 per litre, that’s not ideal. What would be ideal is a stainless steel thermal store and flue. If I were to go down that (expensive) route, I would be as well buying a commercially available wood burning water heater and adding my own heat exchanger, if needed.

A final, final point is that I still intend to build a solar water heater to be used in conjunction with the wood burning heater. Time defeated me during this summer’s build (2021) but it’s something I hope to return to in future.

In a previous post I dismissed the idea of a solar water heater as being too complicated and expensive.

However, it is still an appealing idea. Simple, clean, free, effective and ideally suited to make use of the sunny French climate where the the tiny house is located.

Unlike a ‘normal’ house with a ‘normal’ family, my hot water requirements are minimal. Using a Portashower, I know I only need 5 litres of water per shower. The tiny house can only sleep three people (and mostly it would be just me). So I could probably get by with 20 litres of hot water per day, or even less.

Commercial Solar Water Heater Systems

The typical solar arrangement (shown below) with a roof mounted solar panel feeding a large hot water cylinder would be way over the top for my tiny house.

Even a simpler ground mounted solar panel with a thermosiphon header tank is too big and too expensive. For example, this 200 litre system, from the French company Solaris, costs nearly 1300 Euros.

DIY Systems

I know that people have been making DIY solar water collectors for years. Typically these consist of copper or plastic pipes in a grid or coil arrangement, painted black and placed in a glazed insulated box. Hot water from these collectors is normally transferred to a tank – either by pump or thermosiphon.

There is plenty of information out there on DIY solar water builds, such as the one shown below from instructables.com.

Going down the DIY panel route has advantages. I could build something small enough to meet my needs and I wouldn’t have to pay the shipping costs of a bulky commercially available panel. But what about the header tank? I don’t need 200 litres. What are the options?

I considered lots of possibilities but the one that seemed most promising was based on an insulated plastic cool box. I could easily get one with about 20 litres capacity (perfect!) and I could probably fit a tap to drain off the hot water as needed. The problem then is that I’d then have to manually top up the panel with cold water, unless I fitted a float valve plumbed to a cold water feed. So even now things are starting to get a little bit more complicated.

Is there an even simpler way?

My thoughts then turned to the old idea of the solar batch water heater. At their simplest, these are basically a “tank in the sun”. Typically they consist of a recycled water cylinder, painted black, put in a glazed and insulated box and plumbed in-line with the domestic water heater. Thus the cylinder is both the solar collector and the storage vessel. The hot water coming from the cylinder, into the domestic water system, may require some ‘topping up’ with heat to get it up to temperature, but useful energy savings can still be made.

OK, so the batch water heaters are cheap and simple, but I dont want a giant cylinder in my garden with much more water than I need.

But what if I could make a normal solar collector with a large enough water capacity that it effectively became a batch heater? I.e. no need for a separate tank.

Remember, I only need 5 litres of hot water at any one time.

With this in mind, I came up with the following design:

Mini Batch Water Heater

There is nothing revolutionary in my design. The important point is that the 22mm copper pipe is large enough to give a total water volume of 5 litres for a panel that is around 1 square metre. It might not heat up as quickly as a normal collector (which has a lower water volume) but that’s not too important. If it can get the water to 45 deg C in 30 mins or so, that will be absolutely fine.

I wont go into too much detail regarding the construction. The 22mm copper pipe would be soldered together using standard elbows and tees. Ideally it would be clipped onto a thin metal sheet of some kind in order to increase the amount of heat absorbed, but I think this is optional. The whole lot is then sprayed matt black.

The retaining frame is made from wood and insulated at the rear to reduce heat loss. I think that anything can be used for the insulator but I suspect a sheet of foam would be cheap, effective and easy to get hold of.

As for the glazing, I wouldn’t be too keen on using real glass. Besides the weight and fragility, it would also have to be custom cut. I would go for clear, corrugated polycarbonate. To avoid it fogging on the inside, I would seal the ends with tape. A bead of clear silicone would fix it into the frame.

A pressure reducing valve (PRV) might be an idea if there was any chance that high temperatures would cause water expansion and pressure on the solder joints. I guess this wouldn’t be needed if the heater was attached to a garden hose as this could expand slightly. Of course, if the garden hose was disconnected all the water would drain out! So maybe a non return valve would be a good idea, in which case the PRV might be needed after all.

Some wooden legs either side of the frame would allow the panel to face the sun at, say, 45 degrees. If I was being really clever I could make a cradle so that the panel could be rotated to any angle.

If I wanted to store a greater quantity of hot water then I could fill up an insulated container 3 or 4 times during the afternoon, giving me some for the evening.

Should the panel prove to be effective, it might actually be worth connecting it to the water system of the tiny house. It could conceivably be fixed onto one of the exterior walls of the house and hinged so it would swing outwards to collect the sun.

To conclude – solar is back! If I make it to France this summer (2021) and I finish all the other jobs I have to do, then I will have a go at making the mini batch water heater. Stay tuned to see if it works…

Something which had taxed my brain for a while was the tiny house hot water system.

I considered lots of options. As ever, I wanted something simple, cheap and environmentally sound.

A back boiler on a wood burner could have been used to heat a hot water tank by a simple thermosiphon arrangement. Unfortunately, there’s not much room inside the tiny house for a hot water tank. Neither would it be practical in the summer, when it’s already 25-30 deg C inside, to have to light a stove in order to take a shower!

Solar water heating systems seemed to be complicated and expensive and, again, I don’t have much room for hot water storage tanks.

I considered a cheap gas powered water heater, the sort that gives hot water on demand. Apart from the gas consumption and it’s lack of eco-credentials, I was wary of it breaking down in a few years and the likely hassle of obtaining spare parts. Not to mention having to install the thing on the wall inside the cabin (no space!) and have a flue vented to outside.

In the end I decided to build a wood fired water heater that could be located outside the cabin but connected to the cabin’s plumbing system.

This would solve the problem of overheating inside the cabin, it wouldn’t require bulky hot water tanks indoors and it should be relatively cheap, simple and efficient. Nor would I need to install a twin wall flue inside the cabin, with all the expense and hassle that would entail.

Regarding the design, I got some inspiration from my storm kettle (a.k.a Kelly or Ghillie Kettle). The principle is that water is contained in a jacket surrounding a central chimney heated by a firebox beneath. Presumably it is this large heat transfer surface area which accounts for its effectiveness. In use it can bring 1.5 litres of water to the boil in just a few minutes using a handful of dry sticks.

The sketch below shows my design idea of the tiny house hot water heater. It’s based on two salvaged butane gas bottles, one on top of the other. The upper bottle contains the hot water, through which the flue passes. Sitting in the water is a coil of copper pipe that will act as a heat exchanger. The lower bottle is used as the firebox.

The insulation would serve three purposes. Firstly, it would reduce heat loss from the hot water, keeping it hotter for longer. Secondly, it would reduce heat loss from the sides of the firebox, thereby increasing the amount of heat transferred to the water via the top of the firebox and flue. Thirdly, it would improve safety by reducing the temperature of the external surfaces.

The design could be improved further by adding other elements, such as :

• Legs to raise the heater to a more comfortable height.
• A thermometer to measure the water temperature.
• A pressure relief valve and/or expansion vessel on the copper pipe leaving the heater.
• A grate to allow for good airflow under the fuel.

My main concern with this design concerns the heat exchanger. Would it be effective? In other words, would the cold water pick up enough heat on it’s passage through the coil to be of any use?

I did some research and, based on comments in various forums, it didn’t look too promising.

It’s likely that the flow rate of the hot water from the heater would be low. This might be fine for filling up a washbasin or kitchen sink but it’s unlikely to be good enough for a normal shower running from a mixer valve.

So was there any point in continuing with this design? How else could I make a working shower system? ‘

On previous building trips I had used a 5 litre garden sprayer such as the one shown below as an improvised shower. I removed the spray head and cut the tube so that just a little stub was left. This was necessary to improve the flow rate but gave a stream of water rather than a spray.

The last modification to the sprayer was a coat of matt black paint. After this, it functioned pretty well as a ‘solar shower’. By leaving it in the sun in the afternoon, I could have a hot shower at the end of the dirty and sweaty day of building! If I needed a shower at other times (or if I had guests) I simply put a kettle of hot water in the sprayer, topped it up with cold water and it was perfect.

So, getting back to the point, could I use a similar system for my indoor shower room? If so, I wouldn’t need a great flow rate from the wood burning water heater. I’d just need to be able to fill the sprayer with 5 litres of moderately warm water. It should also be possible to fill the sprayer repeatedly so that guests could have a shower too.

With that decided, I wondered if there was a better sprayer system on the market? One with a proper spray head?

After a bit of searching I found that Hozelock make a “Portashower” which has a 7 litre capacity, a good spray head and is pretty cheap (about £30). The reviews are very positive so I have ordered one and plan to use it this summer (2021).

While searching for the Portashower I also came across lots of electrically powered camping showers. Some of these have rechargeable batteries and others run from a 12v cigarette lighter socket.

As I have a 12v leisure battery in the house, I wondered if one of these systems could also work as a permanent installation in the shower room? In this case I could have hot and cold taps in or near the shower cubicle. These would be used to fill a bucket with warm water which would then be pumped electrically to give a shower. This might be a bit more convenient than the portashower (no pumping and screwing/unscrewing of lid) and the shower could be longer (10 litre bucket).

So, as a plan B, I bought a Ring RS1 portable 12v shower. At around £20, this is even cheaper than the Portashower and also has good reviews. I chose the Ring system as they are a well known brand but if you look on eBay or Amazon there are plenty of similar, inexpensive systems.

So now I have the shower problem solved … let’s get back to the wood burning water heater design.

I’m fairly confident that the heater will be able to produce 20 litres of hot water quite quickly. If this turns out to be true then the next question will be the design of the heat exchanger coil.

I calculated that a copper coil of 14mm internal diameter and a length of 20m long would have a volume of 3 litres. Mixing these 3 litres of hot water with another 3 litres of cold water will be enough for a shower.

Obviously, if there is still burning fuel in the firebox, the three litres of fresh cold water in the coil would also start to be heated. So while the coil would probably be poor in terms of providing constant hot water (probably wont pick up enough heat relative to the water flow) it should be able to provide hot water in batches.

Quite how easy it would be to neatly make a 20m coil of relatively large diameter copper pipe and then stuff it into a small cylinder is another matter! I suspect it would be a real challenge (and quite expensive if it goes wrong). So I will probably make a smaller, cheaper and easier coil in order to carry out a test. Here in the UK I can get a 10 metre coil of 10mm soft copper microbore tube for under £30, so that will be my starting point.

What if, after testing, the heat exchanger idea turns out to be completely impractical? Time for another plan B! In this case I could directly remove some of the heated water to fill up the shower sprayer, either via a drain tap or a simple hand pump.

The obvious problem with the steel cylinder is that of rusty bathing water. To mitigate this I could try coating the inside of the cylinder with some high temperature paint, such as engine enamel. Would it work? I don’t know, but it seems reasonable. The water in contact with the paint can’t get hotter than 100 deg C, so hopefully the paint itself wouldn’t get much hotter than this, despite the high temperatures on the other side of the metal. Another plus point is that I wouldn’t have to heat the water to near boiling point. Instead, a temperature of 40-45 deg C would be enough and this could hopefully be achieved with a less intense fire in the lower half of the heater.

So, to conclude this blog post, I have (a) a wood burning water heater design, (b) two methods of using the hot water for showering and (c) some contingency plans if the heat exchanger doesn’t work.

At the time or writing I have scavenged two old gas cylinders and have removed the valves and purged them with water. I’ve also bought some 75mm steel tube for the flue, some 30mm steel tube for the legs and some steel flat bar, nuts, bolts and washers. The next step is to build the heater and see how it performs.

I’ll report back in the blogpost “Free Hot Showers! Part 2: Water heater build and test”.