Sustainable heating using wood fuel at Coed Cae B&B, Dolgellau, Snowdonia, Wales

Heating with wood fuel at Coed Cae B&B

Last Updated Autumn 2014

The Coed Cae B&B journey to sustainable heating

Spring 2008
I have become a bore!
There are some who would say that there is nothing new here, and that the desire to share understanding beyond what others may consider ...err... interesting has long been evident. A key difference is that now I can see it in myself. I have become a bore.

In particular a heating bore. Well not so much heating, as the peculiarities of trying to marry thermal storage technology to solar heating circuitry whilst maintaining ...and there I go again.

We have been on a journey with our heating system; one that changed us irrevocably. It was a journey from ignorance to understanding, from naivety to enlightenment, from cold to warm, from fear of plumbing to being able to draw a heating circuit for the whole system. In short, from normal to boring. Like I said.

The Plan

The seeds of the project were germinated in the dirt of our old heating system, and since this was a solid fuel Rayburn which produced dust by the kilogram this is no allegorical reference. I have to say I had a grudging respect for the old range, for all its drawbacks. It gave a real character to the kitchen, which is pretty much the hub of our lives, and provided a warm retreat through the cold winter months. The problems really came to light during spring and autumn, when we needed the heating on, but it wasn't that cold. At this time the kitchen became unbearably hot, and guests would see us serving breakfast in shorts and t-shirts, sweat lashing out, whilst they were reaching for jumpers. Another slight problem was that it didn't heat the house, and what heat it did give out peaked about 3 hours after the last fuelling. Typically this meant that the house was at its warmest at 2:00 a.m and 10:00 a.m. which was far from ideal.

Nell finds a source of wood for the new log boilerWe finally conceded that we needed to upgrade the system and started looking at the alternatives. Oil didn't seem such a good long-term investment, and around this time Russia had been turning of the gas pipeline to further political aims, so we started thinking about some alternative solutions. Ground-source heating was discounted because we are pretty much stood on solid rock, so we looked at wood based systems. Pellet boilers seemed to offer the best controllability but required quite a bit of space, and worst of all, the pellets were at that time being imported from Poland. Out of the frying pan etc etc. In the end we opted for a log-burning furnace, sacrificing some controllability for low fuel miles. After all, we are surrounded by trees and forestry plantations. Another added benefit was that the furnace would need lighting every time it was to be used. Coming from a long-line of near-clinical pyromaniacs this was a certain bonus, and the unseemly struggles to get to it first at family gathering needs to be seen to be believed.

Paying for it

There were no two ways about it, this was to be an expensive venture. As well as the heating hardware, we needed to build a boiler house to enclose it, and a series of dry log stores for the fuel. Fortunately at the time, various streams of funding were available to help with the installation of carbon-neutral technology. We had been working on the principle of installing a system to which we could add solar water heating at a later date, but in the end bundled the lot together on the application. The form filling for a business application was quite a task, it being the same form as a health trust would fill in to convert their hospitals, and we failed on our first attempt. Given that you are only permitted two application attempts this was a blow, but we addressed the problems, reapplied and were successful. The grant only applied to the specific technology (i.e. the boiler and panels, but not the boiler house or the plumbing required to attach to an existing heating circuit) and covered about half of the cost.


One of the stipulations of the grant was that installation be carried out by an accredited engineer, and by chance we noticed that one of them The Green Heat Team was based about 4 miles up the road in Dolgellau. We were aware that they were a fairly new business with only a few installations under their belt, but mindful of the leaps of faith given to us over the years and the potential benefits of dealing with a local firm, we decided to go with them. We had scheduled the installation for late autumn, it being a quiet time for the B&B, and the switch from Rayburn to log boiler occurred just before Christmas. On commissioning the new system, it was immediately obvious that things were not going as planned.

The lack of background experience of the Green heat team had been a concern, but turned out to be a red herring; the technology itself, or the way we were using it, was sufficiently new that when we ran into problems even the importers of the hardware were floundering and giving different answers each time we spoke. In the end, most technical problems were ironed out around the kitchen table with ourselves, Andy from GHT and large piles of scrap paper. In order to make any kind of valid contribution to the discussion we had to get up to speed on conventional heating systems, then bin much of what we had learned as we got to grips with the new technology. I think that it was around this point we became boring.

Christmas was spent with clip-board and/or thermometer in hand checking various parameters, or trying to explain the system to anyone prepared to listen (in itself a dwindling supply). Because we were effectively making it up as we went along, it took time changing settings and logging the effects on the system. Slowly but surely we ironed out the problems, redesigning bits along the way when required until the system was fully active. Not withstanding the pressure of needing the place heated for B&B guests whilst working around an unfamiliar system, the whole process of design, installation and troubleshooting has been strangely satisfying, and to be honest, I think we ended up with a better system than we initially specified. To a large part that would be because the original design didn't work, but if it had worked, what we ended up with is better.

Details of the system itself, problems experienced along the way, and various redesigns are detailed below lest you become as nerdy as us!

The Heating System

The heating system was designed around thermal store technology. In this set-up a huge insulated Akvaterm accumulator tank  containing 1500 litres of water acts as a heat store that can be added to or drawn upon, much like a rechargeable battery.

Heat In
Heat can be added to the store in a number of different ways, depending on the technology chosen. In our system the primary heat source is an Atmos 18 log boiler. The boiler is connected to the accumulator tank by a Laddomat 21 unit; a sophisticated set of valves designed to charge the accumulator tank to its full temperature while saving the boiler. In operation it creates an expanding layer of hot water at the top of the tank, and the production and maintenance of such stratification is key to efficient use of thermal store technology. The manufacturers have a rather nifty PowerPoint presentation which explains the process in more detail.
We had originally planned to run a solar heating circuit (via a heat exchange coil) through the colder strata at the bottom of the accumulator tank to help during the summer months.

Heat Out
Domestic hot water (i.e for the hot taps, DHW) is drawn from a mains-feed and is heated to temperature by being passed through a heat exchange coil in the hot strata at the top of the tank. Water for the radiators is drawn directly from the tank and mixed to the required temperature using return water via a Thermomatic K constant heat regulator.

So far so good. Or not, as the case may be.


  1. Whilst firing the boiler the tank stratified as expected with hot at the top and cold at the bottom, but the minute the heating switched on the whole tank got mixed up, so instead of half a tank at 70'C (useable temperate) we ended up with a full tank at 35'C (unuseable). As a result we ended up with no heating and no DHW.
  2. The change to a pressured DHW system meant that the temperature regulator valve in our shower no longer worked, and some of the joints in the existing pipework started leaking.
  3. The solar circuitry only activates when the panel temperature (on the roof) is higher than the coil temperature (at the bottom of the tank) The de-stratification of the tank meant that the bottom of the tank was warm, which in turn meant that the solar panels were never likely to operate.

When the stratification was breaking down, there seemed very little help to be found from the equipment suppliers. A particular low point (both for us and the GHT) was when we hit the answer “well, storage technology doesn’t really work that well on radiator systems, it is much better for under-floor heating which work at a lower temperature”. Great! Nobody was putting that in their adverts! Anyway, at this point, with no intention of re-plumbing the entire house and with a strong stubborn streak, we set about MAKING it work.

But first I need to acquaint you with the heating circuit plumbing into the storage tank. The hot water is drawn from the top and returns to the bottom. In an ideal world where stratification is maintained, hot water is drawn off at the top, loses its heat in the house, and returns cold to the bottom. If warm water is returned to the bottom of the tank, it messes things up pretty quickly. We figured that the hot water needed as much an opportunity to lose its heat as possible, and the longer it was in the heating circuit, the more heat loss would occur. With this in mind we set the circulation pump on its slowest setting and it instantly improved things. However, the heat loss ultimately depends on the demands being placed on it by the system, and if most of the radiators are turned off, the water may still return hot. There is a further feature to the plumbing; the outflow and return pipes are bridged by a shunt so that the pipework resembles a letter H turned on its side. The shunt is controlled by a ThermOmatic K valve which redirects hot return water through the shunt and back around the heating circuit rather than allowing it into the bottom of the tank. This again minimises the disturbance to the tank strata.

There was a whole set of issues relating to this valve, mainly associated with where to place the temperature probe. The received wisdom was to have the probe stabilising the outflow temperature. We batted this one around until our heads hurt, and, no longer placing much value on received wisdom, tried various positions and settings. We currently have it on the return pipe, set at 55'C, so if the water returns above that temp, it gets sent around the system again. The idea behind this is that when the hotter water is used up, the whole tank still contains serviceable domestic hot water (DHW), which is important to our B&B guests. However, this still left a lot of warm water at the bottom of the tank. To make better use of this, and to further stabilize the stratification, the DHW was re-routed through a heat-exchange coil at the bottom of the tank before it passed through the normal coil at the top. In this way the cold mains-feed water was preheated by the cooler water at the bottom of the tank then topped up by the hot water at the top. This constantly works to re-establish the stratification by taking heat primarily from the bottom of the tank.

Because the solar was getting little opportunity to contribute to the accumulator tank, we decided to change tack and use it to further pre-heat the DHW. This meant a separate dedicated tank (250 litre, I think) for the solar heating. The DHW was sent through a heat exchange at the top of the solar tank before heading off to the accumulator tank. By being first in the pre-heat sequence, and therefore exposed to cold water every time a hot tap is turned on, the heat in the solar tank is quickly removed, and thus the tank is kept relatively cool. This means that the panels have to heat up less before the system is activated, and so even on fairly overcast winter’s days the panels are contributing.

The system as described is fine for when the heating is running (Winter), but we needed to anticipate other situations. For example on hot summers days the solar system may be producing enough hot water on its own, and we would not want the water to be cooled by passing through a cold accumulator tank (a situation termed “Summer A”). Equally, if during the summer months, we fired the boiler to top up the hot water, we would not want the solar water being cooled by the bottom of the tank (“Summer B”). These needs led to much head-scratching and we couldn’t come up with an effective way of automating the three useful permutations: solar only, solar plus top of the accumulator, solar plus bottom and top of accumulator. In the end we put in mechanical valves and will have to manage this ourselves.

Wood boiler and accumulator tank plumbing map

The problem with the shower was resolved by the installation of flow reducing washers in the pipe work, and the leaky joint problem (arguably the most stressful of all) was resolved by waiting for damp patches to appear, and watching in horror as water lapped around the computer on the office floor. The system still loses pressure over time, so I'm not convinced there isn't another leak out there somewhere.

Ah well, it can't all be brilliant!

Update November 2008

As we roll towards our first full winter (which arguably started early September), a few issues are worth adding into the pot.

Atmos Wood Boiler litFirstly, we are changing the boiler, upgrading the 18 kW for a 25kW Atmos. The boiler sizing calculations that we initially used were based on a domestic residence, with fairly static occupation levels, making management predictable and consistent. The problem is that we run a B&B which is heated as required, so the arrival of unexpected guests can mean that we go from unheated to full demand at very short notice. The Atmos 18 kW model is at the bottom end of the rating for a 1500l accumulator tank, meaning that it takes several loads to bring the tank to full temperature, and we were finding that the 18 kW, though adequate in predictable circumstances, could not keep up with these sudden demands which leave the tank needing replenishment. Another added bonus is that the maximum log size is longer for the 25 kW, meaning less handling, and fewer cuts means less wood lost as sawdust.

The Thermomatic K probe is currently on the outflow and set at about 60'C. Having messed around with this in various settings, it seems like on the outflow works best if you have a high tank temperature, whilst on the return seems to make the most of a lower temperature tank situation. I guess you pays your money and you takes your choice!

Another nasty surprise came in the form of our wood fuel stores. The plan was that we hold a years worth of fuel (about 30 cubic metres) on site and we work through it in simple rotation. However, we have recently discovered that much of the oak which has been seasoning undercover for 12 months is still too wet to burn. Fortunately, we have some pine from the same time which is fully seasoned, but this has meant digging into the (mixed) log store in various positions to locate it. Bang goes our simple store rotation, along with any notion that a year is enough to season oak. I am currently logging some large oak trunks which have been cut for at least a couple of years, to the extent that the bark has mostly fallen away and the outer couple of centimeters are rotten in places. Yet when I split these logs, I am constantly sprayed in the face with sap from the sound wood beneath. Irrespective of the felling time, it seems like the seasoning process only really gets going once the logs have been split. In answer to these problems, we have built a further large log-store, taking us towards a two year stock, and are working hard to fill it ASAP. Additionally, I have been making a conscious effort to split logs smaller, and to split smaller logs, giving it all a better chance of seasoning quickly.

Splitting logs for the wood boilerUse old car tyres to help contain split oak logsOn a more positive note, I was given what can only be described as a Top-Tip for splitting wood by a chap working for Coed Cymru. We were discussing the simple satisfaction of splitting logs with a maul, but bemoaning the constant bending and lifting to place logs back onto the block following each stroke. The answer is very simple: stand two old car tyres one on top of the other, drill some holes through the sidewalls and wire together. If your log length is short, stand the tyres on a chopping block, but I have found this not to be required for longer logs. Stack wood for splitting into the tyres with enough space to move laterally when split, but not so much that they can slump sideways. Work around the tyre splitting as you go. This trick is great for splitting smaller diameter logs which would be impossible to balance on a block. On a cautionary note, my favourite splitter is on a hickory handle, and I quickly noted damage to the shaft, presumably caused by outside logs hitting it when trying to split the middle ones. I swapped over to a less effective but fiberglass handled reserve splitter and things have been fine. Now all I need to do is find a replacement fiberglass handle for the best one and I'll be laughing.

Lesser Horseshoe BatAnd finally.... Over the summer, in increasing numbers, some new guests moved in to the open fronted boiler house, and we are now proud to host a small colony of lesser Horseshoe bats. As a result many of our guests have had the chance to get their first good look at these fascinating animals. The bats were probably initially attracted by the moths fluttering around the outside light, but have taken up residence apparently unperturbed by the light going on and off, or the random assortments of wet jackets that have been drying there this summer. They seem to have adapted to the new build very effectively, and dangle from the roofing membrane with apparent ease. I don’t suppose that Horseshoe bats are that rare in Snowdonia, but it seems unusual for them to live so close to an area in constant use, and are so easy to see.

My understanding is that they will probably leave us for a quieter hibernation spot over winter, and we look forward to seeing them again next summer.

Update Spring 2011

Boiler and tank still working fine and problem free. Interest in the project remains high, and we managed to come in as runners-up in the sustainable business category of last years Welsh National Tourism Awards.

The price of wood started to climb steadily around here last year (I guess in line with other heating costs, and more people looking to wood as a cheaper alternative). We were fortunate to find ourselves with a lump of inheritance money to invest (and yet nobody died....the best way!) so we bought ourselves a plot of woodland nearby. This means we are now self-sufficient for wood, and have been able to sell of some surplus.

A couple of things we'd have done differently if we were starting from scratch again:
We run a B&B, so we can be heating the whole house one day, and just our bedroom the next. We tapped directly into the existing plumbing circuits when we installed the boiler, but with hindsight we should have started afresh and made a more modular system where whole loops could be isolated. Although we turn off the radiators in empty rooms, the loss of heat from just the pipe work is still significant.

The system works at a lower temperature than say an oil fired boiler. The radiators therefore need to be over-specced to give effective heating.

Log store - logs and kindling stored for burning next yearThe nature of the boiler system is that it is not ideally suited to radiators (though that is what we have). Given more time and resources, it might have given us a more efficient system if we had got under floor heating on the ground floor. Not easy with stone and concrete floors and low ceilings.

It took us a while to get our wood storage correct, and the problem is that you only get to realise your mistakes in a years time when the wood hasn't seasoned properly. It is a vicious circle: wet wood means that you end up burning more than you should, which means you eat into your stores faster than you planned, which means less time for the wood to season, which means wetter wood. Our mistake was that our stores were too densely packed, so the air did not circulate and the wood just mouldered rather than dried. We now leave about a quarter log length empty between each row, which seems to do the job, although over the years we have had to dig out a couple of chickens who got stuck down the gap. This of course means you need more storage for the same volume of wood.

Our biggest advice to people starting off down the boiler route is to sort your wood out NOW. There is a reason why coal and oil are popular fuels, and that is that they are comparatively dense energy sources. We currently burn just short of 40 cubic meters a year, and that is a big pile of wood. The store that I first thought would be (by eye) plenty big enough for our needs, turned out to be about 15 cubic meters when measured.

Update Autumn 2014

It’s all got a lot less touchy-feely and a lot more form-filly, red-tapey since the last update. I blame the politicians for transforming me from enthusiastic-amateur to what now seems to be professional heating-bore status (with government backing).

After 6 years of trouble-free heating we decided to once again change the boiler. The reasons for this were two-fold: firstly to improve reaction time by fitting a bigger capacity boiler, and secondly to access the Renewable Heat Incentive (RHI) scheme. This second reason feels slightly less noble than some of our other motives along the way, but if the government is prepared to pay an incentive for organisations to burn biomass fuel, then it makes business-sense to be involved as we are doing it anyway. Our old set-up would not qualify for the scheme, hence the need for a new installation.

For the non-domestic RHI at least, you have to have the installation completed before you can apply to join the scheme, a situation that led to one or two sleepless nights worrying that we would end up with an unnecessary installation (and corresponding bill) if the application was rejected. Anyway after a good few rounds of batting the application back and forwards, we were finally accepted in late summer.

Heat-Meter-a source of much joy for the confirmed boiler-nerdAn unexpected bonus on the boiler-nerd front was the requirement to install a heat meter which records the kWh of heat produced by the boiler. To the government, this small piece of kit simply records the heat production on which they pay out, but to the boiler-bore it is the key to a whole new world of graphs and charts. On the plus side it has already led to a quite startling discovery about boiler performance. We were burning fuel from a new source and were quite surprised and not a little disappointed about how little heat we were getting, with the figures down around 65% of what my calculations predicted (yes I know). The boiler manuals suggest that the ash is cleared every day, but our reality has been a little more ‘when I remember.’ When I did remember, I found the ash compartment quite full. Once cleared out, and with the same fuel batch the next day, figures were back up at 98% of predicted. Seemingly we were losing about a third of our heat output because the secondary chamber was clogged! We have become much more methodical about ash now. Undoubtedly there are other variables at play too, so it might not be that simple. For example heat flow (measured between the input and output to the thermal store) will be influenced by the Laddomat valve, which is in turn influenced by the heat of the water in the tank. The system may well be more efficient heating up a cold tank than topping up a warm tank. Watch this space!

In anticipation of future directions regarding RHI fuel supply, we are now registered as Producer-Traders on the Biomass Suppliers List. It seems likely that at some point in the future, those on the RHI scheme will need to obtain their fuel from a BSL supplier, so this was essential if we were to continue supplying our own fuel needs from our own woods. As a bonus, we will be able to sell any surplus to other domestic and commercial RHI participants. In a strange way this regulation might actually be a good thing for the countryside by making effective woodland management a viable business.

Coed Cae Bed & Breakfast on the Mawddach Estuary near Dolgellau, Snowdonia, Wales

Web Design © 2005 - 2017 Hubmaker