DIY Guide to
Ground Source Heating
Why you should read this before doing the ground works
This is more of a diary of one couple’s adventure with ground source heating than an authoritative manual, but it is written in the hope that you will avoid some of my mistakes and consider some of my methods to make your heating system as efficient as it can be.
We placed an order with Ice Energy for a heat pump and received a 92 page manual for the installer of which 3 pages related to the ground works for our project.
I was hoping to do most of the work myself so wanted to plan the following steps (some I found out I should have planned!)
Hire of a digger – how long?
Purchase and delivery of sand
Purchase of other items listed by Ice Energy as essential to the project.
Planning the pipe route
Planning the digging route
Planning the pipe testing
Getting access for the digger
Technique for digging required trench
Technique for laying pipe
Technique for application of sand
Time and Cost
Preparing the manifold
Restoring the garden
Whatever your background and skills I feel sure you will be interested in considering some of these points before you start.
We purchased Marchwood because it has lovely views and great potential for renovation. We wanted to find a way to document the difficulties of an average family making an ordinary house zero carbon. By zero carbon we mean that if everyone did the same UK could achieve its 2050 targets for virtually zero residential carbon emissions. This definition does rely on the government getting its act together and producing all electricity from renewable sources.
We did not go the route of micro generation because it can only form a very small part of the nation’s energy needs and in our view is a relatively expensive dead end.
We did not go the pellet boiler route for two reasons. Firstly we understand there is not enough wood in the UK to support millions of pellet boilers. Secondly there may in the end be a ban on burning anything which would make the boiler obsolete.
Ground source heating is not cheap but it is a viable way of harvesting the sun’s energy stored in the ground each year. It is well tried in Scandinavia and Germany and well suited to many parts of the UK. Ice Energy have installed 2000 heat pumps in the UK but it is still little understood by the general public, local builders, plumbers and, I suspect, by government.
The real battle is for the efficient generation and use of energy. I find that salesmen and engineers alike are quite vague and cautious when you ask how efficient your heat pump will be. The standard Coefficient of Performance (COP) depends on the efficiency of the ground loop and the efficiency of the distribution system in the house I am told.
I want to initiate COP wars in which heat pump users vie with each other for the highest efficiencies and we all learn how to do things better.
As result I have incorporated some ideas of my own which you will read about later if you wish.
Part of planning is to make sure you are physically capable of undertaking the task. The 200 metre coils are quite heavy. The delivery driver passed me the 100 metre coil to carry to the back of the house but delivered both the 200 metre coils himself. After he had gone I checked the weight and found I could roll them along level ground with reasonable ease. This proved to be the most demanding part of the job so the rest was possible.
The other factor is the number of working hours in the day and the week. When hiring a digger at £200 per week your schedule and fitness must allow for 120 hours work in about 2 to 3 weeks. Driving the digger is no more demanding than driving a car but rather more interesting. I did not find any inclination to go to sleep at the wheel! Barrowing sand and fitting the pipe is more demanding. As you start doing it you may find you can benefit from some help with that, as I did.
Most texts are written by experts and assume more knowledge than I have, and this is no exception. We have a small garden 26 metres by 13 metres and I had drawn up a plan during my discussions with Ice Energy. Their surveyor then attended the property and said I had done his work for him.
He confirmed my plan, answered a few questions I had, and went away.
I then read the Pump Installation Manual and started to understand what was needed.
The diagram showed a pattern of loops in a straight trench with a return pipe along the other side of the trench attached with cable ties to each loop. The result looked so neat in the photograph and a real piece of cake!
I paid the next instalment of the purchase price and the first surprise arrived.
The 2 coils of pipe 200 metres each long were much heavier than I expected and much less flexible. I was later to learn the significance of this. The third coil was for the connections.
Planning the Route
I had done quite a lot of research to find out whether slinky collectors were suitable for my sized garden. At the Buildbase Centre in Swindon an expert recommended an air source heat pump instead because it would be cheaper to install and my garden barely had the area needed. He said I should not be drawing more that 40 watts per square metre from the ground or the sun in the Summer would not be able to replenish the heat I had used.
This came as a surprise to me because I thought the heat came from the ground. Even with a borehole the heat all has to be replenished by the sun.
I submitted one or two designs to Ice Energy and their surveyor eventually approved the one illustrated with 200 metres of pipe in 80 metres of trench with the trenches 2 metres apart.
Studying the manual again I became aware of what would be happening under the ground. There were warnings to insulate the cold pipe running from the heat pump very well so that it did not cool the warmer water returning from the ground. This would be giving away the precious heat just before it entered the building safely.
As you can see my plan is based on 2 U shaped trenches each containing 200 metres of pipe. Under these circumstances it appeared we could do better than the Ice Energy model as in that the return pipe is attached to each outgoing loop and must loose a little heat at each contact point. I asked Ice Energy whether I could return from the other end of the U trench directly and they approved.
I did not straight away realise that this would mean the loops in the trench would be closer together as the way back was shorter.
I planned a start point near the house on the South trench so that spoil could be stacked beside it leaving a narrow path for the barrowing. There would be less space at the corners so spoil would be stacked on the farmer’s field margin at the back of the property. Without this extra space it would have been more difficult but not impossible..
Most of my planning worked out well and I was blessed that the mistakes I made did not prove critical. For example, I had not noticed that the garden tapered by 1 metre towards the house!
I found I could dig between a bush and a young copper beech tree without encountering large roots so this lengthened my North circuit to compensate for other losses. I also adjusted the plan somewhat to make the trenches simpler nearer to the house. It is better to dig as much as possible moving backwards over firm ground and then digging the connecting trenches from the side to a lesser depth. I have a test point to make sure the temperature is not lowered significantly by an insulated pipe from the heat pump going through.
This illustrates what happened in fact.
Buying the Sand
40 tons of sand seemed a lot to have to barrow from the front garden to the back so I asked if it was essential to have 100 mm under the pipe as well as 100mm over the pipe. The Ice Energy engineer explained that the sand was not to protect the pipe but to ensure a good contact between soil and pipe without any voids. Voids he explained could cause lower efficiency and could cause the ground to heave. The most important step was to use a lot of water when covering the pipe to ensure the best contact.
I purchased 20 tons of sand, allowing for 100mm over the pipe, and resolved to buy more if I ran short.
I got 2 quotes locally. One was from a local coal, sand and gravel merchant and the other from H. R. Smith, an aggregate supplier, so I could compare these with the Wickes bag price. The prices were £500, £396 and £750 and a local farmer later confirmed that I had made the right choice. Even so, sand is not to be wasted!
Delivery from a 20 ton truck proved difficult. I had planned the width into the drive OK but I had forgotten that it needed about 7 metres height to tip its load. The driver managed after 2 attempts without any damage to trees and telephone wires!
I did find out later that a cheaper grade of sand than builder’s sand is available but I’m not sure if it can be used.
Hire of a Digger – How long?
My only previous experience of digger hire was a couple of days to reshape a garden. I thought a week would be sufficient. I booked a week and then found that according to my favourite weather forecast, Windguru, about an inch of rain would fall in the first 5 days. This looked too much to cope with. I could imagine sliding on the mud into the ditch I had dug, and wondered how much it cost to recover a digger from a hole! More about that later.
The week proved glorious, with very little rain, so the second week I went ahead even though the forecast was bad. The machine was a 1 ton Kubota U15-3 from Greenplant in Wheatley and cost £196 plus VAT for the week. It has the advantage of shrinking its tracks to 965mm for going down our 1 metre sideway but I wanted to get it delivered to the back garden first as I had gained permission from the farmer behind.
The digger arrived and we had the choice of a steep bank from a ditch into the field behind or a chug round the field margin fro a quarter of a mile away.
The delivery guy was most helpful and tried the steep bank for me. It was about 1 in 1 and he tried it backwards with the bucket lowered behind him. It got most of the way up but just failed at the crest. So he loaded the digger on the trailer again and took it to the field gate.. There he unloaded, took the trailer over the rough entrance, and then reloaded for the trip round the field, a final unload, and a brief demonstration. What helpful service!
The only obstruction underground I expected was the oil pipe from the tank to the house. I expected this to run direct to the house path and then along to the kitchen entrance.
I changed the bucket to the narrowest one and dug a communication ditch from the position of the manhole for the manifold to the nearest slinky trench.
Wow! Almost the second scoop I hit the oil pipe which went diagonally in a curve from the end of the tank to the entry point for the house. Thankfully it bent but did not sever!
Starting again a bit further along the line I was going great guns when the rain came!
It rained most of the next 3 days, and I also noticed that, in my enthusiasm I had dug the narrow communication trench too long and needed to widen it. Progress was so slow that I wondered whether it would take more than a month.
Choice of Bucket
I changed to the next size of bucket and made better progress. I often found the clay soil did not fall out of the bucket onto the spoil and I had to get down and clear it with a spade. At last the first of the 4 trenches was completed in about a week.
Later I changed to the largest digging bucket (600mm one with claws on) and made even better progress doing a metre every half hour. The soil seemed not to stick in the larger bucket
At first it was difficult to maintain the width of the trench to the metre required at the bottom. The corners tended to round and the trench narrow without constant reworking and checking. Then I started using the skew foot control to offset the arm to the right or left depending on the side I was excavating. This means the scoop automatically excavates right into the corner at the bottom of the trench and reworking is not needed.
Planning the spoil is an art too as you have to have access for a barrow of sand either to right or to the left of each trench. It is often necessary to push the top off the spoil pile as the bucket releases the next scoop full. Quite fun!
Dick came across the road and gave me a hand. As a farmer he has had a lot of digger experience, although the controls on the Kuboto work the opposite way to a JCB. He surprised me by backfilling the trenches very well with the small scraper blade on the front of the Kuboto.
He also explained that sinking in a trench should not be a problem as you can prop the digger off the ground by forcing the bucket down and then add spoil under the tracks to enable the digger to climb out. I only had to do this once but it works very well.
The placing of the spoil from the trenches must be planned to ensure access to the trench with barrows of sand, ease of backfilling and ensuring the digger is not encircled by trenches when you need it elsewhere, perhaps to load the sand onto the barrow. Particularly near the house this can be difficult where trenches link together.
If you can place the topsoil separately from the subsoil then this will help to restore the garden to its former glory. If only one spoil row is possible then why not place the topsoil further from the trench and the subsoil nearer. In this way you can use the scraper to backfill the base of the trench and place the topsoil last with the bucket.
Don’t worry too much as a good rotavation afterwards should spread enough top soil from the 2 metres between the trenches across the trenched areas. (This proved rather optimistic as you will read!)
Technique for laying the pipe
Don’t be fooled by the pictures. That pipe is stiffer and thicker in relation to a 1 metre trench and can put up a real struggle. It is heavy to handle and, like its namesake slinky toy, once it gets tangled it is hard to fathom how to untangle it.
My first attempt was rather poor. I rather lost control of the slinky in the first trench. I tried to follow the instructions not to fully unroll the coil but even so it does not uncoil very evenly into the trench.
Take great care not to kink the pipe or set up tensions that may kink later under the ground. If a kink should occur Ice Energy say you must cut out the damaged pipe and make a proper join. Fortunately I avoided this. I discovered that you can use the spring in the pipe to anchor it in place by cutting into the side of the trench at an angle below where the loop touches it. The spring in the pipe then secures it to the floor of the trench.
I allowed too much space between some of the coils and then had to double up on others to use all of the pipe by the end of the U shaped trench.
Turning the corners was difficult and I put the pipe in whichever way it would go without stressing it too much. This meant figures of eight and long overlapping loops where necessary.
For the second U trench I was more methodical. I dragged the coil of pipe to the field at the end of the garden where the curve of the U will be. I had noticed that trouble starts from the ends so I tied the end in the middle of the coil to a spade in the ground and fed the loops out of the centre of the coil without unrolling so that they stood in a line on the field.
I secured the last 4 loops together so that the other end would not cause trouble.
There were 60 loops so I planned 28 for the first leg of the U and marked the 28th loop.
I gathered the 28 loops back into a coil carefully ensuring that they remained in sequence and untangled and tied them with soft rope for ease of handling. The 28th loop I pegged down with a fork.
I could then heave the bunch of loops into the trench, release a few from the rope, and move down the trench paying them out one by one.
I found it was better to arrange the loops on their side in the trench ready to be flattened and fitted later.
Having got to the end they needed to be evened out. Care was taken that the end was long enough to reach the manifold position with a metre to spare.
Then comes the fitting. I found this easier without sand in the hole. It is easier with some help, but if alone, roughly fit the coils in the trench using feet to hold them down, tying a few of the cross-overs with cable ties and undercutting the sides where necessary.
Once the length is in place go back and improve the fit to the floor of the trench by removing bumps in the floor and further undercutting at the sides. Use ties on all cross-overs to make the whole very secure.
Technique for applying the sand
The sand is to improve the contact between the pipe and the ground so I expect it to be important in getting the best efficiency from the system. It is also expensive and heavy to barrow so you don’t want to use more than you need.
I applied the sand over the pipe in sufficient quantity to tread down all round the pipe and fully hide it. This usually meant the middle space was covered but not always to the same depth.
It was difficult to divide the sand piles into 4 evenly so I found that by the time I had reached the 4th trench I had too much sand over. I used this both under and over the pipe.
I wanted to use the digger to load the sand barrows so I reduced the width of the tracks so that it would go down the sideway to the front garden. Then very gingerly, with about 2cms each side, I crawled slowly through. A branch of a tree I had not noticed was the only casualty that time! Later with a bit more confidence I managed to damage the neighbour’s fence.
Adjusting the width of the tracks is best done with the digger suspended between its scraper at the front and bucket at the back. It then widens and narrows with ease.
Barrowing the sand is hard work and at 74 I want to use the most energy efficient method. Firstly keep the sand as dry as possible as wet sand seems twice the weight. Secondly don’t over fill the barrow. If you have to struggle it is probably too full. I manage a builder’s barrow about ¾ full.
Wheel the barrow with the handles at arms length by just leaning forwards. If it does not roll easily forward then you have over filled. You will wheel many times over the same track so it is worth while to fill any dips and clear any bumps.
When near the trench it is better for the barrow to fall in than for you to follow it. Let go if in doubt.
It sounds obvious but following these tips makes it more fun anyway.
When pressing the sand around the pipes you may get some help from the grandchildren if you can agree the price!
Technique for Backfilling
Before backfilling use the trench for running armoured cable you will need for garden lighting, pumps for rainwater use, garden water connections and anything else you can think of.
The backfilling method will depend on the location of the spoil piles. First backfill any trenches you will need to cross to reach the main spoil piles.
The first 15 to 30 cms of fill should be watered and trampled to make a really muddy base to give good contact with the pipe. Once more than 30 cms has been put in the trench it may be difficult to get the mud trampled well right down to the sand. After this back fill the rest of the trench. Working in 3 metre sections seemed to work well for me single handed.
The scraper blade of the digger is good for this but the first 30 cms is probably easier done with the shovel of the digger. Fill the trench to slightly above the original level and use the weight of the digger to roll it down. Don’t get the surface soil too wet or this will hamper the work.
If the digger gets stuck or sinks too far into a hole then it is usually easy to get out by lifting one end or both with the scraper and the bucket so that spoil can be shovelled into the space.
During backfilling I introduced a couple of variants of my own. I sank 7 lengths of rainwater pipe at various points so that I could take temperature readings under the ground later. I purchased a digital thermometer with a 1.5 metre probe for £6.50 so that I could do this easily. These pipes will be decorated with solar lights to conceal their purpose.
The rainwater pipes could also be useful to enable the trench to be flooded should additional thermal contact with the pipe be needed during operation. Hopefully the hard packed floor of the trench would carry a proportion of the water along the coiled slinky pipe.
I also buried 50 metres of garden hose in each trench with the beginning and end at a strategic point in the garden. I plan to use these loops to circulate water underground to warm the soil in the Summer. They may work as part of a solar water feature.
Having completed the backfilling I took a further 4 hours of digger time to landscape the garden. I lowered a trampoline into the ground to make it less obtrusive and used the excavated soil to build a crescent shaped bank to hide the trampoline and attract the eye from the fence into the view beyond.
A pipe tester seemed difficult to hire. I engaged a local plumber to come with his tester and he asked for a couple of days notice. When I called him with half a loop buried for a test he said his tester was now broken and he was not expecting to replace it. He said that testing with the mains at 3 bar was almost as good.
So I filled and pressured the pipe as described in the Ice Energy instructions letting the water flow for 10 minutes to get rid of the air in the coils. It certainly held pressure for an hour or so. When released a jet of water issued for 20 seconds or so. This indicated to me that there was still air trapped in the coils so I contacted Ice Energy who said their installer would deal with that when commissioning.
I wanted to test the way Ice Energy had recommended but I had difficulty finding a hire place who could supply a pressure test kit. I have eventually found Johnson Buildbase in Cowley who have recovered theirs from one of their branches and have reserved it for me.
This proved an interesting experience with no instructions except some diagrams on the side of the appliance and the Ice Energy requirement that I should test at 4 bar. I filled the tester with water, freshly filled the pipe loop from the mains, and attached the tester. I closed valve 2 which directed the water pumped back into the tank and pumped away for a few minutes. The gauge did not register any change in pressure. Another few minutes pumping and the gauge started to move from the end stop.
Pumping seemed to get harder but to make more difference as the pressure steadily rose. When the gauge showed 2 atmospheres pumping became quite hard and I stopped, closed valve 1 and recorded the pressures every 5 minutes as instructed by Ice Energy.
Then I looked at the gauge more carefully. It was marked in 3 units of pressure, one of which was psi or pounds per square inch. I had reached about 300 psi which I realised was 20 atmospheres! I had tested to 5 times the pressure required by accident! I was very thankful the pipe had not burst! There was no indication on the gauge that the 1 bar line meant 10 bar. Beware!
Preparing the Manifold
The instructions say I have to purchase the components for the manifold but do not give a list. However in the box of bits I have 6 gate valves and junctions which look like suitable components. However they do not seem to be complete enough to make 2 manifolds. I will phone Ice Energy next week.
Meanwhile I have purchased 30 engineering bricks, cement and a manhole cover.
Ice Energy helpline have now emailed a schematic for assembly of the manifolds from the components they delivered and putting them together was easy. Two surplus gate valves had confused me.
When the manifolds were assembled it was clear that the output pipe was at right angles to the input pipes which was an alteration from the pictures in the manual. No problem but helpful to know early on.
Restoring the Garden
I hired a rotavator for the day for £38 plus VAT and set about breaking up the packed earth left by the digger operations. I expected to find myself driving a lawnmower type beast and was quite surprised. The machine was very heavy and my first attempt was quite unsuccessful.
I was shown how to start up and how to engage the forward drive.
A dead man’s hand safety device cuts the motor out if you let go of the left side handle.
I tried this in the back garden myself. Started the engine, engaged forward drive, engaged the blades and gently released the clutch.
The rotavator leaped forward hardly touching the ground and left me running to keep up. Very fast but the results were minimal! I couple of high speed runs and I had to stop and think.
It turned out that the forward drive was almost engaged but not quite. It could be coaxed in with a click just as the clutch started to bite. Then we moved off at a controlled pace and the depth of the dig could be regulated by more or less weight on the handles.
The ground was much harder than I expected, especially in some patches, and after I had tilled an inch or three I was amazed at the proportion of stones. Where was the top soil? Where even was the clay I had experienced during the past 2 years of gardening? All I had left was sub-soil and stones.
The rotavator completed its work in about 3 hours with only one mishap. On one section I noticed my spirit level close to the line of tilling.
As one might with a mower, I guided the rotavator with my left hand and reached down with my right to pick up the level. Just then the rotavator hit a stone and bounced sideways. The right handle hit the side of my head quite a whack.
I let go of everything and the machine came to a halt. I was dazed for a moment and wondered why. Then I noticed my glasses had gone!
I found them intact, for which I was very grateful, but I must get them re-adjusted at SpecSavers as they are not the same as before.
So nothing more than a small bruise for my adventure, but I will remember to keep both hands firmly on the handles every second of the job in future. It’s rather strange that the machine is plastered with safety does and don’ts but there is no hint of this danger.
The next job was to prepare for some grass seed. It is 10th May so if I can get some seed down before I go away it might germinate before the hot weather arrives.
I tried to rake the stones aside but only raised more from below. Some were the size of eggs as you can see.
The most successful method seemed to be to use a shovel to scrape the stones from the surface very lightly leaving the soil and small stones behind. Once a small pile of the larger stones was gathered they could be shovelled into the wheel barrow.
Some top soil from the front garden was added where there was no soil evident. The results were poor but I scattered the seed in hope. I may have to order a load of topsoil when the main building work is done anyway.
Time and Cost
The work so far took much longer than I had expected. I booked the digger for one week and have just returned it after 16 days. I worked 118 hours in total and have still the testing and manifold to do. It was great fun and I feel much fitter than for a long time. I had 15 hours of help for which I paid £210. This was mainly to keep the digger working through some of my meal breaks when I was concerned at the time the work was taking.
The total of 153 hours broke down approximately into 40 hours digging, 20 hours pipe fitting, 37 hours barrowing sand, 32 hours back filling and 24 hours landscaping.
Best of all I have the satisfaction of knowing that corners were not cut and I have the most efficient installation possible in my garden with current understanding of the physics. Hopefully my measurements will enable an even more efficient installation in the future.
Pressure Test Kit
Rainwater Downpipes Wickes 5
Garden Hose 100 mts
B&Q Luxury Grass Seed 20kg
Hope your job goes well!