In 2007, with dairy farmers struggling in times of fluctuating milk prices, Kemble Farm owner, Colin Rank, struck upon the idea of building an anaerobic digester to power the farm from the readily available slurry and provide himself with a secondary income. With the bit between his teeth, he hopped over to Germany to see how small-scale systems were being run over there, sold a couple of houses on his land – the farm being asset rich, and cash poor – got a 32 per cent grant from the Department of Energy and Climate Change (DECC) and, within a year, had a brand new anaerobic digestion (AD) plant running on the farm.

Seven years later, the farm is now owned by the Bathurst Estate, there’s a dedicated manager for the AD plant, and it’s become a place others flock to, to see how it’s done.

I’m shown around the plant by AD Manager Miles Bishop and, speaking to him, you’d think he’d been studying the technology for years. It turns out he’s learned on the job, after years in farming generally and running the grain stores, to which he “draw[s] parallels with the biogas plant because it’s all about moving things around”. He admits, however, it hasn’t all been plain sailing: “I won’t lie, it’s been a steep learning curve. If there’s a silver lining on break downs it’s that the more you have, the more you take things apart and fix them, the more you understand how it works and what the relationship is between various parts.”

Bishop starts each day with a walk around the plant to make sure there’s nothing amiss. “If there’s a rotten egg smell, chances are there’s a gas leak. So look, listen and smell.” And while you can definitely smell the slurry, which makes up the bulk of the feed (around 50 tonnes a day from 650 cows!) it isn’t the only ingredient. Five tonnes of farm-grown maize and one tonne of ‘imported’ high-energy glycerol are also added, to increase energy outputs.

The tour kicks off with the maize hopper, something that has become a nemesis to the plant’s operator. It gets filled twice a day with around about 2.5 tonnes. Every hour, it spills maize into a sub hopper where it is augured into the digester. As Bishop explains, the solid/liquid interface can give you problems if the liquid bursts back into the hopper. It can be a time-consuming part of Bishop’s day when it does go wrong. “Over the last six months, this is something I have done a lot of work on because if it fills up with liquid and trips out the level sensor you have to take the end plate off and dig the whole sorry mess out and start again. It got to be quite regular and really quite annoying, especially at the weekends!”

Not to be outdone by a machine, he seems to have cracked it with a few modifications to the timing and structure: “I’ve kept the door permanently open, changed the programme so the mixer only comes on when it wants to feed the hopper, and modified the knife sections on the central auger by welding an angle on there to act as a push out bar, and that helps to push the maize out in the sub hopper.”

Of course, it’s all a learning process, and since working on it full time, Bishop says he’s found that the secret is consistency: “It is a living organism in effect, a tin cow if you will, because it works in a similar way to a cow and when things are stopping and starting it is never very good for it. However, if you can have the same amount of feedstock going in at regular intervals, well, I know it might sound daft, but the bacteria become accustomed to it and they seem to thrive.”

Left: The exhaust stack, AKA the ‘giant patio heater’ is used to disperse excess gas
Right: The farm-grown maize is fed into the hopper at the start of the process

The mix stays inside the digester for around 30 days, coming out in a continuous, as opposed to batch, feed. For every 50 tonnes in, around 48 tonnes of digestate comes out. The farm used to spread the slurry straight to land, but the digestate from AD – both the liquid and the solid stuff – is of a superior quality: “The digestion process improves the nitrogen availability and it’s slightly less smelly. People will say it doesn’t smell at all after digestion, which is not strictly true, because it still has a nose to it, but it is less offensive.”

Following the pipes on from the digester on our walk-around, we get to the huge spherical cylinder, which holds the gas. If the plant stopped producing gas all of a sudden, the cylinder would hold enough to run for about two hours. “The holder just acts a sort of buffer between where the gas is produced in the digester and where it is used in the CHP with the exhaust stack”, explains Bishop.

If for some reason, however, there is a surge in the amount of gas produced there are other options: “You can’t switch the digester off because it is a living thing, and you have to do something with the gas, but you can tail gas production off by reducing the feedstock, though in the short term, you’ll just flare it off in what is essentially a giant patio heater!”

One of the problems most farm AD plants have, especially with slurry as a feedstock, is the production of hydrogen sulphide, which results from the breakdown of organic matter in the absence of oxygen and has that characteristic rotten egg smell. As hydrogen sulphide is a corrosive and poisonous gas, ferrous chloride is injected into the digester slurry, which reacts to form less harmful iron sulphide salt. The gas also undergoes carbon filtration through four stainless steel cubes just before it is fed into the combined heat and power (CHP) engine.

This engine is found at the centre of the plant, turning the biogas into heat and electricity. The heat is used in the dairy unit, to maintain the digester’s temperature, thereby ensuring optimum viscosity (if it gets too cold it gets very thick and too hard to pump) and to heat the water in the three herdsmen houses at the top of the road. Because transporting heat is very costly and there isn’t another use for it on the farm (like greenhouses or chicken sheds), the rest of the heat is simply vented into the atmosphere.

The plant was limited to producing 300 kilowatts (kW) of electricity by its locality to grid connections – more power would require installing three miles of additional, “hideously expensive” cabling. “Of the electricity we produce, we use about a third on site and export the rest”, explains Bishop. For the plant’s first full year, the farm received £72 per megawatt (MW) and in 2010 it received £35/MW for the electricity it exported. This price takes into consideration the double Renewables Obligation Certificates (ROCs) the plant receives. However, if it was built now, the subsidy would be unavailable. And since the government’s decision to limit double ROCs to projects over 5MW, Bishop suggests farms are out of the equation as 5MW is “beyond the scale of most farm ADs”. With government recently deciding not to review feed-in tariffs for small-scale AD (despite promises to the contrary), Defra’s celebrated £3-million loan scheme for on-farm AD (announced last year and administered by WRAP) might not find many takers…

After the walk around, Bishop takes his daily manual log readings, on top of all the automatic data logging: “Everything I can record I do record, which is an invaluable source of data not only for seeing how well you are performing, but also how much you’ve exported, how many tonnes of maize you’ve used, how many metres of slurry you’ve used. In that way, you can form relationships between what you’re putting in and what’s coming out. If it’s being done automatically you don’t really think about it and you might not notice something, but [you will] if you’re writing it down – you’re having to think about it.”

Then there’s routine maintenance to do, like cleaning out the steel screen inside the digestate separator. “That clogs up with limescale so that has to come out, and that’s quite a laborious and cumbersome task… Then there’s changing the carbon in the carbon filters, maintenance on the engine, the CHP, oil changes, the list goes on.”

In 2008, it was estimated that it would take seven years until the digester was making money, and despite a few hiccups – “partly down to not knowing enough about the plant, partly down to a few little design features, and a lot down to not getting enough attention” – it’s still on track for 2015. “Now I’m here full time, we’ve really improved the generation figures… Many plant builders will tell you that it’s totally automated, but in practice, like many things, it never quite works out like that… It still does things that I cannot quite provide an explanation for. One day we’ll have a sudden surge of gas production, even though nothing has changed in terms of input. Another day the gas production can fall off, but again, nothing’s changed. Because it’s a complicated mix of biological and mechanical, it’s a bit of a dark art.”

It’s no surprise then that other local farms have got onboard, with at least three more in the area adding AD facilities, despite the scale and complexity of the technology. No doubt this is in part due to Bishop’s willingness to be a sounding board: “I give people the tour and tell them the pitfalls. You can ring the plant manufacturers or the experts and they’ll say, ‘Oh, that shouldn’t happen’, but speak to people who are actually running the plants and you get the real feel for what’s going on and how you can improve things. I’ll tell people things you can’t get from a handbook but you can only get from getting your hands dirty.”