*This is the second post in a series on DH. The first can be found here.*

As I mentioned in my previous post, cost of heat on DH schemes is directly tied to system efficiency. The more efficient the system, the less fuel is needed to meet the heat requirements of the customers. And of course the reverse is also true: lower efficiency means higher cost of heat. This relationship between efficiency and cost is hugely important: it’s real cash, coming from residents to pay their heating bills and from the landlord or ESCO to pay the fuel bill.

In fact, I’d go as far as saying that **efficiency is the single most important issue for DH schemes.** This post explains why efficiency matters so much.

At its simplest, DH can be thought of as having one input (gas, for example) and one output (useful heat as metered in the dwellings). In between the fuel input and the heat output is a single number representing efficiency. In practice, this efficiency depends on a range of factors including level of demand on the system, flow and return temps, control strategy, type of plant, etc. But ignore all this for a second and just imagine we’ve got a black box with a single efficiency number associated with it. Distilled to this level, you can define the cost of heat in the following way:

*cost of heat = cost of fuel / efficiency of overall system*

That’s a very simple way of thinking about heat cost.

Now let’s divide our one black box (i.e. the efficiency of the overall system) into two boxes: plant and network. The plant that produces the heat – boilers, CHP, etc. – has an associated efficiency (again, in practice this efficiency will vary depending on a range of factors but let’s distil this down to one number for now). And the distribution network (the pipes) will have losses. So instead of the single efficiency figure in the equation above, we’ve now got two efficiency figures, and our formula looks like this:

*cost of heat = cost of fuel / (plant efficiency x (1 – network losses))
*

The reason for breaking our original single black box into two parts like this is because we want to focus on the network efficiency, which is the tricky bit. For plant efficiencies you can refer to CHP manufacturers’ information or boiler databases for example. But there’s no such source of information for networks – each one is unique.

Here’s a worked example, keeping plant as simple as possible. Let’s assume we’ve got a central gas boiler with an efficiency of 88%. If we assume that gas costs us 3.5p per kWh, then the price of heat *as it exits the boiler* is 3.98p per kWh (3.5p / 88%).

But that price isn’t particularly useful to us since we’ve got to get the heat from the boiler to the flats via the network. The amount of heat lost from the water as we move it across the network has a big impact on the cost per kWh of the delivered heat (all that lost heat still has to be paid for since it cost money to generate in the first place). The way the cost of heat varies with network losses is shown in the graph above.

You can see that the cost of heat increases with the network losses, but not in a linear way. At first the price increases on a fairly shallow line, then gradually becomes steeper at higher losses. The difference between 20% and 60% losses is enough to double the cost of heat to the resident! This is a massive impact, especially at a time when fuel prices are rising.

Hopefully this is enough to convince you that **efficiency should be the number one priority of anyone providing heat over a DH network**. In the upcoming posts I’ll focus quite a bit on how systems end up with poor efficiency or how to achieve the highest efficiencies we can.

Next post: Getting it wrong part 1 – Design

## Leave a Reply