When integrating combined heat and power (CHP) into new-build and retrofit projects, the most common approach is to use it in series with boilers. This offers a consistent way for the CHP to see load, working as a preheat for boilers when demand is beyond the CHP’s reach. It can also mitigate the risk of a high return being fed back to the CHP.
However, during certain load conditions, a series connection can remove the boilers’ ability to condense and decrease their efficiency.
We can understand this better with a node-by-node analysis of a typical CHP and boiler set-up, with the CHP sized at approximately 10% of an assumed 400kW peak load (see Figure 1).
Figure 1: CHP in series – full load
This design assumes a flow of 70°C and return of 40°C, which is within the condensing bracket for both boilers and CHP (where the CHP can condense). As the CHP sees a portion of the 400kW load, the 70°C flow from the CHP will increase the boiler return water temperature to 43.3°C at node A.
The smaller temperature difference, or ΔT, reduces the load on the gas boilers, decreasing the firing rate and cutting gas consumption. However, the rise in return water from 40°C to 43.3°C will also affect the boilers’ condensing operation, lowering overall boiler efficiency by approximately 2%.
Using the same baseline conditions as in Figure 1, we can see in Figure 2 an assumed part load of 88kW. This example shows the boilers receiving return water at 55°C generated at node A. The higher temperature restricts the boilers’ ability to fully condense, reducing efficiency by up to 7%.
Figure 2: CHP in series – part load
In some cases, a minor rise in return temperature at full load can be accepted, as the CHP mass flow rate is small compared to the system return volume. During part load, however, the rise could drastically impact the efficiency of the major source of heat.
As a CHP unit only delivers savings when it is running, carrying out a feasibility study and checking the heat and power demand profiles is critical, as is ensuring that heat is fully used without dumping into the atmosphere.
However, for condensing arrangements, the boilers must also be able to operate in condensing mode. This cannot be achieved practically or fully in preheat-style series designs, as a result of return water blending. For this, a modified parallel arrangement would be required.
By introducing this, we enable hydronic separation between different heat generation devices (see Figures 3 and 4), and separate the heat load from generation via a low loss header with flow and return headers attached.
Figure 3: CHP modified parallel arrangement – full load
Figure 4: CHP modified parallel arrangement – part load
Applying the same nodal analysis as in Figures 1 and 2, under the same system conditions, there is no preheat to the return water back to the boilers. Both the CHP and boilers are receiving 40°C return and maximising efficiency via condensing operation.
During load transition, there may be a period where the combined primary flow of the boiler(s) and CHP is more than system load. In the low loss header forward flow then occurs, reducing the boilers’ ΔT initially prior to them modulating. However, the CHP will continue to condense, isolated from the high temperature return. Based on this design, the main advantage is that, at part load conditions, the additional heat required by the boiler to meet system demands is fully condensing, giving maximum efficiency.
It is important to note that the system philosophy must be qualified prior to application on new and existing systems. As a modified parallel arrangement assumes variable flow on both primary and secondary, with close control load ΔT, older, non-condensing systems with retrofitted CHP will benefit more from a series arrangement. Careful analysis of nodes, temperatures and flow rates is needed to fully understand where the CHP will best fit to improve all heat generation efficiency.
We can also explore the benefits of a thermal store to deliver more heat for longer. Both two-port and four-port stores offer a means of releasing heat stored via charge from the CHP. In a four-pipe store, the CHP flow constantly interfaces with the thermal mass in the tank, while a two-pipe store allows the CHP to bypass the tank straight to load.
In both cases, the discharge pump (shown after the thermal store on Figure 4) can control actively if the tank is charging, discharging or in equilibrium. It does this by adjusting flow rate while observing temperatures in the tank as the thermocline moves.
Most manufacturers will recommend a four-pipe configuration, as the heat source always interacts with the thermal store – this works best under steady state loads. It also adds another layer of protection for the CHP and allows for more flexibility with tank pipework. But where applicable, a two-pipe configuration can be used to allow direct online flow for the CHP at the expense of a slightly more constrained installation.
In both examples, this methodology allows for full condensing operation of the CHP and boilers throughout the entire load range without sacrificing the efficiency of either.
This article describes how we approach CHP and boiler design for optimal efficiency, but it is important that the following are considered prior to application: is the overall system condensing? Does the system have a fixed or variable flow rate? Are the boilers condensing or non-condensing? Do the boilers have fixed or variable speed pumps and what controls them?
While there is no ‘one-size-fits-all’ solution with CHP, applying the right methods at the right time will set you up for project success.
About the author
Ryan Kirkwood is specification manager at Remeha. Part two of this series, published in spring 2020, will explore thermal store sizing and controls