Heat recovery ventilation (HRV) is one of the usual techniques (next to demand controlled) to reduce the energy impact of ventilation in buildings. For a given air change rate, the energy savings of HRV are in the first place dependent of the heat-exchanger efficiency, usually measured in standardized laboratory conditions. However, many other factors can have an impact on the overall system performance in practice. Through three different projects in the last years, BBRI had the opportunity to monitor about 15 systems during several months, allowing to (try to) evaluate their performance in real operation.
Before going into the details of the measurement campaigns, this paper gives a short overview of the various factor that can negatively affect the actual energy recovery compared to what could be expected from the heat-exchanger efficiency only. This overview will help to interpret and explain the monitoring results.
What concerns the monitoring campaigns, the temperature in the 4 flows of the heat exchangers were measured with a few minutes time step during several months. The frequency and the duration of the measurement allowed to compute the supply and exhaust efficiencies and derive statistics (mean efficiency, …), but also to observe more localized events like the triggering of frost-protection systems. In the two first campaigns, these temperatures were measured directly within the ventilation groups at the vicinity of the heat exchanger so that the measurements were not impacted by the heat released by the fans. For the last measurement campaign, the temperatures were measured in the ducts just before and after the ventilation unit. The outcomes of this measurement campaign are twofold:
- For most installations, the average efficiencies of the ventilation units (as computed in the Belgian EP calculation) are between 70% and 90%. These are quite in line with the declared product values (as published on www.epbd.be).
- Some of the results in the first two campaigns showed that a non-negligible temperature heterogeneity exists in some of the fluxes leading to biased evaluation of the efficiency. Furthermore, the monitored data of the third campaign showed less noise of unexplainable variations than in the first two campaigns. Even if the sample was quite reduced for this last campaign (5 installations), the trend seems quite clear and we would thus recommend this measurement method for future measurement campaigns.