Establishment of Leakage Rates of Mobile Air Conditioners in Buses and Coaches

January 2007
Prepared for the European Commissions (DG Environment) ENV.C.1/SER/2005/0091r
Winfried Schwarz

On behalf of the European Commission, this study establishes empirically the annual leakage rate of mobile air conditioners (MACs) in buses and coaches for the use-phase of the vehicles. In buses, refills are carried out in relatively short service intervals to compensate for leakages whatever their nature. Such refills are recorded over a sufficiently long time and in appropriate detail in Sweden where annual inspection is mandatory for every installation with a refrigerant charge of HFCs more than 3 kg. The refill-based approach of this study consists in the analysis of approximately 2,000 report forms on inspections of mobile air conditioners (MACs) of buses and coaches in a specialised garage in South-East Sweden over the 1996-2005 period. 

Equating refills with emissions, the following conclusions on the annual leakage rate can be drawn.

Over ten years, a total of 1,267 annual inspections were carried out on 309 different MACs in buses. For the ten-year period taken as a whole, the leakage rate per annually checked bus MAC averages 2.34 kg ± 0.26 kg. This leakage rate as a percentage of the full refrigerant charge amounts to 23%. These values still disregard any differences in type or design of the buses, and particularly in time of MAC installation. 

The most important results of the analysis are:

Further analysis reveals significantly higher leakage rates for articulated compared to single buses, and for gas compared to diesel driven buses. There were differences in leakage rate between particular bus and MAC makes, however, they depend on differences in leak tightness by bus design and fuel type. 

The analysis of age dependency splits the bus sample into three categories. MACs installed between1995-1999 show 2.1 kg/a within their first five inspection years and 2.9 kg/a in the next five years. New MACs, installed ex 2000 exhibit a leakage rate of only 0.9 kg/a over the 2001-2005 inspection period. These three figures indicate a double age-dependency. The leakage rates grow as the MAC systems age, and modern MACs are less leaky than older ones. This evidence might lend some support to the argument of MAC makers that there have been substantial improvements in leak tightness due to replacement of flexible hoses and to better connections. 

As regards coaches, the average 1996-2005 leakage rate of the 308 annual inspections on 115 vehicles is 1.5 kg/a. This low value can be explained by the fact that all coaches are diesel-driven solo vehicles. The corresponding figure of MACs in solo diesel buses is of the same magnitude (1.25 kg/a). 

A closer look at the data reveals differences between MAC types. The integrated split MACs show almost three times higher leakage rates than rooftop MACs. Higher leakage rates by makes of coaches were also found; those with higher leakage rates had higher shares of integrated split MACs. 

In bus MACs like coach MACs there is an increase in the annual leakage rate as the MAC systems age. In contrast, the leak rate in new coach MACs (av. leakage rate 1.2 kg/a) in their first five years of operation is similar to the leak rate of older coach systems. 

It can be concluded from a statistical analysis of the recorded refill data that the average leakage rate of new MACs in diesel driven coaches is of the order of 1 kg/a (1.20 ± 0.74 kg/a) and is of the same magnitude as leak rates from MACs of new buses with diesel drive, with 0.92 ± 0.40 kg. The percentage leakage rates are 13.3% and 13.7%, respectively. 

One of the most important results of this study is that it is possible to reduce the annual leakage rates of modern bus and coach MACs to approximately 1 kg, equivalent to 13% of the full refrigerant charge. These values enable the European Commission to make projections of bus specific emission levels of HFC-134a beyond 2010 if no additional policies and measures are undertaken. 

This report is structured as follows: Chapter I explains the "refill-based" approach of this study, and Chapter II presents the selection criteria for an EU representative survey that the sample should fulfil, pursuant to the specifications set by the EC. Chapter III describes the Swedish bus sample and discusses how close it comes to those selection criteria. Chapters IV and V present and broadly discuss the statistical results of the analysis of the recorded data for both buses and coaches. Annex I of this report tries to estimate to what extent the refill-based approach fails to cover the total of regular leakage. In Annex II issues relating to the accuracy of the recorded refill-data are discussed, and in Annex III the contribution of individual MAC components to the annual leakage rates is analysed.