This paper contains the recommendations of the high dose rate (HDR) brachytherapy working party of the UK Institute of Physics and Engineering in Medicine (IPEM). The recommendations consist of a Code of Practice (COP) for the UK for measuring the reference air kerma rate (RAKR) of HDR 192Ir brachytherapy sources. In 2004, the National Physical Laboratory (NPL) commissioned a primary standard for the realization of RAKR of HDR 192Ir brachytherapy sources. This has meant that it is now possible to calibrate ionization chambers directly traceable to an air kerma standard using an 192Ir source (Sander and Nutbrown 2006 NPL Report DQL-RD 004 (Teddington: NPL) http://publications.npl.co.uk). In order to use the source specification in terms of either RAKR, (ICRU 1985 ICRU Report No 38 (Washington, DC: ICRU); ICRU 1997 ICRU Report No 58 (Bethesda, MD: ICRU)), or air kerma strength, SK (Nath et al 1995 Med. Phys. 22 209–34), it has been necessary to develop algorithms that can calculate the dose at any point around brachytherapy sources within the patient tissues. The AAPM TG-43 protocol (Nath et al 1995 Med. Phys. 22 209–34) and the 2004 update TG-43U1 (Rivard et al 2004 Med. Phys. 31 633–74) have been developed more fully than any other protocol and are widely used in commercial treatment planning systems. Since the TG-43 formalism uses the quantity air kerma strength, whereas this COP uses RAKR, a unit conversion from RAKR to air kerma strength was included in the appendix to this COP. It is recommended that the measured RAKR determined with a calibrated well chamber traceable to the NPL 192Ir primary standard is used in the treatment planning system. The measurement uncertainty in the source calibration based on the system described in this COP has been reduced considerably compared to other methods based on interpolation techniques.