The goal of the old carbon project continues to be the measurement of ¹⁴C/C in natural methane at levels well below 10^-18 by Accelerator Mass Spectrometry (AMS). Applications include selection of materials for scintillation counters suitable for neutrino and dark matter studies. It has already been shown that the high-energy of AMS is capable of eliminating interference from abundant molecular fragments from the molecules ¹³CH^- and ¹²CH₂^-, that inevitably accompany the low flux of ¹⁴C^-. In addition, the residual carbon isotopes from a low level continuum, inherent in the use of present tandem accelerators, can be eliminated. In this work we describe the studies in progress to understand the origin of the real ¹⁴C atoms coming from the ion source region, excluding sample pre-treatment and preparation which will be considered later. There are three interconnected sources of such ¹⁴C atoms. 1) The residual gas of the vacuum system which contributes due to contemporary hydrocarbons incident on the target, a substantial fraction of which convert to ¹⁴C^- ions. 2) The target support and surroundings will inevitably be contaminated with carbon. 3) The sputtering beam itself will also contain carbon compounds with an abundance depending upon the nature of the ion source. The work reported here will describe experiments that endeavour to clarify these problems further. A test bed for the study of the composition of the Cs⁺ beam has been set up to measure the beam composition over a wide range of masses to extend the earlier work below mass 133 to higher masses. The carbon content of ion source materials is observable and variable. It is at a low level and the ¹⁴C content indicates that it is not contemporary. The relative contribution of vacuum hydrocarbons, Cs⁺ beam contamination and intrinsic carbon material and its age all have to be sorted out. The getter role of the sputtered material has been observed and this serves to reduce the effect of the vacuum hydrocarbons in some ion sources. This in turn suggests that an ion source not used routinely for ¹⁴C dating will be suitable if the carbon in the target support material is old enough or, if thick enough, old carbon layers are deposited. At present the Cs⁺ beam carbon contamination appears to be low enough from masses below 133 but the region above shows components >> 1ppm and these are being studied.