Energy loss of charged particles in a gas has been estimated by a computer simulation technique with a view of studying primary cluster counting as a mean of the particle identification. Primary electron yield and secondary multiplication have been estimated through a detailed calculation of the atomic processes. Results are presented for pure argon in terms of cluster size distributions and relativistic rise of the number of clusters: the latter reaches a plateau at a 3' of 200 and of level of 1.32 times minimum ionization. Limitations to detector resolution due to electronic dead time and longitudinal diffusion of the electrons drifting in the gas are presented. Such limitations are introduced in a comparison between the performance in particle separation of detectors based on energy loss, making use of pulse height measurement or cluster counting.

Experimental results on the influence of angle and cell geometry on pulse height measurements in a multiwire proportional chamber (MWPC) are presented. Two track orientations are mainly considered: (a) tracks parallel to the wire plane and (b) tracks traversing the gap with one wire touched. In case (a), the positive induced pulse effect is analyzed: pulse heights have to be corrected by factors of 1.3-1.7 and wire inefficiency can be found up to 20%, depending on the cell geometry. The saturation effect starts earlier in case (a), but the influence of the angle between track and wire when approaching 90 ° is moderate. On the contrary, this influence is very effective in case (b). A detailed study of this phenomenon is then presented with a description of possible mechanisms involved.