Because of their unique properties of high strength, excellent hardenability and good corrosion resistance beta titanium alloys have been the subject of much attention. Recently, since cost is an increasingly important parameter, a LCB (low cost beta) titanium alloy(Ti6.8Mo-4.5Fe-1.5A1) has been developed to be used in non-aerospace new technology as high performance substitutes for more classical material. Metastable β Ti alloys are typically known to precipitate additional phases (α andω phases) during thermomechanical treatments. The morphology, size and distribution of these precipitates determine in large part the mechanical properties of the alloy [1]. It is known thatω precipitates can be formed during quenching by a diffusionless martensitic mechanism (ωath) and during aging by a diffusion controlled process (ωiso) [2]. Nevertheless, the parameters allowing the control of ωiso phase crystals density are still unclear [3]. Therefore, it seems necessary to control the appearance of ω phase particles in β titanium alloys and particularly to understand the nucleation mechanism. The direct interest of such a knowledge is crucial since it was recently reported that the ωiso precipitates could behave as nucleation sites of an extremely fine and dispersed α phase at intermediate aging temperatures leading to higher yield strength (β + α) duplex alloys [4–6]. In this present paper a method based on an experimental new approach of the β→ω transformation is described. This method could allow an experimental control of ωiso dispersion. The β phase decomposition during heating was studied on Ti-6.8Mo-4.5Fe-1.A1 water quenched from the β field (1123 K). This method was developed by studying the electrical resistivity changes of samples during subsequent heat treatments. The electrical resistance measurements were made by a computer driven four-point technique. R/Ri (Ri is the initial resistance) is plotted as a function of time and temperature. In this way, microstructural evolutions and transformations kinetics can be investigated [7, 8]. It was reported that the metastable β titanium alloys had after quenching a resistivity increasing with cooling and decreasing with heating. This effect named NTD (negative dependence temperature) was ascribed to the ωath precipitation during quenching and reversed in an entirely reversible way and without hysteresis during the subsequent heat treatment [8, 9]. In addition, it is known that an ωiso phase also hexagonal could precipitate during aging, implying diffusional phenomena. We choose to consider the TCR (temperature coefficient of resistivity= d(R/Ri)/dT ) between 343 and 383 K as a parameter for monitoring microstructural changes during heating and cooling of Ti-6.8Mo4.5Fe-1.A1 samples. A first series of resistometric experiments (Fig. 1) allowed us to notice that cycles carried out up to 423 K (v= 10 K/min) display a total recovery of TCR (TCR recovery ratio= 100× TCRcooling/TCRheating) showing that the reaction ωath⇔β is completely reversible in good agreement with previous results [8]. No hysteresis was noticed during these thermal cycles below 423 K. Above 423 K, the recovery of our coefficient is only partial which can be connected to irreversible microstructural changes. The assumption that the precipitation of the ωiso prevents the reprecipitation of ωath during cooling is quite well established now [9]. Nevertheless, the involved mechanisms binding the two types of particles remains to be clarified. In order to investigate the state of irreversible microstructural changes, resistometric experiments were carried out. Quenched samples on one hand, and subsequent cold rolled samples (at different reduction rates) on the other hand were aged for 18 ks at 473 K and 523 K. TCR (measured during heating and cooling under conditions quoted above) and TCR recovery ratio were then graphically evaluated. During the course of this work, extensive characterization using transmission electron microscopy and X-ray analysis confirmed that irreversible microstructural changes above 423 K could be associated with ωiso precipitation (with ellipsoidal morphology) in the β matrix. Figs 2 and 3 suggest that rate of rolling has little influence on the TCR recovery ratio evolution. Thus, although the applied stress is thought to create lattice defects such as vacancy clusters and misfit lattices acting as potential nucleation sites, the ωiso