Feeding of high-concentrate diets to cattle increases the risk of subacute ruminal acidosis (SARA). Indwelling wireless pH sensors have become popular as an early diagnostic tool for SARA recently. However, the recommended pH thresholds of SARA derive from measurements taken from free-rumen liquid (FRL) in the ventral rumen, and not from the reticulum, where the mentioned sensors are located. The aim of this study was to evaluate commercially available indwelling pH boli for the accuracy and precision in predicting ruminal pH of FRL and the particle-associated rumen liquid (PARL) under varying dietary concentrate levels. An additional aim was to define SARA pH thresholds of indwelled pH boli, which can be used for SARA diagnostics. The experiment was conducted with eight nonlactating rumen cannulated Holstein cows fed 0% or 65% concentrate. Data showed that the mean pH of indwelled boli was consistently higher than in FRL and PARL (pH 6.62 ± 0.02, 6.43 ± 0.02 and 6.18 ±0.02, respectively) across feeding regimens. Interestingly, the diurnal differences in pH among indwelled boli, FRL and PARL became greater during concentrate feeding, especially at 8 h after the morning feeding, suggesting that with high-concentrate diets a particular adjustment of reticular sensor pH vs. ruminal pH in FRL and PARL is needed. The concordance correlation coefficient analysis, representing the reproducibility of the bolus measurements, was high for bolus-FRL (0.733) and moderate for bolus-PARL (0.510) associations. Furthermore, the quantitative relationship of the pH in FRL and PARL to the pH of the boli was described by linear regression analysis. The study determined that the common SARA threshold of pH 5.8 in FRL corresponds to a bolus pH of 6.0.