Certain applications of laser blow‐off impurity injection, such as ion temperature measurement and spectral line identification, may require the ratio of injected ions‐to‐plasma electron density to exceed 0.001. To achieve these concentrations we have relied on injection of micron‐sized clusters as well as monatomic species. In these cases the injected impurity can no longer be regarded as a nonperturbing trace sourced at the plasma edge. We therefore have studied the variation of plasma parameters in PLT and PDX as a function of the amount of injected impurities, the species mix, and the type of impurity. We find that for ohmically‐heated plasmas, impurity injection can increase ?e more than 4×1012 cm−3 without causing disruptions. During Ge‐injection experiments where Δne/ne ∠0.05, Te initially rises ∠5% on axis and decreases ∠20% at the edge. The plasma current drops 1–3 % and the loop voltage and radiated power both double. Varying the amount of injected impurities changes the peaking and decay times of both high and low impurity charge states by 10–50 %. Analysis of this in terms of altered temperature profiles, impurity penetration profiles and impurity transport is presented. These studies suggest criteria for performing nonperturbing transport experiments.