A method is described for generating three-dimensional, time-dependent turbulent inflow data for simulations of spatially-developing boundary layers. The approach involves extracting instantaneous velocity data from an auxiliary simulation of a zero pressure gradient boundary layer. This auxiliary simulation generates its own inflow conditions through a sequence of operations where the velocity field at a downstream station is rescaled and reintroduced at the inlet. This method is a variant of the Spalart method, optimized to convert an existing inflow-outflow code into an inflow generation device with a single subroutine. The method produces realistic turbulent boundary layers with statistics agreeing with experimental data and direct simulations. It is used to provide inflow conditions for a large eddy simulation (LES) of a spatially evolving boundary layer with Reynolds numbers 1530–2150. The results from the LES are compared with other simulations using approximate inflow conditions. The proposed method is shown to be highly accurate with minimal adjustment near the inlet, unlike other methods that produce transient errors persisting downstream. The method is also extended for non-zero pressure gradients. The paper surveys existing inflow generation techniques, including the random fluctuation method, which requires a lengthy development section and is less accurate. The proposed method is more efficient and accurate, producing realistic inflow data with minimal computational overhead. The method is validated by comparing results with experimental data and other simulations, showing good agreement in boundary layer thickness, displacement thickness, momentum thickness, and shape factor. The results demonstrate that the proposed method produces accurate inflow data with minimal development section, making it suitable for spatially evolving boundary layer simulations.A method is described for generating three-dimensional, time-dependent turbulent inflow data for simulations of spatially-developing boundary layers. The approach involves extracting instantaneous velocity data from an auxiliary simulation of a zero pressure gradient boundary layer. This auxiliary simulation generates its own inflow conditions through a sequence of operations where the velocity field at a downstream station is rescaled and reintroduced at the inlet. This method is a variant of the Spalart method, optimized to convert an existing inflow-outflow code into an inflow generation device with a single subroutine. The method produces realistic turbulent boundary layers with statistics agreeing with experimental data and direct simulations. It is used to provide inflow conditions for a large eddy simulation (LES) of a spatially evolving boundary layer with Reynolds numbers 1530–2150. The results from the LES are compared with other simulations using approximate inflow conditions. The proposed method is shown to be highly accurate with minimal adjustment near the inlet, unlike other methods that produce transient errors persisting downstream. The method is also extended for non-zero pressure gradients. The paper surveys existing inflow generation techniques, including the random fluctuation method, which requires a lengthy development section and is less accurate. The proposed method is more efficient and accurate, producing realistic inflow data with minimal computational overhead. The method is validated by comparing results with experimental data and other simulations, showing good agreement in boundary layer thickness, displacement thickness, momentum thickness, and shape factor. The results demonstrate that the proposed method produces accurate inflow data with minimal development section, making it suitable for spatially evolving boundary layer simulations.