Still seeing some pressure jumps at compressible inlet-wall nodes

[bigfooted]

This is a simulation of the flow through a channel with a bend, using the compressible solver, taken from the website:
https://turbmodels.larc.nasa.gov/smitscurve_val.html

There are stills some peaks in pressure and temperature at the inlet (it has total conditions prescribed).
In ::BC_Inlet, we have this code to circumvent it but it doesn't seem to be sufficient. Anything else we can do to prevent these jumps?

        /*--- Non-dim. the inputs if necessary. ---*/

        P_Total /= config->GetPressure_Ref();
        T_Total /= config->GetTemperature_Ref();

        if (geometry->nodes->GetViscousBoundary(iPoint)) {
          P_Total = nodes->GetPressure(iPoint);
          T_Total = nodes->GetTemperature(iPoint);
        }