Oke, nevermind. The solution came to my mind as soon as I posted this:
Of course if the expression is a string, the float value of my variable has to be converted to a string.
pCalcQ.Expression = ‘inputs[0].PointData[“U_Normal”]*inputs[1].PointData[“density”]*inputs[2].PointData[“H”]-’ + str(hRef)
Works fine.
Cheers, nopech