[aero777]

Hi,

I have used an inviscid compressible flow model for carrying out an analysis of the flow-field.

I have plotted the variation of the Mach number with distance along the nozzle from predictions of 1D theory as well as Fluent's results at the centreline and the wall.

My predictions were that the 1D theory would more closely correlate to Fluent's result at the centreline but the curve is in fact almost halfway between the distance of the 2 curves by Fluent - nothing too conclusive.

My main Q is : What accounts for the differences in the plotted curves?

Attached is a typical graph; position x=0 refers to the throat.

[narendra.gadwal]

http://instruct1.cit.cornell.edu/cou...zzle/step4.htm

[aero777]

Originally Posted by narendra.gadwal

http://instruct1.cit.cornell.edu/cou...zzle/step4.htm

I am aware of this page. Is it down to the choice of solver or numerical errors from the grid generation?

[aerogt3]

Originally Posted by aero777

I am aware of this page. Is it down to the choice of solver or numerical errors from the grid generation?

I would expect very little error from the grid.... a structured 1-D grid should be quite good quality, no?

[narendra.gadwal]

Why don't you run the simulation considering turbulence ?
Viscosity can also appear because of flow. (Turbulent viscosity)