transient boundary condition in fluent

[sps]

hello to all
this time we are using the fluent for CFD analysis of the intake manifold.... the problem i am encountering is that how to decide the boundary condition at the outlet of the runner lengths in the case of natural aspirated engine.. as this flow at the outlet is transient i.e. it is varying with the crank angle. can any one help me to decide the outlet boundary conditions of the runner in the case of fluent.. i will be highly thankful..

[Simon Dingle]

Hello,

If this is the first time that you are using CFD software, then I would strongly recommend that you simplify the problem to a steady-state analysis and start by investigating the throttle/restrictor/diffuser section without the manifold.

A transient simulation of the entire manifold will take a huge amount of time and computational power to complete, even without any kind of validation.

If you are set on a transient simulation then you need to use an LES simulation and I recommend thorougly reading the Fluent help manual on the topic, it is genuinely useful.

[TorqueWrench]

Second to what simon said.

I do most of our simulation work, both transient and steady state, in CFDesign. Three ways to do transient I have seen. First (and easiest if you are doing anything with GT Power) is to run a coupled simulation with your engine simulation suite (have to use Fluent or two other programs though). Second, I have seen people assign pressures and temperature to the inlet and outlet pressure transiently and compare mass flow rate to what the model says. Third, I have seen people assign mass/volumetric flows and temperatures transiently and let the program determine your pressures. In theory they all work great, some are just easier than others to do.

The whole manifold will take a long time to do to tell you the truth. I know the time step sizes required to do anything with the restrictor in it are tiny due to the high velocities found in this area. I think one of the models I ran where the flow peaked at 300m/s had a time step of 5E-4. I know CFDesign recommends a time step that results in a particle traveling 5% of the way along the model's length, so it gets quite small quite quickly. While this may be required by the program for the restrictor, it really isn't needed for the rest of the model.

As with what Simon said, try some steady state simulations first. I did a good chunk of our intake manifold steady state due to the reduction in computation time and resources it needed and have been told by some people in industry it turned out pretty good. Come up with a few good ones steady state and flow the best transiently. You are going to need to do a steady state one anyone to correlate your model in the end.

[sps]

well if i start my steady state analysis first then what outlet bundary conditions should follow at the valve side of the manifold. i.e if i use inlet pressure of the manifold 80 kpa ( this pressure is known from the analysis of the restrictor in steady state) then what value will be chosen for pressure at the outlet ( i.e at the valve side of the manifold) ?????

[arun1]

Originally posted by sps:
well if i start my steady state analysis first then what outlet bundary conditions should follow at the valve side of the manifold. i.e if i use inlet pressure of the manifold 80 kpa ( this pressure is known from the analysis of the restrictor in steady state) then what value will be chosen for pressure at the outlet ( i.e at the valve side of the manifold) ?????

[arun1]

hii everybody, i am doing transient analysis for intake manifold 4 cylinder in fluent, i am not getting how create events & profile,
it has 4 runners. the boundary condition is that 1st runner open at 180 crank angle & 2nd,3rd, 4th opens at 360,440, 720 of crank angle respectively.
please can any one tell me easy method to solve & steps, procedures in fluent

[mmendes]

Im hoping torquewrench or somebody else can chime in here and help me out.

Im looking for someone to point me in the right direction for the settings to use for an IM simulation. Ive already figured out how to import pressure drop profiles for each runner.

Initially I tried folowing the instrucitons in the fluent manual for transient compressible flow but I can never view any animations to make sure the prssures are dropping in the runners like I want.

I already have a few designs done in steady state that I like but would like to see a transient simulation to see what direction to drive my final design towards.

Thanks
Marc