CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Cylinder head port problems (

Jon Reynolds March 23, 2006 09:38

Cylinder head port problems
Hello all,

This is my first post so please be gentle.

I am a (mature) student studying Motorsport Engineering in Swansea, UK in my second year working on my main project.

My project, which will span into my final year to make the basis of my dissertation is based on analysis of flow through cylinder heads.

I have modelled the ports and valves and cylinder of a particular Jaguar engine and have successfully used a combination of Solid Edge for modelling, I-deas for meshing and Fluent for "CFD'ing".

My first series of tests were with the inlet valves in an open position (obviously :)) at a lift of 10mm (0.400"). This worked well and solved no least with laminar flow.

But now all I have changed is simply the position of the valves. I have positioned them in a low lift position of 2mm (0.080") without altering any other parameters and now it will not solve!

A few issues i am trying to establish solutions to...

1) my initial working results were a little higher than the flowbench result which I immediately put down to the afct that my modelled walls were of course very smooth and did not mimic the rougher cast walls of the real thing, so we tried using various turbulence models in the "viscous" options and setting a wall roughness, butnone of these would solve and just gave huge velocity figures which were just way out.

So i settled for using laminar flow with a fine mesh. This was close but about 10% too high. But I settled for it as I need to get on with modifying the design and proving that changes in design of valve and port can bemeasured in CFD and then proved on the bench.

So any suggestions to a more accurate method would be greatly appreciated.

2)Any ideas as to why changing nothing except the position of the valve would cause it to not solve any more...very infuriating.

sorry for the lengthy post but I had a bit of detail to cover.

Many thanks in advance for any help.

Jon Reynolds.

p.s. here is a sample of pics taken from my results so far

All times are GMT -4. The time now is 04:56.