(Morrigan, as a synonym of Morgan of Arthurian legend)

Tuesday, 18 August 2009

Throttle body calcs-4

This is a second attempt at creating a fluent analysis of a throttle body, this time utilising MSC Patran to produce the volume mesh. I now believe that for straight forward meshes that Patran has greater capability than GAMBIT.

Second attempt throttle body - Rhino3D

The throttle body model was generated in sections, this will allow quick production of the varying throttle angle models to be run through. That is once i work out what i am going to be looking for.

Throttle volume meshed in Patran

Meshing coincident volumes in Patran, when you have more than six sides requires a little ingenuity. Because the software cannot recognise that the faces of the volume parts meet, it will not produce the same mesh on both. Therefore you have to create a surface mesh on one face. Mesh the volume, then disassociate it and associate it with the adjoining face, before meshing the adjoining volume. In this way you can create matching volume faces which will equivalence into a larger volume. In this way the model is split into a top section of the air horn, then throttle butterfly, then a profile change from circular to oval, and the final section for mounting the fuel injector.

Spliting / Guiding surfaces for volumes

This allows great control of the meshing process. GAMBIT will now only be used for applying the face properties to the finished volumes. Only TET4 meshes are able to be read from a patran neutral '.out' file, into GAMBIT.

Mesh in Fluent

The Fluent run is progressing smoothly, with a steady reduction of continuity residuals. And the results are in...
Fluent output

Fluent output - looking down the bore, (vortices rather like an owl)

Fluent output

Sunday, 16 August 2009

Throttle body calcs-3

Decided to have a play, and get a CFD model of a throttle body put together.

Based on the head inlet area of 15.5cm^2 the throttle body needs to have a diameter of roughly 45mm, allowing for a 5mm shaft for the butterfly valve.

First throttle body test

Close up detail

After several hours playing around trying to mesh the model, I gave up and dropped a dimension, as 2D runs a are much easier to build and run. Generally the numbers should be ignored, but the flow lines are usually pretty indicative.

CFD Throttle @ 5250rpm - 45deg opening

Fluent close-up

But there is scope for setting up some runns with varying throttle opening, inlet pressures. Then can develop injector positions to get the best mixing arrangement.

Saturday, 8 August 2009

Throttle body calcs-2

Well i have extended my calcs for the throttle body to include the variation of both engine rpm and also the throttle angle. This gives the total mass of air inducted by the piston during one cycle. Usefull for showing the inducted volume limitations, I need to update it for the actual area of the inlet on the head, which is the main bottle neck.

The methodology is somewhat brute-force, This pic uses 1000 throttle angle points, 1000 rpm points and 1000 piston angle points. So 1x10^9 calculations, but that is what octal core workstations are for...

Engine speed 0-5250rpm, throttle opening from 0-90deg

edit: One interesting point to note is that idle RPM is 800, this is well clear of ~400rpm where the mass inducted reaches its highest for throttle angle vs RPM.

Still finishing off the garages, need to;
  1. Repair the roof - DONE
  2. Replace the smashed sky-light -DONE
  3. Board up the broken windows - DONE
  4. Fabricate a front & hang a door - DONE
  5. Move spare parts, tyres and stock out of main workshop - DONE
  6. Re-arrange the tool cupboard - DONE
  7. Build a work bench - DONE
  8. Apply a second coat of floor paint - DONE
  9. Install wood burning stove and fabricate chimney - DONE

Saturday, 1 August 2009

Throttle body calcs-1

Got side tracked with building a calc tool for evaluating the effect of a throttle body. From first principles I've used the displacement of the piston to find the volume, then pressure, mass flow, new mass in cylinder and then velocity through the throttle orifice.

The calcs need to be checked through, but the order of magnitude is about correct.

5250rpm throttle opening from 1deg to 90deg
Volume (cm3)
Velocity (m/s)
Pressure (bar)
Mass (grams)