Inlet Manifold Design - Triple Injectors
Turbo Mini Forums

Sorted.

In the end i managed to get the Hole Wizard to punch a couple of holes through the plenum wall, so i did not need to change the assembly at all.

Here is a section view:



Ready for the CFD now

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

I'm struggling to get a working copy of Ansys so I thought I would see what the Solidworks CFD feature would make of it.

The main restriction of the Solidworks CFD is that it only allows one inlet and one outlet, so I had to put a bung in one of the runners. Also it will not do transients.

Anyway first attempt with a flow of 0.5 m3/sec:



I'll have another play tomorrow with the other port and do some velocity checks. However, my first impresion is that there is not a problem with the trumpet being very close to the inlet.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

very clever stuff, I wish I has the time to play and learn solid works better

Team www.sheepspeed.com Racing

No offence Paul but to my untrained eye it just looks like a child scribbled all over it with a blue crayon.

top job Paul!!!

hope we will see it running soon :)

Nothing special!

Great work Paul, what addition of Solidworks are you using, is the CFD an add on package like Cosmos?

1972 998 TURBO SLEEPER

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

cheers Paul

1972 998 TURBO SLEEPER

Ok, I've now got flows equivalent to about 9000rpm on a 1275, so a little higher than reality.

Two steady state runs on either port:





If you think about it, the manifold will see a very quick alternation between the two scenarios seen above.

However, we are not seeing any presure pulse data yet. I'm hoping that the other software will allow transient conditions and show pressure data.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Interesting data.

Since there seems to be a significant velocity difference within the inlet, do you think each cylinder pair will have different VE behaviour? One inlet would seem to favor the inside cylinder while the other favors the outside one. Or is it simply irrelevant with little overlap?

Jean

http://www.jbperf.com/

The velocity profile in the ports is biased towards the side away from the inlet. By the time it has got to the valve area it should have even out. So I'm not sure if it is significant enough to warrant changing anything.

I would imagine that if I shifted the inlet to the centre, for example, then the velocity would be higher on the outside of each port.

I've also ran the model at 0.03m3/sec which is equivalent to around 6000rpm and the velocity distribution is still much the same as above.

Unless I made the body of the plenum much bigger to lose the flow inertia within the plenum, then that is unliley to change.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Paul I didn't read the entire thread, just took a look at your Solidworks sims.
I remember that usually this kind of runners work the best when the trumpets are close to the floor. You'll find a lot of old F1 or Aston plenum examples.
The plenum works better when its section decreases toward the farthest runner, this creates a more balanced flow between runners, otherwise air misses the first runner, makes the best in the middle and does not go to the farthest. You'll find most of stock plenums of inline engines are like that.

Edited by alpa on 10th Jun, 2010.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

If something is worth doing, it's worth doing half of.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Well, it looks like there may be benefits in adding the taper. A couple of iterative analises:





You can clearly see that the velocity profile in the port is much better.

Umm....

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Surely, unless you can run a simulation of the two runners flowing at the same time - yes, I know that doesn't actually happen, but unless you simulate it - the effect of the taper on air favouring one runner or the other, can't be show.

I think the taper is more to do with them each getting the same amount of air rather than the flow distribution once it's in the runner.

Schrödinger's cat - so which one am I ???

As I said ^^^ this particular version of the software will only allow one inlet and one outlet plus it does not does transient simulations. So it has limited functionality.

However it does show some very useful data.

To do this properly will involve transient simulations with pulse flow on each outlet.

The purpose of the taper is to slow the flow in the plenum to negate any inertia effects. It is the inertia at the inlet that is causing an uneven flow distribution in the port.

The last simulation clearly shows an improvement, albeit a snapshot in time, when all the flow from the inlet is going into the single port.

I'll try and get the other software sorted.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

This is what we need to do:

http://www.youtube.com/watch?v=X3ecq8MNfk0&NR=1

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I've made the taper a bit longer:

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

neat - if you fancy trying something else - put the inlet feeding into the middle of the two ports, front on or underneath

JK

If Carling made Mini engines
it would probably be like this one!

Paul, what you are basically prooving is that the intake path should not change air speed because every change is a source of reflections (impedance) .
Also you can not use a throttle as small as the port, because transients would become very slow when you'll need to fill the plenum. You can not use a large one, because it's a huge speed change. Solution is somewhere in the middle with smooth section changes (no more that 15% per cm if I remember).

Edited by alpa on 11th Jun, 2010.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."