CFD Head Study
Turbo Mini Forums

Following on from my CFD studies on inlet manifolds, I thought that as it's now too cold to spend much time in the workshop, I could at least attempt to do a CFD study on a 5 port head using Solidworks Flow Simulation to do a full transient analysis.

I've got a number of standard leaded 998 A+ heads that are heading for the scrappy unless I can find a use for them. So, the question is: What will it take to make them flow as well as a 12G295 and how much of the work can I do on my 3-Axis CNC Mill.

First job is to measure up the standard ports. Started with a bit of chopping:

Inlet port


Exhaust port


Interesting that the inlet port is mainly crcular in cross section throughout, whereas the exhaust port maintains a retangular cross section for most of its length. I never knew that.

Next job is to work out how to cut the ports in solidworks. The challenge will be the fillet radius on the divider.

Edited by Paul S on 7th Dec, 2012.

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."

Im going to keep my eye on this one, I use Ansys fluent for my work, if you require any help give me a shout.

I expect it will take a while to model.

Good luck

What you need is a laser scanner company..... which my colleagues son is just about to set up Suspect the rates might be a little prohibitive though!

Watching this with interest though.

Steve
DSN Classics Ltd
www.dsnclassics.co.uk

Did you section the end exhaust ports as well ?

Metric is for people who can't do fractions.

That is the end exhaust port cut vertically. I just cut the siamese port horiz.

I've sorted the method for radius on the divider. Didn't realise that it was possible to select such an odd profile and fillet it :)

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."

Work in progress on the inlet port:



Still need to add the valve so that the exit flow is around the valve rather than just the stem.

It obviously does not like the guide boss, but we knew that.

I don't know that I'll be able to do the full transient analysis as that will need moving valves.

Edited by Paul S on 8th Dec, 2012.

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."

No progress today, but managed to work out how to find out the flow rate.

The model is now based on a 25"H20 pressure drop, so I can relate to flow bench numbers. (picture above was about 15"H20)

This is flowing 0.0178 m3/sec which is about 37.5 cfm. Not a lot but given that there is not enough room to stick ones index finger past the valve guide, I'm feeling confident with that.

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 don't know about solidworks flow, but in Ansys you can setup moving zones for your transient analysis which could be used to move the valve up and down.

I can do Motion Studies of a moving valve, but can't integrate that with the Flow Simulation.

The FS just allows for a single mesh that is updated for every geometric 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."

this will be interesting

Put a static valve in for now at 9mm lift. Added a few fillets and chamfers to replicate the casting and re-ran the analysis:


Now flowing 31.6 cfm at 25"H20.

I probably should actually put the combustion chamber on the end now, just to get a more accurate result, but the restrictions are around the valve guide, so that's where to start with the mods.

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."

A "Cut Plot" shows a bit more:

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."

Rimflows?

I run a supercharger and I don't care the TB is on the wrong side.
VEMS + 12 PSI + Liquid Intercooler = Small Bore FUN!

Pressure "Cut Plot" shows that the main restriction is the valve guide/boss:


Odd how it behaves on the short side radius, not as expected.

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."

My head hurts again!

Yep mine too,

In my mind all this must be complete bollox as there is no way I can comprehend it lol.

1/4 Mile 14.3secs 96Mph Terminal 10psi of boost.


Fibreglass Parts? - http://www.tdkracing.co.uk/
Split Rims? - http://www.force-racing.co.uk/

I thought a standard 998 casting would look alot worse than that on flow.

I find CFD study quite interesting... just wish i knew more about it to delve and understand it

David.

Flow at full lift only shows how good the port is. Lift up to ~1.5mm shows how bad the valve is.

Edited by Sprocket on 12th Dec, 2012.

Not much progress as I'm having to do CFD type work for a living this week. I did find a few cock-ups on what I have done, so I'll update the results as soon as.

It's an amazing tool. We are all surrounded everyday with things designed using CFD. Now we can actally find out more about engine components than the original designers knew.

Edited by Paul S on 12th Dec, 2012.

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."

What's your residuals like? The pressure spike at the base of the valve could be due to poor meshing or mesh type? I would expect it to elevate but that looks high. Not that I've got anything to back it up with in terms of cylinder head simulations. I'm more Venturi's and pipework.

Air/fluid flow dynamics are interesting. I dont undersyand it all but the graphics do help


I'd be interested in seeing the results when you get to low lift levels, similar to Sprockets suggestion

yep, the first flow issue will be the guide boss,

after that you'll be looking at seat/throat geometry.... can you model the seat?

Yes, the seat edges can be filleted with whatever radius and I can move the valve nearer the seat.

I did build another model with an increase on the port diameter from 24mm to 28mm and fitted a larger valve. The port flow went down!!!!!

Of course, I'd made the classic mistake of not enlarging the combustion chamber and the valve was shrouded. So I opened the notional combustion chamber up and the gain was nearly 50% more.

So as well as the port being important at full lift, the combustion chamber shape is equally significant. Maybe BL realised this and introduced the open chambers on the 12G295 to go with the larger port.

I really do need to accurately represent the combustion chamber shape in the model before going any further. But that is stretching my CAD skills.

I'm more interested in what, if anything, can be done WRT charge robbing, than simple port and valve seat mods.

The conditions in the port at valve opening and closing are very different for the inner and outer cylinder. So why do we make the ports the same? Just thinking out loud.

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."

Soooo..... bigger inlet valves on the centre cyl;inders then! lol