Simulated Intake Manifold

[DBFIU]

Ok the results for that manifold are:

Cyl 1: massflow = 0.24747 lbm/s
Cyl 2: massflow = 0.25693 lbm/s
Cyl 3: massflow = 0.24650 lbm/s
Cyl 4: massflow = 0.24912 lbm/s

Worst deviation is from Cyl 2 to Cyl 3. Cyl 2 is getting a straight shot of flow and cyl 3 a big recirculation zone right in front the of the horn effectively reducing that flow a little bit, you can clearly see that here on this surface streamline plot.




It's about a 4.2% deviation between those 2 cylinders which doesn't sound like a lot, but it is.

Imagine if you had to tune this descrepancy out of an engine, one cylinder basically making 4.2% more power than the other. You'd never know!

Or, one cylinder is getting 4.2% LESS FUEL than the other. Basically, this is not good; you get the point.

[inlineS54B32]

This is awesome information - one of the best posts I have read... I have a question about TB 6 being farther away/not getting as much air. On an engine like the s54, do the independent throttle bodies actually help with this? Meaning, I would imagine that BMW could "tune" the throttle bodies (not sure if they are independent though?) to minimize this effect?

I have read on forums that the ITBs don't really do much - but I have a hard time believing BMW/M would do it for no reason. I guess the same question comes into play for valve based throttles (is there still a main plate to control airflow at low speeds?) - could they theoretically tune the opening of the valve to help out cylinder #6?

Always was curious as to what the independent throttle bodies are helping out the most - I know that the M engines I have "tried" feel MUCH more responsive that a "normal" engine... Is this one of the main contributors to the great response?

Again, very cool stuff, looking forward to more of this.

[DBFIU]

This is awesome information - one of the best posts I have read... I have a question about TB 6 being farther away/not getting as much air. On an engine like the s54, do the independent throttle bodies actually help with this? Meaning, I would imagine that BMW could "tune" the throttle bodies (not sure if they are independent though?) to minimize this effect?

I have read on forums that the ITBs don't really do much - but I have a hard time believing BMW/M would do it for no reason. I guess the same question comes into play for valve based throttles (is there still a main plate to control airflow at low speeds?) - could they theoretically tune the opening of the valve to help out cylinder #6?

Always was curious as to what the independent throttle bodies are helping out the most - I know that the M engines I have "tried" feel MUCH more responsive that a "normal" engine... Is this one of the main contributors to the great response?

Again, very cool stuff, looking forward to more of this.

That is in essence why some cars have ITBs, they should each be able to be individually controlled to keep this phenomena from happening. This makes for a well balanced engine, reduced vibrations and much better load handling capability. I am not sure if the S65 or S85 has the ability to control each ITB separately, maybe someone with experience in that platform can answer that question.

[DBFIU]

Guys remember how cylinder 2 and 3 showed the largest deviations of flows?

Want to see why?

Look at these streamline plots going through each cylinder on the XZ plane.

Notice how Cyl 2 has a good straight shot of flow going to it, cylinders 1 and 3 show big zones of recirculation (the big blue blotches right in front of the horns). These are what I need to get rid of. If you want to squeeze every ounce of power out of your engine, you will have to go through this process.

[fundahl]

Maybe changing the angle of the throttle body would help? Perhaps design an oval-ed throttle body/plenum entry?

You get a big, straight, high velocity area towards cyl. 2 & 3 with the current throttle body position it seems.

[DBFIU]

Maybe changing the angle of the throttle body would help? Perhaps design an oval-ed throttle body/plenum entry?

You get a big, straight, high velocity area towards cyl. 2 & 3 with the current throttle body position it seems.

Changing the angle will help. So will adding a taper in the manifold as you get farther away from the TB to help keep velocity up.

[PEI330Ci]

If you can 3D scan a stock N54 manifold that would be appreciated!!!

Do you mean destructive scanning? (Slicing?)

[DBFIU]

Do you mean destructive scanning? (Slicing?)

The only way to get the geometry of the volume inside the manifold is to slice it up and scan each piece then assemble it, or fill it with foam and try and pull the foam out without destroying it... Yea basically destructive lol

[PEI330Ci]

The only way to get the geometry of the volume inside the manifold is to slice it up and scan each piece then assemble it, or fill it with foam and try and pull the foam out without destroying it... Yea basically destructive lol

In terms that maybe others can relate to: "reverse stereo lithography"

[Sticky]

There should have to be less spreading, because not everybody shares an equal amount of useful info.....

It's coming damn it...

[black bnr32]

you might be able to 3d scan the outside of it as a surface and use an 'offset' function in your 3d modeling software to generate the 'walls' of the part

[DBFIU]

you might be able to 3d scan the outside of it as a surface and use an 'offset' function in your 3d modeling software to generate the 'walls' of the part

Yea, but that may not necessarily be the shape of the inside.

[Itsbrokeagain]

I'll see if I can dig up an intake manifold from someone...I know someone who broke the nipple off it and replaced it with a whole new one. I'll see if he still has it.


What if you offset the TB more towards the front, say 2-3" before it opens up into the plenum? Would that help disperse the main flow pattern or just make it worse by increasing the velocity and actually resulting in MORE air hitting cylinder 2?

[DBFIU]

I'll see if I can dig up an intake manifold from someone...I know someone who broke the nipple off it and replaced it with a whole new one. I'll see if he still has it.


What if you offset the TB more towards the front, say 2-3" before it opens up into the plenum? Would that help disperse the main flow pattern or just make it worse by increasing the velocity and actually resulting in MORE air hitting cylinder 2?

If you can get a stock manifold I might be able to simulate it. The only problem is getting the correct geometry for the inside surfaces (I have to model the negative of the manifold, the volume has to be accurate).

One way to fix the problem with my test manifold is to move the TB a little farther away from the primaries. If you do this though you have to keep the plenum volum constant by reducing the height of the manifold. Changing the plenum volume changes the helmholtz resonant frequency and the also shifts your powerband around.


Edit: I may end up using these 'test' manifold pics for my website

[mycoupe]

If you can get a stock manifold I might be able to simulate it. The only problem is getting the correct geometry for the inside surfaces (I have to model the negative of the manifold, the volume has to be accurate).

One way to fix the problem with my test manifold is to move the TB a little farther away from the primaries. If you do this though you have to keep the plenum volum constant by reducing the height of the manifold. Changing the plenum volume changes the helmholtz resonant frequency and the also shifts your powerband around.


Edit: I may end up using these 'test' manifold pics for my website

I know this thread is old and long abandoned, but I would love to get it back up and going.

As for the scan of the factory manifold. A point cloud from an x-ray could work. The manifold might need to be filled with something that would show up, but we have NDE work done at my place of employment daily. I could talk to some people and see if they would have the time to scan my manifold, or point me in a better direction.

[mycoupe]

I also use NX 7.5 at work, almost daily. So, if you need someone to tackle the design/modeling side of things to lessen the burden, I am ALL IN.

[lulz_m3]

Dbfiu sold his 135, got a mustang boss 302 and hasn't returned to this forum since.

[Itsbrokeagain]

Dbfiu sold his 135, got a mustang boss 302 and hasn't returned to this forum since.

I just realized this...he was awfully quiet, guess that explains it!