CFD/CAD/Thread

[hoffman900]

Let's design some shapes to compare.
Define a location and angle at the start and end.
One can be 2 straight sections with whatever blend radius you choose.
The other can be the least tension curved shape.

Post some dimensions and we can test it.

Jon,

Let’s just do it like an exhaust port (but on the intake). Try a 3” radius off the intake manifold / head interface and do one at a 6” radius.

[SchmidtMotorWorks]

Jon,

Let’s just do it like an exhaust port (but on the intake). Try a 3” radius off the intake manifold / head interface and do one at a 6” radius.

Can you make an image in paint or something?

[hoffman900]

Jon,

Let’s just do it like an exhaust port (but on the intake). Try a 3” radius off the intake manifold / head interface and do one at a 6” radius.

Can you make an image in paint or something?

I was just thinking like a manifold runner, but a tight bend vs a larger bend.

I guess my point is to illustrate how a bend in a manifold is going to require a different in head port shape.

[JC565Ford]

In the test I did in the other thread, the curved duct did flow better than the mitered and blended duct.

Have you seen those cheap Chinese fabricated manifolds with the straight ducts?

s-l500.jpg

In this case that's redundant, don't you think? That is a good example though, but you'd probably have a hard time finding a worse/better example

[SchmidtMotorWorks]

I was just thinking like a manifold runner, but a tight bend vs a larger bend.

I guess my point is to illustrate how a bend in a manifold is going to require a different in head port shape.

Didn't we cover that in the other thread?

The curved duct flowed better.

I can't think of any well developed application where efficient flow is critical that uses a mitered rather than curved shape to transition from one direction to another.

Even HVAC and plumbing work uses a formula to compute pressure drop based on the radius of the turn.

[SchmidtMotorWorks]

In this case that's redundant, don't you think? That is a good example though, but you'd probably have a hard time finding a worse/better example

Testing that was one of the few times that a crowd gathered in the dyno area at Edelbrock.
I don't remember the numbers but do remember a lot of laughing.

That is not to say that the Chinese don't make some good castings.
A lot of hi-end Japanese and German cars castings are made in China.

At Edelbrock there was an SBC head made in China that we were looking to copy some ideas from.
They were molding the port cores diagonally and then assembling the cores, that way they didn't have any molding draft taper.
Edelbrock didn't want to do it, too expensive.

[bob460]

Jon what's your thoughts on this BBC intake.........

https://scontent.fmel7-1.fna.fbcdn.net/ ... e=647B3BCD

[JC565Ford]

I had an interesting discussion with a CFD engineer.

He said that of course the curved duct will flow more.
and
If the fuel is dropping out in either design, the solution would be to have smaller fuel drops.

Is that always possible with carbs ?

Annular discharge boosters help. but they can only do so much.

[hoffman900]

I had an interesting discussion with a CFD engineer.

He said that of course the curved duct will flow more.
and
If the fuel is dropping out in either design, the solution would be to have smaller fuel drops.

Is that always possible with carbs ?

Annular discharge boosters help. but they can only do so much.

You’re ultimately limited by the fact that the fuel in the float bowl is at atmospheric pressure and the difference is that vs whatever the pressure is in the carb booster.

A “low pressure” fuel injector might be at 5bar (which a lot of aftermarket systems are), but the port injectors Formula One used prior to the current PU’s were cap at a rules limited 100bar (and they made power with ever increasing amounts) and figure a gasoline DI injector is around 150 bar.

At the end of the day fuel being forced through 12+ orfices injectors at 100 bar plus, plus whatever the port pressure is at that point, so there is a delta P component, is going to have a larger Joule-Thomson effect.


FYI Honda had a 24 hole injector at 100 bar, and an average droplet size of 17 um, at 100 L/h flow rate, and was worth 20hp over what they had before. Droplet size on a carburetor, that I kind find, ranges 50-200um depending.


Carburetors only make good WOT power because they rely on the plenum to beat it into submission, and often they’re compared against systems where the injectors are 6-8” closer to the valve. If you look at the Ilmor Indy Car engines from the early 1980s up until they went to DI, they injected with injectors through the plenum roof aimed at the bellmouth. Without rules dictating position, Pro Stock builders would absolutely do the same.

I’m really surprised we haven’t seen people in the aftermarket use the current DI pumps to drive high pressure port injectors, like F1 did. Cosworth was testing 200 bar on the dyno and it worked great for them (these were the 20,000 rpm engines, so besides evaporative cooling, they needed to deliver 110hp / cyl worth of fuel in 166x a second, and have it so said fuel could have stable combustion in an almost 4in bore).


Some cool work they did:

D4E19504-18E4-48F5-B3B7-3928B6DF1316.jpeg

[JC565Ford]

Carburetors only make good WOT power because they rely on the plenum to beat it into submission, and often they’re compared against systems where the injectors are 6-8” closer to the valve. If you look at the Ilmor Indy Car engines from the early 1980s up until they went to DI, they injected with injectors through the plenum roof aimed at the bellmouth. Without rules dictating position, Pro Stock builders would absolutely do the same.

I’m really surprised we haven’t seen people in the aftermarket use the current DI pumps to drive high pressure port injectors, like F1 did. Cosworth was testing 200 bar on the dyno and it worked great for them (these were the 20,000 rpm engines, so besides evaporative cooling, they needed to deliver 110hp / cyl worth of fuel in 166x a second, and have it so said fuel could have stable combustion in an almost 4in bore).

Some day I'd like to test that Holley Carb/injection system (The Sniper) on something:
https://www.holley.com/products/fuel_sy ... injection/

[hoffman900]

Carburetors only make good WOT power because they rely on the plenum to beat it into submission, and often they’re compared against systems where the injectors are 6-8” closer to the valve. If you look at the Ilmor Indy Car engines from the early 1980s up until they went to DI, they injected with injectors through the plenum roof aimed at the bellmouth. Without rules dictating position, Pro Stock builders would absolutely do the same.

I’m really surprised we haven’t seen people in the aftermarket use the current DI pumps to drive high pressure port injectors, like F1 did. Cosworth was testing 200 bar on the dyno and it worked great for them (these were the 20,000 rpm engines, so besides evaporative cooling, they needed to deliver 110hp / cyl worth of fuel in 166x a second, and have it so said fuel could have stable combustion in an almost 4in bore).

Some day I'd like to test that Holley Carb/injection system (The Sniper) on something:
https://www.holley.com/products/fuel_sy ... injection/

These just seem like a worst example of both.

With a proper fuel injection system you now can put the intake runners wherever that maximizes mass flow and wave dynamics (and package better in regards to cog and aero), and then can inject the fuel at much higher pressure wherever you want; from in the cylinder to all the way at the end of the bellmouth, and time it however you want.

Another Honda example:

514CD98D-51D9-466D-95C7-7DB9B0AD885E.jpeg

A27E2324-9502-492C-AA55-CDBDD753DC49.jpeg

[SchmidtMotorWorks]

Jon what's your thoughts on this BBC intake.........

https://scontent.fmel7-1.fna.fbcdn.net/ ... e=647B3BCD

The injector placement and throttles look likely to have fuel suspension and distribution problems.

[bob460]

Jon what's your thoughts on this BBC intake.........

https://scontent.fmel7-1.fna.fbcdn.net/ ... e=647B3BCD

The injector placement and throttles look likely to have fuel suspension and distribution problems.

It's going on a BBC 500ci pump gas deal 900-1000hp.......should be interesting to see how it works out.

[In-Tech]

Hiya,
The suspension and distribution will be a huge problem. It could easily work at a "particular" rpm based on what I can see by the huge plenum hoping for pulses with a certain amount of overlap to the intake and hopefully EGR from overlap with induced IAT heat. Good luck

I am no expert, the runners look half decent, the entry leaves a lot to be desired unless you plan on a massive IAT change and density with meth. Definitely not trying to be a jerk. 1000hp on 91 is easy with the right overlap cycle to cool, of course the exhaust overlap to cool the chamber.

[JC565Ford]

Those box cross rams work really well on the 426 hemi. If he sized it correctly for the planned 500" engine and rpm range it could work really well.

He knows a bit about building high performance engines, so I suspect he did his homework.