I think that soon enough, if not now, there will be enough computing power and technology to do a finite element model on a whole motor, so that we can predict the airflow thorugh the entire motor on a molecular level.

 

I think you'd be looking at computational fluid dynamics as opposed to finite element analysis, but either way...with more computing power comes a better understanding. Then again, those computer models are only as good as the person coding it in...

 

Sorry to bring an old one back to the top, but seems there is some good material here. A couple of questions...
I've read before about port speeds going sonic and disrupting airflow. I'm assuming (uh oh) reshaping those areas of the port with reducing local velocity in mind is the answer, thereby increasing net port flow. I think most of us hot rodders have mistakenly assumed that porting overall is about increasing velocity. Is it better to think of porting as balancing airflow throughout the port?
How close has 'wet flow' technology come to showing the virtues of quality airflow. Is this a viable alternative to standard dry bench numbers? Or are both methods merely representative of what is thought to occur with airflow as a whole?

 

I think that you ask some good questions. You can go too small in cross section or too big. It is better to err on the small side of course, but it is best to get it right. Runner cross section has a lot to do with the overall car/engine combo. Small cubes, small cam, low rpm's, all lean you towards a small cc/high velocity runner. And vice versa. A short track car that needs a lot of acceleration out of a turn, an automatic street car (read a 346 Camaro with an automatic tranny), or a Comp Eliminator car that is hooked up solid all want a lot of velocity so the engine pulls up quick when the engine is grunted down. A car that is "traction challenged" can benefit from a larger lower velocity intake port. I've made circle track heads so small that the engine had too much power out of a turn and the driver couldn't control it and had to open them back up. Same thing happens with a drag car. Intake runner cross section is something that a head porter can use to tune the head to the car/engine combination to achieve better track times.


On wet flow...the most important thing that I have seen that improves things is unshrouding the intake valve. This has a twofold benefit. Unshrouding the intake valve improves fuel distribution in the chamber, and ups the flow volume as well. It's a win/win modification. I don't belive in swirl or rely on it in a performance application. To even get it you have to basically kill most of the flow coming off of the short turn so that the air column coming off the back side of the valve seat can be dominant and make a full turn around the cylinder without being hindered by the flow from the short turn. It is far better to put the air/fuel mixture into the chamber correctly in the first place so you don't have to rely on flow robbing drawbacks as swirl. Swirl is for a low rpm daily driver, not a high VE racing engine.

 

Let's see how well I've picked up on Head Porting. Maybe Dennis can check
over my reply.

For a given piston speed at a given RPM range, you will want to ensure the
port is flowing at the most efficient rate while keeping all areas of the port
turbulent/sonic free.

Start off with a scenario in which you have tuned the cylinder head to work
at 6500 RPM at a piston speed of xxx.x feet per second. All areas of the port
are flowing well and the cross section is acceptable for these depressions and
piston speeds.

Now, bump up the RPM to 7500 RPM. The piston speed is faster and the
port air speed should increase. If the cross section at any point in the port
causes flow to go sonic, the fix may be to enlarge the cross section at that
specific point to get the speed sub-sonic and pick up flow once again.
The solution may also include shaping the port to control turbulence.

Take that same head and put it back on a 6500 RPM peak. Piston speeds
are slower, and the port has changed in localized areas (cross section and/or
shape). The air speeds in those areas are not going to be comparable (likely slower) and
the engine may lose power, and lose E.T./MPH on the strip even though the port
has not gone turbulent, or sonic.

I guess the same adage applies:

"all parts of the motor must be tuned for the target RPM intended"

I hope that helps, and I hope my thoughts are accurate based on what I
have read.

 

According to your logic, heads should be ported according to C.I. and intended rpm range. Seems to be heading in the right direction to me.

If you take a given head and increase the C.I beneath it, the port will go sonic and become turbulent at an earlier rpm. Correct??

 

yeah i agree.

i thought i knew stuff before reading this thread, but some of you are freakin' mechanical engineers. all i know is that you can't just throw high flowing heads and a monster cam on a short block and expect it to be right. there's only one head/cam/c.i.d. ratio that's right, and a billion wrong ones. seems like there should be a formula for figuring out the right head/cam combo for a given cubic inch. . . . .quench effect is where it's at....

 

I'm sure there is one, but it would have to be on a sliding scale with rpm range figured in. It is all about airflow. From what I have been able to learn, at a given rpm the ci determines the amount of airflow required.

If the head choice were to remain static, a 346 would require more rpms to develop the same horsepower that a 427 does. The potential is there for the same hp, but the rest of the package must be optimized for the difference in displacement.

I am glad there are other people out there that understand this **** so I can just pay them

 

I don't know about that; I'm only going from literature and not practical
dyno testing, track trials, or flow bench data.

To answer that definitively would require all of the above and years of
hands-on.

If I were to guess based on everything being equal besides stroke, I would
think your statement is correct. With the faster piston speeds at lower RPM,
I would expect the port to go turbulent/sonic sooner. However, now you're
leaving the valve timing aspect on the table...ICL, IVC, valve lift and valve
train stability at lower RPM (improved).

With an increase in bore, there might be other variables which would change
even more such as valve (un)shrouding, air flow in and around the chamber,
quench ratio.

This is the point where I bow out gracefully and wait for the boys with the
million dollar labs and letters behind their names to step in.

 

To expand a little bit on number 2 statement, realize that what he was saying is that port shape plays a more important role then flow numbers. It only takes 350cfm to make 700hp. Why others dont make the horsepower is simple. They arent utilizing the right valve events to fill the cylinder to its capacity. In other words, the combination of valvetrain and heads to camshaft is not optimized.

Every car has its own specific needs for that customers goals. You port and shape the ports to those needs and cam around the numbers.

 

Aside from being extremely expensive- people are generally retarded. You'd have to explain things to them, and most people just want simple answers to complicated questions, no necessarily the best or correct answers. Ref: Politics.

You'd basically need to put an MAF in the intake runner of a working engine if you wanted to take it to that level. And really... you can do that by putting it on the dyno if you want comparitive numbers.