Port texture and velocity

[mag2555]

A textured port/ runner wall surface is not a cure all for all wet flow issues in regards to the wet flow changing to a vapor.

Fuel being in vaporized state is a good thing until in coming air temps or running temps get to the point where too much fuel is now changed into a vapor, and that vapor now being a gas takes up room that should be had for air and this condition brings on a drop in power of course.

[4vpc]

Does it matter what finish you put on an exhaust port anyhow? As soon as you turn the key it's covered in soot and it just gets thicker the more you run it.

[mag2555]

Not to be a wise ass, but if your Exh ports are getting covered in soot then you have a big time oil control issue!
A lot of Exh ports in race heads have too great of a area expansion rate as compared to the valve bowl Throat size, so carbon build up/ ruffness may only be a issue in the valve bowl.

To that end the last thing I do after the normal polish job is to go back in with a 320 grit sanding roll to get the needed mirror like finish.
Carbon scale can be a flow factor, not a smooth covering of soot.

[70GS455]

Intake ports really only need to be textured on the short side radius. The idea is that texture creates a boundary layer (changing the Reynolds number), helping to keep the flow attached as it rounds the SSR, making the turn into the cylinder. This is why a dimpled golf ball flies farther than smooth

[SpeierRacingHeads]

Intake ports really only need to be textured on the short side radius. The idea is that texture creates a boundary layer (changing the Reynolds number), helping to keep the flow attached as it rounds the SSR, making the turn into the cylinder. This is why a dimpled golf ball flies farther than smooth

totally wrong. short sides need polished because the layer is fast.

[KnightEngines]

Intake ports really only need to be textured on the short side radius. The idea is that texture creates a boundary layer (changing the Reynolds number), helping to keep the flow attached as it rounds the SSR, making the turn into the cylinder. This is why a dimpled golf ball flies farther than smooth

totally wrong. short sides need polished because the layer is fast.

2nd that.
I'll leave a whole port carbide finished, but sand the turns.

In fact, I just did that today on a near max effort pair of heads.
Without sanding the turn they get turbulent, separate & flow backs up 8-10cfm at .750" lift, sand the turns & stable out to .850".

[BradH]

... short sides need polished because the layer is fast.

2nd that.
I'll leave a whole port carbide finished, but sand the turns.

In fact, I just did that today on a near max effort pair of heads.
Without sanding the turn they get turbulent, separate & flow backs up 8-10cfm at .750" lift, sand the turns & stable out to .850".

Anything specific w/ the sanding? Basic side-to-side (crossways) or inline w/ the flow? What grit rolls are used?

Don Terrill recommended using something like 36-40 grit strips hand-sanded in the direction of flow to create "riblets" on the short turn. That was something intended to reduce the drag in the boundary layer, if I understand the reasoning correctly. What I never researched is whether that type of surface texture also reduced the tendency for the flow to separate, as well.

[BradH]

Intake ports really only need to be textured on the short side radius. The idea is that texture creates a boundary layer (changing the Reynolds number), helping to keep the flow attached as it rounds the SSR, making the turn into the cylinder. This is why a dimpled golf ball flies farther than smooth

Interesting...

Chad S. and Knight Engines both disagree.

Two different research papers I've just looked through seem to reach opposing conclusions. I've tried to not take the individual quotes from each out of context.

1. https://www.cambridge.org/core/journal ... 8CFB1033FB

"... Compared with a smooth-wall case, streamline detachment occurs earlier and the separation region is substantially larger for the rough-wall case, due to the momentum deficit caused by the roughness. The adverse pressure gradient decreases the form drag, so that the point where the wall stress vanishes does not coincide with the detachment of the flow from the surface."

2. https://jeb.biologists.org/content/jexb ... 6.full.pdf

"... The high-resolution boundary layer measurements show that the flow over rough wings is indeed laminar at low angle-of-attack and Reynolds number, but becomes turbulent at higher values. In contrast, the boundary layer over the smooth wing forms open laminar separation bubbles (enclosed separation zones) that extend beyond the trailing edge. The boundary layer dynamics of the smooth surface varies nonlinear as a function of angle-of-attack and Reynolds number, whereas the rough surface boasts more consistent turbulent boundary layer dynamics."

The second one has a lot of interesting info that combines different surface textures / angle of attack / Reynolds number scenarios. The angle of attack and Reynolds number variables themselves appear to have as much influence on the results as the surface textures.

Almost sounds like another big "It depends..." answer.

[digger]

Intake ports really only need to be textured on the short side radius. The idea is that texture creates a boundary layer (changing the Reynolds number), helping to keep the flow attached as it rounds the SSR, making the turn into the cylinder. This is why a dimpled golf ball flies farther than smooth

Interesting...

Chad S. and Knight Engines both disagree.

Two different research papers I've just looked through seem to reach opposing conclusions. I've tried to not take the individual quotes from each out of context.

1. https://www.cambridge.org/core/journal ... 8CFB1033FB

"... Compared with a smooth-wall case, streamline detachment occurs earlier and the separation region is substantially larger for the rough-wall case, due to the momentum deficit caused by the roughness. The adverse pressure gradient decreases the form drag, so that the point where the wall stress vanishes does not coincide with the detachment of the flow from the surface."

2. https://jeb.biologists.org/content/jexb ... 6.full.pdf

"... The high-resolution boundary layer measurements show that the flow over rough wings is indeed laminar at low angle-of-attack and Reynolds number, but becomes turbulent at higher values. In contrast, the boundary layer over the smooth wing forms open laminar separation bubbles (enclosed separation zones) that extend beyond the trailing edge. The boundary layer dynamics of the smooth surface varies nonlinear as a function of angle-of-attack and Reynolds number, whereas the rough surface boasts more consistent turbulent boundary layer dynamics."

The second one has a lot of interesting info that combines different surface textures / angle of attack / Reynolds number scenarios. The angle of attack and Reynolds number variables themselves appear to have as much influence on the results as the surface textures.

Almost sounds like another big "It depends..." answer.

IMO there is a difference between flow through a pipe vs over an object. Also one of those papers is low Reynolds number i.e laminar flow

i see it like this for turbulent flow in a pipe which might be totally wrong, if the surface is smoother the flow has less tendency to stay attached but also boundary layer is thinner so the port acts like it is larger slowing down the speed which counters it.

flow over a wing or cylindrical object is not "constrained" so there is no effect of the change is size like with a pipe, it just stays less attached in the smooth case.

so i always thought if the rough port helped power in part it was because the size was too big. Obviously there is a fuel wetting thing to consider. it should be relatively straight forward looking at air consumption and BSFC on the dyno to gain a fairly decent understanding

[KnightEngines]

... short sides need polished because the layer is fast.

2nd that.
I'll leave a whole port carbide finished, but sand the turns.

In fact, I just did that today on a near max effort pair of heads.
Without sanding the turn they get turbulent, separate & flow backs up 8-10cfm at .750" lift, sand the turns & stable out to .850".

Anything specific w/ the sanding? Basic side-to-side (crossways) or inline w/ the flow? What grit rolls are used?

Don Terrill recommended using something like 36-40 grit strips hand-sanded in the direction of flow to create "riblets" on the short turn. That was something intended to reduce the drag in the boundary layer, if I understand the reasoning correctly. What I never researched is whether that type of surface texture also reduced the tendency for the flow to separate, as well.

Side to side sanding roll with 80 grit for the most part.
For twitchy turns I'll use a strip of 80 grit tape fed through the port & pulled back & forth over the turn to polish it.

[Erland Cox]

Where the flow is slow, the boundary layer is thick.
Where the flow is fast, the boundary layer is thin.
Texturing is used to stop the boundary layer from growing so it is not needed where it is thin over the crown.
But past the turn where the flow slows down texturing helps to turn the flow.
Sometimes leaving the bottom cut sharp also helps the air turn.
Sometimes smoothing in the bottom cut works better.

Erland