44 replies to this topic

[malbear]

My friend "dave" and I have spent the afternoon arguing over the appropriate finish for intake and exhaust ports. He favours the mirror finish and I favour the matt or sharkskin aproach. I can see how a mirror finish may be of benefit within the combustion chamber to limit thermal losses
but a matt finnish would aid flow and help fuel vaporisation from the shower injectors above the intake trumpets. anybody like to step up to the podium and put in their 5c ? Has anybody seen the insides of the current crop of V8 F1 engines?

[J. Edlund]

Originally posted by malbear
My friend "dave" and I have spent the afternoon arguing over the appropriate finish for intake and exhaust ports. He favours the mirror finish and I favour the matt or sharkskin aproach. I can see how a mirror finish may be of benefit within the combustion chamber to limit thermal losses
but a matt finnish would aid flow and help fuel vaporisation from the shower injectors above the intake trumpets. anybody like to step up to the podium and put in their 5c ? Has anybody seen the insides of the current crop of V8 F1 engines?

a matt finnish would aid flow

Quite the opposite actually. The mirror finish will aid flow (reduce pressure loss). The only case where a "rough" finish aids flow is where flow separation rather than skin friction is the issue.

"help fuel vaporisation"

Possebly, haven't seen any facts that either support or disprove it. With good fuel injectors I doubt that it will affect anything though. With direct fuel injection it isn't a problem at all.

"Has anybody seen the insides of the current crop of V8 F1 engines?"

Haven't seen the latest F1 V8 engines but most racing heads tend to use a rather high finish.

In the exhaust ports a mirror finish will also make it harder for soot to stuck to the walls, and reduce the heat transfer from the gas to the walls.

[malbear]

why do golf balls have dimples?

[sblick]

Golf balls have dimples for aerodynamics IIRC. A spherical object with a smooth skin doesn't travel very far, and is more apt to take crazy turns if struck incorrectly (spin causing hook or fade). The dimples control the flow seperation off the back of the ball. The dimples do not translate to the environment you are thinking of. Seperate the spherical environment from a tubular shaped environment. Air flows over a ball and is unconstrained in its environment. Air flows through the intake runner (tubular) and into the head of the engine and is constrained to the shape of the runner.

[malbear]

wouldn't it be better if the valve stem valve guide and port floor had dimples so that the air flow around these adhered to the shape instead of breaking into vortices behind these objects in the flow path

http://www.aerospace...ics/q0215.shtml
http://www.aerospace...ics/q0215.shtml

[GeorgeTheCar]

The flight of a gofl ball and flow through a runner or port are two completely different aerodynamic regeimes.

A golf ball needs to fly straight and long and the dimples, interacting with the rotation of the ball, assit that goal.

A ruuuner and port have to flow the maximum amount of air inspite of having their space invaded by the presence of a valve stem and other elelments as well as being interrupted by the valve action up to 10,000 times a minute. Empirical eveidence show that polishing the runners and other exposed surfaces enhance flow and output.

[Canuck]

To date I have not heard that a polished intake runner is of any benefit, and is in fact of significant detriment to performance on typical engines. Biggest cause of performance loss being fuel separation/puddling. OTOH, as has already been pointed out, polished exhaust ports seem to work rather well.

I have never seen an all-out engine with polished intake ports beyond the point of fuel entry but I won't claim that it ain't so.

[GeorgeTheCar]

I lack sufficient imagination to think that at the engine speeds in modern engines fuel has enough time to pool, not to mention anything about the pressure in the fuel injections systems.

I also remember an article somwhere about someone, Smokey Yunick I think, who developed pultrusion to gain the goal of smoother runners for more power.

To understand the issue fully we also need to recognize that a production engine which has to operate under a broader set of conditions, includiing lower speeds, closed throttle and colder temperatures might have need of a different set of conditions to preculde puddling and other issues.

[McGuire]

I have seen enough on the wet flow bench and dyno to convince me that a relatively rough intake port finish makes more power by promoting fuel breakup. By "relatively rough" I mean an intermediate paper-roll finish, decidedly coarse in feel, or CNC with little or no cleaning up.

(I say "breakup" only because I dislike the terms "vaporization" and "atomization" so much. In reality we are dealing not so much with fogs and droplets but big nasty globs and strings of liquid fuel, whether we are dealing with individual port injectors or carburetors.)

My first look into a good wet flow bench was an eye-opener. One of the first things you discover is that no matter how and where the fuel is introduced to the intake port, the first thing it will do is head straight for the port walls, as if drawn there. But if the wall is relatively rough, it will help kick the adhered fuel out into the airflow again, especially at the port turn radii. (The wet flow guys like to call these areas "shear points." A slick finish creates a relatively dead "shear point" -- the fuel just hangs there.

... in general, one of the main things wet flow study illustrates is that fuel is a whole lot heavier than air... but that's a subject in itself.

I don't think surface finish makes much if any difference at all on the exhaust.

And as J. Edlund notes, with GDI it won't make any difference in fuel delivery on the intakes either.

[malbear]

Hank s Guys
so the consenus seems to be polish the exhaust port to keep carbon deposits at bay for a period and only polish the intake if there is no fuel present. I wonder what efect a set of reed valves in the intake would have on kicking any puddling of fuel and breaking it up. Has any body tried it with a wet flowbench.
when my wife jibs me for not using that little 4 letter term of endearment I say that it is the empathy in the conversation that matters not the actual content so hank yous
I wonder if it is sometimes the triumph of asthetics over practicality that drives the polishing like chopper riders and 4 wheel drivers with tires that never see dirt.

[dosco]

Originally posted by McGuire
By "relatively rough" I mean an intermediate paper-roll finish, decidedly coarse in feel, or CNC with little or no cleaning up.

Say 60 to 125 microinches RMS?

[Halfwitt]

Originally posted by malbear
Hank s Guys
so the consenus seems to be polish the exhaust port to keep carbon deposits at bay for a period and only polish the intake if there is no fuel present. I wonder what efect a set of reed valves in the intake would have on kicking any puddling of fuel and breaking it up. Has any body tried it with a wet flowbench.
when my wife jibs me for not using that little 4 letter term of endearment I say that it is the empathy in the conversation that matters not the actual content so hank yous
I wonder if it is sometimes the triumph of asthetics over practicality that drives the polishing like chopper riders and 4 wheel drivers with tires that never see dirt.

I've seen inside the intakes of some quite recent F1 heads, and while they're not mirror-polished, they are pretty smooth, and these are engines with overhead injectors.

Reeds in the intake would cause more problems than they solve. For a start the wave reflections which determine the optimum intake length would be arriving at the wrong place and at the wrong time. Someone did a one-off motorcycle engine about 15 years ago like this, and I think it wasn't very succesful (Performance Bikes magazine?)

[Sharman]

Two points, there is a hell of a difference between a producrtion engine and a racing engine . The one is designed for ease of manufacture and the other without much regard to that factor.
One of the most successful engine tuners was Danny Richmomd of Downton Engineering of blessed memory. The B type BMC they did for me was mirror finish induction side and combustion chamber and realtively rough on exhaust. Mind you I'm talking nearly 50 years ago.
George Mangoletsi said to me "if you can get enough in and burn it all, the waste will get out somehow".

[malbear]

I think that Honda did a 4stroke big single dirt bike with reeds on the intake and Ive heard reports that the riders used to take them out to get more top end performance but lose a little on the low rev performance XR500 1980

[Greg Locock]

There was a series of photos around showing what really happens in a typical carbied intake, using a perspex intake. The term globule is misleading. PUDDLE would be a better word. The fuel lies there with occasional droplets being whisked away by the passing breeze.

[Canuck]

Interesting that there's so much difference of opinion on this issue. I've always been led to believe that a polished intake tract was the hallmark of someone that didn't truly understand what they were up to. However having been humbled on here a number of times, I tend to shy away from many absolutes these days. Given McGuire's experience on the wet bench, Greg's observations and my own experience, I'm going to continue leaving the intake tracts unpolished on my own work.

[McGuire]

Originally posted by Canuck
Interesting that there's so much difference of opinion on this issue.

Yep, it's one of those subjects engine builders have been debating since the dawn of time.

But in recent years the technical capability has progressed to the point that many of these old disputes are gradually being put to rest one way or the other: Art is displaced by science. That makes this sort of an interesting time to be around, all things considered.

[desmo]

The older ('90s) Ferrari F1 intakes I poked my fingers into had a surface finish I'd guesstimate to be 100-150 grit in American nomenclature.

[cheapracer]

My Mate was Serco's (Brisbane, Australia, Wiseco pistons etc.) port & polish - flow bench man for 21 years and he says rough (re Desmo 100-150 grit finish comment) was best for flow but polish sold the heads.

[cheapracer]

Sorry I should mention I'm commenting on inlet tract.

I know some tuners who swear for half/half, polished roof - rough floor.

Polished exhaust always.

[Stefan_VTi]

I too found that all walls with wet flow should be a bit rough (I do 240/280 grit, in flow direction), the other walls can be made to mirror polish for a bit better flow and less carbon buildup.

On a production car I would look at the carbon pattern from the stock injector, as you can see where the fuel starts to contact the intake runner.

Would the step that is commonly used in intake component mating surfaces also promote the pickup of fuel droplets 'rolling' along the intake wall? (I would think so)

[NRoshier]

Georgethecar makes a valid point. In some research I did for a magazine article there were strong hints from well known engine tuning companies that port finish (for a racing engine) was dependent on port shape and intake velocity...and was not polished.

[shaun979]

I've never seen an intake port finished to higher than 120 grit, with nearly everything 80-100 grit. 80 grit cartridge rolls run at high speed and low pressure can finish up like standard 120 grit. I'm not sure where the 150, 240, 280 grit numbers are coming from? The last two numbers would literally be like a mirror!

[malbear]

Dam I ordered a box of 150 grit last week
I quess I can always hand finish with 80 grit
My new head design ( 500 Yam ) is flowing 178 cfm exhaust and 270 cfm intake rough diegrind finish Ill let you know the 150 grit finish results.

[McGuire]

Originally posted by shaun979
I've never seen an intake port finished to higher than 120 grit, with nearly everything 80-100 grit.

Back in the Day, intake and exhaust ports were polished to a mirror finish -- hence the term "ported and polished." To many it was the sign of a quality job.

Looking back I can see how this thinking developed. Guys were fumbling around in the dark with very little idea of what really worked and few ways to truly quantify it, relatively speaking, but everyone was trying as hard as they could.

[Sharman]

McGuire
That goes back to what i said in a earlier post. I did, at the same time, mention that it was nearly 50 years ago
JF

[cheapracer]

Originally posted by McGuire


Back in the Day, intake and exhaust ports were polished to a mirror finish -- hence the term "ported and polished." To many it was the sign of a quality job.

Looking back I can see how this thinking developed. Guys were fumbling around in the dark with very little idea of what really worked and few ways to truly quantify it, relatively speaking, but everyone was trying as hard as they could.

Flow benches have changed a lot of thinking and its amazing just how many ports they show up that actually need metal ADDED and the port made SMALLER.

[malbear]

well I smoothed out the ports with the 150 grit rolls and then hand finished with some 100grit and retested on my www.flowbenchoz.com/ The results show only a slight improvement from the rough finish. Exhaust 179.5 CFM and intake 271 CFM at 28 inches. A bit of an improvement on the standard head of 155 CFM exhaust and 182.5 CFM intake
http://www.sixstroke.com/index.html

[desmo]

As mentioned by McGuire earlier, it seems the advantages of a rough intake port finish are more or less invisible to a dry flow bench.

[Canuck]

The proof as they say, is in the pudding. How is your throttle response? Mileage? Acceleration?

[ray b]

so why not direct inject instead of port inject fuel?
strictly F-1 or other full race motors
I under stand costs but in F-1 costs are not as big a deal as power
if this would give a little extra power it sure looks do able
even have an odd idea of useing hot spots to vaporize and cool
useing direct injection so why is this not done

btw I agree semi ruff is better for flow then mirror smooth

[desmo]

Haven't Audi done DI spark for Le Mans? I think so

I was given to understand there were some inherant problems trying to get F1-type rpms to work with DI.

[Engineguy]

DI is banned in the F1 rules.

[J. Edlund]

Originally posted by desmo
Haven't Audi done DI spark for Le Mans? I think so

I was given to understand there were some inherant problems trying to get F1-type rpms to work with DI.

Mario Illien mentioned in an interview that if it hadn't been banned, they probably would have had a working direct injection system in F1.

Originally posted by ray b
btw I agree semi ruff is better for flow then mirror smooth

Actually, mirror smooth is better for flow but worse for fuel brakeup. This means that there is some ideal solution between a very rough surface and a very smooth surface.

[malbear]

I think that the problem with DI in F1 would be that at very high RPM above 8000 or more is that the injectors cannot keep pace with the engine revs and need to be at 100% duty cycle . meaning that they would be injecting during the exhaust and expansion . But a shower injector above the inlet trumpet can go to 100% duty cycle and vary the pressure to vary the fuel supplyto keep it at 12.5 to 1 AFR. as long as the inlet runners are finished to 100 gritt The cooling efect of the fuel also increases the charge density

[McGuire]

Originally posted by J. Edlund



Actually, mirror smooth is better for flow but worse for fuel brakeup. This means that there is some ideal solution between a very rough surface and a very smooth surface.

Naturally, a specific surface finish has been identified and quanitified by people working on it... but that is not my info to blab around. Without talking out of school I think it is ok to say that it's pretty darn rough compared to recent practice.

[ray b]

has anyone seen the audi or other diesel racecar ports
as they are latest hi-teck DI that should be the test
of current air only flow in heads

[malbear]

[IMG]http://fueleconomyti.../powrelynz1.jpg[IMG]

How about this [URL]http://fueleconomyti...elynz1.jpg]port[URL]

[Bill Sherwood]

A flowbench won't give you the full picture, as they don't recreate the very high airspeeds the inlet flow gets to.
And yes, a slightly rough surface is the way to go.

[cheapracer]

...which, Bill, leads to sonics, an art in itself. I used to whip up a few expansion chambers and can say that the math doesnt always work in the real world.

I recall a 70's Renault Le Mans class winning racer (4 cyl, 2 litre) having inlet tracts over 2 feet long - against all the rules.

[NRoshier]

if it is working then it is showing that this particular engine uses different rules or that the rules are not properly understood.

[cosworth bdg]

Every brand of engine is different....

[cheapracer]

Originally posted by NRoshier
if it is working then it is showing that this particular engine uses different rules or that the rules are not properly understood.

Sorry, your refering to the Renault engine?

If so it won at Le Man (class).

Inlet tracts so long get narrower and narrower down the tract as surface tension intervenes, not suitable for a high rpm engine - at least thats the thought rule!

[Stefan_VTi]

They balance wall friction (shorter port, less surface friction) to packaging & harmonics (intake runners that long probably use a lower order harmonics for better pulse strength). My guess would be they found that for that particular set of rules and engine design decisions it was best to go for improved harmonics instead of lower wall friction, package weight and size.

[Christiaan]

Guys, I may be speaking out of my ass here but I seem to remember something about a rough pipe better flow charactoristics than a smooth pipe. I was in uni about 10years ago, I don't quite remember. If this is one of those things engine builders have been argueing for so long, I am intigued as to why there has been no resolution. Whats so difficult about conducting an experiment