CNC Mill Mark Removal in Intake Port

-
E

Earlie A

Well-Known Member
Joined
Aug 22, 2023
Messages
529
Reaction score
677
Location
TN Hills
This topic came up in another post so I'm just starting a new post on this topic. The question was asked by @dusterbing the peace if testing had been done comparing a CNC ported ribbed finish to a smoothed/sanded finish. Here's one initial test on a factory CNC'd Speedmaster intake port. As a first test, the CNC ribs are smoothed out with a 36 grit cartridge roll. This is a fairly rough finish. In fact, the CNC finish feels smoother that the 36 grit. This CNC finish has a toolpath step length of about 0.040", which is pretty tight. Because the CNC cutting tool is a ball of about 3/8" to 1/2" diameter, the wider the toolpath step length, the rougher the washboard effect. Wider tool paths would make the valley to mountain top dimension greater, which would create a rougher texture.

Here's the results. I want to go on and smooth it down some more, especially in certain areas. Interested in the comments first.

IMG_2644.jpg


IMG_2643.jpg


IMG_2645.jpg


IMG_2646.jpg
 
Wow, you work fast!!!

Interesting to see first pass with a really rough grit; essentially the same as the “washboard” (that term fits perfectly) finish.

Is the test with the valve job as shipped from SM?

Many, many, thanks for doing this!!!
 
other camps want smooth exhaust ports but rough intake runners to aide fuel / air mixing?
 
So with multi port fuel injection and the injector just above the valve… there is no time for fuel atomization at high rpm. And anyone who wants to tell me the length of the runner and head port is enough at 4,000rpm to “mix” and properly atomize fuel and air… I don’t buy it. Will rough runners help? maybe, but explain sheet metal intakes? Manufactures trying to get more fuel miles from a gallon of gasoline do not polish or touch intakes - at all. If there were value they would.

SN: great work by Early A. I appreciate the energy.
 
ir3333 said:
other camps want smooth exhaust ports but rough intake runners to aide fuel / air mixing?
Click to expand...
That is my understanding as well. Really hoping these guys with dyno/track and engine building experience want to share. I have some thoughts on how rough is enough and other thoughts on roughness in specific areas, but my knowledge is mostly based on the flow bench and reading.
 
Last edited:
Syleng1 said:
So with multi port fuel injection and the injector just above the valve… there is no time for fuel atomization at high rpm. And anyone who wants to tell me the length of the runner and head port is enough at 4,000rpm to “mix” and properly atomize fuel and air… I don’t buy it. Will rough runners help? maybe, but explain sheet metal intakes? Manufactures trying to get more fuel miles from a gallon of gasoline do not polish or touch intakes - at all. If there were value they would.

SN: great work by Early A. I appreciate the energy.
Click to expand...
Here's my understanding from some Darin Morgan videos: Intake manifold for dry flow in a MPFI system would be smooth. Downstream of the injectors needs to be rough, 60 grit or rougher. The interesting thing is that upstream of the injectors (for some distance I can't remember) needs to be rough as well. This is because flow is not always in one direction.

For wet flow intake manifolds and runners, 36 grit. For heads, 60 grit or rougher. As velocity goes up, surfaces can be smoother. Since intake manifolds/runners are usually lower velocity than head ports, go with rougher texture.
 
dusterbing the peace said:
Wow, you work fast!!!

Interesting to see first pass with a really rough grit; essentially the same as the “washboard” (that term fits perfectly) finish.

Is the test with the valve job as shipped from SM?

Many, many, thanks for doing this!!!
Click to expand...
Valve job is factory. My disappointment with this head is the large volume (199cc) and the large throat (93% on a 45 degree seat). The throat was the shocker. I think it's losing a lot of flow throughout the range because of that.
 
Earlie A said:
As velocity goes up, surfaces can be smoother.
Click to expand...
What are your thoughts on going smoother (like 120 grit or more) to see if there’s any noticeable effect on flow numbers…fuel mixing concerns aside. My thinking is along the lines of a high revving stroker, where the velocity is actually there to induce mixing vs chuffing around on the streets. I know your next post mentions the oversized aspects of the port, so there may not be much else you can do without touching the valve and seat.

Thank you again for all the testing and info you provide!
 
Earlie A said:
Here's my understanding from some Darin Morgan videos: Intake manifold for dry flow in a MPFI system would be smooth. Downstream of the injectors needs to be rough, 60 grit or rougher. The interesting thing is that upstream of the injectors (for some distance I can't remember) needs to be rough as well. This is because flow is not always in one direction.

For wet flow intake manifolds and runners, 36 grit. For heads, 60 grit or rougher. As velocity goes up, surfaces can be smoother. Since intake manifolds/runners are usually lower velocity than head ports, go with rougher texture.
Click to expand...


If you can (or if you even care to do it) go to Drag Boss Garage on YouTube and look through is Darin Morgan videos.

I can’t remember which one it was but the topic came up.

In that video he said for a wet (carb) intake it should have a rough finish everywhere. If it was injected you use a rough finish 3 inches above the injector to the manifold face.

Chad Spierer only does a burr finish on his intakes. And on some of his ports.

Larry Meaux was interviewed about 20 years ago. I’ve posted the link to it many times. He says he is doing a burr finish everywhere. The rougher the better. That’s the intake manifold, the intake port, chamber and exhaust port.

And he said he found power doing it every time.

What I found was when the step over got big enough the port would go so turbulent you couldn’t read the manometer.

When I was testing burr finishes I never could get it rough enough to duplicate that. And for the most part the burr marks run parallel to the port where the step over runs perpendicular.
 
Earlie A said:
Valve job is factory. My disappointment with this head is the large volume (199cc) and the large throat (93% on a 45 degree seat). The throat was the shocker. I think it's losing a lot of flow throughout the range because of that.
Click to expand...
That head with the short S.S. and massive throat is basically unfixable. I find anything over 90% may look good on the bench but underperform on the dyno and even more so in real world part throttle conditions. 90.5% MAX---93%=JUNK.

As for surface finish what you are really doing is altering boundary layer thickness--I leave everything I do AS Carbided, polishing is a complete waste of time and only makes dust. Will customers complain? Probably. If they don't trust your judgement/expertise you don't want them as a customer anyways. I also wouldn't follow what the flowbench says chasing flownumbers based on surface finish. Which finish makes more power?

As for fuel mixing--If your carb/EFI tune is not atomizing the fuel and you are relying on the surface finish in the port to re-atomize the liquid fuel--your problem isn't surface finish--its the fuelmixer/tuneup. J.Rob
 
It was years ago i read flow test results on a number of intakes from one of the major manufacturers. The intakes with the highest flow did not make the most power. I have no hands on experience but the rough intake port was for better mixing at lower rpms .
 
RAMM said:
That head with the short S.S. and massive throat is basically unfixable. I find anything over 90% may look good on the bench but underperform on the dyno and even more so in real world part throttle conditions. 90.5% MAX---93%=JUNK.

As for surface finish what you are really doing is altering boundary layer thickness--I leave everything I do AS Carbided, polishing is a complete waste of time and only makes dust. Will customers complain? Probably. If they don't trust your judgement/expertise you don't want them as a customer anyways. I also wouldn't follow what the flowbench says chasing flownumbers based on surface finish. Which finish makes more power?

As for fuel mixing--If your carb/EFI tune is not atomizing the fuel and you are relying on the surface finish in the port to re-atomize the liquid fuel--your problem isn't surface finish--its the fuelmixer/tuneup. J.Rob
Click to expand...
Would going to a 2.08 valve help? I think it would only bring the throat to about 90%, which sounds too big still...otherwise it sounds like SM just has a program to hog out the ports for show, not go. Sounds like money would be better spent (at least on SM heads, and in search of higher end power) to get the non-CNC versions in the hands of an experienced person...or pony up for some better heads altogether?
 
dusterbing the peace said:
Would going to a 2.08 valve help? I think it would only bring the throat to about 90%, which sounds too big still...otherwise it sounds like SM just has a program to hog out the ports for show, not go. Sounds like money would be better spent (at least on SM heads, and in search of higher end power) to get the non-CNC versions in the hands of an experienced person...or pony up for some better heads altogether?
Click to expand...
Going to a bigger valve will help but its not really the right way because if those heads are like what I have seen the throat is chopped completely straight down. A 2.100" may even be the best "fix" but then the rest of the port probably won't play well with a larger valve. A smaller seat ring and some welding or good Belzona grade epoxy would be the right way. J.Rob
 
I’ve been blessed with customers that leave the port finish up to me. Exhaust ports get a sanding roll finish. The intake ports get a burr finish. Not a bent burr finish but a common everyday burr finish. I’ve seen some beautifully finished ports and they look great in pictures and probably get the head porter some business but to me that’s oldschool thinking. A good head porter can gain flow on any cnc’d head but it’s from material removal not just buffing a ribbed finish. As far as the large throat size I’ve never seen it what I considered oversized on the non cnc’d Speedmaster heads but it’s at max size in my eyes. If I get 290 cfm out of a 2.02 valve I’m satisfied because these heads beg for a 2.055 valve size so the head porter can do the needed valve seat work.
 
Imo, the “correct” way to evaluate changes to the intake port surface finish is on the dyno, not the flow bench.
And I don’t work on anything where doing that kind of R&D is cost effective.

I’ve seen what I felt were pretty impressive results from smooth, and rough, and cnc finished ports.

I don’t feel like a different surface is going “fix” a bad port.

If I think of some of the better combos I’ve built, where I ported the heads myself, those heads had the smooth finish.

There was a thread about this subject on YB a few years ago, and I remember Chris Uratchko commenting that one of his best builds had smooth ports.
 
What I did this morning with the cartridge roll took 15 minutes for 1 port and was almost foolproof. Sandpaper is just not that aggressive. If Joe average put a little Gorilla tape over the valve seats to protect them and was careful with the sander, there is no reason a 60 grit or 36 grit finish could not be done at home. The carbide burr is a different animal. It can get away from you and do damage very quickly. Best left to the professionals. I'm not advocating the sanded finish, just saying it's an option for us average guys where the burr finish is not.

I still want to sand the SSR to a smoother finish just to see what it does on the bench.

Thanks for all the input. What a great group of responses.
 
I appreciate your efforts. A wet flow runner needs help keeping or returning raw fuel to an atomized state. The worse the runner, the more help you need to avoid having raw fuel run down into the port. Total CFM flow being effected by surface finish would be incredibly small if any. Quality of the mixture far outweighs any loss in flow there might be. I run a “bad” intake, you can see where I’m just slamming raw fuel onto the plenum floor. I need to remix that fuel. Very unscientific, but I take the fact that the fuel staining peeters out before getting an inch out of the plenum as a sign my rough plenum floor is helping.
 
One more thought………
If trying to determine if/how much the surface finish may impact flow numbers on the bench…….
I feel that the “test piece” has to be one without inherent flaws to its shape.
One that doesn’t have a compromised form, but does have a well developed valve seat & chamber area.
No meaningful flow turbulence or flow regression until well past the usable lift range.

And even then, I’m not sure how accurate the flow results would be in determining how the different finishes impact the power output.
 
RAMM said:
.... may look good on the bench but underperform on the dyno and even more so in real world part throttle conditions.
Click to expand...
thank you for this. 1000x thank you for succinctly putting it in the words that everybody can understand.
 
Unfortunately until someone's does the dyno and track testing with a wide variety of combinations to figure it out it's probably just best guesstimation what works.
 
Here's the final two tests just for fun. First, sand entire port with 60 grit cartridge roll. Second, hand sand the SSR only with 180 grit cloth backed sandpaper.

The 180 grit hand sand made a difference. I got similar results a few months ago when I hand polished the SSR. As was mentioned earlier in the post, each time the port is sanded the shape changes a little. When you sand with a cartridge roll on a convex curve it is difficult not to leave a series of cascading flat surfaces because the cartridge roll is straight. Hand sanding the SSR as the final step may have just smoothed out the straights. So the slight gain could be more because of shape than texture.

One other quick thought about flow testing. It is tough not to let human bias slip into testing. My preconceived notion about this test was that the 180 grit hand sanding would help. So while I really tried not to do anything to make the 180 grit win, I found myself taking a little more time checking the details to make sure it did not lose. For instance, I have a wood entry plate that I use on the entrance to the port. It does not fit exactly right because it is a generic plate. There is always a little clay that has to be used to smooth things out. Did I take a little more time and do a better job smoothing the clay out on the final test? No such thing as completely unbiased testing. No such thing as completely objective humans.

One final thought. IMO, 1 cfm change from test to test can be ignored. 2 cfm change might be real. 3 cfm change and something is really going on. A small change to the entry plate can easily cause 3-4 cfm change.

IMG_2648.jpg


IMG_2647.jpg
 
So what I see is the cnc finish produced the best numbers. Since I started used Killer carbide burrs I could probably count the times I’ve used a double cut burr on 4-5 fingers on an intake port. Burr finish entrance to exit for me.
 
pittsburghracer said:
So what I see is the cnc finish produced the best numbers. Since I started used Killer carbide burrs I could probably count the times I’ve used a double cut burr on 4-5 fingers on an intake port. Burr finish entrance to exit for me.
Click to expand...
You're right. Or, looking at it another way, don't let me touch your ports with a cartridge roll!
 
Earlie A said:
You're right. Or, looking at it another way, don't let me touch your ports with a cartridge roll!
Click to expand...


The funny thing is a sanding roll let’s you see all the dips and gullies you have created with your burrs and you would think it would let you create the perfect shape
 
My initial thoughts about that series of tests is……there needs to be 2 more tests.

Baseline another untouched port, then do nothing but sand the SSR.
 
You must log in or register to reply here.
-

Back
Top