BACK to the links page

The information used on the page belongs to Scott Potter. I found part of the article while searching on google. I tracked him down and he very kindly sent me the entire document. His Alfa related web site still only shows part of the document.

He also added that the 3.0 valves work well in the 2.5 head with similar porting techniques.

Scott now specializes in another Italian beast; the mighty Laverda motocycle. He can be found in the "rolling hills of central Texas" and on the web at

Thank you Scott!

Street Porting the 2.5 Alfa Romeo V6 Cylinder Head

This is a step-by-step guide to porting the standard 2.5 head as fitted to the GTV-6, Milano and 75 models utilizing the stock valve sizes.

First, some basics lifted from the Superflow 110 Operating Manual:

"The horsepower of an engine is directly proportional to the amount of air drawn into the cylinder and retained until ignition occurs. By reducing the air flow resistance of the intake and exhaust tract, cylinder filling is improved and engine horsepower increased directly."

"Total flow through the engine is ultimately determined by the valve diameters. Valve size is limited by the diameter of the cylinder bore. Generally, hemi head engines permit valves up to .57 times the bore diameter due to extra space in the combustion chamber. "

"Tests indicate there is generally no power improvement as long as exhaust flow is greater than 60% of intake flow. This would dictate an exhaust valve diameter of .77 to .80 times as large as the intake valve."[1]

So let's look at one of our favorites the 2.5 V6: 88mm bore, 41 x 36.5mm valves, intake manifold inside diameter ~34.67mm.

41mm intake is only .465 times bore diameter.

41mm x .85 = 34.85mm optimum venturi size.

36.5mm exhaust is .89 times intake valve diameter.

So our intake is too small, but efficiently sized and our exhaust is too big.

Let's look at what we can do to improve the efficiency of the existing valve size combination.

Besides the basics of a clean well lit workspace and a supply of compressed air (this does give my tired little 5hp 60 gallon compressor a workout) here's what I use:

BTW you can use an electric die grinder

Once the head is stripped of moving parts and cleaned you can see the valve pockets appear to have been roughed out at the factory , no doubt by some one in a hurry, prior to the installation of valve seats. Sometimes the pockets line up with the seats and sometime times they don't and I have seen them opened up larger than the seat insert.

The intakes tend to be especially dismal as the designed port does not allow a straight shot to the valve like the old four cylinder heads. The rough shaping of the casting along with the fellow in a hurry on the assembly line conspire against a smooth transition around this bend with turbulence and separation. As flow losses occur from changes in direction and decreases in velocity I want to smooth the directional flow and keep the velocity high. There is free power just waiting to be unleashed with just a little bit of time and conscientious effort. Follow me as I go along.

First I rough out the pocket with the ½" burr. Easy does it as the coarse flutes can remove a lot of material in short order.

I follow with a 70 degree cutter to open the inside of the valve seat to increase the net valve area [2]. The inlet side I open from 1.44 (36.6mm), yours may vary, to 1.525" (38.7mm). Cool, huh? It's like putting in a 2mm larger valve. The exhaust side is opened from 1.218" (30.9mm), again, yours may vary, to 1.30" (33mm). Another 2mm! For those of you into building headers, 1 ½" head pipes are perfect. The second image following gives a good view of the cut to open the seat.

One more time with the ½" and I do a rough cut on the inside of the kink of the inlet port to give it the general shape of a curve. Again, not too much as I'm not hogging this area out, just reshaping the curve. I'll get my inside dividers out and set them to 85% of my inlet valve area so I don't go too far as I don't want to kill my port velocity.

Why 85%?[3] In the classic formula for maximum flow, the ideal intake system would have a single carb throat per cylinder with a slide plate throttle and a venturi equal to .85 times the intake valve diameter. Below the venturi, the carb bore should gradually open up to the size of the intake valve at the manifold entrance and gradually taper down to .85 times the intake valve diameter at a point about ½" or 13mm below the valve seat.

Notice that 85% of the inlet net valve area is the same size as the exhaust net valve area. Again, tests indicate there is generally no power improvement as long as exhaust flow is greater than 60% of intake flow. This would dictate and exhaust valve diameter of .77 to .80 times as large as the intake valve.

As can be seen, this exhaust is oversized in relation to the intake. Work here will be limited to smoothing out the rough work done on the assembly line and giving the entire port a consistent finish. Even with a larger inlet valve the exhaust needs very little work and no changes to the port dimensions past the pocket or bowl.

In fact our optimal inlet valve size in relation to the cylinder bore works out like this: 88mm x .57 = 50.17mm; 50mm x .77 = 38.5mm. These should be the optimum sizes, but I doubt there's room for a 50mm valve seat. Yeah, I know, you're thinking about the spark plug hole…it can always be reduced for a 10mm plug, gaining valve seat insert room, but I'd be concerned with the seat protruding into the combustion chamber. Using a 46mm intake would give us a good relationship with our exhaust valve and maybe a seat will fit without fouling the combustion chamber. I haven't tried this yet, but it is known the 44mm inlet valve and seat from a 3.0 litre will work, but that is for another article. Let's get back to the business of porting, shall we?

Next I chuck the 3/8" flame burr. The long shank lets me get well into the port from either end. Now I smooth out my rough cut. The fine flutes of the flame leave a good finish on the port walls. The kink is made into a smooth contour watching to not open the throat more than 85%. The point of the flame allows me to get close to the guide. The guides need not be removed unless they are worn enough to require replacement. It is easy to see the 70 degree cut of the inside of the valve seat in this view.

One of the most critical areas for optimal flow is known as the short side radius. It is here is where the bottom of the port turns into the seat and the incoming flow must make this turn with out detaching from the walls and going straight interrupting the rest of the incoming flow. This should be a nice smooth radius, Very little needs to be removed here, just a smooth blending of the radius to allow the flow to gradually make the turn, reducing flow loss. This is difficult to photograph. The short side is on the right in this image. it is easier to feel than to see.

The port walls can be gently opened where they meet the floor of the port along the short side radius, widening the radius in to a slight "D" shape where the straight back of the "D" is the floor and the short side radius.

If you are keeping the guides, you are nearly done here. This stage of the porting and this finish of the port walls is a quantum leap over the factory job. You could go straight to a valve job and feel good in the knowledge you have made a noticeable difference. Skip on ahead to the matching of the inlet face. From here on the returns are less for the time involved.

If the guides are junk, and these are, it is time to knock them out. A big mallet and a proper sized driver (tool A.3.0134 or similar turned on a lathe) makes easy work of it. No heating of the head is necessary, drive them out from the seat side. Refer to the factory workshop manual for more details.

Once the guides are out I use the ¾" ball to smooth around the guide opening. I also use the ball to smooth out any casting marks present along the the port wall without changing the shape of the port wall. The diameter of the ball and the double cut flutes enhances the overall quality of finish in the port. I smooth the entire port without major material removal and I will periodically check with my dividers to keep my nominal dimensions intact through out this port and match the remainder to this one.

Next I address the inlet face at the manifold where there is another example of a rough job from the assembly line. Using the ½" burr I blend the port wall to the inlet face without changing the opening of the inlet face. By the way, I have seen cases where the manifold sits askew of the port, check with an inlet gasket and scribe a new port opening if necessary and blend to this new mark. Use a spray dye or marker on the inlet face then lay a gasket in place and scribe a line along the inner gasket edge.

Note the port dimensions from inlet face to throat are basically unchanged. This keeps the velocity high in the port, the throat acting as a venturi to speed up the airflow to the bowl before passing through the seat.

Now the sanding drums are used to give a consistent surface finish to the ports. And there you have it. Install guides and a 3 angle valve job in the next installment.

Note: the greatest flow loss in the inlet port is due to the expansion of the air out of the valve. This makes the area from ½" below the valve to ½" above the valve the most critical area. The valve seat has a considerable effect on the flow.

[1] Superflow 110 Operating Manual. (back)

[2] For the purpose of this article the Net Valve Area is defined as the inside diameter of the 45 degree angle in the valve seat. (back)

[3] For an interesting argument using a 65% cross section in 4 valve engines I direct you to (back)

BACK to the links page