Project E7, Part III
Intake Port

In Part II of Project E7, we made some significant improvements in flow on the intake port with some simple and basic porting. Trust me, there was still a lot of air in that intake port.

Flow testing was done on a micro flow bench from Flow Performance. A 4.5 hp shop vac supplied the air and all CFM values have been calculated to 28" water column by the Flow Performance FP1 electronic flow bench processor. Even using a shop vacuum, the sensitivity of this flow bench clearly displayed even the smallest changes made to the port or valve on this project as you can see below in the chart.

Before and after photos of the E7TE valve pocket. You can't really see clearly, but the wall on the left side of the pocket, the outboard wall, has been cut back beyond the valve seat on ported example.

I wish I could tell you that using superior intellect and skill, I was able to produce the outstanding results that you will see here. Unfortunately, this is not the case. I dabbled and fussed over this intake port for quite awhile, making hardly any progress. Then I started looking at a GT40P head I have and began to wonder if I could copy the GT40P port profiles onto the E7. The moment I started to copy the GT40P port profiles, big gains began to appear. So, for the rest of Project E7 Part III, we will mostly be dissecting and copying a GT40P intake port.

Over 50 flow tests were performed for Part III alone, but I am only going to chart the ones that made gains or losses.

Lift Stock Part II 1 2 3 4 5 6 7 8 9 10 11 12 Gain
.1" 58 67 69 68 69 66 68 62 61 63 61 62 65 64 +7
.2" 112 119 119 121 122 121 121 118 118 119 117 118 119 121 +7
.3" 141 153 152 154 157 154 156 157 154 153 154 153 154 154 +13
.4" 153 169 173 178 176 175 176 179 181 176 180 179 180 178 +27
.5" 159 180 180 184 185 185 187 185 187 195 203 202 202 200 +43
.6" 164 188 188 189 191 193 194 193 193 205 210 212 211 207 +47
AUC 70.5 78.2 78.7 79.95 80.45 79.75 80.5 79.75 79.75 81.15 82 82 82.55 82.05
CFMs Calculated to 28 Inches of Water.
The first column in the chart above, lists the valve lift heights for each row of data, which is CFMs (Cubic Feet per Minute) of air. The column labeled 'Stock' lists the flow volumes for the stock intake port before any work was done. The column labeled 'Part II' lists the results from Project E7 Part II, were some basic porting was done. The rest of the columns denote by number, the following porting procedures.
The AUC, or Area Under the Curve, is sort of an average flow for all lifts.

None of these procedures are being recommended. The dimensions given in the following procedures are what I achieved on this particular casting. You should not assume that any dimensions listed in the following procedures can be repeated without structural failure.

[No. 1]
Here, I begin copying the GT40P side wall profiles by removing more material from the outboard side of the bowl and the inside radius runner wall leading up to the pocket. You can see, from the rather crude drawing above, how the GT40P intake runner is not as straight as the E7.

[No. 2]
I removed some material across the entire roof.

[No. 3]
More material is removed from the outboard side of the bowl. I have cut under the valve seat at this point.

[No. 4]
Even further cuts are made into the outboard wall of the bowl and more material is removed from the inside radius wall of the runner leading to the bowl. The GT40P intake bowl measured about .760" from the valve stem to the outboard wall of the bowl.In this example, I cut the bowl wall back to about .720" at which point I broke through. The valve guide was shortened to 2.150", the same length as the GT40P guide. While taking a bit off the top of the guide, itself, does not improve flow, it does make it much easier to get the guide as thin as possible.

[No. 5]
I push the outside radius wall out near the flange, enhancing the S profile of the port. Most all lifts see an improvement.

[No. 6]
The stock valve from Part II that has 3 angles cut into it is replaced by another untouched stock valve. Notice that while the low lift numbers dropped, .4" improved.

[No. 7]
The stock valve has about a 19 degree back cut cast into its face. The upper edge of this backcut can be seen as the thin light line around the outer edge of the face of the stock valve in the photo below. Taking the advice of a subscriber to this site, this cast backcut was recut to make its edges more pronunced. .4" and .5" saw further improvements. (Thanks Triston)

[No. 8]
After many more grind and flow sessions (cut & suck) it seemed I had reached the limits of my ability to make this port flow any better. This is when I started to really look at a GT40P head I had in the shop. Why does this head flow so much better. The valve isn't that much bigger. I had made good improvements by copying the curvatures of the walls and deeply undercutting the valve seat on the outboard wall of the bowl. So what is it? This is when I realized how much shallower the short side radius on the GT40P head was compared to the E7. The floor of the GT40P intake port does not rise nearly as high as the E7 before it turns down into the throat. Also, the GT40P short side radius has a much larger radius than the E7. Where the GT40P radius makes a gracfull arch into the throat, the E7 ramps up, then takes a rather steep dive into the throat.
Normally, I would never touch this area, However, since I did have an example, the GT40P, to work from I lowered the floor a bit and increased the radius on the short side radius of the E7. This, effectivly, increases the 'Window'. The window is the area that you can see straight through the port unobstructed. In other words, it is the area that when you look through the port, you can see daylight at the other end.
Lowering the floor and increasing the short side radius had dramatic results at high lifts, as you could imagine.

[No. 9]
Seeing the dramatic effects of lowering the floor and working the short side radius, I measured the height of the GT40P port at the top, or apex, of the short side radius to the roof on the GT40P head. Hard to do, but I came up with approx. 1.340". So I worked the E7 floor and short side radius to this same dimension, trying to copy the exact profile of the GT40P short side radius in the process. Again, the results were dramatic. .5" is now well over 200 cfm.

Here is a profile of the GT40P intake port floor and
short side radius referenced to the head deck. It may
not be exact since I taped pieces of paper together to
form it, but each of the small sqaures is .100" x .100".
So the total height is about .8" and length about 3.50"

[No. 10]
Not wanting to push my luck with the short side radius, I began work on the valve. The stock valve already had its as-cast 19 degree backcut recut to sharpen the edges. I applied a 25 degree backcut, .020" wide between the 19 degree and 45 degree seat cut.

[No. 11]
I then widened the 25 degree backcut to .030", then to .040". Each time the lower lift numbers would come up a bit, while the higher lift numbers held.

[No. 12]
With the 25 degree backcut increased to .045" wide, things began to change. The high lift numbers are really falling off. I would say that this port likes the 25 degree backcut at .030" to .040". The 25 degree cut seemed to hold the high lift numbers longer, allowing the low lift numbers to come up a bit before the high lift numbers began to drop. The customary 30 degree backcut, while raising the low lift numbers, would start dropping the high lift numbers almost immediatly.

Notes of Interest

Final port volume was 144.6 CCs.

For some cross section photos of the GT40P ports check Sean Meldrums Fords GT40p Cylinder Head web page.

I tried many valve angles, some with 3 angles, some with even 5 angles cut into it. I tried reshaping the face, polishing the face, even tried "The Golfball". This port seemed to be happiest with a near stock valve and a 25 degree backcut with the as-cast surface on the face.

The pushrod hole was sleeved so that the entrance could be widened and the inside radius wall could be straightened. This was not neccessary and did nothing to improve flow on this example. Perhaps a bigger valve could take advantage of this procedure.

In Part II of this project, I had increased the port entrance opening to 1.010" wide. At that time I had estimated that the walls were still about .070" thick. I was wrong. When I cut the wall back to install the sleeve, I found that the wall was actually .050" thick.

Any work done to the inboard wall of the bowl, the divider between the intake and exhaust bowl, showed very little flow improvements.

It is interesting to note how the valve can be 'tailored' to the port to acheive the highest AUC. It is also interesting to note how sensitive the flow is to subtle changes in the valve angles.

I tried lowering the floor even more, and obtained 213.9 CFM @ .6" valve lift, but the low lift numbers were very poor, and the AUC dropped below 80 CFM.

More good stuff!