SHORTSIDE SECRETS
It's just the reality of any conventional port that the flow path has to make a turn from the runner down toward the valve into the cylinder. Moving air, like it or not, wants to travel in a straight line. The form at the turn is critical to how much air will ultimately make it through the runner, around the bend, and past the valves into the engine. The turn-in from the floor of the port down toward the valve seat is referred to as the shortside turn, while the opposite side (going down from the roof) is the long-side. The long-side, as the name implies, has a longer flow path, and naturally has a larger radius of curvature. The shortside is by definition a much sharper turn, and just like making a sharp turn around a corner in a car, going too fast will make it lose the turn and hit the wall. Similarly, air going around the shortside will reach a velocity at which it will miss the turn, or separate.
The velocity of air moving through the port is primarily a function of how much air it is moving vs. the port's volume (more accurately, its cross-sectional area). As the port moves more air, the velocity goes up; eventually getting so fast it blows the shortside turn and separates. When that happens, the port will typically stall, flowing no more or even less air no matter how much further the valve is opened. How much air will get through before the shortside separates and the port stalls will depend on the shape of this critical part of the port. The shortside form is a key limitation on the ultimate flow potential of a head. All ports aren't created equal, and some have a decent shortside form while others are handicapped in this area. A large radius turn represents a better flow path than a tight one. Some heads are compromised in how much material there is to work with, while others are generously endowed. Race heads are often designed with raised intake ports and a large meaty radius to help get the air around the bend without separating.
Reworking the shortside when porting at its most basic level begins with removing any sharp edges. An edge or ridge will trip the air prematurely, causing the air to separate at the short turn, costing flow. More involved modifications attempt to rework the turn to provide as large and gentle a radius as possible, given the amount of material there is to work with. Generally, a rounded concave radius is the best form. For high-lift flow, the short turn should not project in toward the valve stem, but rather needs to be laid back toward the floor. Water lurks under the shortside of most heads, so beware.
 The short turn is critical...  The short turn is critical to peak flow. It's not uncommon for a port to hardly gain anything in high-lift flow until the short turn is worked, with the tendency being to stall at a certain max flow level. This intake shortside was cleaned-up as shown in the port at the left, netting about 10percent in peak flow. Note the rounded, laid-back shape. On the right is the raw shortside before porting. |  What works for the intake...  What works for the intake side also works at the exhaust: a large-radius, rounded form. A head reworked to this point--machined, bowl blended, and with a shortside rework--will nearly always show substantially improved flow throughout the entire flow curve. |  The Vortec Chevy head has...  The Vortec Chevy head has one of the best basic short-turn forms of any production iron head. Ported, like these 272/217-cfm examples, an almost ideal form can be achieved. Notice how the rounded shape of the seat is held around the corner to meet the floor of the runner, in a nice radiused sweep. |
POCKET O' POWER
It's no secret that a great deal of flow typically lives in the port bowl, hiding under the valves and around the valve guides, just waiting to be harvested by the carbide. Reworking this area, accessible from the chamber side of the port, is referred to as pocket porting. There's a broad range of what one guy or the next will call a pocket-ported set of heads, but the idea here is to smooth-out or enlarge the bowl area and long side of the port adjacent to the valve guide boss, and even re-profile the guideboss itself. Don't go overboard when enlarging the bowl, however, or a flow separation at the seat will be the likely outcome.
 Here we have a good example...  Here we have a good example of a worthy pocket-porting job, defined as blending the "pocket" area under the valve. The bowl area is worked to create a smooth transition into the roof and port runner, and the valve guide is profiled to a streamlined shape. There's water all around the bowl, so don't go overboard with the cutter. |  Sometimes in the pursuit of...  Sometimes in the pursuit of airflow, greed can get the best of any porter, and the tendency is to go too big in some places. Nowhere is the price to pay higher than going too big in the port throat, the point of constriction just below the valve seat. Make the throat too big, and the venturi effect is ruined, and usually the flow will be too. Keep the intake port throat no larger than 90percent of the valve diameter, and the exhaust throat down around 85percent. |  Here is a similar before and...  Here is a similar before and after of the exhaust port. The idea is to make the transitions all the way from the previously blended area of the bowl into the port runner as seamless as possible. |
THE RUNNER
The port runner is the "straight" part of the port before it turns into the bowl towards the valve. The runner is usually the best flowing part of the port, however virtually any pushrod V-8 has its major limitation in the port area at the point were the runner is pinched-in to clear the pushrods. This is particularly true of intake ports on inline-valve engines. The runners in most heads start off relatively tall and narrow, and then broaden out and get shorter as the runner approaches the bowl. The runner cross-sectional area is tied to some major aspects of how the engine will perform, with peak torque rpm at a given displacement closely tied to minimal cross-sectional area. If the port is too small, the engine will tend to run out of steam prematurely. In contrast, a runner that is too large will soften the bottom end.
Very few head porters will go through the effort of making a study of cross-sectional area, even though there are major power gains to be had by fully exploring these characteristics. Most ports are cast with fairly dramatic changes in cross-sectional area along the length of the runner. When runner cross section changes, the air is forced to speed up and slow down, and that costs energy and flow. Some simple measurements with a set of dividers and calipers to determine the height and width of the runner at various points of constriction can "map" the port, giving a good indication of "tight" areas needing attention.
Whether or not opening up the runner will gain flow depends upon how well the rest of the port is doing. If the shortside and bowl are putting a cap on peak flow, working the runner will gain nothing. A flow system is only as good as the weakest link, and as the rest of the port starts to move some air, the pinch point becomes a limiting factor. In the upper leagues of high-performance engine building, getting enough cross-sectional area in the pinch becomes a major factor in getting high rpm output out of an inline valve engine. A cross sectional limitation here will drive up port velocities to the upper limits of port flow, putting a cap on peak torque rpm, and ultimately high-rpm horsepower.
 How far you can cut back the...  How far you can cut back the pushrod pinch before breaking through? Here's where the Helgesen E-tool is used to gauge the remaining metal thickness. The gap at the open end of the tool is exactly twice as thick as the remaining metal. |  It pays to take some time...  It pays to take some time to study the runner before digging in with the carbide. The typical constriction points in a runner are the pushrod pinch, or bulges to accommodate head bolts or oil passages. Work to even out the cross sectional variations, and make the runner as straight as the architecture will allow, cutting for sweeping curves where necessary, rather than abrupt directional changes. |  Metal thickness in many areas...  Metal thickness in many areas of the head cannot be gauged without a sonic tester, so it's best to be conservative. Other areas can be gauged, such as the remaining roof thickness at the spring seat of this head, measured here with a set of outside dividers. |