Michael Galimi
December 1, 2006

Last month, we started on an unusually large small-block Ford engine. The short-block was built in collaboration with Rich Groh Racing and Pro Power Performance Parts, and the massive Windsor is being constructed with complete street worthiness in mind.

To accomplish the goal, the parties involved came up with a 445ci solution to our problem. They took a World Products Man O' War block and bored the cylinders to 4.155 inches. The Man O' War block is one of the few on the Ford market capable of handling a huge bore without filling in the cooling passages with concrete or block fill for added strength.

They also stuffed a Prime One 4.100-inch crankshaft into the crankcase to jump the cubes even more. The short-block is sure to handle almost anything we throw at it-within reason, of course. The short-block is the foundation on which we will build a street beast that will hang with even the most hard-core '03-'04 Cobras and other nasty street machines lurking around our neighborhood. This mill will also benefit from a Vortech YSi-Trim supercharger that will breathe 12-14 psi of boost into the Edelbrock Super Victor intake manifold.

Some may argue that running such a long stroke puts undue stress on the piston skirts due to side load. Side load can come from a short connecting rod coupled with a long-stroke crankshaft. There is an angle that pushes excessively on one side of the piston. With those theories out and about in the engine-building community, why would anyone want to build an engine with such a big stroker crankshaft? It is easily explained: "Side load comes from poor rod-to-stroke ratios. What is an acceptable ratio before there is a problem? The rod-to-stroke ratio of this combination is 1.51:1. To put that into perspective, there are OE engines that have worse rod-to-stroke ratios than the one in this engine," says Dale Metlika of Pro Power. He discussed with us the finer points of side loading and explained that our engine buildup was not excessive.

Metlika also brought up the importance of oiling and how it helps prevent excessive wear in a short rod, long-stroke engine. For those who read Part 1, you may remember us discussing the finer points of oiling with this engine. One of the features Groh incorporated was a medium plateau hone. It was meant to promote better oil retention on the cylinder walls. The bore surface was flat, but there are microscopic grooves to help retain oil on the walls. A piston does not rub against the wall. Rather, it glides on a film of oil, so the piston is technically not smashing into the side of the bore with every up and down motion. The Prime One/Pro Power pistons also have provisions for oil retention on the skirts, so there is no shortage of oil as the pistons move up and down. Another hot topic is ring wear because of side load. Pro Power includes rings in its stroker kits that are beefy and long lasting.

Groh used a 4.155-inch bore on his flow bench because that is the bore size of our engine. It is important that all flow data is achieved with the correct bore of the engine the heads are scheduled to go on. The bigger the bore, the more air will flow through the head. The intake valve is unshrouded because the cylinder wall is farther away from the valve.

There are harsh realities to face when attempting to run a 750-plus horsepower engine on public roadways. Immediately, what comes to mind are steep gear ratios and ultra-loose torque converters (for those running an automatic transmission). Others envision lumpy cams like the ones found in Pro 5.0 cars. This large Windsor keeps those fears at bay. The reason is that we are not twisting this superstar to the moon. Thanks to the large cubes, we can keep the revs relatively low (6,000-6,500), ultimately keeping the aforementioned drivetrain components tame and sane.

Nasty cams that snarl at idle and come alive at 8,000 rpm need not apply for a spot in this engine program. Those types of cams also require ridiculous valvespring pressure-a no-no for the street unless you want to adjust the valves before each cruise night and replace the springs on a regular basis. An added benefit to low-rpm street engines is that bearing and ring life will be improved because the bearings and rings will not be pounded due to severe rpm.

Running big cubes isn't the end of the story. This month, we tackle the induction system and show what it takes to supply the air and deliver the fuel to these large cylinders. We were back at Rich Groh Racing, and in our hands were lots of goodies. There were challenges to be tackled and hurdles to jump over to get to this point. Thankfully, we are here to tell the tale of airflow.

Two major pieces of the induction puzzle are the cylinder heads and camshaft. They work together in an effort to whisk the air/fuel mixture into the cylinders and do their part to help get the spent gases out. We know that it is easy to get caught up in the numbers game when selecting these parts. Big lift flow numbers are nice for bragging to your friends, but what we're after is good average flow. A large cam may sound mean and snarl at the competition, but it can present problems such as reduced vacuum at idle (important with power assisted brakes), and it requires a big-stall torque converter. Valvetrain maintenance jumps up considerably as well, as we mentioned earlier.

TFS Street Heat heads were selected for this bruiser. The Street Heat history goes back to the early '90s, but this tried and true design has been slightly tweaked over the years by the TFS folks, and it is still one of the most effective inline cylinder heads on the market today. Even when other companies set out to copy these heads, they still fall short in achieving the great airflow and horsepower capabilities of the Street Heat heads. The heads have truly stood the test of time.

Anyone can hog out a port and get it to flow copious amounts of air. The real story is in the velocity of that air as it rushes into the port. Groh uses an air-speed probe to test air velocity. For example, these TFS heads flowed 290 cfm at 0.500-inch lift with an air speed of 340 ft/sec. According to Groh, the air was moving fast through the port, but he cautioned that speed is only part of the story. Air swirls and tumbles around in the port. He also said too much air speed would create a dead spot on the short turn, which is a no-no for performance. Porting heads can be considered an art form, and there is no particular way to do things. Every head porter does something different to arrive at the same result.

We picked a set that was CNC-ported in-house at TFS and features 2.08-inch intake valves and 1.60-inch exhaust valves. They came complete and ready to bolt on, but we chose to pop off the valvesprings and use a different set. Groh selected stiffer springs from Comp Cams to match the camshaft he designed for the engine. One thing we forgot to mention to TFS was that we were running larger pushrods. One phone call and TFS swapped out their standard 51/416-inch guide plates to the larger 31/48-inch stuff.

Groh carefully ran the TFS Street Heat heads on his flow bench to gather data. The flow numbers tell only half the story, as Groh also evaluated air speed (or velocity) to get an idea on how well the volume of air moves through the head. The flow numbers are an essential part in the camshaft design. According to Groh, the lower lift numbers are decent and this cylinder head shines in the peak numbers area. "It's a good cylinder head with good peak numbers. For a street car, I would work the chamber to get better low-lift numbers, but then the high-lift numbers might suffer. Like everything in racing, preparation is full of compromises," Groh says. The out-of-the-box heads were fine to go on our project engine. Of course, there is more to be had with these cylinder heads, but we were content with the flow capabilities of the set we had in our hands.

In addition to getting the cfm results, Groh inspected the port velocity using a special probe. The probe provided the airflow data in feet per second. "This gives me an idea of the speed of the air," he says. "The air moved through this head at 340 ft/sec at 0.500-inch valve lift and moved 290 cfm. It is a very fast cylinder head and will fill the cylinder quickly. There is no industry standard for the air speed and its relationship to flow. Knowing what works and doesn't work is based on experience."

The cam is key in this application because of its giant cubic inches and supercharged nature. That is not to say an off-the-shelf grind wouldn't work, but the custom-designed cam will help get the most out of our engine. It is a custom hydraulic roller piece designed by Groh, ordered from Pro Power, and cut by Comp Cams.

Maximum valve lift comes in at 0.570 inch on the intake side and 0.575 inch for the exhaust. The duration at 0.050 inch checks in at 242 degrees for the intake and 248 degrees on the exhaust side. Groh put the lobe separation at 115 degrees. "That is a fast lobe, so it will idle nicely and use the air from the cylinder head and supercharger," Groh says. Some people may question the smooth idle, but Groh assured us the idle and street worthiness will be just fine with the lobe he picked out.


The Comp Cams valvesprings were specially selected for this engine combination. They are required due to the fast opening and closing of the lobes. The valves will not float with these springs in place, even as the engine soars to 6,500 rpm or so. If we had used titanium valves and retainers and triple springs, we could have turned the engine to 7,200 rpm without floating the valves. Unfortunately, more spring pressure leads to more maintenance and increased valvetrain heat-which will wreak havoc on valvetrain survival rates.

"The fast lobes will work in this engine because it is a low-rpm engine," Groh says. "The fast lobe will also help smooth out the idle quality. We widened the overlap to get the street idle quality. Plus, the cam is opening the exhaust valves earlier because of the supercharged nature of this engine. Some people may think 115 degrees [of] lobe separation is wide, but outside of the Ford world, this type of cam works great. I have used plenty of these with success in LS1 engines as well as other street Ford engines.

We spilled the beans on most of the specs, but there is more to the story. Groh prefers to keep the rest of the information under a veil of secrecy-specifically, the type of lobes he selected for this camshaft. It is not uncommon to get this request from engine builders, as it is proprietary information.

The rest of the valvetrain is from Comp, and it supports the healthy camshaft we had ground for the project. The rocker arms are Pro Magnum and have a 1.6:1 ratio, while the springs are from Comp Cams. The pushrods came in at a lengthy 8.600 inches and are of the hardened variety. They are also the beefier 3/8-inch pieces. It is important the pushrods are hardened to eliminate the possibility of flexing and/or galling. They will be subjected to the harshness of trying to open and close valves under extreme cylinder pressures. The thicker, hardened pushrods will not budge under those conditions.

With the camshaft, valvetrain, and cylinder heads in place, we are nearing the end of our project 445ci engine. Next, we plan on finishing the buildup and conducting some tests on an engine dyno.

Steady Flow Of Data
Cylinder-head flow is widely misunderstood in the high-performance world. Magazines are accused of promoting only the big flow numbers, but if you take the time to read, you'll see this couldn't be further from the truth. Today, most editors understand the importance of overall flow and openly talk about the big picture, not just peak numbers. With that said, many enthusiasts still make mistakes when evaluating cylinder heads because they base everything on flow. If you spent a few days at Rich Groh Racing or any professional engine shop, you would see that just slapping a cylinder head onto the flow bench and letting it rip isn't the proper way to evaluate heads. It is a process that has been standardized, and testing procedures are replicated exactly each and every time a head is tested.

It all starts with the flow bench. For many, the SuperFlow SF-600 is the bench of choice. When ordered at RGR, it was specifically calibrated for the shop location. The reason for that is the flow bench readings at a head shop in the mountains of Colorado will certainly have different readings from a flow bench at sea level. With the calibration in place, Superflow provided specific flow plates to test the flow bench to ensure accuracy.

Another useful piece of equipment that has helped bring more accuracy to the cylinder head flowing segment of the industry is FlowCom. In the early years, head porters relied on the "test tubes" on the flow bench, and they eyeballed and scribbled down the results. These days, a digital readout is provided that ensures 100 percent accuracy in the readings.

On the exhaust side of the TFS heads, the ports are 95 cc and flow a maximum of 251 cfm at 0.750-inch lift. Our camshaft lifts the exhaust valves a maximum of 0.575 inch. These heads flow 230 cfm at 0.550 inch and 238 cfm at 0.600 inch through the 1.600-inch exhaust valve.

Once you have a bench that is calibrated properly, it is time to set up cylinder heads. Test equipment is sensitive to how the test subject is attached to the tool. Problems can occur if there are air leaks, as an inadequate seal will allow the results to be skewed. The most common areas for air leaks are the valves, head gaskets, and spark plug hole. Groh even said he runs the rocker arm studs when he flows the heads. He proved his point by flowing the head with the rocker arm studs in place and without. By not running the studs, the heads were down by 8 cfm. The cfm dropped because air entered the port and disrupted the airflow pattern.

These heads were flowed in 0.050-inch increments using step blocks to simulate the valve lift. Groh set up a micrometer on top of the valve and calibrated it. He used the blocks to open the valve to the predetermined lift and he took measurements. He also double-checked the micrometer so it was always set up perpendicular to the valve tip. If it is not lined up properly, the results will come out higher and be less accurate.

It's all about consistent and accurate setup techniques.

Photo Gallery

View Photo Gallery