Any time you push the boundaries of engine output, you know there will need to be some dyno testing, and that's exactly the case here. We had some boundaries to work within, so that the end result was a sane build for the average mid-budget hot rodder. These boundaries included pump gas, a minimum of 350 lb-ft at 2,500 rpm, and a near lope-free idle of no more than 850 rpm. As for output, our goal was to see if 600 hp from a 350 block running on pump gas was even feasible. In the process, we would look at some newer speed parts, plus some tried-and-tested parts. All this, it should be said, will be attempted without the aid of a blower or nitrous.

400 Cubes With A 350 Block!
It's doubtful any of us need to be reminded that more cubes means a stouter engine, but it's been some 35 years since GM made the 400-cube version of the small-block. These tended to have an overheating and cracking problem, which has resulted in a scarcity of usable ones. The same cannot be said of the 350 block. It's about as reliable as a block can get and there are still millions available to the hot rodder. Stretching a 350 to 383 inches has been a common practice for the best part of 20 years. The question here is, can we go more? In terms of bore diameter, there is not much room to go more than 30 over, although a sonic tester can find blocks good for as much as 60 over. The combination to get a 383 is to bore the block 30 over (to 4.030 inches) and use a 3.75-inch stroker crank. The stroke increase over stock is 0.27-inch. Crowding this into a 350 block normally means cutting more rod clearance into the bottom of the bores and the sides of the crankcase. The amount needed to do a 383 sometimes compromises the block, and the clearancing goes into the water jacket.

Because the possibility of finding water is not that remote of a reality, the 3.75-inch stroke has long been considered the sensible limit of a viable 350-based stroker. But hot rodding, by its very nature, is all about pushing boundaries. Working with crank and rod manufacturer K1 Technologies, Lloyd McLeary of T&L Engines in Stanfield, North Carolina, has developed a procedure to reliably get a 4-inch stroke into a 350 block to produce 408 inches.

The process of producing 408 inches starts with a block selection procedure that involves both visual and sonic inspections. The sonic testing is the most critical aspect of a 408's success. Only about 50 percent of the good blocks already pre-selected for high-output builds will safely take the 4-inch stroke. Even so, it's close, as there is not much room to maneuver here. A typical performance rod on a 4-inch stroker crank will regularly break through into the water jacket. A key factor in making all this work are the modifications T&L does to the K1 rods. They make two moves to give the outer rod-bolt head more room: First, the bolt platform of the rod is machined about .025-inch lower, then the end of the rod bolt is machined .025-inch shorter. This, in conjunction with a block selected on the basis of sonic test results, gets the job done. The final inspection is to pressure-test the block to make certain it is watertight at the ground clearance points.

Going to 408 inches in an effort to achieve our 600hp goal might look like a big expense, but a stock factory crank at this power level will be well on the way to its limit. This makes a stout aftermarket crank more or less essential to the engine's ultimate survival. If a crank is needed, then it may as well be a stroker crank, as this will contribute to the output as well as beefing up the bottom end.

The piston requirements to meet the needs of an engine such as this are a little more critical than normal. First, to ease the reciprocating loads brought about by a stroke increase of over half an inch, we need to go for a tough piston that is also light, and these two factors are, for the most part, diametrically opposed. Fortunately, Mahle has a piston that fits the bill really well. Not only is it light, but this high-tech piston features thinner rings that are also narrower in plan view. This makes them able to seal up better with less ring drag, and that's an important issue on a stroker motor.

At this point, we have a short-block based on a 350 block that displaces 408 inches. You could build this short-block at home if you wanted by ordering all the parts from the relevant suppliers and getting a fully prepped and pressure-tested block from T&L, but quite frankly, it would cost you at least as much as getting the built short-block from T&L. The cost for an assembled block and balanced rotating assembly with a roller cam of your choice installed (we will get to that later) is around $3,200.

At this point, our 350 upgrade is looking good. We have 58 cubes more than stock, but that will not be of much use if the heads can't supply the air to feed those extra cubes.

Cylinder Heads
When AFR introduced their 195cc port Eliminator heads, it was without any great fanfare, so attention from our direction was somewhat limited. But rumors of great results with these heads (typical street price under $1,500) kept cropping up, so it looked like time to investigate. That was a year ago. Since then, we have sat in on dyno tests on four occasions and the output figures achieved were strong. For an air-hungry 408, some top-notch heads would be needed if results were to reflect the investment of time and money in the bottom end. These AFR heads looked like they would fit the bill and do so at a great price.

The AFR Eliminator heads feature 2.05-/1.60-inch valves and are CNC ported. As a pioneer in the field of both aftermarket cylinder heads and CNC porting, AFR has over a quarter century of experience to draw on. The Eliminator heads are a culmination of this experience, and represent some serious fine tuning of the 23-degree Chevy heads.

AFR has always been very strong in the cost department. This is because they CNC port these heads to a finish that delivers the most cost-effective result. If the finish is made finer, the cost escalates dramatically, while power may only change marginally-and it could go either up or down here. If the finish were made coarser, the price would drop only minimally, while the flow and power would drop measurably. But at the end of the day, finish is, at best, a second-order effect. The primary factor toward a functional head is port and chamber shape. Fail here and it matters little how good the surface finish is.

The reality is that shape dictates flow, swirl, port velocity, efficiency, wet-flow characteristics, and more. With any high-performance head, good airflow is essential, and though these AFR heads had that, there proved to be a lot more to their power producing capabilities than just airflow. That said, let's start by taking a look at the airflow as measured on T&L's freshly calibrated bench, and then move on to the other power producing assets.

The nearby graph (opposite page) shows the flow numbers we saw on the T&L bench. These, incidentally, were slightly higher than AFR advertises. Along with good peak flow numbers were fat curves, which were good from the seats all the way up. A point of note is that the bigger the displacement for a given intake valve, the more important low- and mid-lift flow becomes. With a 175-cfm intake flow at .250-inch lift, this AFR head is about 5 to 7 cfm more than we normally see at that lift station.

Along with good flow, the AFR intake ports also generated swirl numbers that typically deliver good results. Additional probing showed good port velocities and velocity distribution. All this adds up to promising results on the dyno.