This month, we continue our 396 stroker build-up for our 1988 Firebird Formula project car--dubbed Magnum TPI. Those following along will remember that we covered the short-block build-up of our 396 at Strope Speed Shop in the September issue of GMHTP. In that installment, we documented Aaron Strope's careful machine work and assembly using a Cola forged 3.875-inch crank and Lunati internals (Pro Mod forged rods, custom 4032 alloy pistons, chromoly rings, bearings, wrist pins and locks). To strengthen the bottom end, we also threw in a set of billet main caps from Pro Gram Engineering and used Royal Purple synthetic assembly lube at all stages.
With a strong bottom end like this, we aren't about to put junk breathing components on top. A stroker small-block needs to breath big-block numbers, or else it will fall flat on its face. Flying seemingly in the face of big power are two other requirements heaped on top: excellent street manners and the ability to pass the strict New Jersey IM240 emission inspection program. If that isn't enough, we would really like to run 11s with pump gas, but we may have to settle for very low 12s. Phew!
Third-gen TPI fans out there already know that making power with a factory TPI is like squeezing water from a stone, so we realized early on that it was folly to even try. Several of you have been quite emphatic on the various internet message boards about us trying the ACCEL SuperRam manifold, with the rest of you urging us to use the TPIS MiniRam. Both are excellent pieces, but we decided to fall on the side of big torque (the MiniRam tends to excel at the upper end of the power band where race cars inhabit). Not that a stroker motor would have any trouble making torque with a MiniRam, it's just that our camshaft selection and planned street gearing favored the torquier SuperRam.
Clearly, we weren't blazing a new trail. Many others over the past seven or eight years have gone before us to point the way, all we had to do was read the burnout marks on the asphalt. Turns out, many of those tracks were aided by cylinder heads from Air Flow Research. Our previous cylinder head flow test (see "Just The Facts," Sept. 2000) revealed that AFR have clearly done their homework, so we ordered a set of emission-legal AFR 190 heads, but with a twist. Needing a larger port, but wanting to live within the letter of the law, we had AFR perform a competition CNC port job, which enlarged the intake port volume to 195 cc.
AFR's 190 head is a direct bolt-on, requiring no special intake manifolds, valvetrain hardware or headers. Rick Sperling of AFR told us that the exhaust ports are raised .125 inch which in most cases is not enough to cause problems with header fitment. Exhaust port volume is increased to 70cc, and the cross section changes from a "D" shape to a square, which accommodates the exhaust flow to a set of 13/4-inch headers (like our SLP units) just perfectly.
After Strope Speed Shop received our tweaked AFR 190s, Aaron Strope massaged them further while porting the SuperRam's runners and plenum area. The result was an intake port that flowed 279 cfm at .500-inch lift without the use of a radiused inlet fixture (see airflow data sidebar). If we calculate a conservative 3 percent flow increase with a radiused inlet, we pass our target airflow by 2 cfm and at .050-inch less lift! Even better, the intake port volume now measures only 198cc after Strope's additional port work. These are heads that are clearly capable of getting us where we want to go.
One of the cornerstones of engine building is the principle that the better the cylinder head, the fewer compromises have to be made in the camshaft. This is a vast oversimplification, but it serves to point out that for our street motor, the only way to achieve our stated goals is to put the most resources into the cylinder heads. In this department we didn't skimp, spending a total of $2724 including all our port work at Stropes ($1924 for the CNC-ported heads and $800 for additional porting and shimming the spring heights). Naturally, if you're willing to settle for less performance or if you can perform much of the work yourself, you'll do better on the price, but this was one area we were unwilling to sacrifice. Perhaps of all the gear we chose, we wrangled with the cylinder heads the most, not on the grounds of performance, but on the grounds of cost. We know many third-genners out there can't afford heads like these, but it's impossible to achieve serious performance without a good set of heads (our last fiasco ought to be proof enough of this). If you only take away one lesson from this story, let this be it!
Excellent cylinder heads sure made camshaft selection a lot easier. For this, we turned to the Competition Cams tech hotline. We fed Comp all our data, including our projected compression ratio and goals of streetability and emissions, and they developed a custom cam profile specifically for our SuperRammed 396 stroker. The split-pattern hydraulic roller cam they suggested specs out to 224*/230* at .050-inch lift while valve lift tops out at .536 inch on the intake and .544 inch on the exhaust (with a 1.6:1 ratio Pro Magnum stainless steel rocker). We also ordered a .10-under double-roller timing set for align-bored blocks, hardened pushrods, anti pump-up hydraulic roller lifters and a hydraulic roller conversion kit which supplied new lifter link bars, a lifter retainer assembly and fresh camshaft thrust plate.
One other key feature that we designed into Magnum TPI's 396 is a complete Evans cooling system. This consists of Evans NPG coolant, a high-flow Evans coolant pump (don't call it a water pump, there's no water!), an Evans radiator and larger coolant hoses. While most cars will benefit from using just the coolant alone, we wanted the maximum benefit in the hope that this will turn into extra power when the calibration is completed. This expectation is not unfounded: similar engines to ours have seen as much as 20 additional horsepower from increased octane tolerance and better thermal efficiency. We'll go into the theory in Part 3, but suffice it to say that Evans' customers have noted significant gains in fuel-injected applications and we think this high-tech solution is completely appropriate for both this application and our readers.
Using our Mr. Gasket Desktop Dyno simulator, we plugged in our engine specs including cylinder head flow numbers and our cam timing from our Competition Cams cam card. According to other users, the one area Desktop Dyno doesn't seem to model very well is the SuperRam intake. SuperRam proponents tell us Desktop Dyno falls far short in its torque estimate and that we should see another 40 or 50 lb.-ft. at peak torque. Even with this attenuating factor, the simulation projected 470 hp at 6000 rpm and 480 lb.-ft. at 4000 rpm using the "tuned port" induction model with small headers and mufflers. Providing we see the missing torque in the real engine, we should easily go 12-ohs with the proper gearing and torque converter.
The optimal programming will probably push Magnum TPI into the 11s, thanks to a new Gen VII DFI with wide-band tuning control. For the time being however, we're soldiering on with the stock ECM and a custom computer chip. It will be interesting to see just how much power the new DFI will be worth over an expertly-programmed PROM chip with a stock ECM and mass air flow meter. All that's coming up in the January and March 2002 issues of GMHTP, so keep your eye peeled here for details. For now, let's watch Aaron Strope perform the top-end assembly of our new 396 at Strope Speed Shop.