Muscle Mustangs & Fast Fords
1993 Ford SVT Cobra Project Stolen Goods - Dyno Tested, Owner Approved
Project Stolen Goods Hits The Dyno And Then The Road.
If You've been following the modification of our resident '93 Cobra, Project Stolen Goods, you're probably as ready as we are to know that it's finally running, driving, and turning dynamometer rollers.
Last month, we fired up Stolen Goods for the first time in more than a decade, making sure everything was functioning as it should. After a 15-minute break-in period, we changed the oil and filter and poured in a bottle of D.S.S. Racing's oil additive. We did the same thing again after another 30 minutes.The Meziere electric water pump and Flex-a-lite electric fan worked flaw-lessly, and we had plenty of oil pressure, so it was time to get the car on the dyno.
In order to optimize the combination and get horsepower and torque numbers, we took Stolen Goods to HP Performance in Orange Park, Florida, where our Cobra was strapped to the company's Dynojet. Before we let loose with the throttle, however, HP's Tony Gonyon hooked up his datalogging equipment to see what the engine was doing while he was tweaking the A9L ECM with a custom SCT chip. The original X3Z processor had long since been separated from SG, so a standard 5.0, five-speed ECM would work just fine.
Professional Mass Air Systems did a great job with the mass airflow calibration, which was tuned to match the 38-lb/hr fuel injectors we used. Mass airflow sensor calibration, however, has to be a middle-of-the-road approach to suit the many different combinations out there. Considering that, and the fact that manual fuel pressure and timing adjustments are one dimensional in their function and only a small part of the tuning equation, we needed to optimize our Boss 347 bullet using an aftermarket piggyback-style computer chip.
We began by setting the base fuel pressure at 40 psi with the vacuum line off and the base timing at 12 degrees. Being able to datalog what's going on in the engine is extremely helpful when you're trying to nail down the proper tune-up, and we realized there were several changes that needed to be made to the program. Because of the larger injectors, the cranking fuel and air tables needed to be altered to allow more air and less fuel for an easier startup. Knowing that we had moved the O2 sensors way downstream, Gonyon delayed the O2 read time. With shorty headers, the O2 sensors heat up quickly, and then they start reading the exhaust. With the O2s placed further away from the exhaust ports in a long-tube configuration such as ours, they need more time to heat up so they can accurately read the exhaust information.
Other changes to the idle and drive air tables which control the IAC solenoid were made, along with a slight reduction in cruise timing to reduce a bit of light throttle bucking that we noticed. Lastly, Gonyon altered the fuel and spark tables to create his own timing ramp and fuel compensation, and he also raised the factory rev limiter to 6,400.
With the tuning finished, the Boss 347 was cranked up, and a full pull was made in Fourth gear. The numbers came up at 342 rwhp and 368 lb-ft of torque. Given the humid temperatures and relative lack of break-in miles, we expect to see a bit more power down the road and during cooler weather. All in all, we got exactly what we wanted-a very torquey motor that, depending on what estimate you use to calculate the drivetrain loss, makes right at about 400 hp at the crank. More importantly, the car is a blast to drive.
A few tuning sessions out on the road followed and all was deemed well except for the rearend whine that we noticed. After disassembling the rear, we found that the pinion bearing had failed, which caused the backlash to be off. Our plan was to get the car on the alignment rack and then get a few street miles around town before returning to HP for a new gearset.
A few days later, we pulled out of the dealership where our friend George Xenos had just performed the alignment. After mildly easing through Second and Third, we suddenly heard a loud tapping from underneath and in front of the car. A roadside audible assessment pointed to a pushrod tapping the valve cover, and that was later confirmed once we had the intake and valve cover off. The reason it was hitting was because the rocker-arm stud had snapped right above the nut. Our contact at Summit Racing had a replace-ment stud shipped in short order.
The following week, we trekked back to HP Performance and installed the new stud and adjusted the valve. The new ring-and-pinion went in (same 3.55:1 ratio as before); then, we strapped the Cobra to the dyno once more just to see how things were. Once the run was stopped, Stolen Goods and its Boss 347 engine had improved power output to 358 hp and 386 lb-ft of torque. The tune-up was still dead-on, so no changes were deemed necessary.
Now that the snake was back in action, it was time to burn rubber, so we set off for Gainesville Raceway in Gainesville, Florida, home of the NHRA Gatornationals and the first outing for our recently finished project car. The two-hour drive on I-75 went off without a hitch, but it rained the whole time and the track called it right before we got there.
After stopping for a bite to eat, we noticed that the throttle seemed to be hesitant, as if the engine was loading up with extra fuel. We thought the plugs might possibly be fouled. After about 40 minutes of interstate driving, the Cobra was down on horsepower, and it definitely wasn't running properly. We pulled over at the nearest rest stop to check things out.
Your author had experienced symptoms similar to this in another project car, and it turned out to be a wet mass air meter and harness. After surveying the meter and air filter on Stolen Goods, it was immediately apparent that both were well soaked in water. Since the car wasn't making any odd noises and the gauges checked out, we wrapped the meter and harness with a towel and completed our journey home.
We picked up a new set of NGK BKR6E spark plugs, but before we swapped them in, we pulled the meter, air filter, and harness to make sure everything was clean and dry. We also cleaned the mass air sensor's element in case the water on the air filter had brought in any oil from the filter. A quick twist of the key and we could still tell the engine wasn't up to snuff, so we swapped out the plugs. Interestingly enough, the front four plugs were perfect, while the rear four were gas-fouled. We also noticed that we weren't getting a positive snap when we put the spark plug wires on, so a little WD-40 was in order for the boots. Time for a test run.
The engine still wasn't running correctly, so we pulled the plugs again to find that the fuel-distribution problem was gone, but the No. 7 plug had not fired at all. After pulling the valve cover, we realized that the very same intake rocker arm stud had snapped again, but this time the rocker wedged itself sideways, which restricted the pushrod's movement and explains why we didn't hear the pushrod tapping the valve cover.
Since we had a deadline to meet, we pirated a rocker-arm stud from another car and reassembled everything. Unfortunately, Stolen Goods still ran poorly, and pulling the plugs again revealed that the No. 7 cylinder still wasn't firing. After consulting with engine builder Tom Naegele at D.S.S. Racing, we were looking at the possibility that the pushrods were too long and the rocker didn't have enough travel, which ended up binding the assembly, and the stud took the brunt of the force applied.
On Naegele's recommendation, we used some White-Out tape and our Comp Cams pushrod checker tool to see where the roller was riding on the valve tip. He pointed out that having the roller centered is only part of the adjustment equation, as you also need to take into account the amount of sweep across the tip as well as where the sweep centers itself.
With the 6.85-inch pushrods we were using, we noticed the sweep to be located towards the exhaust side of the head. Running the assembly with the checker tool at 6.75 and 6.70 inches moved the sweep towards the intake where it needed to be. We also noticed that the width of the sweep decreased as the pushrod length was shortened. This is a better scenario for the valve as it doesn't create as much side-loading force.
At this point, we're looking at ordering a shorter set of pushrods, but we also need to perform a leak-down test to make sure the intake valve has not been damaged in any way. We performed a compression test and found no compression whatsoever in cylinder No. 7, which may be a result of using the incorrect rocker-arm stud and/or incorrect pushrod length. Either way, we'll have this mystery sorted out for the next installment, and we'll fill you in on the details.
Now that we've had a taste of what Stolen Goods has to offer, we're looking forward to getting the snake slithering on the nearest track.