Neil Van Oppre
December 8, 2006
We finally answered the age-old question of whether a stick is quicker than an automatic with this C4-to-Tremec comparison in Project Pro Tree Mustang.

It would be impossible to count the number of times we've been asked what the performance difference would be when switching from a five-speed to a C4, or vice versa. The general consensus has always been that the five-speed-equipped car would be quicker and faster, all things being equal. But the fundamental question remained: how much quicker?

We set out to answer that question by doing a direct comparison between a full-competition C4 from Transmission Specialties and a Pro-Shifted Tremec 3550 built by Hanlon Motorsports. The C4 was equipped with a drag-race-friendly, 4,700-stall speed, 8-inch Dura Sprag converter, an ATI chromoly SFI flexplate, and a transbrake. The Tremec was fronted by a Ford Racing Performance Parts heavy-duty clutch and billet-steel flywheel. The only items changed were the flywheel, trans, and driveshaft. No changes were made to the engine combination, and the initial dragstrip tests were performed using the exact launch and shift rpm. The reason for the driveshaft change was simply because in the past we had always run the stick with the custom Dana driveshaft that came with the car.

Our assumption was that it was stronger, and it clearly fit better with the stick, meaning the yoke didn't hang out of the tailshaft as far as the OEM driveshaft did. The Dana piece is 311/42 inches in diameter versus 3 inches for the OEM unit, yet the factory driveshaft outweighs the Dana by 2-3 pounds. While even 3 pounds of rotational weight can make a difference, we don't believe it's dramatic enough to make a noticeable one on the timeslip or a chassis dyno.

The five-speed is a Pro-Shifted Tremec built by Hanlon Motorsports. It features face-tooth engagement and was assembled using 3550 internals and a Second Generation case (additional ribbing). A Pro-5.0 shifter grabs the gears. Whether you're using a T5 or a Tremec- in synchronized or Pro-Shifted form-you can expect to see similar results compared to a C4.

Our tests were conducted in summer weather, all at the same track-Old Bridge Township Raceway Park near Englishtown, New Jersey. The weather varied during our time there, but we were able to test each transmission under similar conditions. Since we were using a vehicle of known quantity, it was easy to compare our tests with prior runs. This helped put our results into proper perspective.

It's also worth mentioning that the tests were conducted during weekly test and tune nights where the track conditions varied due to weather, oil downs, and street-tired cars affecting the racing surface. This only helped to shed light on what it would be like to live with each transmission in the real world. As an added bonus, we were able to dyno the car with each transmission, which not only confirmed some of our expectations, but surprised us as well.

The testbed was our own Project Pro Tree Mustang ("Value-Minded Mustang," June '06), a naturally aspirated, 302-powered combination representative of many street/strip cars. PTM uses a stock roller short-block with mild '71 351 Windsor cast-iron heads and shorty headers. Prior to this transmission comparison, we swapped the FRPP E-303 cam for a Ford X-303, and port matched the Weiand Stealth intake to the heads (we also used Fel-Pro 1250 gaskets). For those following this project, the cam swap made essentially no change, which we attribute to the limitations of the OEM heads. More on that in a future issue.

Our C4 and 8-inch converter are from Transmission Specialties. This is a full competition unit featuring a deep pan, a heavy-duty servo, and a transbrake. It was mated to our mild 302 by an ATI Chromoly SFI flywheel. We've used this transmission for six years without an overhaul, and it still shifts as well as the day we got it.

Our first order of business was to baseline the C4 combo on a chassis dyno. We took the car to PRA Solutions, a division of RIPP Modifications in Staten Island, New York. There, Perry Papadopoulos helped us find the best timing and jetting for our new cam with the C4. I was personally surprised when we made 302 rwhp, considering I didn't believe this engine made much more than that at the flywheel. Nonetheless, this tuning session netted us a full tenth over what we had been running previously with the X-303 cam. At the track, the car ran several 12-teens in some oppressive heat and a best of 12.09 at almost 109 mph.

We were confident the five-speed would put the car into the 11s in the heat, and it didn't let us down. Right off the trailer, we ran an 11.99 at 111 mph. The 60-foot time dropped to an all-time best of 1.54, and that was with the exact same 4,700-rpm launch and 6,200-rpm shift point we had been using with the C4. The only change that was made in deference to the five-speed was the shock settings. Knowing the stick would shock the driveline (and our short-sidewalled 26x10 slicks), we reverted back to the same rear shock settings we had used in the past with a hard-launching stick car-full tight. We also tightened up the front struts to try to control the antici-pated bounce from the big wheelies we planned to pull. Our current strut situation is a bit of a compromise. When one of our old struts blew out before a big race, we were pressed for time, so we installed a new set of adjustable Koni Sport struts. Although they're infinitely adjustable, they're not intended for drag racing. The valving is the same for compression and rebound, which wasn't an issue with the automatic since we ran them soft. But now if we tighten them enough to eliminate the bounce through the 60-foot clocks and after the 1-2 shift, we run the risk of reducing weight transfer to the point where we may compromise traction. This situation is further compounded by the soft spring rate of our four-cylinder front springs. So far, the results have been acceptable, but we'll need to address this if we want to maximize our results.

We dyno'd the car with each transmission at PRA Solutions using their eddy-current dyno. Who would have thought the C4 would record more rear-wheel horsepower than a five-speed? Well, it did-by 12 hp. But since horsepower is a calculated expression of work over a period of time, and we observed the engine accelerated in high gear more quickly with the automatic, that result shouldn't be all that surprising. What matters most is the change in torque output, and there the five-speed shined.

We tried launching harder than 4,700 on our first outing, but it didn't improve anything, and the car appeared to hook (see comparison chart). We've disregarded that run as an anomaly. On the last run of our first outing, we managed to run an 11.92, and anyone who witnessed that run must have thought the driver's right arm and left leg had fallen off. Thanks to some well-worn lower control arm bushings, the left rear shock broke in two almost at the exact moment of the 1-2 shift. As the clutch got kicked, the shock was already separating, which induced a nosedive of embarrassing proportions. As the shift was completed, the car attempted to pull another wheelie as the rear bumper made an attempt at scraping the ground. It's fun to watch on video, but yours truly felt like an idiot behind the wheel. At least we got a decent timeslip for our troubles.

Back home, we installed new bushings in the LCAs, as well as Strange adjustable shocks. We started out with the rear shocks set on position 6 and tightened the front struts another 11/42 turn for our next outing. We also added 0.5 psi in the slicks. The changes apparently worked as the car ran an 11.93 right off the trailer at a 2,400-foot corrected elevation. We followed that up with an 11.88 thanks to a 5,500-rpm launch, but after that the track was inconsistent.

While we found no surprises when we took the five-speed to the track, we were surprised by a portion of the results on the dyno. Naturally, we were going to be able to record torque numbers at a lower rpm because we didn't have to contend with a slipping torque converter. And we expected to see higher overall torque output with the stick. But we didn't expect to see less rwhp (see graph).

Dyno pull No. 013, in red, represents the C4 compared to the broader power band of the five-speed, in blue. That fat torque curve (27 lb-ft more), when coupled with the higher gear ratios of the five-speed, resulted in almost 3-tenths quicker e.t.'s and a 4-mph increase in trap speed.

It was particularly surprising because RIPP uses an eddy-current dyno from Dyno Dynamics. These units have a reputation for delivering consistent results. We had made no changes to the car other than the stick. The dyno was loaded the same way it was for the C4 tests, and the air was even a little better at the start. Just to be sure we weren't crazy, we double-checked everything, ran it with and without fans blowing over the engine, and played with the tune-up. The same timing and jetting that made the best power with the C4 also made the best power with the five-speed. The results were certainly consistent, we just saw different numbers than we expected. But as we like to say around the MM&FF offices, dynos-like gauges and weather stations-are just reference tools. It's best to consult them to observe changes and let the real-world results speak for themselves.

So we did. The five-speed has proven to be between 2 and 3 tenths quicker and 4 mph faster than the C4 using the same launch and shift rpm. We suspect there's a bit more e.t. available-if we can tune the suspension to handle a harder launch on the 26x10 slick. If you're wondering why we didn't try bigger tires, a 28x10.5 slick would lower our trap rpm by 500, not to mention slow the engine's acceleration in every gear. This would probably slow the car. Even if we gained better 60-foot times with consistent 6,000-rpm launches, at best we might break even. Our opinion is that PTM would run quicker using a little more gear and the small tire. And it would certainly need a gear change to accommodate a taller tire. But those changes don't fit into our immediate plans. For now we'll stick with this gear and tire combo.

Even more impressive is the five-speed has turned in numbers at 2,000-2,400 feet that the C4 only produced once in mineshaft air (500-700 feet below sea level). Using the formula that every change in Density Altitude is worth 0.01 in e.t., the C4 combo was running exactly what it should have in the heat prior to the trans swap. It would have run 12-flat to mid-11.90s at sea level or below. Using this same formula, we expect the five-speed is capable of running 11.70s in better air.

We called in a few favors and swapped the transmissions at home in the driveway. The entire job can be handled with hand tools and a floor jack, but it helps to have an extra pair of hands. The Tremec was installed along with an FRPP billet-steel flywheel, a heavy-duty clutch, and a Lakewood blowproof bellhousing.

Another thing to consider is that there would be an even greater difference if we were compar-ing a five-speed to a C4 with a 10 or 11-inch converter. With a stall-speed range of 3,000-3,500 (10-inch converter) or worse, 2,000-2,500 which would be typical in a street/strip car, a Mustang outfitted with a mild engine like ours just couldn't accelerate as hard as with the 8-inch race converter we used. You can expect to see similar results if using a T5 or T45 in comparison to a three-speed automatic. An AOD should be compared to these results only if you are using a unit that has been modified to bypass the lockup function and it has increased stall speed. An OEM AOD that releases the converter in high gear is an unfair comparison to a C4.

You shouldn't expect a Pro-Shifted five-speed to be noticeably quicker than a synchronized box. Its main benefit is reliability. It would be hard to miss a gear with a Pro-Shifted trans, but they're not suited to street driving.

Shown here is the C4 installed versus the five-speed. The C4 versus five-speed swap is a direct bolt-in using the same transmission mount and crossmember. You can even use the same driveshaft. If you don't already have an adjustable double-hump crossmember from FRPP, you can grind the welds off your fixed unit so that the mount can be slid forward to accommodate the automatic, or rearward for the Tremec.

Without a doubt, the C4 is easier to live with. In all the time we ran that transmission, whether in competition or test and tune, we never did anything except set the pressure in the tires and the airbag, and make sure the fluid levels were right. It wasn't necessary to give the car a second thought. While it left hard with the transbrake, it always hooked consistently. You don't always have that luxury with the stick. Unless you over-tire the car, or use a tune-up that guarantees adequate traction, you'll be far more susceptible to changes in track conditions. This is the smart approach for racing consistency, but it will leave some e.t. on the table. An adjustable suspension and a logbook full of good data will go a long way toward success with a five-speed.

Another issue to consider is driver consis-tency. It's easy to pull gears with an automatic. Repeatability is critical with a stick if you want to be consistent.

Now, if you want to talk about the fun factor, the stick wins hands down. The satisfaction of winning a race driving a stick car is far greater than driving an automatic. It's common knowl-edge that dropping the clutch and banging gears is all too much fun.

We have a couple more thoughts on the dyno results. The C4 made 302 hp at the wheels versus 290 for the stick. But the torque was 300 lb-ft for the stick versus 273 for the C4. Naturally, the five-speed is going to show a broader powerband on the dyno, and the increased torque output coupled with the gear ratios in the manual are what accelerate the car down the track quicker. But it still bugged us that the car appeared to suffer more driveline loss with the five-speed.

Ultimately, we don't think it boils down to the transmissions themselves. Other than this test, we're not aware of anyone ever trying a stick and an automatic in the same car on the same dyno. So, we really have nothing similar to compare our results to, and in the spirit of letting the results tell the story, we've come up with the following theory. On the rollers there was a dramatic difference in the way the engine accelerated up to 6,200 once the converter locked up over 4,700 rpm. With the five-speed, the engine accelerated slower in high gear between 4,000 and 6,200 rpm. We attribute this to the extra weight of the billet-steel flywheel and clutch versus the lighter chromoly flexplate and converter combo. Frankly, we have no idea of the actual rotational weights of each transmissions internals. Sure, the five-speed has roller bearings riding on a mainshaft compared to the non-rollerized planetaries in the C4, but the Tremec is a large, heavy box. Who's to say it's actually lighter internally? It sure was heavier when we hoisted it into the car on our backs.

One of the few changes necessary to install a five-speed in your Mustang when replacing an automatic is the installation of a new roller pilot bearing to support the input shaft of the trans. Be sure to wipe out the excess grease before installing your flywheel. Natur-ally, if you're swapping from stick to automatic, you'll have to remove this bearing from the crank so the pilot on your converter will fit.

What we're thinking is this: On the track the engine works less to accelerate the car with the five-speed, but on the dyno it works harder to spin the engine from 4,000-6,200. Conversely, the engine works harder to spin the C4 down the track, but has an easier time of it on the dyno.

Where we truly see the most dramatic difference with the five-speed on the track is in high gear. The exact same engine pulls much harder in high gear with the stick. That was evident on the very first pass.

Ultimately, does any of this theory matter? Not really, but since we got ourselves into this mess by adding the aspect of before and after dyno tests, we felt we needed to try to explain it. The bottom line is that the car is quicker with the five-speed, and we're lovin' it.

Gear Ratio - Comparison Chart
Overall First-gear ratio with 4.27 rear gear: 10.50
Tech Specs - Test Car at a Glance
Car'84Mustang: Project Pro Tree {{{Mustang}}}
EngineOEM '91 5.0 H.O. short-block
Cylinder Heads'71 351W, 1.{{{90}}} intake, 1.60 exhaust
CamshaftFRPP X-303
CarbCarb Shop 750 Holley
IntakeWeiand {{{Stealth}}}, port-matched, with 1-in four-hole spacer
Timing40 deg total
ExhaustEqual-length shorties, 2.5-in H-pipe, Flowmaster three-chamber mufflers
RearDana 44 with Detroit Locker
Gears4.27 Precision Gear
Suspension (front)Koni adjustable struts and four-cylinder coils
Suspension (rear)Koni adjustable shocks, Moroso Trick springs, Steeda aluminum lower control arms, FRPP upper arms, airbag
LubricantsAmsoil 10W-30, engine; Castrol Type F, C4; Redline Lightweight Shockproof, Tremec; Amsoil 75W-90 gear lube,rear
TiresP165 radials, front; {{{M}}}/T 26x10, rear