The first generation of turbocharged Mopars were not part of the executives’ original plan, but they were tough, well-engineered, efficient, and, with the Turbo IV of the 1990s, innovative enough that modern turbos follow in their path. All were four-cylinders, from the same family, made in two displacements (one was mildly stroked). The more powerful of two designs sold in 1984-89 cars nearly doubled the power of the setup it was based on.
One intelligent decision, pushed hard by engineer Pete Hagenbuch, was selling all the turbo engines with multiple-port fuel injection in the US, unique among Chrysler engines of the day. The company leaders hated to spend money, but they could be pushed into doing so for a premium engine.
Two of the designs for the 1980s were dubbed “Turbo I.” These were the original 2.2 turbo, with 142 horsepower at launch, and the longer-stroke 2.5 liter version of the same engine. The third version, dubbed Turbo II, used a charge air cooler (which nearly everyone calls an intercooler, so we will too); these produced 174 horsepower, and as of 1988, had black manifolds to distinguish them from the others.
For the 1990s, buyers could get a “Turbo III,” whose primary differences were not so much the turbo setup but a modified engine with dual overhead cams, operating four valves per cylinder. Pushing out more than one horsepower per cc, it produced 224 horsepower, with high torque as well, and featured then-unusual returnless fuel injection.
A final design, the Turbo IV, used a variable nozzle design jointly developed by Chrysler and Garret. This was meant to sharply reduce turbo lag, and by all accounts it did so — but that’s a story, like that of the Turbo III, for Mopar Turbos of the 1990s.
There were also an M4S concept car and the jointly developed Chrysler TC by Maserati, which used Cosworth heads, but we’re not getting into those rather specialized items.
Though Chrysler set its sights on turbochargers, lead engine performance man Pete Hagenbuch thought superchargers were the way to go; he believed supercharging was ideal for street engines, while turbos were best for racing. He never got authorization to build even a test supercharged 2.2, though. Pete argued that turbochargers produced too much heat for normal driving; he believed he could overcome friction losses on superchargers within six months.
Technical notes for all the turbocharged four-cylinders
The 2.2 and 2.5 liter engines were noninterference designs; the valves and pistons did not share any space. If (when) the timing belt broke, there was usually no damage from pistons hitting valves.
Originally, there were different cast-iron blocks for the 2.2 and 2.5, both with siamesed bores, a short-skirt design, an internal oil pump, and a partial open deck. Weight was dealt with by milling.
Since boost was based on exhaust volume, more load brought more boost, until the wastegate lever opened and let exhaust bypass the turbocharger. Engine computers monitored numerous sensors (engine revs, knock sensor, wheel speed, intake temperature, battery temperature, throttle position, exhaust oxygen, coolant temperature, and manifold pressure; and, after 1984, air conditioner and electric fan status) and adjusted boost (after 1984), fuel, and spark as needed.
Chrysler did not use airflow sensors at the time, because engineers felt they were both too expensive and not reliable enough; instead, they used a manifold pressure (MAP) sensor to tell the computer how much vacuum or boost existed. Periodically, the computer would also measure ambient air pressure with the MAP sensor. There was no actual load sensor; that was computed. After 1984, again, overboost was allowed for up to ten seconds.
Fresh oil cooled the turbocharger bearings, while a water jacket brought antifreeze around the bearings and the rest of the turbocharger. There was some air cooling as well, through ambient air movement.
While the Turbo I used cast crankshafts, all but Shelby Turbo IIs had forged cranks, as did the later Turbo III and Turbo IV.
The 2.2 and 2.5, in brief
The 2.2 liter engine was developed and built entirely by Chrysler, despite some rumors that Mitsubishi was involved. It was also a clean-sheet design, not a variant of the slant six, as other rumors claim; it did share some measurements with that venerable engine, to cut tooling costs a little.
Engineering started in the mid-1970s; the original specifications were for a single-overhead cam 2.0 liter design, with iron block and head, which would power a car based on the European Chrysler Horizon (whose design had already begun). It quickly became apparent they would need another, heavier car, to be the Reliant and Aries, and the displacement was pushed up to 2.2 liters.
Engineers considered maintenance, warming the intake and carburetor with exhaust gases for faster warm-up, protection against crashes, and making the engine easier to install alongside Volkswagen four-cylinders in the Horizon/Omni assembly line. Cost, durability, performance, and maintenance were all key factors in the design and engineering; turbocharging and a long-stroke version were not. It was made in Trenton, Michigan, in an area once used for big-block V8s, appearing in the 1981 cars. There were computer-controlled test stands for checking engine parameters—a cold test stand where the stand powered the engine, and a hot test stand where the engine itself provided power. In 1981, the 2.2 provided 84 horsepower and 111 pound-feet of torque, with a two-barrel carburetor; that was good for the time. For 1983, changes including revised intake and exhaust ports and higher compression yielded 94 horsepower and 117 pound-feet of torque; with 100-horsepower and even 107-horsepower variants. One million engines were built in three years.
By the time the first turbo engines were made, a single fuel injector had replaced the carburetor in some cars, with output at 99 horsepower and 121 pound-feet.
Turbo I: the first-ever Chrysler turbo engine for cars
Chrysler produced its first-ever factory-made turbocharged engine for the 1984 Dodge Daytona; it would reach other cars later. The Daytona, the second American front-drive turbo car ever, had a 2.2 liter four-cylinder, producing 142 net horsepower at 5,600 rpm and 160 lb-ft of torque at 3,200 rpm. That might not seem like much, but consider the day; the Mitsubishi 3-liter V6 turned in around the same numbers, and the standard 2.2 liter four-cylinder only made around 100 horsepower and 130 pound-feet of torque (rounded). Fuel economy was superior to the V6 and competitive with the base four-cylinder.
For this first engine, Chrysler used Garrett Research’s T-3 turbocharger unit, which had a built in, mechanically controlled wastegate* and could reach 7.2 psi of boost. The wastegate acted on the difference between throttle body vacuum and compressor output, bypassing the turbine when pressure got too high. The housing itself was iron, though the turbine was aluminum; the bearings were lubricated by oil under pressure. To prevent heat loss, the turbocharger was mounted as close as possible to the exhaust manifold.
* The “wastegate” diverts part of the exhaust stream into the turbocharger housing, where it spins a turbine around; another part of the turbine pushes air into the engine. It is essentially an exhaust-powered fan which, by packing more air into each cylinder, effectively enlarges the cylinder, allowing a four-cylinder to have six-cylinder-like output when needed. It may seem as though, if you’re pumping six cylinders’ worth of air into the engine, that you may as well just have a six-cylinder; but when extra power is not needed, the wastegate lets all the exhaust through instead of diverting it, and the engine runs with its actual size. In some cases, e.g. the modern Fiat 1.4 liter “FIRE” engine or the new 1.3 liter “GSE” engine, the turbocharger can double the power of the engine when power is needed—while the engine runs on a trickle of fuel at other times.
Chrysler did not just toss a turbocharger onto a standard engine. Again at Pete Hagenbuch’s insistence, they built in a detonation (knock) sensor so if people used lower-octane fuel, it would not destroy the engine; when the computer saw detonation, it dialed back on the ignition timing. (Pete had to fight to get premium fuel approved for the engine at all; it did make a substantial impact on power output.) The fuel injection system, engineered and made by Chrysler at its space-electronics plant in Huntsville, Alabama, had far higher fuel pressure than past systems, leading to fuel leaks in early cars as they got older. Transmission choices were the A-525 manual transmission and the three-speed automatic.
Engineers lowered the compression by using special pistons, which for performance reasons were lightweight and strutless. Durability measures included stronger valves, springs, rockers, rods, and cam, with select-fit bearings and better rings. The engine was topped by a die-cast aluminum head cover; Pete Hagenbuch and legendary Viper engine-man Dick Winkles wanted a rounded top for the engine, for better airflow, but the Daytona team wanted a flat top with ribs for looks, and they won out. Chances are the effects on performance were small.
Pete Hagenbuch, the man responsible for the turbo engines’ performance, told me (when I was at allpar.com) that “We had no one in-house who knew much more than the very basics. It was pretty much learn as you go. The electronics folks at Chrysler were not any better off…We learned a lot about turbocharging and, yes, the 2.2 responded to everything we did.”
Mitsubishi was absolutely not involved in the design and engineering of the 2.2 or 2.2 turbo, aside from selling their turbochargers to Chrysler in later years.
Turbo I – 2.5 liter
The original 2.2 turbo produced 142 horsepower. The original 2.2 liter Turbo I was replaced in the 1989 cars by a 2.5 liter version which had better torque right off idle; a car with this engine covered 10% more distance in the first five seconds than in the Turbo I. This was especially important with minivans, which for the first time had an engine more powerful than the not-very-strong 2.6 liter Mitsubishi four-cylinder (the base engine was a single-injector 2.2 liter Chrysler four-cylinder). The turbo engine was mainly a placeholder for a Mitsubishi V6, partly because Mitsubishi was having a rough time supplying all the engines Chrysler wanted; but with a stick-shift, it made the minivan move quite well. (It was poorly matched to the three-speed automatic, partly because the automatic’s shift points were not tuned to the engine.)
The 2.5 liter engine had counter-rotating balance shafts, a clever addition to an engine never meant to have them; that smoothed it down a bit.
The Turbo I had a new block, shared with the Turbo II engine. This had diagonal coolant passages drilled between cylinders (lining up with passages in the head) for better cooling. There were also new aluminum-alloy pistons, with cast-in steel struts to control expansion; a dished crown to get the right compression ratio; and a high-hardness iron crank, which is not interchangeable with the 2.2 crank, because of clearance issues. Power output was 150 horsepower (at 4,800 rpm), with 180 pound-feet of torque. As with all Chrysler’s four-cylinder turbo engines, it required premium fuel, but a knock sensor made it forgiving of regular. The turbocharger assembly had some minor changes in the 2.5 for higher airflow, as well as a new oval (rather than round) air cleaner.
This engine was quite satisfactory in cars, but was overworked in the minivan; it worked best with the efficient Getrag manual transmissions. Because the automatics were three-speeds, it was not particularly well suited to use in automatic transmission cars, though most buyers preferred automatics. Chrysler would not have a more efficient four-speed automatic until 1989, and that transmission would never be hooked up to a turbo engine.
Turbo II (1987 to 1989)
Originally, the engineer in charge of the Turbo II setup was Dick Winkles, later to take charge of Viper engines (after a stint in charge of four-cylinder racing). Testing and final development started out in May 1985.
Some early intercooled engines were run by Carroll Shelby’s 1986 Omni GLH-S, but the hardware and software work was done by Chrysler; that should be no surprise, since Shelby’s operation was relatively small, and a production engine needs work in emissions, fuel-system and spark calibration, hardware durability, electronics, and supplier coordination. Shelby’s early engine lacked the bottom-end toughening of the Chrysler version, which had roughly the same output; torque on Shelby’s own intercooled engines was reduced somewhat, ostensibly for transmission durability. Chrysler’s version had a stronger crank than Shelby’s, and a one-piece intake; both had 1984-85 Turbo I rods, Mahle pistons, larger injectors.
In essence, the Turbo II was a Turbo I with a charge air cooler (intercooler); because this brought power up to 174 horsepower and 200 pound-feet of torque, the engineers made numerous changes for reliability, including a forged crank, small tubes between the cylinder bores at the top deck for cooling, and heavier-duty manual transmissions (not part of the engine, but part of the powertrain package). The Turbo II shared connecting rods with the 2.5 liter Turbo I engine; they were double weight sorted for balance.
One can ask how the intercooler (charge air cooler) increased peak power so much; the answer was by dropping the air temperature by up to 120°F, so boost could peak at 12 psi rather than 7.2 psi. Colder air is denser air; denser air has more oxygen to burn, so more fuel can be injected. Again, it was a matter of taking the same space in each cylinder and making it seem as though it was larger—adding more air and fuel.
The Turbo II was used on the 1987 Shelby Z; the 1988 Lancer ES (optional) and TC automatic (standard); and, finally, the 1989 Daytona Shelby and C/S, Shelby Lancer, LeBaron GTS, and LeBaron (optional). It was a sturdy, powerful, reliable engine.
The Turbo IV was mainly distinctive because of the variable geometry turbocharger, developed by Chrysler and Garrett together; Dick Winkles was deeply involved in the project. The Turbo II was originally to have the variable-geometry turbocharger, but it was not ready in time—some would say it was not ready when it was launched in 1989. Kim Lyon and Stuart Davis were also deeply involved in this project, the last Pete Hagenbuch was involved in.
While the Turbo III was probably the most advanced system, overall, the Turbo IV (which, confusingly enough, appeared earlier in time) had the most advanced turbocharger, boasting variable-nozzle technology (VNT). This increased boost at lower engine speeds, fighting turbo lag. The top horsepower was the same as in the Turbo II, while torque rose from 200 to 210 pound-feet with the new system; but the engine reached boost in half the time of the Turbo II, giving the engine a broader torque curve, and it was smoother, not starting out weak and then slamming the driver with the seat-back as the Turbo III did.
Instead of a wastegate, the VNT system had a solenoid-driven actuator which moved the vanes (based on the computer, which, as in past engines, used manifold pressure as its cue). One ring moved the 12 vanes to adjust the flow of gases to the turbine; the vanes pivoted around the turbine wheel.
Chrysler also put in balance shafts to smooth the idle out (in 1990, all the turbo 2.2s would have them), as well as sequential multiple-port fuel injection (rather than having fuel splashed against closed valves, as in the past) and a new crankcase ventilation (PCV) system with an improved oil separator to fight leaks.
The Turbo IV was used in very few cars—the Shelby CSX (but not the CSX-T); the Chrysler LeBaron coupe and convertible; and one model of the Dodge Daytona. An early flaw gave the design a poor reputation for reliability. According to Burke Brown, speaking with me for Allpar, they dealt with clogging problems by flipping the nozzles back and forth a few times to clean them whenever the system was started.
Year by year changes
Chrysler dropped the mechanical wastegate control after just a year, replacing it with computer control. The engineers cleverly programmed in calibrated pressure limits, and the computer activated or shut off the turbine via the wastegate actuator solenoid. At higher boost pressures, a spring pushed on a rod to open the wastegate, allowing exhaust gases to leave without spinning the turbine. The new system allowed for short bursts of 9 psi boost, with better detonation control; horsepower rose from 142 horsepower to 146. Numerous other changes were made, some as running changes (e.g. lightweight connecting rods, anti-drainback oil valve, and 11 mm head bolts); most were done for durability (high temperature timing belt, material added to the head and block at the oil transfer hole, a revised oil pump, and a late-launch 8-bolt flywheel. A water box was integrated to the bottom of the intake, and the intake valve surface was improved. Unknown to the public, Chrysler started testing a four-valve-per-cylinder version on an AWD Daytona with a 10 psi turbocharger and 225 horsepower.
The 1986 cars gained a new head design which burned fuel faster, which allowed wide-open-throttle spark timing to be backed off (pistons were also modified). That reduced problems of spark knock and sensitivity to low octane. The company also launched a new distributor with fewer parts. Numerous other changes were implemented, again mainly for reliability, including thicker head gaskets, new valve covers designed to prevent splashing, new valve springs, longer intake and exhaust valves, better rod caps, and a common dipstick.
For the 1988 model year, the engineering group achieved higher output from a long-branch intake manifold, which included a tuning effect and better fuel-air ratio distribution; this let the engineers have a more aggressive spark advance curve without detonation (knocking). (The earlier intake was referred to as the “log” manifold.) This new intake was used for both Turbo I and Turbo II; the Turbo II versions had a charge air sensor, which was replaced by computer programming on the Turbo I engines from 1988 onwards.
A smaller Mitsubishi turbo replaced the Garrett T03, largely to spool quicker; torque rose and arrived at a lower engine speed, reaching 170 pound-feet at 2,400 rpm on the Turbo I. The compression ratio was dropped to 8.0:1.
For 1989, again, the 2.5 replaced the 2.2 as the Turbo I engine, and a new common block replaced the old ones for all the 2.2/2.5 engines, turbocharged or not. It had stronger main bearing caps and supports; engineers replaced the 2.5’s long deck with a shorter piston.
Follow the engines further in our 1990s section…
Roads left untaken
Lotus’ Michael Royce wrote to me when I was at Allpar to describe three development programs run by Lotus Engineering. Chrysler Vice President of Engineering Robert Sinclair signed the contract for these in March 1985.
One was for a 16-valve turbocharged engine, program A-522, which became the Turbo III. Another was for a 16-valve, naturally aspirated 2.5 liter engine, which would resolve the engine’s breathing issues and make it smoother, with a higher redline, providing roughly 140 horsepower. This did work out, but was killed due to budget issues in mid-1986. The engine was running in a Sundance or Shadow with a manual transmission, and in emissions testing, when it was dropped.
They dropped the AWD Dodge Daytona program in late 1987, again due to budget shortfalls, though the car reportedly performed and handled as well as the Audi Quattro.
Oddly, the Turbo III, a troublesome and presumably expensive engine, was not dropped and made it to fruition; admittedly, it had the highest power potential.
See Chrysler’s 1990s turbocharged engines