Thunder and Lightning, Part 2: The AE86 Toyota Corolla Levin/Sprinter Trueno

In the eighties, the Toyota Corolla and its Japanese-market Toyota Sprinter sibling switched to FWD, but not without one last fling for the sporty rear-drive coupes. In the second part of our story, we look at the origins and history of the final RWD Corolla — the 1983–1987 AE86 Corolla Levin and Sprinter Trueno — and consider the later history and fate of the Levin and Trueno coupes.

1986 Toyota Corolla Sport SR5 (AE86) - Corolla decklid decal © 2014 Aaron Severson


As we explained in the first installment of our Corolla/Sprinter history, the first (E10) Toyota Corolla was launched in 1966, followed in 1968 by the Sprinter coupe, which in Japan subsequently became a distinct (albeit still Corolla-based) model line sold through a separate Toyota Auto dealer network rather than through Corolla Stores.

In 1972, Toyota added sportier versions of the second-generation (E20) Corolla and Sprinter coupes, dubbed Corolla Levin and Sprinter Trueno, powered by the Yamaha-developed DOHC 2T-G engine. The TE27 Levin and Trueno were not widely exported, but proved successful enough (both in sales and on the racetrack) to continue through the third generation (the 1974–1979 E30/E40/E50/E60 Corolla and Sprinter) and into the all-new fourth generation (E70), launched in Japan in March 1979.

1973 Toyota Corolla Levin (TE27) front 3q © 2013 TTTNIS (PD CC0 1.0 - modified 2014 by Aaron Severson)

The first TE27 Corolla Levin and Sprinter Trueno coupes were introduced in March 1972. Their stiffer suspension, bigger brakes, and wider wheels and tires (which necessitated the bolt-on fender flares seen here) were available on the U.S. Corolla SR-5, but the JDM cars’ DOHC 2T-G engine was not. (Photo: “1973 Toyota Corolla Levin” © 2013 TTTNIS; dedicated to the public domain by the photographer under a Creative Commons CC0 1.0 Universal Public Domain Dedication, resized and modified (further obscured bystanders) 2014 by Aaron Severson)

The 1979–1983 TE71 Corolla and Sprinter DOHC models (now available in most Corolla/Sprinter body styles, not just coupes) represented something of a departure from the decidedly non-athletic specifications of other Corollas. The twin-cam TE71 models were still RWD cars with live axles and recirculating ball steering, but featured a faster steering ratio, a stiffened suspension with front and rear anti-roll bars, four-wheel disc brakes, a five-speed gearbox with a shorter axle ratio, and the 1,588 cc (97 cu. in.) DOHC 2T-GEU engine with electronic fuel injection — still a novelty in that class.

The DOHC Corollas and Sprinters sold in only modest numbers in Japan, were not officially exported to the U.S., and appeared in only a few European markets (and then generally using carburetors rather than fuel injection). However, the twin-cam cars were quite successful in sedan racing and other forms of motorsport, spicing up a popular but bland subcompact known more for sensible, reliable transportation than sporty flair.


Work on the fifth-generation E80 Corolla and Sprinter began in the spring of 1979, soon after the launch of the fourth-generation cars.

At that time, the future of the Corolla was the center of a heated internal debate. The issue was not whether the Corolla and Sprinter should continue — by that time, their combined sales were around a million units a year worldwide — but whether they should remain RWD or switch to front-wheel drive. By then, front-engine, front-wheel-drive (FF) layouts predominated for smaller B-segment cars, but there was as yet no clear consensus in the C-segment. The Nissan Sunny (a.k.a. Datsun 210), the Corolla’s chief rival in the Japanese market, had for the moment retained rear-wheel drive, but the Honda Civic was FF, as were many of the French, German, and Italian C-segment cars with which the Corolla now competed in Europe.

1982 Toyota Tercel DX (AL10) side © 2012 happy days photos and fun/free photos (CC BY 2.0 Generic)

The 1978–1982 AL10 Tercel, sold in some markets as the Corolla Tercel, was Toyota’s first FWD production car and introduced the A-system engine that later became a mainstay of the larger Corolla and Sprinter lines. Like the Corolla, the JDM Tercel had a “twin” sold through a separate dealer network: the Corsa, sold through Toyopet Stores. The subsequent AL20 Tercel gained a third sibling, the Corolla II, which, naturally enough, was sold through Corolla Stores. (Photo: “1982 Toyota Tercel Coupe de luxe” © 2012 happy days photos and fun/free photos; used under a Creative Commons Attribution 2.0 Generic license)

Toyota approached the FWD question cautiously. The company’s first FF production cars, the AL10 Tercel and its Corsa twin, had bowed only about six months earlier, supplementing but not replacing the similarly priced RWD Starlet. While front-wheel drive offered clear packaging advantages as well as the prospect of better ride quality (thanks to the elimination of the heavy live axle), the Corolla design team was understandably wary of making such a dramatic switch with the company’s bestselling car.

It’s easy to criticize such conservatism, but there’s no denying that cars that sell as well as the Corolla did in the late seventies put their makers in a tricky position: If subsequent generations evolve too little, customers may desert you for newer, more modern products, but changing too much may leave the golden goose with a nasty cough. Compounding the dilemma was the fact that the Corolla and Sprinter enjoyed a quite startling level of customer loyalty. Not only did a lot of people buy Corollas, many of those customers — in Japan, some three out of four — came back for seconds, which made it particularly crucial to avoid alienating existing customers.

Even if buyers proved amenable to idea of a FWD Corolla, making that switch with such a high-volume product promised to be very expensive. The retooling costs alone were estimated at more than ¥120 billion, roughly $600 million at 1979 exchange rates and a good deal more than Toyota had originally paid to build the Takaoka plant where most Corollas and Sprinters were assembled. That was a daunting amount of money even for a company as large as Toyota and did not include the actual development costs of the new model.

1984 Toyota Corolla DX five-door (E80) side © 2010 James Benjamin Bleeker (CC BY-SA 3.0 Unported)

Although the RWD coupes look sleeker, the FWD five-door actually has the lowest drag coefficient of any E80 Corolla or Sprinter: a claimed 0.34, compared to 0.40 for the four-door FWD sedans or 0.35 for the three-door coupe. (Photo: “1984 Corolla DX at 544,204 Miles (875,811 Kilometers)” © 2010 James Benjamin Bleeker; resized 2014 by Aaron Severson and used under a Creative Commons Attribution-ShareAlike 3.0 Unported license)

By the end of the year, Toyota’s leadership had conceded that an FF Corolla would shortly become a commercial necessity, but that still didn’t address the tooling cost problem. The eventual compromise was to make the transition in stages. Sedans and hatchbacks were given first priority, since they appealed to family buyers who put a premium on interior space and comfort. The coupes, for which packaging efficiency wasn’t a major selling point, would remain RWD for another model cycle, as would the comparatively low-volume station wagons and vans. We don’t imagine this was really any cheaper in the long run, especially accounting for inflation, but it did obviate the need to swallow the whole expense in one gulp. Toyota would take a similar approach with the larger Carina/Corona/Celica platform, retaining rear-wheel drive for certain models for several years after the rest switched to front-wheel drive.


The bifurcated fifth-generation Corolla and Sprinter lines debuted in Japan in May 1983 and went on sale in major export markets later that year as 1984 models, although the E70 sedans continued in some export markets for at least another year.

1987 Toyota Corolla (AE82) four-door sedan front 3q © 2011 dave_7 (CC BY 2.0 Generic)

North American E80 Corolla sedans were 166.3 inches (4,225 mm) long, 3.5 inches (90 mm) longer than JDM Corollas of this generation. This 1987 Canadian car presumably has the carbureted 1,587 cc (97 cu. in.) 4A-CL engine, which didn’t have a direct equivalent in Japan; in the Japanese market, all FF Corollas and Sprinter 1600s all had fuel injection by this time. (Photo: “1987 Toyota Corolla” © 2011 dave_7; used under a Creative Commons Attribution 2.0 Generic license)

Although all the new cars were still called Corollas and Sprinters, those badges now encompassed several distinct chassis:

  • Four-door sedans and five-door hatchbacks rode the new FF E80 platform, which featured transverse engines, rack-and-pinion steering, and fully independent suspension with MacPherson struts front and rear. All FF Corollas and Sprinters except the diesel now used variations of the OHC A-system engine previously introduced on the Tercel, offered in 1300 (1,295 cc/79 cu. in.), 1500 (1,452 cc/89 cu. in.), and 1600 (1,587 cc/97 cu. in.) forms. FF cars carried the chassis codes AE80 (for 1300s), AE81 (for 1500s), AE82 (for 1600s), or CE80 (for diesels).
  • Coupes, offered in two-door notchback and three-door Liftback body styles, retained rear-wheel drive and rode an updated version of the previous live axle chassis, but shared the FF cars’ rack-and-pinion steering and A-system engines. The RWD coupes carried the chassis codes AE85 (for 1500s) and AE86 (for 1600s).
  • Corolla station wagons and vans got a mild update later in 1983, but retained the E70 body and chassis with RWD and Hotchkiss drive rear suspension. In many markets, the wagon and van also retained the older pushrod K- and T-system engines, which were not offered on the E80 cars.
  • The Sprinter Carib station wagon actually rode the platform of the smaller AL20 Tercel/Corsa/Corolla II, whose longitudinal engine layout better lent itself to 4WD.
  • In late 1984, Toyota added an additional FF line with three- and five-door hatchback body styles. Those cars, dubbed Corolla FX in Japan and North America, were based on the FF sedan platform and shared the same chassis codes, but had sportier styling.
1987-88 Toyota Corolla FX16 GT-S three-door (AE82) front © 2009 IFCAR (PD - modified 2014 by Aaron Severson)

Introduced in Japan in October 1984, the three-door Corolla FX hatchback was originally intended mostly for the European market (although the FX name wasn’t used there for trademark reasons), but the three-door’s late introduction proved a commercial miscalculation in Japan, briefly costing Toyota some market share. This is a 1987 or 1988 U.S.-spec Corolla FX16 GT-S, powered by a transversely mounted version of the 4A-GE engine making 108 hp SAE (81 kW). (Photo: “Toyota Corolla FX16 GT-S front” © 2009 IFCAR; released into the public domain by the photographer, resized and modified (reduced glare) 2014 by Aaron Severson)

1986-88 Chevrolet Nova five-door front 3q © 2011 dave_7 (CC BY 2.0 Generic)

The E80 Sprinter also became the basis of the 1986–1989 Chevrolet Nova, built by GM and Toyota’s jointly owned NUMMI plant in Fremont, California. The Nova had different exterior panels than the JDM Sprinter, a unique grille, and the obligatory U.S.-spec bumpers. Most Novas used the same carbureted 1,587 cc (97 cu. in.) 4A-CL engine as U.S.-market FWD Corolla sedans. (Photo: “Chevrolet Nova” © 2011 dave_7; used under a Creative Commons Attribution 2.0 Generic license)


Add a Comment
  1. Wow! What a great and detailed history you have written. Thanks.

  2. This was truly well worth the wait! Very comprehensive article – the AE101 Levin and Trueno shared a dashboard with a 4 door hardtop called either the Sprinter Marino or the Corolla Ceres, some of which also carried the 20V engine. A friend of mine ran a few of these cars, and the early manual ones definitely needed a 6-speed gearbox – you also had to be a skilled driver to rein in some of the torque steer. The boy racers graduated from the Suzuki Swift GTi to these in the late 1990s when they became available as foreign used cars in the Barbados market. There are still a lot of them around in varying stages of tune – it’s become increasingly difficult to find one that’s completely standard.

    1. Thanks, Dave. It hadn’t occurred to me that the Levin and Trueno dash was the same as the Ceres/Marino, but I looked it up and you’re quite right.

      I don’t think a lot of my U.S. or European readers will probably have heard of the Ceres and Marino, so I added a photo. Four-door “hardtops” like these were very popular in Japan for a number of years, going back to the late seventies. Most were not actually pillarless hardtops in the traditional sense, but they approximated that look by concealing the B-pillars (designing them to be thin in profile, painting them black, and putting them behind the door windows rather than between them), using frameless door glass, and sometimes adopting a more coupe-like roofline. A couple of examples with which readers may be more familiar include the ’90s Integra four-doors, the Lexus ES250 (which I believe was based on the JDM Camry Prominent hardtop), and the last U.S.-market Mazda 929.

      The Corolla Ceres and Sprinter Marino were offered in three grades (F Type, X Type, and G Type) that approximated the AE100/AE101 Levin/Trueno S, SJ, and GT grades in engines and equipment. There wasn’t an equivalent to the GT APEX or GT-Z, so as far as I know you couldn’t get a Ceres or Marino with Super Strut or the 4A-GZE engine, but as you mention, the G Type did have four-wheel discs and the 4A-GE TWINCAM20 engine. You could also order a sports package that included the coupes’ spoilers and other cosmetic bits, so a properly equipped Ceres G looked and performed a lot like an AE101 Levin GT.

  3. It looks to me that the Super Strut suspension worked by simply divorcing the strut from the steering, as with Ford’s RevoKnuckle on the Focus ST and GM’s HiPer strut on the Regal GS/Insignia and that the extra lower arms allow better steering geometry and reduced offset, a la BMW’s double pivot and the lower-half of Audi’s four link design. It probably also induces camber gain, as you speculate.

    1. You’re right — I hadn’t previously looked closely at the layouts of the RevoKnuckle or HiPerStrut (or whatever Renault calls theirs), but the basic principles look to be very similar, just executed a little differently in each case.

      1. I’ve been studying these systems more closely and Super Strut was actually somewhat more complex than the current HiPer Strut, RevoKnuckle, and PerfoHub. The newer setups have the relocated steering axis and reduced spindle height, but Toyota also did something very complex with the way the strut extension is pivoted to the rear lower arm to allow more camber gain. The GM, Ford, and Renault setups give a little more camber gain, but looking at the way they’re set up, I’m reasonably confident that it’s not as much as Super Strut provided. The tradeoff is that they’re also not as complex or as expensive and will hopefully be more reliable.

  4. Love this site! I love the Levins from start to finish. I own an AE111 Levin with the 20 valve blacktop and man isn’t it fun to drive!

  5. A great article.

    Several of the AE86 Levins that competed in the Australian Touring Car Championship in the mid to late 80’s are now racing in historic touring car racing – they are great little cars!

  6. My mostly stock 4A-C powered USDM AE86 with the manual transmission (slightly modified intake/vaccum setup and rear muffler delete ,keeping catalytic converter) has hit GPS Verified 101mph on flat ground, I’m sure if it was a little fresher (mine has 230k miles) it might hit 105. That’s getting close to 5,000rpm in 5th which is right when peak power starts to fall off, so anything beyond that would be stretching it.

  7. Thanks, I enjoyed the article… I’ve had my AE86 Trueno for 23 years now… Handling in stock form was pretty scary. Not sure how they got the factory JDM model to 2090lbs. Mine fully stripped for racing with no interior and sweating every gram is still 2030lbs.

    1. I’m a little puzzled by the JDM curb weights myself; I’m not sure what Japanese vehicle regulations require in terms of fuel and fluids in those calculations, but the quoted figures seem consistently lower than what I’d expect from a fully equipped and fueled car. (That’s a frequent issue with manufacturer curb weights regardless of country of origin.) Unfortunately, there’s not much to be done about it short of putting the car on a scale or finding some reasonably neutral third party who has. For that reason, I tend to find factory weight figures of most use for comparison purposes — for instance, the weight difference between trim levels.

  8. On the subject of the Toyota A engine, aside from Toyota reputedly drawing inspiration from the Cosworth BSA (some go as far as to say it is a reliable copy), does any relation exist between the Toyota A and Daihatsu C-Series engines?

    Preexisting ties between Toyota and Daihatsu notwithstanding, it is difficult to ignore the fact both engines appeared roughly at the same time in 1977-1978, feature cast-iron blocks with alloy-heads and belt-drives, the 1-litre Daihatsu CB and 1.3-litre Toyota 2A virtually share the same 76mm bore with Daihatsu even making use of Toyota’s lean-burn design system.

    1. I’m not familiar with the Daihatsu engine, but Toyota owned a big chunk of Daihatsu by then, so it’s not unlikely. It should be noted that the lean-burn system Toyota used in the late seventies (in the JDM 12T engine and others) was actually a Honda design used under license, so while that in no way contradicts your theory, it’s not probative either. Also, as I understand it, the early Toyota A-system engines (of which I think the first was the 1.5-liter 1A in the late seventies Tercel) ended up being somewhat troublesome and requiring some further design changes (details of which I do not know) to yield the later 3A and 4A engines of the eighties, so if the Daihatsu CB is related, I don’t know where it falls in that development sequence.

      As for the Cosworth comparisons, I’m leery of those. That engineers at Toyota (and/or Yamaha) were familiar with the Cosworth twin-cam engines is not unlikely, but Cosworth did not invent belt-driven OHCs (as discussed in the Pontiac OHC six article), and a great many of Toyota’s seventies engines were SOHC or DOHC with aluminum heads on iron blocks, so there was a lot of prior Toyota experience in many of those areas. Likewise at Yamaha, which had designed the heads for many of Toyota’s DOHC engines; Toyota’s official information on the 4A-GE is somewhat vague about Yamaha’s involvement with that engine, but Yamaha had done the heads for the 2T-GE, 18R-GE, and the later 1G-GE six, inter alia.

      I’ve noticed that British sources tend to be particularly insistent that any Japanese invention or design of any merit must necessarily be a copy of some (implicitly superior) prior British or European design, which I find frustrating, to say the least. It isn’t a strictly yea-or-nay question, since of course some Japanese automakers of the fifties and early sixties did use British technology, and at times hired European designers or consultants, but Toyota, in particular, was and is an enormous company which by the seventies had tremendous depth of engineering and manufacturing resources. While they were by no means adverse to licensing outside technology — the aforementioned Honda lean-burn system for one, Bosch D-Jetronic and L-Jetronic electronic injection for another — doing so was a matter of expedience rather than competency. So, I tend to take the instance that “the 2T-G was a copy of the Lotus Twin-Cam” or “the 4A-GE was a copy of the Cosworth BDA” with a grain of salt.

      If I were an engineer, I might be able to provide a more detailed comparison between the Cosworth and Toyota engines in combustion chamber design and so forth. Lacking that skill, I will say instead that the development brief for the A-system engine was to replace the existing K- and T-system engines with an engine family better-suited to eighties emissions standards, which by 1980 were as strict in Japan as in the U.S. There were twin-cam performance versions of the A-system, but on a numerical basis, its primary applications were in mildly tuned, frequently carbureted SOHC form, for which the high-strung Cosworth was probably not the most useful of models.

      1. The Daihatsu C-Series is largely associated with the Daihatsu Charade including potent 1-litre turbocharged engines from the 68 hp Charade Turbo and 101+ hp Charade GTti.

        Given the ties it is difficult to imagine Daihatsu going with a completely clean-sheet design for the Charade as both the car and engine appeared in late-1977, followed in 1978 by the Toyota A powered Toyota Tercel.

        1. It’s possible; as I said, I don’t know. FWIW, the 1987 Jikayousha buyer’s guide I have identifies the four-cylinder engines in the (later-generation) Charade as the 2A-U and 3A-U, which are Toyota engine codes (1.3 and 1.5-liters respectively) and presumably just Toyota engines.

      2. Belt driven DOHC engines were built concurrently by Fiat/Lancia, Cosworth, and apparently by several Japanese companies. I don’t doubt for a minute that I have missed many other makers from other countries who also utilised belt drive for DOHC engines.
        Who built the first one, and the first offered in mainstream cars I have no idea.
        But I think it was convergent evolution that brought them about. It was no secret that DOHC was a tremendous help in building an efficient engine. The problem designers and engineers faced was reliability and cost for production engines. The appearance of toothed belts was a useful tool in addressing both concerns.
        It seems technology has now caught up with the problems enclosed chains had, more cars are appearing with chain driven valve gear, improvements in lubrication and probably the metals used mean that the timing chain on a well maintained modern engine will probably last the life of the car.


        1. The Pontiac OHC 6 article talks at some length about the history of belt-driven OHC engines, of which Glas was the first series production manufacturer. Pontiac did a great deal of work on developing timing belts suitable for more powerful engines, culminating in the 1966–69 Pontiac 215/230 cu. in. SOHC six and a variety of experimental V-8 engines. The belt was quieter than a chain or gears as well as being cheaper, which added to its appeal.

          The eventual limitation, of course, is that with transverse engine/FWD layouts, or even longitudinal RWD ones where the engine bay is cluttered with plumbing and accessories, changing the belt is a bear of a job, and with some valvegear layouts, a broken or slipping belt can mean serious engine damage. As expectations have increased regarding what “the life of the car” ought to mean, that’s become harder to accept, to the point that even Honda (long a proponent of belt-driven cams) has switched to chains in the past decade or so.

Leave a Reply

Your email address will not be published. Required fields are marked *

Comments may be moderated. Submitting a comment signifies your acceptance of our Comment Policy — please read it first! You must be at least 18 to comment. PLEASE DON'T SUBMIT COPYRIGHTED CONTENT YOU AREN'T AUTHORIZED TO USE!