In early 1986, initial development of the upcoming Corvette supercar’s 5.7-liter, dual-overhead-camshaft LT5 V-8 was already underway at Lotus Engineering in Hethel, England. Meanwhile, at General Motors’ Warren Tech Center in Michigan, decision makers were homing in on the best way to construct the high-tech powerplant.
Roy Midgely, Chief Engineer for 90-degree V-Engines; Russ Gee, Director of Chevrolet Powertrain Engineering and Dick Donnelly, Chevrolet Engine Manufacturing Manager, quickly dismissed using a GM engine plant for several reasons. First, these plants were already struggling to meet demand. Second, the LT5 would be a low-volume project unsuited for high-volume facilities. Third, GM’s plants could not produce the engine’s complex aluminum castings. And finally, they were not capable of achieving the necessary level of quality.
In short, the project would require an outside contractor. The team decided to target a U.S. manufacturer with which GM already had a working relationship: MerCruiser in Stillwater, Oklahoma, which had been “marine-izing” Chevy engines for decades. As a world leader in the computer-numeric-controlled (CNC) machining of complex aluminum castings, MerCruiser was a natural fit for low-volume, high-quality, high-performance engine programs. On March 20, 1986, the two corporations cut a deal to build LT5s. Just 21 months later, on December 24, 1987, the first MerCruiser-built LT5 ran on a dynamometer at the Stillwater plant.
In June of the following year, the engine debuted during a 1989 Chevy-model preview held at Riverside International Raceway in California. During an introductory Friday-evening press conference at the Riverside Convention Center, Midgley delivered a briefing on the LT5 while MerCruiser Assembly Manager Chris Allen and Chevrolet Technician Ron Opszynski assembled one of the engines at stage right. As Midgley finished, Opszynski fired the engine and let it idle. Naturally, this demonstration would become the talk of the preview season in ’88.
The next day, media reps were allowed to examine LT5 engine parts at the racetrack. I was impressed with the quality of the machining, and decided I need to visit the facility where the engine was made.
The following December, I toured MerCruiser’s Stillwater facility as part of a story for Road & Track. It was immediately obvious from the bright lighting, low noise level and spotlessly clean work areas that this was no typical engine plant. The LT5 area’s 21,000 square feet were split into machining and assembly. Machining operations were performed in a temperature-controlled room filled with Cincinnati Milacron T10 three-axis machining centers, a cam-bore machine, a straddle mill and other equipment. All of it operated by CNC, with a computer controlling both the cutting tool(s) and how the parts were positioned during the process.
CNC was employed because it greatly increases both production rates and part quality. The T10s were state-of-art mills with automatic tool changing and the ability to position parts to within .0004-inch within a 26-inch cube. Each machining step was then repeated to within .0001-inch of the previous mark.
LT5 engine blocks comprised two parts: an aluminum cylinder case and an aluminum ladder frame with a cast-in-place, iron main-bearing cap girdle. The two were serialized prior to machining and were treated as a unit. Once they were installed in a T10, it took about 45 minutes for the machine to drill all the holes and oil passages, thread the bolt holes, surface the block decks, bore the cylinder-liner locations and machine the remaining flat surfaces. Two subsequent steps used different automated tools to machine the block for crankshaft-counterweight accuracy and line-bore the main bearing bores.
The LT5-assembly process was so technically exacting that MerCruiser was forced to develop new manufacturing technologies. The most impressive example was a tool invented by LT5 Project Manager Terry Stinson to machine the engine’s unorthodox (for a domestic V-8) camshaft bores. Stinson’s creation boasted a tolerance of less than .001-inch and cost $750,000 (around $1.5 million in 2016 dollars)—an astonishing sum for a machine performing a single operation Machining LT5 cam bores without tool chatter, within the specified tolerance and in a production environment, was a global industry first that earned Stinson the Society of Manufacturing Engineers’ (SME) “Young Manufacturing Engineer of the Year” award in 1989.
MerCruiser quality control was a step above what was common in engine plants of the day. In the plant’s temperature- and humidity-controlled Metrology Department were two bench-mounted digital micrometers accurate to one millionth of an inch, along with other metrology devices such as real-time X-ray/video inspection units and fiber-optic bore scopes. All LT5 measuring tools were periodically checked for accuracy on this equipment.
On a recurring basis, finished parts were set up on a coordinate-measuring machine (CMM). Mounted on a six-foot-square steel platform was a moving gantry fitted with a computer-controlled arm that could move vertically. At the arm’s end was a probe that rotated 360 degrees horizontally and moved in a 105-degree vertical arc, allowing the tip to touch any point on the part being examined. The coordinates of each point contacted by the probe was fed to a computer and compared to a “perfect model” stored in memory. In just 45 minutes, MerCruiser’s CMM could check an entire LT5 block—a total of 823 data points. Done manually, the task took 60 man-hours.
“Statistical process control” (SPC) tied all QC efforts together. Joe Carroll, General Manager for Quality Control at MerCruiser, Stillwater, defined SPC as “…using statistics generated by the process to control the process and ensure its stability.”
While GM had been studying the use of SPC for its own facilities, the practice was fully deployed at MerCruiser. Every LT5 machining operation and quality-control activity generated data that was fed to a central computer that produced information used to monitor quality on a piece-by-piece basis. This made it possible to detect unacceptable quality trends before they became detrimental to the process.
At the time of my tour, the LT5 operation’s average QC envelope was 70 percent of the allowable tolerance. For example, that meant that, although the block’s main-bearing bore tolerance was .0004-inch, in practice, it was being held within .00028-inch. MerCruiser’s eventual goal was 30 percent of the allowable tolerance.
The Human Factor
Excepting some head-assembly and engine-balancing tasks, the LT5’s constituent parts were assembled by hand. Consequently, people were key to the project’s success. MerCruiser employees working on the engine were known as the “LT5 Gang.” Their average age was 28.5 years, and 40 percent were women. Atypical of automotive-assembly techs, a high percentage were college educated, thanks to the proximity of Oklahoma State University’s Meridian Technology Center.
One didn’t just lay down a union card on the HR director’s desk and say, “I want to work on the LT5.” First off, there was no union at MerCruiser. Qualifications and experience dictated whether or not one made it into that special department. All LT5 positions were aspirational and went to people already employed at Stillwater. One worker wanted an LT5 job so badly, he took a two-level drop in title and pay to get it.
Applicants had to complete 40 hours of SPC classes. Those applying for a machining role also had to take an 80-hour CNC course. Everyone working the LT5 line cross trained for other tasks. In the words of Jim Cunningham, MerCruiser’s LT5 project manager, “That way, every worker will be able to build the engine. This promotes knowledge of the product and of how individual assembly-line operations interface with each other.”
Assembly used a work-station system. Unique engine stands designed by local supplier RK Machine allowed technicians to easily rotate engines 360 degrees and also hold them in position. The first two stations assembled the short-block. The next few assembled the heads, which required another MerCruiser innovation: a machine, designed by engineer Brian White and built in-house, that allowed one person to simultaneously install all 16 valves, valve springs, retainers and valve locks on a head.
The heads and cams were added to the engine halfway through the job. Another interesting solution to a manufacturing challenge was the process used to apply an anaerobic sealer to the cam covers. This sealer had to be applied in a specific, consistent thickness. The solution came from the apparel industry: MerCruiser used the technique T-shirt makers used to silk-screen designs onto fabric. It worked flawlessly for applying the sealer in exactly the right thickness.
The last stations installed the ignition hardware, induction system and accessories. Then, the engine was connected to a computer that pinpointed any electrical anomalies.
The final QC step was a dynamometer test consisting of initial fire-up, final balancing and break-in schedules, followed by a full-throttle run to verify that each LT5 produced its rated power. Fluorescent dye was added to fuel and oil, so UV light could detect fluid leaks.
During my visit, MerCruiser was building one “pilot engine” per day, though the system was structured for a maximum of 25 daily. (The actual number would reach as high as 22 at the height of LT5 production.) Production got fully underway July 13, 1989, and a total of 6,939 production LT5 engines were manufactured over the next few years. This included 17 “crate motors” for warranty replacements and over-the-counter sales. The tally does not include the roughly 80 development engines built by Lotus, or any of the development or pilot engines manufactured by MerCruiser. The total number of LT5s built is unknown, but may be as high as 7,020.
GM attributed the LT5’s demise in 1995 to the engine’s inability to meet upcoming OBD-II standards, but that wasn’t quite true. A so-called “third generation” LT5 (following the ’90-’92 375-hp and ’93-’95 405-hp versions) was already under development in mid-1991 and slated to power the ’95 ZR-1. Equipped with variable valve timing and other emissions-reduction measures, this new, 475-hp mill would have met the new federal standards.
But around that time the deck was starting to stack against the LT5 for other reasons. The engine was costly and, at 592 pounds, heavy. For 1991, GM restyled the rear fascia of all Corvettes to match the ZR-1’s previously exclusive look, making it difficult to distinguish the base model from its pricier kin. The ZR-1’s sales subsequently cratered, and the engine’s business case became weak.
Furthermore, even an OBD-II-compliant LT5 would not have fit under the upcoming C5’s hood without modifications to the car. And finally, politics likely played a role in the LT5’s demise. Some inside GM were embarrassed at having been upstaged by Lotus’ engineering and MerCruiser’s quality. The easiest way to mitigate that embarrassment was to end the program. In September of 1991, third-gen LT5 development at Lotus was halted; MerCruiser production, meanwhile, ended on November 23, 1993. While the LT5 became Corvette history with the production of the final C4 ZR-1 on April 28, 1995, its influence on GM engine manufacturing is undeniable.
For example, MerCruiser’s James Chen developed “picture process sheets” to increase consistency in machining and assembly. “PPSs” were at every station to show the operator or technician how each task was to be completed; this enabled everyone to work from the same instructions. Starting in 1994, GM rolled out PPS to its own engine plants.
When GM launched the all-new “Gen III” engine family in the mid-’90s—including the C5 Corvette’s LS1—it leveraged LT5 manufacturing ideas to do so. Some of the CNC machines from Stillwater went to GM’s Romulus, Michigan, plant to be used on the early LS1s while permanent manufacturing lines were built for the new engine. Machine tools and processes used to install valve guides and seats into heads at MerCruiser were later employed for Gen III head production at Romulus; they remained in use for 20 years.
SPC, partially responsible for MerCruiser’s impressive quality-control envelopes, was slowly adopted throughout GM’s engine plants. It would go on to become a key factor in making GM Global Propulsions Systems a leader in 21st Century engine technology.
When LT5 production began, there were no coordinate-measuring machines at GM. Noting how these were employed at MerCruiser, GM’s Inspection Department began installing them in the early 1990s. By the 2000s, they were in GM engine plants worldwide.
Earlier I talked about people having much to do with the LT5’s quality. After the end of production at MerCruiser, a few of the LT5 Gang went to work at GM. One was Gary Cline, MerCruiser’s last LT5 Project Manager. In 1995, he began a 13-year career at GM, retiring in 2008 as Manufacturing Engineering Integration Manager for block, crankshaft and cylinder-head production at the Romulus; St. Catherine’s, Ontario; and Silao, Mexico, engine plants. What Cline learned in his years at MerCruiser building LT5s influenced what he did making parts for Gen III and Gen IV V-8s. Later, perhaps fearing he would be bored in retirement, he returned to engine manufacturing as a consultant to GM working on the LT1 and LT4 Gen V small-blocks used in the C7.
For the last 25 years, Cline has also served as the unofficial keeper of the LT5 flame. He appears at Corvette events with his LT5 slide show and organizes the LT5 Christmas party each year in Detroit, which brings together many of the program’s key players.