Photos Courtesy of Honeywell Garrett/Tim Coltey
As the 5.0 Mustang became cool and popular in the late ‘80s and early ‘90s a scary notion of “catalog horsepower” became the norm. A relatively stock Mustang with bolt-ons could eclipse the 350-horsepower barrier by simply whipping out your calculator and adding up a bunch parts from whatever mail-order catalog arrived that week.
It was an era that predated the popularity and accuracy of chassis dyno testing while engine dynos were exclusive to high-end testing and big-budget racers. It is no surprise that horsepower-happy promotions blindly threw bold claims out there, and sadly, a lot of people believed them.
Times have changed considerably since those early days and most aftermarket upgrades seem have been well-vetted by the manufacturers before the parts hit the market. That’s because the enthusiasts today are armed with the ability to test products on engine or chassis dynos with relative ease. And there is that thing called the federal government who will bust a company for false advertising.
Every manufacturer is equipped with the ability to test their products through special test stands regulated by SAE and dyno testing of all types. Most companies go to the extreme to validate their claims and — sometimes — magazine editors get invited to peek behind the curtain and witness the testing firsthand. That is what happened this past summer when Honeywell Garrett and Modular Motorsports Racing invited us to tag along as they put the Garrett GTX5020R Gen II through its paces on a Coyote engine.
We know the MMR guys from NMCA racing where they bring two VP Racing Fuels Xtreme Pro Mod Mustangs to play with the other boosted and nitrous-injected racecars. The company has a California-based engine shop that specializes in Ford Modular and Coyote crate engines and performance parts. Recently, the MMR crew installed a SuperFlow engine dyno, and as Garrett-sponsored racers, they offered the use of the dyno to the turbo company.
“One of the things that Garrett wanted us to do was to max out the GTX5020R turbo, as a way to verify their ratings on the turbochargers that they bench tested in-house,” Mark Luton, of MMR, said. Once the dyno was installed, Tim Coltey of Garrett loaded up the truck with 88mm and 76mm versions of the GTX5020R Gen II and headed to MMR’s shop in Camarillo, California.
The test engine wasn’t a MMR Gen X billet bullet, but instead they opted to use the company’s popular MMR 2000 long-block assembly. Beginning with a ’11-’14 Coyote engine block, MMR knocks out the factory sleeves and adds an aftermarket set to ensure durability. The forged pistons are designed by MMR and built by Manley, which are fastened to MMR billet connecting rods with ARP 2000 bolts. Surprisingly a forged factory crankshaft has been selected, and according to Luton, durability hasn’t been a problem. The larger 3.700-inch bore combined with the stock-stroke crankshaft brings displacement to 5.2 liters.
Oiling is important when you are shooting for 2,000 horsepower and MMR uses an internal oil pump setup, complete with its own billet gears. The sheetmetal oil pan is custom by MMR and uses a highly modified factory pickup inside. Flipping the engine over shows it wears a set of GT350 cylinder heads that were enhanced by a custom MMR CNC-porting program. The factory valves remain in use but MMR valvesprings and retainers keep them under control. The MMR Stage III turbo cams, which are ground by COMP, are bolted in place and the factory GT350 followers and lifters are utilized.
Sealing the heads is critical when the goal is to cram over 40 psi of boost into the engine. Holding the heads down is the job of ARP head studs, which are the standard size that you’d normally find in ’11-’14 Coyote applications. The GT350 heads were dry-decked, meaning the water passages were sealed off to help durability and prevent water from pushing out between the heads and block in the event of a head-gasket failure. MMR also installs O-ring grooves in both the engine block and cylinder heads. Stainless steel rings, which are slightly offset to help lock them in place, are installed in the O-ring grooves and a set of copper head gaskets are sandwiched in place.
The dyno testing served a dual-purpose — first to validate Garrett’s turbocharger ratings and second, to give MMR a chance to evaluate its brand-new billet intake manifold. The prototype manifold has since been released and it was seen here for the first-time. This billet intake is available with single or dual injector runners and an air-to-water intercooler can be installed in the main plenum area. This test saw the use of a Wilson Manifolds 90mm throttle body bolted to the front of the intake.
The SuperFlow dyno at MMR is equipped with a FuelTech FT600 engine management system, which is hard-wired into the dyno cell. The engine shop uses that engine management on both racecars so it made sense to carryover the system to their test facility. Another carryover to the dyno is the use of two sets of fuel injectors — one set of eight Billet Atomizer 270 lb/hr injectors is paired with eight Precision 550 lb/hr injectors. According to Luton the duty cycle was only 40 percent on this test but MMR used the start-up maps and fuel curves from the racecar program, making it convenient. The Brisk spark plugs get their electrical energy through a FuelTech FT Spark box that uses eight coils. MMR’s Greg Seth-Hunter was responsible for tuning the engine with the FuelTech FT600 and he only need a few pulls to get it all sorted.
The engine bolts to the dyno easily and the custom headers were made in-house at MMR, along with the crossover and cold-side pipe. It is a simple setup that can fit in any late-model Mustang. Run exclusively on methanol, the engine didn’t require an intercooler. All testing was done with the TiAL wastegate closed using a stiff spring and C02 pressure, allowing the turbocharger to spin as hard as it could go. Luton also installed a TiAL blow-off valve on the discharge pipe to help relieve pressure when the throttle blade closed.
Using an engine acceleration rate of 600 rpm/second the dyno runs offered excellent resolution for tuning and data analysis. Luton noted, the engine accelerated from 500 horsepower up to 1,200 horsepower in less than a second. As a result, the dyno quickly attempted to get control of the acceleration rate before it went on to make a complete pull, making this combination tricky to run. Each pull saw the engine go to over 8,000 rpm but the peak power was below that number — regardless of turbocharger size — and each result is reported in standard correction factor.
As for the results, the 88mm turbo cranked out 1,944 horsepower at 7,500 rpm and the torque needle twisted to 1,532 lb-ft. The boost reading at peak horsepower was 42.4 psi while ignition timing peaked at 20 degrees. Luton believed that if they had more time, he could’ve coaxed 2,000 or more horsepower from the 88mm setup with more ignition timing and tuning. However, 1,944 horsepower is hardly anything to be ashamed about. There was a satisfaction in the room that the 2,050-horsepower rating on the turbocharger is definitely not overstated.
Despite the big power from the 88mm unit, there was still the GTX5020R Gen II 76mm sitting on the shelf. Since both turbochargers utilize a T6 exhaust flange and the same compressor cover, the turbos swapped over without any issues. The smaller impeller chucked out 1,550 ponies at 7,200 rpm and the maximum torque reading was 1,483 lb-ft.
Interestingly enough, the manifold boost pressure at peak horsepower with the 76mm turbo was 40.7 psi, just barely less than 2 psi less than the 88mm turbo. To dispel bad information — 2 psi isn’t worth almost 400 horsepower! It has to do with efficiency of the compressors and Luton and Seth-Hunter pointed to the data on the FuelTech logs.
The 88mm turbo saw a manifold air inlet temperature of 241 degrees while the 76mm turbo had a rather hot 335 degrees. The air temperature is important because it shows efficiency — a less efficient setup will create turbulence in the air as it is compressed, which increases the air temperature. With the 76mm using a closed wastegate, it most likely moved the impeller speed into an inefficient section of the compressor map. Bench racing the data, lowering the boost would probably bring the impeller back into an efficient operating territory and lower the inlet air temperatures to a more acceptable temperature. But again, 1,550 horsepower most likely far exceeds what most people thought a 76mm could produce.
We also asked Luton about the peak rpm of both engines, there are NMRA Coyote Stock cars that will rev to 8,000 rpm or more in outlaw trim, but this engine only peaked at 7,500 and 7,200 rpm. He said it had to do with the airflow from the turbo, a similar setup in their Pro Modified racecars but with twin Garrett 88mm turbochargers will pull clean to 10,000 rpm. The single 88mm was running it’s hardest by 7,500, so that is why the engine peaked at that rpm.
Regardless of the bench racing comments, the 76mm did churn out an impressive 1,550 horsepower — combined that engine setup with a good working Mustang on a set of 275 radial tires and it would be a hard car to beat. And if there were a neighborhood bully that needed to be humiliated, simply swap over to the GTX5020R Gen II 88mm turbo and we doubt the competition would be running their mouths anymore.
A set of mufflers might have netted “40 horsepower” 25 years ago, but today just one turbo really will get you close to 2,000 horsepower.