The problem was simple: at 711 horsepower, oil temperature was getting too high and oil pressure was falling off. That is how you start worrying about bearings. We baseline-tested the L5P with fresh conventional 15W-40, then swapped from the 2017-2019 oil cooler to the 2020 GM cooler and ran the same 600-horsepower, 3,300-rpm load. The newer cooler cut oil temperature in the pan by about 9 degrees and dropped cooler-out temperature by about 10 degrees, which pushed oil pressure up about 5.5 psi. Then we changed to AMSOIL 15W-40 synthetic. Cooling stayed about the same, but pressure increased another 3.5 psi at the same temperature. The bigger point is that heat and viscosity control are what keep pressure up under abuse. With the newer cooler, fresh oil, and synthetic in the engine, the L5P had the oil control needed to keep pushing toward 800 horsepower.
In this episode of Killing a Duramax, Gale Banks focuses on the engine's oiling system before pushing the combination any harder. The previous session reached 711 horsepower, but oil temperature climbed and oil pressure fell enough to raise concern about bearing survival. To address that, the team replaces the existing oil cooler with GM's newer 2020-spec unit and switches from conventional diesel oil to a synthetic 15W-40. The goal is straightforward: improve heat rejection, maintain oil pressure under sustained load, and build confidence before attempting an 800-horsepower test.
Banks explains that the engine had accumulated 92 hours on the existing oil package, which he considered overdue for replacement given the abuse this dyno program imposes. He compares those 92 hours to roughly 10,000 miles of extreme operation, essentially the equivalent of climbing a 40 percent grade at 120 to 130 mph-something a stock truck could never do in the real world, but which the dynamometer can simulate continuously. That kind of punishment had clearly sheared the oil down and contributed to the pressure loss seen in earlier testing.
For the baseline refresh, the team installed fresh conventional 15W-40 CK-4 oil. Banks notes that CK-4 is the current American Petroleum Institute standard for high-speed diesel engine oil. With fresh conventional oil in place and the older cooler still represented in the comparison set, they could establish a clean baseline before evaluating the newer GM cooler and then the synthetic oil.
The test method was designed to hold the engine at a severe but repeatable operating point: 600 horsepower at 3,300 rpm. Banks emphasizes how unnatural that speed is for this engine. In stock form, the Duramax makes its rated power around 2,800 rpm, and in a pickup the governor would cut fuel before it could ever sustain 3,300 rpm. On the dyno, however, they can hold that speed and load long enough to stabilize temperatures and expose weaknesses in the oiling system.
The team monitored several key channels during each pull. They watched oil temperature in the pan, oil temperature leaving the cooler and entering the engine's main oil gallery, and oil pressure throughout the run. They also tracked supporting engine data such as turbine inlet temperature, turbo speed, and air-fuel ratio to ensure the operating point remained consistent from test to test. During the loaded runs, turbine inlet temperature was around 1,250 to 1,260 degrees, turbo speed was about 97,000 rpm, and the air-fuel ratio settled in the low-20s to mid-20s depending on measurement point. The objective was not a peak-power sweep but a stabilized thermal punishment test.
With fresh conventional 15W-40 installed, the team first ran the engine using the earlier 2017-2019 L5P oil cooler configuration as the baseline reference. They brought the engine up from idle, built heat deliberately, and held the load long enough for oil temperature to stabilize. Banks repeatedly stresses that the point was to cook the oil package under sustained stress, not just make a quick pull.
This baseline established the thermal and pressure behavior of a fresh conventional oil package before any hardware changes. It also provided a direct comparison against the previously used 92-hour oil, which had already shown substantial viscosity loss from shear. That older oil had produced notably lower pressure, so the fresh-oil baseline helped separate the effect of oil age from the effect of cooler design.
After the baseline, Mike swapped in GM's 2020 L5P oil cooler, which Banks had previously highlighted as clearly superior to the 2017-2019 version. He notes that GM did not significantly change the engine's advertised output for 2020-the rating remained 445 horsepower with only about a 5 lb-ft torque increase-but the truck's towing capability increased substantially. In Banks' view, that points directly to improved thermal management rather than a power increase. The real issue is not how much power the engine can make briefly, but how long it can survive at the limit while dissipating heat through the oil and coolant systems.
According to Banks, the entire cooling package in the nose of the truck was redesigned for 2020, and the oil cooler is part of that broader improvement. In this test, the newer cooler was evaluated first with the same conventional 15W-40 oil so that the cooler alone could be isolated as a variable.
Once the data from the conventional-oil runs had been captured, the team drained the engine and filled it with AMSOIL Max-Duty Diesel Oil 15W-40, also carrying the CK-4 specification. They then repeated the same 600-horsepower, 3,300-rpm stabilized run with the 2020 cooler still installed. Banks was particularly interested in whether the synthetic oil would improve pressure retention compared with conventional oil under the same thermal conditions.
He also comments on the difference between merely meeting an oil specification and materially exceeding it. While some oils satisfy the CK-4 requirement at the minimum acceptable level, he says the AMSOIL product exceeds the standard by a wide margin in several respects, especially in cylinder-wall scuffing resistance. Although that characteristic was not the direct focus of this test, it reinforced his interest in using the synthetic oil for future higher-power work.
The temperature data showed a clear benefit from the newer GM cooler. With fresh conventional oil and the 2017-2019 cooler, oil temperature reached 270 degrees in the pan and 246 degrees at the cooler outlet, where the oil entered the main gallery in the block. After switching to the 2020 cooler while keeping the same conventional 15W-40 oil, pan temperature dropped to 261.9 degrees-effectively 262 degrees-and the temperature entering the block fell to about 236 degrees.
Banks summarizes that as roughly a 9-degree reduction in pan temperature and a 10-degree reduction at the cooler outlet, all at the same 600 horsepower and 3,300 rpm. He considers that a meaningful improvement and evidence that the 2020 cooler is doing its job. When the team changed from conventional oil to the AMSOIL synthetic, cooling performance remained virtually identical, indicating that the temperature reduction came from the cooler upgrade rather than from the oil type itself.
The oil-pressure results were even more important than the temperature drop. Banks first compares the fresh-oil tests to the previously used 92-hour oil. With that heavily punished old oil, pressure in this test series was only 60.7 psi, and he notes that in some earlier operation he had begun seeing readings in the high 50s. Simply replacing that worn conventional oil with fresh conventional oil raised pressure by almost 7 psi, confirming that the old oil had sheared down significantly.
With fresh conventional oil and the older cooler, pressure was 67.6 psi. Swapping only the oil cooler to the 2020 unit increased pressure to 73.1 psi, a gain of about 5.5 psi. Then, with the 2020 cooler still in place, changing to AMSOIL synthetic raised pressure again to 76.6 psi. Banks points out that this synthetic-oil run occurred at essentially the same oil temperature as the conventional-oil run, yet still gained another 3.5 psi. He characterizes that as nearly a 5 percent increase in oil pressure.
Taken together, the total gain from fresh conventional oil in the old cooler to AMSOIL in the new cooler was 9 psi. Compared with the badly sheared 92-hour oil, the improvement was even more dramatic: the engine moved from pressure in the high 50s or low 60s into the mid to high 70s under the same kind of severe load.
Banks concludes that the 2020 GM oil cooler is a worthwhile upgrade for the L5P when the engine is subjected to sustained high-load operation. The newer cooler reduced oil temperature measurably, and that lower temperature translated into better pressure retention. The synthetic AMSOIL 15W-40 did not materially change oil temperature, but it did improve oil pressure further at the same temperature, suggesting better viscosity retention under stress than the conventional oil package.
For this project, that combination of lower temperature and higher pressure is the real outcome. Banks says the results give him enough confidence to continue escalating the build without immediately risking the bearings. With the oiling system now performing better, the next step is to install larger injectors and push the Duramax toward 800 horsepower in the following episode.