Originally Posted By: MrCritical
As for torture tests, I believe Hyundai has run the 2.0 turbo from the Sonata 300 hours at 100% load on a dyno, removed the rev limiter, and then ran 30 hours at 7800 rpm/100% load.
Reportedly, GM did the same with the Northstar engine. A powertrain engineer described another torture test it went though; they would run the engine to full hot temperature, then cycle in super cooled coolant and let it heat that up. They'd do that over and over to simulate intense thermal cycling scenarios.
But the engine still had problems with head bolt threads in the field.
I'm sure simulated torture testing is getting better and better at uncovering eventual problems in the field, but there still seems to be nothing like real field trials. For example, I'm sure the spark plug issue in some recent Ford Modulars, and the valve guide seal problem in earlier Modulars, never reared their heads during testing (or at least I hope they didn't), but were problems in the field nonetheless.