RM750 od Chicony (pak je radši vyřadili):
Overall, the Corsair RM750 failed our load testing today. Now, the data above looks odd as it indicates the unit failed during Test #2 and usually units that fail at some point like that are not able to continue on to tests with greater loads than where it failed. However, this unit failed differently than most other units we have seen. You see, rather than the unit simply tripping the OTP on the unit at some load level where every load greater would as well because there is more heat being produced, this unit has a problem turning its fan on appropriately under select conditions. In Corsair’s great exuberance to produce the quietest power supply on the market, it got WAY too aggressive with the passive mode on this unit and the fan fails to turn on in our load testing even at 50% load at 45C. When this occurs, the unit internally heats up to the point that the OTP trips and the unit shuts down as you would expect from a thermally controlled fan controller. However, when the unit is run at loads greater than 50% the fan turns on even if the temperature is not as high as it was at 50% load. So what does that tell us? Well, two things. First, the fan controller on this unit is not functioning correctly. Second, the fan in this unit is not strictly thermally controlled or load controlled, and it is improperly implemented. Rather, there is some sort of combination of thermal and load determination to when the fan turns on and there certainly are cases where the fan will not turn on even at loads/temperatures greater than what are indicated on the package as well as situations where it should have turned on to prevent an OTP shutdown.

Aktuální modely od CWT používaj tak z poloviny Crapxony GF, já bych na to bez výměny kondenzátorů nehráb, kór kolem těch 50 % když to bude topit jak kamna.

Zdá se, že to postihuje většinu zdrojů od G nahoru, nakonec, je to stejná topologie.