LONG TERM, FULL POWER, THERMAL TORTURE TEST
This is a test we have never run on any PSU before: Long-term, full load test without an exhaust fan in our PSU hotbox. It was mentioned earlier that even running this box without the exhaust fan is not normal. Commercially manufactured PSU load testers have non-defeatable, multiple high speed fans to keep themselves cool. This is a major reason why we use our custom-cobbled load tester: It lets us turn the fans off entirely when measuring or recording PSU noise. So why run this test on the X-400 Fanless? Well, almost on a dare.
In direct discussions about the X-400 Fanless, Seasonic reps were brimming with confidence that their new baby could not only perform just like a fan-cooled 400-watter, but even deliver full power under extremely low airflow conditions, such as one might find in a fanless (or single fan) silent PC. The parts in the X-400 FL are apparently robust enough that if it was fan cooled, it could easily be rated for 700W. This led to our running the normal tests without an exhaust fan. When our Seasonic rep found out the 30 minute duration of the 400W test load, he wanted us to try running it at full power for 24 hours, as is often done for testing of mission-critical server PSUs — which are never fanless.
The challenge was on. So, the X-400 Fanless was set to full load without any exhaust fan in the test hotbox, and left to run as long as possible. It was decided early on that 24 hours continuous was not going to be practical. For safety reasons, the test would be monitored regularly (about once every half hour) by a physically present person. That's your author. The test was begun at about 8 AM with 22~23°C ambient room temperature, and left to run until 11 PM that night, a total of 15 hours. The ambient temperature in the room climbed to 26~27°C by 1 PM, stayed around there for several hours, and only declined down to 24°C at 11 PM.
The temperature in the hotbox directly beneath the PSU ranged 56~60°C. Spot checks with the aforementioned infrared thermometer showed temperatures inside the X-400 as high as 82~84°C; mostly it hovered close to but under 80°C. The efficiency dropped as the PSU heated up. The 444W AC input seen after half an hour rose gradually about 10~12W higher during the hottest part of the day.
The end result after 15 hours: The X-400 Fanless survived the test. Furthermore, it worked perfectly the next day — and ran another 12 hours at the same load and conditions. At the end of the second day, a 1500W hair dryer was brought into play while the PSU was at full load, blowing the hot air down into the open grill cover of the X-400 to see what would happen with thermal overload. In a minute, the PSU shut down with a little click, the Over Temperature Protection finally kicking in. The IR thermometer showed temperatures of over 90°C around the main heatsinks. The AC cord was unplugged, the power switched off for a few minutes, then it was turned back on — with no change to the load settings: The X-400 powered back up without a hitch and kept running. The hair dryer was used several times to force OTP, just to check on the protection function. It worked without a flaw about half a dozen times. The X-400 sample is now powering a new system being set up for use in the lab.
It's hard to imagine any other fanless PSU surviving our long term full power torture test... but it was a valid question that came up, a question that we decided to answer in
Fanless PSU Torture Test Roundup.