After these tests we tried to pull even more power from BFG 800 W. Below you can see the maximum amount of power we could extract from this unit keeping it working with its voltages and electrical noise level within the proper working range. During this test room temperature was of 51º C and the power supply was working at 57º C.
Input | Maximum |
+12V1 | 33 A (396 W) |
+12V2 | 30 A (360 W) |
+5V | 9 A (45 W) |
+3.3 V | 9 A (29.7 W) |
+5VSB | 3 A (15 W) |
-12 V | 0.8 A (9.6 W) |
Total | 855 W |
% Max Load | 106.9% |
AC Power | 1,137 W |
Efficiency | 75.2% |
Here noise level increased to 44.4 mV at +12 V, 31.4 mV at +5 V and 25.2 mV at +3.3 V, which are great numbers.
The problem, however, is that after less than five minutes working under this configuration the power supply completely died. We tried to turn it on with 20% load and nothing. We waited until its temperature dropped to below 30º C and, again, nothing. We killed our BFG 800 W!
After opening the unit we found out that we burned one of the four +12 V rectifiers. Which is strange, as we were pulling 63 A from the 12 V outputs and the theoretical combined limit of the rectifiers was of 160 A, as we saw when we analyzed the secondary of this power supply. Should we categorize this as a bad luck of getting a defective unit?
Anyway, this is bad. A power supply isn’t supposed to burn only because you overloaded it. In fact this is exactly why the over power protection (OPP) exists, to prevent things like this from happening.
Another hypothesis is that this rectifier burned because it was overheated (during this test the power supply housing was at 57º C, so imagine the secondary heatsink temperature) and the power supply over temperature protection (OTP) didn’t kick in. This makes sense as the temperature sensor of this power supply isn’t installed on the secondary heatsink, but inside the +12V coil.
Over current protection (OCP) seemed to be configured at 30 A – even though the power supply label says that the limit for each rail is of 20 A –, as we couldn’t pull more than 30 A from +12V2 input, where we had one of the rails (+12V2) connected alone.
During our tests we could see the speed of the power supply fans changing as the power supply temperature increased. Below 30º C they spin slowly, making almost no noise, and after this temperature they start increasing their speed, also increase noise level.
