Disassembly
_
f
The input filtering starts with a PCB attached to the AC receptacle, which contains an X-Capacitor, a coil, four Y-Capacitors and an MOV. The main PCB adds two X-Capacitors, a coil, and two Y-Capacitors, bringing the total component count to three X-Capacitors, two coils, and six Y-Capacitors. The bridge rectifier is rated at 6A, which is good enough for a 450W product. The primary capacitors are 680µF parts from a brand called FEC, which I have never heard of before. The two switching transistors are 13009s rated at 12A, in a rather dated and inefficient half-bridge configuration. I don’t recognize the manufacturer of these, but their specs are in all likelihood identical to their Fairchild counterparts. Interestingly, 12A switching transistors aren’t often capable of much more than 350W. My guess is that the chunky heat sinks and large fan helped. The 5Vsb rail uses a switching IC, as opposed to the two-transistor circuit used by many other low cost power supplies. These circuits are generally more reliable, and aren’t prone to over-volting. The switching IC used is a Fairchild FSDM311A. The main PWM controller used is an SDC Semiconductors SDC7500.
_
The capacitors used on the secondary side are all made by ChengX, which are of questionable quality, so the long term reliability of this power supply could be a problem. There is no PI filter coil on the 12V output, which is why the ripple was somewhat high.
The 12V rail uses two MBR20150CTP Rectifiers, which are rated at 20A each, so the 12V rail is theoretically capable of up to 40A. The other two rails use one MOSPEC S30D45CS rectifier, which are capable of up to 30A. The monitoring IC is a Richtek R7510A. Since this IC lacks Over Current Protection (OCP) support, this is a single rail power supply.
_
The soldering quality is fairly poor. I was able to find a number of joints with long component legs which were almost shorting on components. Other joints had insufficient solder, and were barely making contact. There was also a trace that was peeling off the PCB.
_
There was no information about the fan’s real manufacturer anywhere on it. Unfortunately, after peeling off the sticker, the bearings were almost completely dry, so I doubt if the fan would have lasted long. It is temperature controlled, but was audible throughout the testing. It was never disturbing, however. The heat sinks are excellent. They are thick, and have plenty of surface to air contact, which is quite possibly part of the reason that the switching transistors managed 450W, when similar transistors have exploded at slightly less than this on other power supplies.
Loading ...