A Look Inside & Component Analysis

Before reading this page, we strongly suggest a look at this article, which will help you understand a PSU's internal components better. Our main tool for the disassembly of the PSU is a Thermaltronics TMT-9000S soldering and rework station. It is of extreme quality and equipped with a matching de-soldering gun. For the identification of tiny parts, we use an Andonstar HDMI Digital Microscope.


This platform looks similar to the CWT GPS, but we were told that it is a different and slightly modified version without modular cables. On the primary side, we find a half-bridge design and an LLC resonant converter, and on the secondary side is a synchronous rectification scheme for the generation of the +12V rail, while the minor rails are handled by a couple of VRMs. The platform is underpopulated and hence doesn't readily reveal the PSU's true capacity. The heatsink on the primary side is rather small, and the +12V heatsinks on the secondary side are tiny, especially if we take into account that the +12V FETs are only cooled by them and the main PCB, with the PSU's chassis not involved in their cooling at all.


The first part of the transient filter starts at the AC receptacle with two Y caps and an X cap. The power cables are also wrapped around a ferrite ring to suppress EMI. The second part of the transient filter is on the main PCB and consists of an X cap, two Y caps after the bridge rectifier, two CM chokes, and an MOV. There is also a CAP004DG on the solder side of the main PCB. This IC is used to block current through the X capacitor's discharge resistors when AC voltage is connected to increase efficiency slightly since no energy is wasted on any bleeding resistors.


The single bridge rectifier, a GBU1506, is on the primary heatsink.


There is an NTC thermistor to suppress large inrush currents. An electromagnetic relay is used to bypass this thermistor when the start-up phase finishes. This relay doesn't only boost efficiency slightly, but allows the thermistor to cool down faster, which increases its efficiency.


The APFC converter uses two Champion GP28S50G FETs and a CREE C3D08060A boost diode. The bulk cap is provided by Nichicon (400 V, 680 uF, 105 °C, GG series, 2000 h @ 105 °C), and its capacity is on the low side for a 750 W PSU.


The PFC controller is a Champion CM6502S; it is supported by a CM03X Green PFC controller.


The primary FETs are two Champion CMS6024s in a half-bridge topology. They are supported by an LLC resonant converter for increased efficiency. The resonant controller is a Champion CM6901 on the PCB's solder side.


The two small heatsinks on the secondary side don't hold any FETs since they are on the other side of the PCB. A total of six International Rectifier IRFH7004TRPBF FETs regulate the +12V rail.


Both electrolytic and polymer caps filter the +12V rail, with the former from Chemi-Con and the latter from FPCAP. All electrolytic caps are rated at 105 °C and belong to Chemi-Con's KY and KZE lines.


Both DC-DC converters are on a vertical board. In total, two UBIQ QM3006Ds and two QM3016D FETs are used, and the single PWM controller is an ANPEC APW7159C IC.


We find two supervisor ICs on the main PCB, a Sytronix ST9S429-PG14 (OCP [2x 12V channels, OVP, UVP, PG) and a Weltrend WD7518D (OCP [2x 12V channels], SCP). Two supervisor ICs had to be installed since this PSU features four +12V rails and CWT for its own reasons didn't want to use a supervisor IC with four +12V channels.


The standby PWM controller is a TinySwitch-LT TNY177PN capable of providing up to 18 W of power with a wide voltage input range if cooled properly.


Soldering quality is good overall. We also found a UTC LM393G on this side, a dual differential comparator that is probably utilized by some of the PSU's protection features.


There are four shunt diodes, an indication that the PSU does indeed have four +12V virtual rails.


The cooling fan is by Martech, and its model number is DF1202512SEMN (120 mm, 12 V, 0.37 A). The BF750F doesn't utilize a semi-passive mode, which isn't necessarily a bad thing since fluid dynamic bearings suffer from more friction during their start-up phase; that is, until the lubricant has been pumped all the way up to the bearing's top.