Part Analysis
| General Data | |
| Manufacturer (OEM) | Seasonic |
| PCB Type | Double-Sided |
| Primary Side | |
| Transient Filter | 4x Y caps, 2x X caps, 2x CM chokes, 1x MOV |
| Inrush Protection | 1x NTC Thermistor MF72 5D-20L (5 ohm) & Relay |
| Bridge Rectifier(s) |
2x GBU1508 (800V, 15A @ 40°C)
|
| APFC MOSFETs |
2x Alpha & Omega AOB160A60L (600V, 15A @ 100°C, Rds(on): 0.16Ohm)
|
| APFC Boost Diode |
1x Infineon IDK08G65C5 (650V, 8A @ 145°C)
|
| Bulk Cap(s) |
1x Nippon Chemi-con (400V, 680uF, 2,000h @ 105°C, KMR)
|
| Main Switchers |
4x Alpha & Omega AOB280A60C (600V, 9A @ 100°C, Rds(on): 0.28Ohm)
|
| Resonant Controller |
Champion CU6901VPA
|
| APFC Controller |
Champion CM6500UNX
|
| Topology |
Primary side: APFC, Full-Bridge & LLC converter
Secondary side: Synchronous Rectification & DC-DC converters |
| Secondary Side | |
| +12V MOSFETs | 4x Nexperia PSMN1R4-40YLD (40V, 214A @ 100°C, Rds(on): 1.4mOhm) |
| 5V & 3.3V | DC-DC Converters: 6x Nexperia PSMN4R0_30YLD (30V, 95A @ 100°C, Rds(on): 4mOhm) PWM Controller(s): Anpec APW7159C |
| Filtering Capacitors | Electrolytic: 3x Nichicon (2-5,000h @ 105°C, HD), 1x Nippon Chemi-Con (1-5,000h @ 105°C, KZE), 5x Rubycon (3-6,000h @ 105°C, YXG) 2x Rubycon (6-10,000h @ 105°C, ZLH) Polymer: 5x Nippon Chemi-Con, 18x FPCAP |
| Supervisor IC | Weltrend WT7527RA (OVP, UVP, OCP, SCP, PG) |
| Fan Controller | Nuvoton M031FB0AE |
| Fan Model | Hong Hua HA13525H12F-Z (135mm, 12V, 0.50A, Fluid Dynamic Bearing Fan) |
| 5VSB | |
| Rectifier |
DK DK5V45R10S (45v, 10mOhm)
|
| Standby PWM Controller |
Excelliance MOS EM8569C
|
Despite the small PCB, plenty of space is available, and the distance between parts is large to allow for enhanced airflow. The large vertical PCBs on the sides allow for an uncrowded PCB. One holds the APFC converter and the main FETs, while the other daughterboard holds both DC-DC converters that generate the minor rails. Seasonic used surface-mounting FETs with tiny heatsinks since the PCB also cools them. This is an interesting design to have a vertical board holding a significant part of the PSU’s primary side. Speaking of the primary side, a full-bridge topology is used for enhanced reliability and increased efficiency. On the secondary side, the 12V FETs are installed on the top PCB side, so they come in direct contact with the fan’s airflow.
Overall, the soldering quality is good, and all the parts that Seasonic used are of high quality, including Nichicon, Rubycon, Chemi-Con, and FPCAP capacitors, along with Alpha & Omega and Nexperia FETs. Moreover, the cooling fan is by Hong Hua, which has been among the most popular fan manufacturers in the PSU market in recent years, and for good reason.
The fan is digitally controlled by an MCU, which is optimal, but you still have to program it accordingly to have a good fan speed profile. Moreover, I would like to see MCUs controlling other parts of the PSU as well.
Bad fan profile settings. Initial fan speed is too high, 940 rpm is loud. It was better before.
I can confirm the fan in GX-850 V4 is way too loud (I have the PSU here). Indeed it starts spinning at too high RPM. The fan curve is absolutely terrible. Even at 360 Watts load during gaming the PSU fan in the GX-850 V4 is by far the loudest component in my PC. I compared the GX-850 V4 ATX PSU against my Seasonic Focus SPX 2021 SPX-750 750W SFX PSU with a 92 mm fan (vs 135 mm fan inside the GX-850 V4) and the small SFX PSU is significantly more quiet, it is almost inaudible below 400 Watts load. The small SPX-750 is absolutely compatible with Silent PC bulds, while the GX-850 V4 is absolutely unacceptable for Silent PC builds. Also the fan curve for the GX-850 V4 is absolutely trash. After closing a video game and going from around 400 Watts down to 80 Watts idle- it takes the fan like 2-3 minutes to stop spinning meanwhile the 92mm fan inside the small SPX-750 SFX psu stops spinning after like 10-15 seconds when closing the game and going from high load back to idle. I suspect it is the “new Seasonic OptiSink Design” that requires much more cooling. While it might have a bigger surface area for cooling in total, it lacks actual physical mass of bigger metal cooling heatsinks. Because of that it needs constant and active cooling compared to solutions with big aluminium heatsinks. All the marketing talk seems to be a bunch of bull, Quote:
“The eightfold increase in heat dissipation keeps the internals of the power supply cooler, eventually leading to improved product quality, long-term reliability, and more efficient operation.
The redesigned layout of the power supply components and its improved thermal performance creates versatility and allows the power supply to be safely installed in any orientation.
The OptiSink design is an eco-friendly solution that not only lowers the costs of material usage and production but also leads to energy savings in the long run.”
What it really means is- it is way cheaper for Seasonic to manufacture and because of that, they make more profit with the new OptiSink design. The downside is that it needs constant active air cooling due to lack of mass from big metal heatsinks, which are way more effective at cooling.
Needless to say I’ve returned the GX-850 V4, it’s absolutely terrible and unacceptable in terms of noise, I kid you not- the small SPX-750 is significantly more quiet. I made video comparisons as well, might upload them to YT later.
https://photo.coolenjoy.co.kr/data/editor/2407/13c9e3cc800d2e5e3c3cf5bafd2b5b87dcd6a2cb.jpg Aris i’ve seen this PSU and find this cybenetics QR on this PSU, did this MAXWELL Duke officialy tested by Cybenetics labs, how it’s perform ?
Nope, I don’t remember testing this unit! Basically I never heard of this brand before. Hmm, I will investigate this.
i’ve seen that the Maxwell Duke 1000W internals looks like 1stplayer NGDP 1000W https://photo.coolenjoy.co.kr/data/editor/2407/647933e3d3d50b746a68c7235da86ad64c5491bb.jpg this is your tearing down the ngdp 1000w https://hwbusters.com/wp-content/uploads/2023/09/top1-2.jpg
This new series is a disappointment