Skip to main content

FSP Hydro X 450 Power Supply Review

After the Hydro G series, FSP decided to release a more affordable power supply called the Hydro X, which targets systems with lower energy needs. This PSU line consists of three units, and in this review we're testing the entry-level 450W model.

A Look Inside And Component Analysis

Parts Description

Before proceeding with this page, we strongly encourage you to a look at our PSUs 101 article, which provides valuable information about PSUs and their operation, allowing you to better understand the components we're about to discuss. Our main tools for disassembling PSUs are a Thermaltronics soldering and rework station, and a Hakko 808 desoldering gun.

Primary Side
Transient Filter4x Y caps, 1x X caps, 2x CM chokes, 1x MOV
Inrush ProtectionNTC Thermistor & Relay
Bridge RectifierGBJ1506 (600V, 15A @ 100 °C)
APFC MOSFETs2x STMicroelectronics STF18N60M2 (650V, 8A @ 100 °C, 0.28 ohm)
APFC Boost Diode1x STMicroelectronics STTH8R06FP (600V, 8A @ 85 °C)
Hold-up CapNippon Chemi-Con (450V, 270uF, 2000h @ 105 °C, KMQ)
Main Switchers2x Toshiba TK10A60W (600V, 9.7A @ 150 °C, 0.327 ohm)
APFC ControllerInfineon ICE2PCS02 Supporting IC: Fairchild KA393AAPFC Disconnect: 1x SEN013DG
Switching ControllerChampion CM6901T2X
TopologyPrimary side: Half-Bridge & LLC Resonant Converter Secondary side: Synchronous Rectification & DC-DC converters
Secondary Side
+12V MOSFETs2x Toshiba TPH1R204PL (U-MOSIX-H Series, 40V, 150A, 1.24 mohm)
5V & 3.3VDC-DC Converters: 6x International Rectifier IRLR8726PbF FETs (30V, 61A @ 100 °C, 5.8 mohm) PWM Controller: ANPEC APW7159C
Filtering CapacitorsElectrolytics: Nippon Chemi-Con (105 C, KY, KZE), Rubycon (105 °C, YXF, YXG) Polymers: Teapo (Taiwan), CapXon (Taiwan)
Supervisor ICSITI PS223 (OVP, UVP, OCP, SCP, OTP )
Fan ModelPower Logic PLA12025B12M (12V, 0.20A, 65.15 CFM, 33.3 dB[A], 1800 RPM, Double Ball-Bearing)
5VSB Circuit
Rectifier1x International Rectifier IRFR1018E
Standby PWM ControllerPower Integrations SC1225K
Image 1 of 4

Image 2 of 4

Image 3 of 4

Image 4 of 4

The PCB is small and under-populated. On the primary side, two small heat sinks cool down the single bridge rectifier, the APFC converter's FETs and boost diode, along with the primary FETs. A synchronous design is used on the secondary side for rectifying the +12V rail; two Toshiba FETs, installed on the solder side of the mainboard, generate this rail. Only two tiny heat sinks are installed above the FETs, since they're mostly cooled by the PSU chassis. The minor rails are regulated by two voltage regulators housed on the same board. The electrolytic caps are sourced from Japanese manufacturers, while the polymer caps are provided by Teapo, a Taiwanese manufacturer with decent-quality products.

A cursory look reveals an assembly that doesn't look particularly nice. Besides the leaning +12V heat sink, we also don't like to see wires running freely over a PSU's mainboard. Because this model is so expensive, we expected FSP to build something that exuded a little more quality inside.

Image 1 of 3

Image 2 of 3

Image 3 of 3

Here's an example of what can happen when cables and wires run freely inside a PSU. The thermistor that provides temperature info to the fan control circuit is attached to a rather short wire, used to communicate with the fan control PCB. This thermistor is normally in close contact with the main transformer. However, in our sample, it was loose inside. This results in inaccurate temperature readings, leading to a constant fan speed throughout the PSU's operational range, regardless of how hot it actually got. Once we ran a first set of tests and discovered this strange issue, we broke apart the unit and easily figured it out. We attached the thermistor securely to the main transformer using lots of glue, and we started the tests over again. The fix was easy for us, but most builders would never go so far (and if they did, they'd void their warranty). Since we had easy access to the thermistor, before we fixed it, we though to test over-temperature protection with a heat gun. In a matter of seconds, the PSU shut down, proving that OTP is indeed alive and kicking.

Image 1 of 5

Image 2 of 5

Image 3 of 5

Image 4 of 5

Image 5 of 5

The small board behind the AC receptacle only holds two Y caps. The other EMI filtering components are on the main PCB; they include two Y caps, a single X one, two CM chokes and an MOV.

Image 1 of 3

Image 2 of 3

Image 3 of 3

As we expect from a contemporary PSU, an NTC thermistor is used for protecting against large inrush currents. This thermistor is bypassed by an electromagnetic relay, allowing it to cool down quickly.

Image 1 of 2

Image 2 of 2

The single bridge rectifier, a GBJ1506, is bolted on a small heat sink. It was wrapped in duct tape that we had to remove in order to identify it.

Image 1 of 6

Image 2 of 6

Image 3 of 6

Image 4 of 6

Image 5 of 6

Image 6 of 6

The APFC converter uses two STMicroelectronics STF18N60M2 FETs and a single STMicroelectronics STTH8R06FP boost diode. The single bulk cap is a Chemi-Con KMQ (450V, 2000h @ 105 °C) with only 270uF capacity, so we suspect that the hold-up time that we will measure won't meet the ATX spec's requirements. Under the APFC's coil, on the mainboard's solder side, is a SEN013DG IC that disconnects the APFC converter when the PSU is in standby mode to reduce energy consumption. This is a clever trick to increase the 5VSB's rail efficiency, while it also reduces vampire power.

Image 1 of 2

Image 2 of 2

A small vertical PCB holds the APFC controller, an Infineon ICE2PCS02, along with a Fairchild KA393A operation amplifier.

Image 1 of 2

Image 2 of 2

The primary switching FETs are two Toshiba TK10A60W arranged into a half-bridge topology. An LLC resonant converter is also used to boost efficiency.

Image 1 of 3

Image 2 of 3

Image 3 of 3

The standby PWM controller is a Power Integrations SC1225K. The FET that regulates this rail is an an IRFR1018E by International Rectifier and it is installed on the solder side of the PSU's PCB. Thanks to the SEN013DG IC, which disconnects the whole APFC converter in standby mode, the 5VSB rail has very high efficiency.

Image 1 of 4

Image 2 of 4

Image 3 of 4

Image 4 of 4

A vertical daughter-board on the secondary side houses the LLC resonant converter, a Champion CM6901T2X, along with an op-amp that's partially hidden by several caps.

Image 1 of 3

Image 2 of 3

Image 3 of 3

The FETs that do the main work on the secondary side are two Toshiba TPH1R204PL (U-MOSIX-H series); they regulate the +12V rail. These FETs are mostly cooled by the chassis since the heat sinks on top of them are too small to handle the task alone. The +12V FETs have eight pins instead of three, so they don't look like regular ones. Four pins are used by the drain, three by the source and the last one is the gate.

Image 1 of 4

Image 2 of 4

Image 3 of 4

Image 4 of 4

The filtering electrolytic caps in the secondary side are provided by Chemi-Con (105 °C, KZE and KY series) and Rubycon (105 °C, ZLH, YXF, YXG). Some polymers are also used for ripple filtering, and they come from Teapo. Polymer caps don't have a problem operating under hard conditions for thousands of hours, since they don't contain any liquid material (electrolyte).

Image 1 of 5

Image 2 of 5

Image 3 of 5

Image 4 of 5

Image 5 of 5

The DC-DC converters that generate the minor rails are installed on a board on the secondary side. Each converter uses three International Rectifier IRLR8726PbF FETs and the common PWM controller is an ANPEC APW7159C IC. These FETs are strong enough to provide 120W combined output on the minor rails. These converters use a mix of Teapo and CapXon polymer caps, for filtering purposes.

Image 1 of 9

Image 2 of 9

Image 3 of 9

Image 4 of 9

Image 5 of 9

Image 6 of 9

Image 7 of 9

Image 8 of 9

Image 9 of 9

Soldering quality is good, however some component leads could be shorter.

Image 1 of 3

Image 2 of 3

Image 3 of 3

A small PCB holds the protections IC, a SITI PS223. This component provides all necessary protections including OTP and OCP for two +12V channels (although this PSU only has one).

Image 1 of 2

Image 2 of 2

The fan is made by Power Logic and its model number is PLA12025B12M (12V, 0.20A, 65.15 CFM, 33.3 dB[A], 1800 RPM). It uses double ball-bearings, so it should last a while. Moreover, the fan profile is relaxed, so it won't make much noise as it spins.

  • powernod
    Do Hydro PSUs work underwater?


    Yeah, they have to work underwater in order to camouflage themselves from Corsair's (PSUs) who dominate the water's surface!! :p
    Or to pass through the Seasonic wall !!:lol:

    ONTOPIC: Decent PSU from FSP, but only just decent!!
    Reply
  • Aris_Mp
    it is hard to enter the US markets with Corsair and EVGA throwing one model after the other and in very competitive prices, however more variation is always welcome. If they lower significantly the price tags on these models then their marketing career will be easier.

    Reply
  • Dark Lord of Tech
    Hard to penetrate the EVGA lineups with their great prices on the higher end units. These need to be lowered to sell.
    Reply
  • turkey3_scratch
    Is there any reason the 3.3V rail shows strange behavior when crossloading? Not that it's bad or anything, but in CL1 with heavy load on the 3.3V rail, the voltage is about 0.10V higher than CL2, where the 3.3V rail has minimal load, and voltage is typically higher.
    Reply
  • turkey3_scratch
    Okay, just finished reading the review. Besides the disappointing transient response of this unit, and the failure to meet hold-up time, this is a good unit. But the price needs to drop to about $60 IMO if it wants to compete well. Currently it's priced the same as the Hydro G 650W, and the Hydro G is a no-brainer choice over this unit.

    It also concerns me a bit that the 5V rail voltage goes to 4.79V in your second transient response test at 50% load. This is all around also a bit disappointing, but it's not a very realistic transient load (unlike 12V which happens always while gaming), but I like the Japanese capacitors, the load regulation is fine, the crossload graphs all show good results; sometimes a lot of units screw up on those. It's nice to be able to see those different load patterns, something other reviewers should try to adopt. Ripple was very nice.
    Reply
  • joz
    Hard to penetrate the EVGA lineups with their great prices on the higher end units. These need to be lowered to sell.

    G2 is love, G2 is life. (G2 550W, about...eight of them....)
    Reply
  • basroil
    FSP almost had an excellent PSU until they screwed up big time on the transient response. Looks like the EVGA G2/ SuperFlower Leadex Gold is still king of the inexpensive PSUs
    Reply
  • Flying-Q
    Please stop referring to quality PSUs with low wattage as 'entry 'level' (in the article subtitle).

    Entry Level (adjective)
    (of a product) suitable for a beginner or first-time user; basic.
    "entry-level computers"

    'Entry level' usually implies smaller feature set. In this instance the feature set of each of the models is the same other than the power output. Current generation computers need less power due to greater efficiency inherent in more recent designs of components.
    Reply
  • turkey3_scratch
    17768395 said:
    FSP almost had an excellent PSU until they screwed up big time on the transient response. Looks like the EVGA G2/ SuperFlower Leadex Gold is still king of the inexpensive PSUs

    I would like other testing sites to start adopting these tests, like Jonnyguru. I wonder how many units that normally pass stuff would fail.

    Another funny thing is FSP just wrote a blog about the importance of transient response. :P

    But also, a 3.3V transient response just doesn't happen in 2016, probably never will. A 5V one is also less common.
    Reply
  • basroil
    17769266 said:
    But also, a 3.3V transient response just doesn't happen in 2016, probably never will. A 5V one is also less common.

    ATX loading specs state a 9A transient on 12V (and 5V might have been there). Considering most modern PSUs are 12V only and then DC-DC for 3.3 and 5V, 12V transients are going to end up affecting the 5v and 3.3V lines too.
    Reply