EVGA SuperNOVA 550 GS Power Supply

PROD LINK: EVGA’s Current Offerings
PRICE: $84.99 @ NewEgg
Price is at the time of testing!

Our first stop on the disassembly trolley is at the TNB fan. This is a sleeve bearing variant with some technology behind it to make it more reliable, so I won’t score against it.

Yep, that’s the same Seasonic based platform we saw in the 650W. Let me just get the guts out, and… well, darn. All of the mainboard screws are loose. All of them. That’s going to be a minor deduction, as this platform requires a good thermal connection between the underside of the board and the heatsink below.

Oh, good God… folks, sometimes Seasonic has an off day when it comes to soldering. Must be that day of the week for us… not only does this look dreadful, but the AC line wire insulation is melted on the other side. On both wires.

It’s time for me to get the heatshrink and insulate those wires. That said, I’ll only do the blue one. Why? Because the only short circuit danger is between the two AC line wires, that’s why. The mainboard layout won’t let the melted insulation portion of the brown wire hit anything critical.

Also, I do need to see wire colors to put them back in properly. You always want the hot leg going through the power switch, and I don’t want to screw that up. As long as I’m confident nothing will short out on the brown wire, it’s all good. And I’m confident.

AC line filtering starts at the receptacle with two Y caps and one X.

Apart from the ghastly job the factory did on the line input wires, everything else is flawless.

Most of the working parts are once again under the mainboard, which is why I’m making such a big deal over the loose mainboard screws. These are the PFC parts – two 65C7225s and a diode.

The main switchers. Two 5R280CEs.

The 12V output parts. Four 2R640s, two S10U45S, and an MP6922A.

More line filtering. Two coils, four more Y caps, and two more X caps.

Here’s where the minor rails come out. No, I am not going to try and pull those VRMs apart.

Here’s the PWM and PFC controller, an NCP1910B65.

Supervisor duty is handled with this here HY-510N. This part is already more capable than the one we saw in the Vero M1 last week, but less capable than it could be. That said, I don’t see much reason to believe this unit has any issues with the protection circuitry… this one does monitor the 12V rail, and the unit clearly handles much higher temperatures. So, I’ll let it be.

There’s our FSQ0165R standby controller, hiding behind the Rubycon main filter cap.

All capacitors on the mainboard are tier one parts… Rubycon and Chemi-Con.

Soldering on the modular board looks flawless.

There are a couple of Enesol caps on the modular board, but this is no problem to me. They’re still polymers, and the critical filtering has already been done on the mainboard anyway.