Now that we know what Corsair's claims of capability are, let's put the power supply to the test.

     The "12V1" load is performed by putting a load on the main ATX connector and a couple of the Peripheral connectors. The "12V2" load is applied to the 8-pin EPS+12V connector. The "12V3" load is applied to peripheral connectors and the PCI-e connector claimed to be on the 12V3 rail.

     As we can see, we were nearly always at or above 80% efficiency. During test one, the efficiency was 79%, but that was at 115V input. No doubt that with 230V input this efficiency would be at least 80%.

     Naturally, the side effect of a power supply that is more efficient is that less power is converted to heat inside the unit, so we see exhaust temperatures that are particularly low.

     The 12V rail is no doubt rock solid. We see a drop of only .18V across the full gamut of tests. Unfortunately, there was also a .16V drop in voltage on the 5V rail. Fortunately, it's very unlikely that you'll need upwards of 20A from a 5V rail.

     As far as fan noise goes, the fan was not even audible until Test 5.

     Other than the occasional .01V drop in voltage, the Corsair HX620W had little concern that it's operating temperatures increase by a whole 10°C! The consistently low delta between intake and exhaust temperatures means another brownie point for the HX620W. All this points to the HX620W's efficiency and the excellent job the fan does of moving the air from the system.

     Speaking of the fan, the fan was audible in the hot box by Test 2. The fan was noticeably audible (a little louder than what could be considered "quiet") during Test 3. Despite this, the fan spun back down as quickly as it spun up once the load was reduced. So in an actual PC, it's not very likely that this fan will spin up and stay spun up the way it did in the hot box, since I'm applying a static load and pumping essentially all of the heat generated by the load tester back into the case.

     Cooling is handled by a 120MM ADDA fan. But the impressive cooling characteristics of this power supply have more to do with the fan controller than it did the fan. I have to say that I never worked with a power supply that adjusted the fan RPM so precisely, to such subtle changes. When temperatures increased, the fan was quick to spin up, but when things cooled back down, the fan spun back down to a lower RPM. The fan never spun any faster than it needed too. I've gotten used to fan controllers that waited "too long" to spin the fan up, and then spun the fan up to full RPM leaving me with "all or nothing" fan noise.

     Good news for those of you with a lop-sided load in their computers... There was no crossload issue at all! With only a 3A load on the 3.3V rail and a 4A load on the 5V rail, the 12V rails all reported over 12V, even with a 16A load on each rail.

Now let's take a look at the O-scope output....

Corsair HX620W	+3.3V	+5V	+12V1	+12V2	+12V3
Test 1
Test 2
Test 3
Test 4
Test 5

     The O-scope is measuring the voltage coming off of the rail as reported by the load tester. Each square North of the zero latitude is +.05V and each square South is -.05V. Each square going across X represents .02ms. So if we see the line jump up one square, while it's going across two squares, that's a .05V jump in voltage in .04ms.

     For the most part the O-scope readings are looking pretty good. Test 5's 12V2 and 12V3 results might warrant a little concern because we're seeing a total swing of .1V (100mV) under this load. But it has to be taken into consideration that this is a 39A load on the 12V rail! That's a pretty heavy load!

Now let's spend some time poking around the inside....

     The back of the modular interface is very clean looking. All of the wires enter the PCB from the right side and the voltages are brought across the board to the respective connector.

     Below are three photos of the inside of the Corsair PSU. To keep things cool, thin plates are used. The plates are cut and "butterflied" outward to help dissipate heat. The large primary capacitor in the center is a 470uF 400V Hitachi AIC.

     The secondary capacitors are all Nippon-Chemicon and seem to be of an appropriate size for a power supply of this caliber.

I believe we've covered most of the bases, so it might be time to just shut up and grade this thing...

     Performance (weight of 40%) gets a 9.5. It's efficient, has power factor correction, puts out as much power as Corsair claims it should, showed a minimal drop on the 12V rails under load, moved air well, was quiet, and seemed to put out clean voltage. The only gripe would be how much the 5V dropped and how much ripple we saw on the 12V rail during test 5.

     Aesthetics (weight of 10%) is 9. The flat black finish alone would have cost this power supply more points if it weren't accented with the red trim. The sleeves on the main power cables looked great and the modular cables make this power supply and your whole computer look awesome

     Value (weight of 30%) score is a 8. At a $169 retail price, it's easy to say that there are power supplies capable of more power for less money, but the Corsair is modular, more efficient than most and has a five year warranty. There's no doubt value in that!

     For functionality (weight of 20%,) I'm giving this power supply a 9.5. Not only is the Corsair HX620W modular, but it's of a "standard" 5.5" deep ATX size. The only thing keeping the HX620W from a 10 is the fact that BOTH 4-pin and 8-pin cables are fixed to the PSU housing, putting the user in the position of having to hide one or the other.

     Overall, that calculates to a score of "9" With only a few minor "flaws" no doubt Corsair has entered the power supply arena with guns blazing.

Is this PSU jonnyGURU recommended?

....Oh yeah! ;-)

SUMMARY:

     No doubt Corsair has put forward a power supply that would suit anyone's needs. We've got 620W of power, coming from a box that's only 5.5" deep, so obstruction in smaller cases is not an issue. The cables are modular, it has active PFC, is very efficient and very quiet.

     In all seriousness, the only major flaw is an aesthetic flaw. Yes, the 5V dropped significantly, but in real life one would not have such a hard-core load on the 5V rail while also loading up the 12 rails. And yes, there was as much as a 100mV ripple on the 12V rails during load test 5, but considering that was a 39A static load, that's not something to be ashamed of. But why would you make a modular power supply and then give the user an extra cable to hide by making the 4-pin and 8-pin EPS+12V cables two separate, fixed cables?

The Good....

• Plenty of juice
• Rock solid rails
• Very efficient
• Active PFC
• Very quiet
• Modular
• Nice cables
• Compact size would fit in any ATX compliant case

The Bad....

• Separate fixed cables for both 4-pin and 8-pin EPS+12V connectors

The Mediocre....

• Some may not like the mix-match of cables (sleeved main cables, ribbon-type peripheral cables)

Addendum:

     Since the review of this power supply, some interesting observations have been made of Seasonic built power supplies claiming to have three rails.

     Below is a picture of the modular interface PCB. This would have you believe that three peripheral connectors and one PCI-e connector were on one rail (labeled 12V3) and the other two peripheral connectors and PCI-e connector were on another rail (12V2.) This would leave one to assume that the 4-pin and 8-pin are on 12V1 perhaps with the main ATX, or the main ATX is on 12V2 with the fewer number of peripheral connectors.

    Looking at the Seasonic main PCB inside the Corsair PSU reveals only two rails, labeled 12V1 and 12V2. There is no third rail. This is illustrated well at Hardware Secrets. Although I can not say that these rails are or are not somehow electronically separated in the PSU's circuitry somewhere, I did find that there was no OCP (over current protection or "limiter") on either of these rails as I was able to load any given connector up 30 to 40A with no drop in voltage, system shut down, etc.

    So it is my opinion that we essentially have a single 12V rail PSU here. Certainly there is nothing wrong with this given the problems high end video cards have had with getting enough power from a single 12V rail when the OCP is set to the typical 240VA limit. But we do lose the advantages of multiple rails such as protection from damage to one rail from a short on another and the simple "filtration" of noise introduced from one rail to another.