IntroductionThis review is one of a series of PSU reviews, if you have not read any of the others you can read all about the test set up and methodology below.
And now... ON TO THE POWER SUPPLY TESTING!
OCZ PowerStream 520ADJ
OCZ started out as a performance memory company. As I stated when I did my ModStream review, when they came out with power supplies, I was both skeptical and curious. After wrapping up my review of the ModStream, I was anxious to get started on the PowerStream. As you can see from the last photo of the ModStream review, I couldn't even wait until the one review was over until I started taking apart the next power supply..
The PowerStream has a reputation for being one of the most powerful power supplies, for the money, on the market. Upon inspection of the insides of the PowerStream 520ADJ and the ModStream 520W one can see that, despite both power supplies being 520W and both being built by Topower, you can see that the layout of the components are different, and just the SIZE of the components is different (everything in the PowerStream being significantly larger.)
Once again we are looking at a titanium finish. OCZ threw me a curve ball by including a green LED fan. At least it wasn't another blue fan, right? The fan grill is dookie gold, which I think flat black would've looked better, but you know what... We've gotten to concerned with looks. It's a damn power supply, right?
Let's take a look at the label on the side of the power supply. We'll be using some of these numbers to full load our power supply. The power stream claims to have 6 individual rails, which tells me that each rail is run off of it's own MOSFET. This doesn't make the rails any less interactive since they all still go through the same rectifiers and tranformers as any other power supply, but what it does mean is that your rails are kept isolated from each other so your power should be cleaner.
|OCZ PowerStream 520W||+3.3V||+5V||+12V||-12V||-5V||+5VSB|
|Max Output Current||28A||40A||
|Max Combined Peak Wattage||92.4A||150W||396W||6W||2.5W||10W|
Hmm... This makes me think there might be some better value in the OCZ 420ADJ! If you look at the specs of a 420ADJ, you'll see that you only lose 3A on the 12V rail. Trust me, you can live with that. The extra wattage for the 520W is made up for on the 5V rail! 30A versus 40A! Woohoo! Let's run out and buy a dual Pentium III! Not.... Wasted power. Another reason to spend close attention to load distribution and not just maximum wattage.
Now let's look at what connectors we get and the quantity of each...
|OCZ ModStream 520W||QUANTITY OF CONNECTORS|
|2 x 2 12V connectors||2**|
|2 x 3 PCIe||1|
|6-pin Xeon/AUX connector||0|
|5.25" Drive connectors||8***|
|3.5" Drive connectors||2|
|SATA Drive power connectors||2|
|Fan only connectors (thermostatically controlled 12V only)||0|
* The ATX connector is 20-pin and has a 4-pin connector that snaps on to the end.
** The two 2x2 connectors can be snapped together as to be used on an Xeon MP motherboard.
*** Two of the cables have EMI filters for use specifically with hard drives.
The ATX cables all have the fish-net covers on them while the PCIe and "HDD Only" have a nice vinyl cover. I know that Tagan power supplies are made by Topower, but I'm starting to wonder if Tagan is a division of Topower or maybe Tagan has a patent on these cable covers because once again I have a power supply that says "Tagan" on it's cables! The other drive cables and the SATA cables are regular four wire cables twisted up really tight.
One of the features of the OCZ PowerStream that really stands out is the ability to adjust your rails into spec. In my opinion, this is nothing more than a gimmick. It's good in the sense that everyone's idle load is different, so you can get the voltage dead on for your particular build, but a well made power supply shouldn't fluctuate to the point where you would NEED to adjust the rails if there's a 10 or even 20A load difference from one PC to the next.
I had fun with them. They're touchy as hell, though! Bareley move the screw one millimeter and you're off by .2V!
I'd hate to leave you hanging but... I have none! That's right. No conclusion. I'm going on strike today. I demand more free products from more manufacturers or I will write no more reviews! And chocolate. I demand chocolate. And not that cheap stuff... Toblerone. I want a stack of Toblerone bars.
Ok, seriously. I ran into some serious problems and I didn't want to grade on a curve or anything until I got some feedback from OCZ. Fortunately, they started a thread over at Anandtech and are fielding questions. I put up a post and hopefully they can give me an explaination to what I experienced.
In all three tests, the power supply did really well... load wise. The rails never fluctuated more than .5% and my load went from 298W to 488W. Power factor was between .68 and .72 (not too bad) and efficiency hovered around 73 to 74%.
The problem reared it's ugly head during test two when the 5V load was greater than the load on the 12V. At this moment, I noticed all of the voltages bouncing back and forth by as much as .2V or MORE! Although the voltages spent most (not all) of their time at one number, this is something that would typically tear up your computer hardware.
So I proceeded to test three. Everything worked great. Voltages were stable. I then cranked the 12V up to the maximum specified amperage as per the label. This put the 12V at 33A while the 5V was at 20A. No problems. I then began cranking up the 5V and, just like clockwork, the rails started bouncing about so frantically, there would be no way you could run a PC off of the power supply. I then proceeded to push the 5V up to 40A and the 12V dropped all of the way down to 10.24V! I could stabilize the power supply by turning the 12V up even higher, but temps were already at 47.3C which is hotter than any other power supply I've tested so far! So THAT wasn't a good solution.
Essentially, I want an explaination of the label. If they said the 12V could do 33A, that's396W. That only leaves 124W for the rest of the rails before the power supply is at 520W. After you take 28.4W away for .5A @ -12V, .5A @ -5V, 3A at 3.3V and 2A for the 5VSB, your left with only 19A on the 5V before you're at 520W!
And it's not that anyone would really load the 12V to 33A and the 5V to 19A for ANY period of time, my concern was the voltage fluctuation with a high 5V and where the heck the numbers on the label come from if you can't even get the 5V up to 30A and stable when there's only a 10A load on the 12V!
In a nutshell, all this would mean that the ModStream actually performed BETTER than the PowerStream! I just refuse to believe that! I know I've graded a few of the power supplies that blew up or had an over-rated label on a bit of a curve, but I would need to whip out a compass to draw the curves I would have to grade this unit on.
I realize that this power supply got rave reviews from other sites, but what did other sites do? What load did they put on each rail... Anyone? Anyone? Beuller? If this is indicitive of this power supply, then we're all blindly walking around praising it because it was able to run Doom 3 Demo on Billy Bob's Prescott rig.
So I'm waiting for someone with OCZ to get back with me. If they tell me this is normal, I'll grade accordingly. If they tell me that it's bad, I'll have to see if they'll swap the unit out. I can't just run out and spend another $150 on another power supply and I can't RMA this one where I got it from because I opened it up. So this review may never have a conclusion....
Reviews, Articles, News, All Reviews...
Reviews, Articles, News...
Career Center, News Users, Login...
A Guru's World, CPU/Memory Watch, SLDeals...
Forums, Register(Free), Todays Discussions...
Search, About Us, Advertise...