When it comes to testing a power supply, there are a lot of choices in how to tackle the job. I have found that while a multimeter can give scads of data, most of us don't have a clue what all these facts and figures mean. Even the elite of the enthusiast crowd doesn't always keep up with the inner workings of a power supply. So what do we do?
Since a majority of our readers tend to fall into the enthusiast realm, why not run the power supply in an AthlonFX based system and then stress it out? We'll begin by starting up a burning session on the CD-RW then go ahead and minimize that. Now if we crank up a long-winded demo of Quake 4, we will have strong power draws coming from the hard drive, memory, graphics and the optical device. Now granted, most of us aren't going to be doing these tasks together, but if our test model can manage to produce good numbers under these situations, it will handle our normal daily chores with ease. Oh, and to make sure that the 12v rail is being properly utilized I'm going to have the power supply hooked up to four fans to provide a constant draw from this source.
All right, we have three primary voltage rails on a power supply but only two are shown above. What gives?
If you'll recall one of the knobs on the back of the power supply was labeled "Memory", you'll know where this is going. To allow the user to adjust the memory voltage, the SinTek 500 SLI adjusts the voltage level to the entire 3.3v power rail. When playing with the rheostat, I measured voltage at 3.29v when set to the lowest level and 3.46v when setting it to its highest. Unfortunately there is no sort of measuring device that comes with the power supply so it will take a little trial and error to set it where you want in the event you don't want full power.
At any rate, what this adjustment does is allow you to get more voltage to those power hungry memory modules like OCZ VX or Mushkin Redline. This can be extremely handy if you have these modules but no motherboard that allows for high voltage levels to the 3.3v rail.
As for the 12v and 5v rails, they performed very well during the testing. The 12v rail fluctuated between 12.02 and 12.04v, so while moving a little, was very stable. The 5v and 3.3v rails didn't move at all. Overall the numbers show a very worthy contender to the PSU market.
While testing, I ran into an issue that may be of concern to some. It has to do with the fan rheostat that is located by the power cord. It worked just fine and you could hear the volume levels change appropriately when adjusting fan speeds, but the draw on the power supply caused some alarm. When I set the fan speed to maximum, voltage along the 12v rail dropped rather sharply to only 11.80v!
Now I realize that fans work off the 12v rail and I do expect levels to drop off a bit, but I didn't expect them to hit bottom like this. Given the 5% rule of thumb, we won't have issues until we hit around 11.4v, but I still feel this is a bit extreme. System stability did not suffer, but this should be a consideration if you are considering this PSU.
For those utilizing PCI-E, the voltage levels did not move around, just the primary 12v rail that flows to items such as hard drives, optical devices, fans, high-end AGP video boards...
Alternatively, when the fans were spinning at max speeds the 5v rail went up to 5.18v. The 3.3v rail fluctuated less than .003v so isn't being figured into the issue.