Testing:

The testing of this power supply will take on a format previously not implemented at Bigbruin.com. After spending a good deal of time analyzing how things are handled at other sites, I decided it was time to move away from the marginally useful method of testing a PSU on a PC, and get some more advanced testing equipment.

While imitation may be the sincerest form of flattery, it can also be somewhat amusing at first. Like watching a little kid attempt to duplicate the moves of their favorite performer... You know what their trying to do, but you also might chuckle at the comical shortcomings. OK, hopefully it isn't that bad, but while I am taking a page from places like jonnyguru.com and hardocp.com, I won't claim to be at that level.



The equipment is all very new to me, and the methodology is still evolving as other equipment is brought on board, so I won't go too deep into the specifics. Basically, the gear to be used for this review consists of four dynamic DC loaders (two rated for 300W each, and two rated for 750W each), a 4-channel USB data logger, a digital multimeter, a 2-channel digital thermometer, a digital infrared thermometer, an AC power monitor, and a heat source. A brief description of each item's role is described below...

The DC loaders will be split up to handle the loads on the main power rails. The 300W units were intended to be used on the 3.3V and 5V rails, while the two 750W units were intended to work together on 12V rails. I say intended, because that was the plan until I realized the 300W loaders didn't care to go down to 3.3V. Luckily this unit has just one 12V rail rated for less than 750W, so one 750W loader will handle 12V, one 750W loader will handle 3.3V, and one 300W loader will handle 5V. -12V and +5VSB will not be considered.

The 4-channel USB data logger is like an oscilloscope, and the channels will be used to monitor and log voltage data on the 3.3V, 5V, and 12V rails. I had initially borrowed an older oscilloscope, but returned it a bit prematurely with too much confidence placed on the capabilities of the USB device. While it is very handy and useful, the main limitation I discovered was that it only goes to two decimal places, making ripple readings down to single digits in terms of milliVolts impossible. Instead, results will be reported in terms like "< 10mV" or "< 30mV", where I can only really say the value was less than 10 or 30mV, for example.

The digital multimeter is mostly for confirmation and setup.

The 2-channel digital thermometer is for monitoring the PSU inlet and outlet temperatures, while the digital infrared thermometer is for setup purposes (and for checking the temperatures of cables and other surfaces to make sure nothing melts or burns).

The AC power monitor will have the PSU plugged in to it in order to monitor AC power data (specifically Wattage to be compared to the DC Wattage to calculate an efficiency value).

The heat source is a compact ceramic space heater about the size of a power supply to be used for testing at elevated temperatures.

While the Ultra Products unit will be the first to use this setup in a published review, I have completed testing on two other units prior to this one, so I am not a complete beginner. One unit will be featured in a review to be released very soon, and the other was an older unit on hand used for initial configuration and for some extreme torture testing.

Having a single 12V rail made testing somewhat easier than a multiple 12V rail unit. Basically, as many 12V leads on the PSU were connected to the loader as possible, to keep the current in any one cable to a minimum. The same was done on the 5V rail. For the 3.3V rail, three wires from the 24-pin motherboard cable were connected.