Testing the Power Supply
Power supplies which employ Power Factor
Correction (PFC, or A-PFC) help ensure efficient conversion of AC to DC voltage,
and reduce the amount of energy wasted as heat. That means a quieter 120mm fan
can be used, and less noise for your ears.
PCSTATS tests
power efficiency with the aid of an Extech Model 380803 Power Analyzer. The
meter is located between the mains 120AC supply and the power supply, and a 120W
dummy load is connected to the power supply being tested. Power jitter is
measured with a Mastech MS8230B Multimeter at the device power connectors. The
test apparatus and power meters are shown below.
The power supply looking device is actually the
120W dummy load. To the right is a standard Voltmeter, and in the background the
Extech Model 380803 Power Analyzer (with readings for Watts, PFC/KHz, source
voltage (AC) and Amperage).
Power Jitter Observations -
Motherboard |
|
Voltage jitter is read by accessing the power
options displayed in the motherboard BIOS, with the system obviously in an
unloaded state. *All voltages were read from the BIOS of an Asus BLITZ EXTREME
motherboard.
Power Jitter Observations
- Motherboard |
Listed Voltages |
Lower Voltage |
Upper Voltage |
Jitter |
+3.3V |
3.25V |
3.28V |
0.03V |
+5.0V |
4.98V |
4.98V |
0V |
+12V |
11.95V |
12.00V |
0.05V |
Voltage readings from the
motherboard are pretty good, there's a bit of jitter but nothing
huge.
Power Jitter Observations - Multimeter
|
|
Voltage jitter was read
by a Mastech MS8230B Multimeter with the system in an unloaded state. During
testing the power supply is was still connected to a full computer and voltage
values.
Power Jitter Observations
- Multi-Meter |
Listed Voltages |
Lower Voltage |
Upper Voltage |
Jitter |
+3.3V |
3.3V |
3.3V |
0V |
+5.0V |
5.0V |
5.0V |
0V |
+12V |
12V |
12V |
0V |
The readings from the multimeter are
rock solid when hooked up to a 120W load. This isn't stressing the power supply
much, but a higher resistance load
was unavailable at the time of this review.
When looking at the Seasonic Power Supply test results we see two values, wattage
and volt-amps. Since it might not be clear what they measure, here's a brief
overview.
The volt-amp (VA) value
is how much real power is being consumed by the power supply being tested to
provide the wattage (W) value. The higher the VA value is, the more electricity
is being used by the power supply. Because no electrical device is 100%
efficient, there will always be some loss when converting AC to DC. The closer
the volt-amps and wattage figures are to each other, the more efficient a power
supply is. This is called the Power Factor: wattage / volt-amps
=PF.
Since we're testing with
a 120W dummy load, the load on the power supply (wattage) should be as close to
this figure as possible. Anything above this load in apparent power describes
the overhead and wasted energy (given off as heat) for the particular power
supply being tested. For the unloaded tests, the wattage and volt-amp figures
should be as close as possible to one another. The lower the figures are, the
less power is being wasted.
Measurements were taken
with an Extech 380803 Power Analyzer.
Seasonic Power Load Tests |
Power Supply
Model |
Wattage |
Test
Type |
Loaded |
Unloaded |
Seasonic SuperTornado |
400 W (A-PFC) |
Active Power |
177 W |
6 W |
Apparent Power |
182 VA |
7 VA |
Vantec iON2 |
350 W |
Active Power |
192 W |
10 W |
Apparent Power |
277 VA |
21 VA |
Antec TruePower 330 |
330 W |
Active Power |
195 W |
22 W |
Apparent Power |
289 VA |
38 VA |
Ultra X-Connect Green UV 500W |
500 W |
Active Power |
193W |
22 W |
Apparent Power |
307 VA |
40 VA |
AOpen Silent Power AO400-12AHN |
400 W |
Active Power |
179W |
9 W |
Apparent Power |
275 VA |
18 VA |
Seasonic S12-430 |
430 W (A-PFC) |
Active Power |
179W |
6W |
Apparent Power |
180VA |
9VA |
PC Power and Cooling Turbo-Cool 510 Express/SLI |
510 W (A-PFC) |
Active Power |
200 W |
29 W |
Apparent Power |
202 VA |
32 VA |
HEC Ace Power 580UB |
580 W |
Active Power |
183 W |
12 W |
Apparent Power |
272 VA |
26 VA |
HEC Win 550UB |
550 W |
Active Power |
184W |
10W |
Apparent Power |
263VA |
22VA |
Akasa PowerPlus AK-P550FF |
550 W (A-PFC) |
Active Power |
194W |
38W |
Apparent Power |
197VA |
40VA |
AOpen Prima Power AO700-12ALN |
700 W (A-PFC) |
Active Power |
172W |
3W |
Apparent Power |
181VA |
9VA |
PC Power & Cooling Turbo-Cool 1KW |
1000 W (A-PFC) |
Active Power |
205W |
25W |
Apparent Power |
207VA |
32VA |
Seasonic S12 600 |
600 W (A-PFC) |
Active Power |
164W |
7W |
Apparent Power |
173VA |
9VA |
Mushkin Enhanced XP-650 |
650 W |
Active Power |
192W |
22W |
Apparent Power |
279VA |
43VA |
Seasonic S12 Energy Plus SS-650HT |
650 W (A-PFC) |
Active Power |
150W |
7W |
Apparent Power |
150VA |
10VA |
Seasonic M12 700W |
700 W (A-PFC) |
Active Power |
161W |
8W |
Apparent Power |
165VA |
11VA |
Corsair HX620W |
620 W (A-PFC) |
Active Power |
171W |
8W |
Apparent Power |
173VA |
12VA |
Zalman
ZM600-HP |
600 W (A-PFC) |
Active Power |
173W |
5W |
Apparent Power |
175VA |
9VA |
HEC Zephyr 650 |
650 W (A-PFC) |
Active Power |
185W |
12W |
Apparent Power |
190VA |
17VA |
GlacialPower GP-PS550BP |
550 W |
Active Power |
178W |
6W |
Apparent Power |
269VA |
17VA |
Cooler Master Real Power Pro 750W |
750 W (A-PFC) |
Active Power |
158W |
6W |
Apparent Power |
163VA |
9VA |
Seasonic S12 II 500W |
500W (A-PFC) |
Active Power |
145W |
7W |
Apparent Power |
148VA |
10VA |
Enermax Infiniti
720W |
720W (A-PFC) |
Active Power |
179W |
12W |
Apparent Power |
181VA |
18VA |
COOLMAX GREEN
POWER CUQ-1200B
|
1200W (A-PFC) |
Active Power |
172W |
9W |
Apparent Power |
193VA |
12VA |
PC Power & Cooling Silencer 750 Quad
|
750 (A-PFC) |
Active Power |
168W |
12W |
Apparent
Power |
175VA |
15VA |
Enermax Galaxy
EGX850EWL |
850 (A-PFC) |
Active Power |
184W |
11W |
Apparent Power |
193VA |
16VA | |
The Enermax Galaxy EGX850EWL
draws a fair amount of power when under no load. The Enermax Galaxy power
supply was not quite able to reach the 80%+ efficiency but that may be
due to the 120W we have available for testing. To adequately test, most power
supplies require at least 20% load, the 120W load we're running only represents
just over 14%.
|