SHARE

Load Tests

We conducted several tests with this power supply, as described in the article Hardware Secrets Power Supply Test Methodology.

First we tested this power supply with five different load patterns, trying to pull around 20%, 40%, 60%, 80%, and 100% of its labeled maximum capacity (actual percentage used listed under “% Max Load”), watching how the reviewed unit behaved under each load. In the table below we list the load patterns we used and the results for each load.

If you add all the power listed for each test, you may find a different value than what is posted under “Total” below. Since each output can vary slightly (e.g., the +5 V output working at +5.10 V), the actual total amount of power being delivered is slightly different than the calculated value. On the “Total” row we are using the real amount of power being delivered, as measured by our load tester.

The +12V1 and +12V2 inputs listed below are the two +12 V independent inputs from our load tester and during all test both were connected to the single +12 V rail present on the power supply.

Input Test 1 Test 2 Test 3 Test 4 Test 5
+12V1 6 A (72 W) 13 A (156 W) 20 A (240 W) 25 A (300 W) 29 A (348 W)
+12V2 6 A (72 W) 12 A (144 W) 17 A (204 W) 25 A (300 W) 29 A (348 W)
+5V 2 A (10 W) 4 A (20 W) 6 A (30 W) 8 A (40 W) 16 A (80 W)
+3.3 V 2 A (6.6 W) 4 A (13.2 W) 6 A (19.8 W) 8 A (26.4 W) 16 A (52.8 W)
+5VSB 1 A (5 W) 1.5 A (7.5 W) 2 A (10 W0 2.5 A (12.5 W) 3 A (15 W)
-12 V 0.5 A (6 W) 0.5 A (6 W) 0.5 A (6 W) 0.5 A (6 W) 0.5 A (6 W)
Total 173.9 W 350.9 W 515.2 W 689.7 W 852.3 W
% Max Load 20.5% 41.3% 60.6% 81.1% 100.3%
Room Temp. 45.9° C 46.3° C 47.2° C 48.9° C 49.7° C
PSU Temp. 47.0° C 48.9° C 50.7° C 55.1° C 57.0° C
Voltage Stability Pass Pass Pass Pass Pass
Ripple and Noise Failed on +5VSB Failed on +5VSB Failed on +5VSB Failed on +5VSB Pass
AC Power (1) 185 W 372 W 554 W 767 W 985 W
Efficiency (1) 94.0% 94.3% 93.0% 89.9% 86.5%
AC Power (2) 191 W 388 W 578 W 787 W 1,009 W
Efficiency (2) 91.0% 90.4% 89.1% 87.6% 84.5%
AC Power (3) 194.5 W 389.8 W 577 W 789 W 1,009 W
Efficiency (3) 89.4% 90.0% 89.3% 87.4% 84.8%
AC Voltage 111.8 V 110.3 V 108.5 V 105.7 V 103.5 V
Power Factor 0.988 0.992 0.995 0.997 0.998
Final Result Pass Pass Pass Pass Pass

We saw really high efficiency numbers with Corsair HX850W, and we were afraid that our load tester would be defective again. We tested all internal components from our load tester, and they were just fine. We also measured AC power using a second wattmeter (a Kill-a-Watt P4400, marked as “2” on the table above), which gave us different readings from our Brand Electronics 4-1850 (marked as “1” on the table above). With both instruments Corsair HX850W presented an impressive efficiency.

Updated 06/24/2009: We re-tested this power supply using our new GWInstek GPM-8212 power meter, which is a precision instrument and provides accuracy of 0.2% and thus presenting the correct readings for AC power and efficiency (results marked as "3" on the table above). Efficiency was very high all the times, peaking 90% when pulling 40% from the labeled power (340 W). With 20% load (170 W) and 60% load (510 W) efficiency was still very high, at 89%. With 80% load (680 W) efficiency was at 87% and at 850 W efficiency dropped to 84.8%, which is still a very good number. We also added the numbers for AC voltage during our tests, an important number as efficiency is directly proportional to AC voltage (the higher AC voltage is, the higher efficiency is). Also, manufacturers usually announce efficiency at 230 V, which usually inflates efficiency numbers. We added power factor (PF) numbers as well. These numbers measure the efficiency of the power supply active PFC circuit. This number should be as close to 1 as possible. This power supply has a practically perfect PF number under full load and very good numbers in all tests, except under light load (20%), where PF was at 0.988, still a good number but not so good as the ones achieved under other load patterns.

Voltage stability was another highlight from HX850W, with all voltages inside 3% of their nominal values(i.e., voltages were closer to their nominal value than needed, as ATX spec allows voltages to be up to 5% from their nominal values, 10% for -12 V). This includes the -12 V output (except during test one, where it was only 0.01 V short of the 3% mark, i.e., -11.63 V instead of -11.64 V).

Even though noise and ripple levels for the main voltages were very low as we will show below, the standby (+5VSB) output had an enormous ripple level during all load patterns but test five. During test one ripple was at 75.6 mV, during test two ripple was at 117.4 mV, during test three ripple was at 106.8 mV and during test four ripple was at 77.8 mV. During test five it decreased to 39.2 mV, which is acceptable. The maximum allowed for this output is 50 mV.

After this review was posted, Corsair tested this power supply using the same load patterns presented on the table above and, with a different equipment, the noise levels for +5VSB were completely different (very low). The only explanation we have is that our equipment was somehow interfering with the results. This way the comments above about the +5VSB output should not be taken at face value.

Corsair HX850W power supplyFigure 17: Ripple at +5VSB output during test two (117.4 mV).

Below you can see the results for test number five. As we always point out, the limits are 120 mV for +12 V and 50 mV for +5 V and +3.3 V and all numbers are peak-to-peak figures.

Corsair HX850W power supplyFigure 18: +12V1 input from load tester at 852.3 W (36.2 mV).

Corsair HX850W power supplyFigure 19: +12V2 input from load tester at 852.3 W (36.8 mV).

Corsair HX850W power supplyFigure 20: +5V rail with power supply delivering 852.3 W (10.2 mV).

Corsair HX850W power supplyFigure 21: +3.3 V rail with power supply delivering 852.3 W (12 mV).

Now let’s see if we could pull more than 850 W from this unit.

1
2
3
4
5
6
7
8
9
10
Gabriel Torres is a Brazilian best-selling ICT expert, with 24 books published. He started his online career in 1996, when he launched Clube do Hardware, which is one of the oldest and largest websites about technology in Brazil. He created Hardware Secrets in 1999 to expand his knowledge outside his home country.