Load Testing The PSU
Testing a power supply is challenging; professional load testers must correctly check the PSU’s behavior.
The first basic test is performed with a simple power supply tester. It doesn’t load the PSU with more than a couple of Watts but helps determine whether the power supply unit is operational. A self-check indicates whether the voltages are at the proper levels. As you can see, there have been no issues with the Phanteks Revolt unit.
The following setup was used for the remainder of the testing:
- Voltcraft VC-870 Digital Multimeter - voltage measurement (+ Fluke 97 scope meter)
- Oscilloscope - GW INSTEK GDS 3154
- DC load - original solution
- Fluke 97 Scope meter (a measurement of voltage and ripple),
- Voltcraft SL-451 decibel meter (volume measurement)
- ACUVIM-IIRF is used to measure active power (input from the socket).
- PCE Instruments PCE-DT 50 tachometer – for the measurement of the rotational speed of the fan.
Measurements are taken only using 230 V input voltage.
Efficiency
Our power supply reviews emulate real-world usage by creating loads that a power-hungry, multi-GPU setup with a powerful CPU would generate. Older components used to require more power, but there has been some press about this. In a typical configuration, reaching over 500-600 Watts of power draw is difficult. Checking efficiency is relatively easy. Let’s take a look at Seasonic Prime's efficiency first.
It looks great even at low loads (e.g., 10%, equaling 160 W), which is quite typical of modern PCs' idle state.
Maximum load
Next is the maximum wattage this unit can hold before switching off.
Maximum Load | Efficiency (in %) |
1845 W | 91.25% |
Efficiency looks excellent, even above 100% load. 1845 Watts is a great value, but you wouldn’t want to come close to that for very long.
ErP Lot 6 Power Off wattage
We looked at the powered-off status (ErP/EuP), productivity mode (when we stress the CPU), and gaming. The lower the wattage, the more efficient the PSU. It is that simple.
ErP Lot 6 Power Off | value in Watts |
1600 W | 0.05 |
This is well below the 0.5 W requirement.
Ripple testing
First up is the ATX12V V2.2 specification for DC output ripple:
ATX12V Ver 2.2 Noise/Ripple Tolerance | |
Output | Ripple (mV p-p) |
+3.3 V | 50 |
+5 V | 50 |
+12 V | 120 |
The PSU achieved the following results:
| AC Ripple (mV p-p) | +3.3 V | +5 V | +12 V |
| 160 W (10%) | 7 | 5 | 10 |
| 400 W (25%) | 8 | 6 | 11 |
| 800 W (50%) | 10 | 5 | 10 |
| 120 W (75%) | 14 | 16 | 18 |
| 1600 W (100%) | 10 | 9 | 16 |
The values are very well within tolerance. Even in the worst-case scenario (100% load), it is not even a quarter of the allowed number, so this is not worrying.
