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

Usually we test power supplies with five different load patterns, trying to pull around 20%, 40%, 60%, 80%, and 100% of its maximum capacity (actual percentage used listed under “% Max Load”), watching how the reviewed unit behaved under each load.

But since we had a bad experience with a different unit from Huntkey and also because this model is labeled by Huntkey as a 650 W product and not a 700 W one, we decided to add other load patterns to our methodology, including other loads between 80% and 100% of the power supply maximum labeled power.

We broke the results down into two tables. On the first table you see the results for loads between 20% and 80%, and on the second table you see the results for loads between 80% and 100%. Below we will explain more about this second table.

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.

+12V2 is the second +12V input from our load tester and during our tests it was connected to the power supply EPS12V – i.e., to the +12V4 rail. The +12V1 input was connected to the +12V2 and +12V3 rails.

Up to 80% load we saw a blue sky, but we were afraid that this power supply would burn or explode if we pulled 700 W from it. So we decided to pull 600 W and 650 W from it before trying to pull the full 700 W. So we have patterns five and six reflecting these loads. Then we have three patterns for the 100% load test. With our first 100% pattern, test number seven, we respected the power limits printed on the power supply label: 509 W for the +12 V outputs and 170 W for the +5 V and +3.3 V outputs combined. But with this pattern the power supply was delivering a little bit less than 700 W, so we decided to increase current on +3.3 V in order to get closer to 700 W (test number eight).

The problem with tests seven and eight is that in order to respect the power supply limits we were pulling a lot of power from +5 V and +3.3 V and not as much power as we wanted from +12 V. This scenario does not reflect a typical computer usage from nowadays, where load is concentrated on +12 V outputs due to the system CPU (ATX12V/EPS12V connectors) and video cards, which are connected to the +12 V line and not to +5 V and +3.3 V ones.

So with test number nine we tested this power supply with its full load the way we like: pulling a lot of power from +12 V outputs and less from +5 V and +3.3 V.

See the results below.

We were impressed to see that Rocketfish 700 W can really deliver its labeled power at 50º C. Not bad at all!

But, as we constantly say, power isn’t everything. You will only achieve efficiency above 80% if you pull up to 60% of this power supply maximum capacity (i.e., below 420 W). With 80% load (560 W) we saw a 78% efficiency, dropping below that on the other load patterns we used for loads between 560 W and 700 W (see table above).

Voltage stability was very good, with all outputs between 3% of their nominal voltage in almost all tests, which is excellent (ATX standard allows voltages to be up to 5% from their nominal values – 10% in the case of the -12 V output).  We only saw voltages outside this 3% range on tests 7, 8 and 9 on +5 V and -12 V, but they were still inside the maximum allowed.

Ripple and noise increased with load. For example, on test one noise at +12V1 input from our load tester was at 17.2 mV, jumping to 92.6 mV during test nine. Even with this increase, noise was inside specs during all tests (i.e., up to 120 mV peak-to-peak at +12 V and up to 50 mV peak-to-peak at +5 V and at +3.3 V).

On the screenshots below we show noise level for test number eight, with our power supply delivering practically 700 W.

click to enlarge
Figure 16: Noise level at +12V1 input from load tester at 695 W (83.6 mV).

click to enlarge
Figure 17: Noise level at +12V2 input from load tester at 695 W (83.2 mV).

click to enlarge
Figure 18: Noise level at +12V2 input from load tester at 695 W (30.6 mV).

click to enlarge
Figure 19: Noise level at +12V2 input from load tester at 695 W (17 mV).

Of course we wanted to see lower values here, especially for a power supply that costs USD 165. Good power supplies are capable of producing far less noise, half of the amount presented by this unit or even less.

Now let’s see if we could pull even more power from this unit and our tests of the power supply protections.