We made several tests with this power supply as described in the article Hardware Secrets Power Supply Test Methodology. All the tests described below were taken with a room temperature between 44º C and 48º C. During our tests the power supply temperature was between 49º C and 53º C.
First we tested this power supply with six different loads patterns, trying to pull around 20%, 40%, 60%, 80% and 100% (two patterns) of its labeled maximum capacity (under “% Max Load” we list the actual percentage that was used), watching how the reviewed unit behaved under each load. On the table below we list the load patterns we used and the results for each load.
For the 100% load testing we used two different load patterns and we need to explain why. The load pattern used on our test five was created in order to simulate a typical PC usage today, with most power being pulled from the 12 V outputs, because the CPU (thru ATX12V and EPS12V connectors) and the video card (thru auxiliary PCI Express connector) are connected to them. This configuration, however, surpassed the maximum specs printed on the product box – 286 W for the +12 V outputs (we were pulling 384 W during this pattern). During this test our power supply exploded (more precisely the switching transistors burned).
Even though we knew that when the switching transistors burn the problem is with the overall power (which is higher than the maximum supported by the power supply) and not with any of the individual power supply outputs being overloaded (because when this happens the secondary rectifiers are the components that burn) we decided to re-test this power supply respecting its maximum limits printed on the product box. To do that we bought a new power supply and reconfigured our load test to use test number six as the test for maximum load for this power supply. The result was what we expected: our second power supply exploded just like the first one since, like we explained, the problem wasn’t on the power configuration of the outputs but with the power supply maximum overall power. This time we taped the test and we will talk more about it below.
+12V2 is the second +12V input of our load tester and on this test it was connected to the power supply EPS12V connector. Since this connector was the only one connected to the power supply +12V2 rail, the +12V1 and +12V2 inputs from our load tester were really connected to the +12V1 and +12V2 virtual rails from the reviewed power supply.
If you add all the powers listed for each test you may find a value different from what posted under “Total” below. Since each output can have a slight variation (e.g. +5 V output working at 5.10 V) the actual total amount of power being delivered is slightly different from the calculated value. On “Total” row we are using the real amount of power being delivered, as measured by our load tester.
| Input | Test 1 | Test 2 | Test 3 | Test 4 | Test 5 | Test 6 |
| +12V1 | 3 A (36 W) | 6.5 A (78 W) | 10 A (120 W) | 13 A (156 W) | 17 A (204 W) | 12 A (144 W) |
| +12V2 | 3 A (36 W) | 6.5 A (78 W) | 9 A (108 W) | 12.5 A (150 W) | 15 A (180 W) | 11.5 A (138 W) |
| +5V | 1 A (5 W) | 2 A (10 W) | 4 A (20 W) | 5 A (25 W) | 6 A (30 W) | 18 A (90 W) |
| +3.3 V | 1 A (3.3 W) | 2 A (6.6 W) | 4 A (13.2 W) | 5 A (16.5 W) | 6 A (19.8 W) | 16.5 A (54.45 W) |
| +5VSB | 1 A (5 W) | 1 A (5 W) | 1 A (5 W) | 1.5 A (7.5 W) | 2 A (10 W) | 2 A (10 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) | 0.5 A (6 W) |
| Total | 90.3 W | 181.4 W | 269.7 W | 356.9 W | 449.9 W | 451 W |
| % Max Load | 20.1% | 40.3% | 59.9% | 79.3% | 99.9% | 100.2% |
| Result | Pass | Pass | Pass | Pass | Fail | Fail |
| Voltage Stability | Pass | Pass | Pass | Pass | Fail | Fail |
| Ripple and Noise | Pass | Pass | Pass | Pass | Fail | Fail |
| AC Power | 112 | 215 | 325 | 485 | Fail | Fail |
| Efficiency | 80.6% | 84.4% | 83.0% | 73.6% | Fail | Fail |
Like we said Huntkey 450 W (LW-6450SG) exploded when we tried to pull 450 W from it using a load pattern that respects the limits printed on the product box (test six) and also using a load pattern that pulls more power from 12 V outputs and less power from 5 V and 3.3 V outputs (test five). This means that this power supply isn’t capable of delivering its labeled power in continuous mode being, as matter of fact, a 360 W power supply.
You can follow our test number six (i.e. our test with the second power supply we bought) on the below video. On this video you first see how our load tester was configured, which is exactly what is described on the table above on the “test 6” column. Then we changed the machine to display the power being pulled by the power supply and, after that, we turn on the CA main power from the power supply. The power supply was initially turned off and delivering only standby voltage (+5VSB). Then we turned on our power supply and you can see it pulling 450 W from our load tester. On top of the load tester there is our thermometer, showing the power supply temperature (above) and the temperature inside our “hot box” (below). It is hard to read it because of the video compression, but when we turned on the power supply temperature inside the box was around 41.5º C, below what we wanted for our methodology (we always test power supplies with a room temperature between 45º C and 50º C). In less than 2 minutes the power supply explodes. Check it out. In order to not being accused of uploading a fake video, we didn’t make any kind of editing and that’s why we didn’t put our logo or any text explaining what is going on.
When we opened the power supply we tested all main components and found out that the two switching transistors and two of their bias resistors had burned, check out on the pictures below the evidences we found.
click to enlarge
Figure 15: Explosion marks on the Green Star 450 W housing.
click to enlarge
Figure 16: Two resistors exploded together with the switching transistors.
Keep in mind that we have already posted reviews of two other power supplies that use the same old design used by this unit (Seventeam ST-420BKV and Kingwin ABT-450MM), but contrary to this Huntkey model these units could deliver their rated power. Also both power supplies shut down when we tried to pull much more power that they can deliver, as their protections kicked in.
Another power supply from this same power range that we have recently reviewed and that couldn't deliver its labeled power was Thermaltake PurePower 430 W NP, which is in fact a 350 W power supply. But this model from Thermaltake didn't explode when we tried to pull 430 W, as the over power protection (OPP) circuit entered in action.
Another clear problem was efficiency. On the product box the manufacturer says that “efficiency exceeds 85% under full load”, which is a lie. On their website, however, they say this power supply has a minimum efficiency of 70% under full load and 50% under 30 W load, which makes more sense. Efficiency was between 80.6% and 84.4% during tests 1, 2 and 3, dropping to 73.6% during test 4, where we were pulling 80% of the maximum labeled power (357 W). We couldn’t measure efficiency under full load as the power supply exploded. So if you pull only up to 270 W this power supply will have an excellent efficiency.
On the tests the power supply worked correctly, voltages were really stable, inside a 3% limit from the nominal voltage – which is really great, as the limit is 5% –, except the -12 V output. This output was at -11.20 V and -11.41 on tests 1 and 2. Even though ATV12V standard sets -12 V tolerance at 10% (which means it can go from -10.80 V to -13.20 V) we would like to see this output closer to the nominal -12 V voltage.
Even though electrical noise was low during tests 1, 2 and 3 (below 12 mV on +5 V and +3.3 V rails and between 33.8 mV and 45.6 mV at +12V1 and between 42.4 mV and 59 mV at +12V2), it increased a lot during test 4, with noise on +12V2 rail reaching 90.4 mV (the limit is 120 mV).
Unfortunately we couldn’t capture screenshots from ripple and noise this time as the power supply exploded.
