|
|
In English
SuomeksiTests
When testing the PSU, it's not so important if there's small differences between the test results and the specifications. But if the power supply fails badly on tests, it tells something about marketing people of the manufacturer.
Tests in this case were made with water cooled resistors. When calculating the proper resistances, the effects of the cables were taken into account.
Will they last 620 watts?
Results
| Voltage (V) | 3.3 V | 5 V | 12 Va | 12 Vb | 12 Vc | -12 V | 5 Vsb |
| Max. current (A) | 24 A | 30 A | 18 A | 18 A | 18 A | 0.8 A | 3A |
| Regulation | +-5% | +-5% | +-3% | +-3% | +-3% | +-10% | +-5% |
| Measured regulation @ max load | -100.00% | -11.80% | 2.41% | 2.41% | 2.41% | -2.17% | -32.00% |
| Measured regulation @ min load | 3.63% | 1.00% | 2.583% | 2.583% | 2.583% | 2.25% | 1.00% |
| Ripple (mV) | 50 | 50 | 120 | 120 | 120 | 120 | 50 |
| Measured ripple @ max load (mV) | - | 120 | 140 | 140 | 140 | 120 | 80 |
| Measured ripple @ min load (mV) | 25 | 40 | 40 | 40 | 40 | 130 | 40 |
| Test load (ohm) | 0.127 | 0.1 | 0.5 | 0.5 | 0.5 | 16 | 2.4 |
| Max load (W) | 170 | 600 | 9.6 | 15 | |||
| Actual test load per line (W) | - | 194.48 | 906.15 | 9.39 | 4.82 | ||
| Total test load | 1114.06 | ||||||
| Total specified maximum load (W) | 620 | ||||||
| Continous load from input (Wrp) | 222 | ||||||
| Max peak load from input (Wrp) | 302 | ||||||
Analysis
Now look at the input and output powers. Do we have a perpetual motion machine over here or just a poor power factor correction? Efficiency of this power supply is about 80% and it takes about 222 watts from the grid. Corsair claims that the HX620W has a power factor better than 0.99 at full load but since that big difference between output and input power, it is impossible.
3.3 volt rail couldn't give the specified maximum current even without additional load. With 0.127 ohm load the power supply just shuts down. But who needs 24 amperes from 3.3 volt line anyway?
So the power supply was overloaded by almost 80 per cent and only 5 Vsb and 3.3 V rails failed, which means it performed very well. Overloading wasn't planned and resistors were calculated for nominal voltages. But it's impossible to calculate the right load resistor for SMPS power supply because the regulation accuracy is not very good and rising of the voltages will also raise the final load. Overloading on other rails didn't affect the voltage in the 5 Vsb line, therefore it's just specified for too high current rating.
We noticed that the HX620W whistles unpleasantly under high overload. This isn't anyways a problem for the end-users because one should never exceed the maximum load in the first place. Fan on the PSU is large and very silent. It kept its silent operation even under overloading.
 | Pages: 1 2 3 4 5 |  |
Content in english!
Sisältö suomeksi!
