Simbadda SB-380W Power Supply Review

Load Testing

Test 1 (115.59W Load – Cold)

Rail Load Voltage Ripple
12V 4.81A 12.02V 86.8mV
5V 5.02A 5.02V 101.6mV
3.3V 9.94A 3.28V 112.6mV
−12V 0A −12.64V 19.2mV
5Vsb 0A 5.03V 11.6mV
AC Power 147.39W
Efficiency 78.73%
Power Factor 0.62

.

Test 2 (205.1W Load – Cold)

Rail Load Voltage Ripple
12V 9.66A 12.07V 104.4mV
5V 9.96A 4.98V 104.6mV
3.3V 9.94A 3.28V 105.6mV
−12V 0.11A −12.93V 69.4mV
5Vsb 1A 4.98V 50.0mV
AC Power 256.37W
Efficiency 80.00%
Power Factor 0.58

.

Test 3 (262.81W Load – Cold)

Rail Load Voltage Ripple
12V 14.45A 12.04V 122.4mV
5V 9.98A 4.99V 99.2mV
3.3V 9.94V 3.28V 107.2mV
−12V 0.11A −13.41V 109.0mV
5Vsb 1A 4.98V 53.2mV
AC Power 329.6W
Efficiency 79.74%
Power Factor 0.58

 

The Simbadda’s voltage regulation was great. The 12V rail was at 12.07V at worst, which is a deviation of about 0.58%. It went as low as 12.02V during test 1, which gives us a variation of 0.05V, or about 0.41%. The 5V rail achieved 0.4% regulation, with a variation of 0.8%. The 3.3V rail stayed at 3.28V throughout the testing, which equates to 0.6% regulation. The −12V rail wasn’t so good, with 11.75% regulation, which is outside of ATX specifications. It’s seldom used these days, however, so it’s unlikely to cause a problem in the real world.

The efficiency, on the other hand, was bad. It peaked at only 80% during Test 2. A similar power supply used in the US with only 110V would have even worse efficiency. This power supply lacks Power Factor Correction (PFC), so the power factor was never any better than 0.62. I attempted to increase the load to around 320W for a fourth test, but the power supply failed almost immediately after.

Rail Test 2 (205.1W) Test 3 (262.81W)
12V    Test3-12v
5V  Test2-5v  Test3-5v
3.3V  Test2-3.3v  Test3-3.3v
−12V  Test2--12v  Test3--12v
5Vsb  Test2-5vsb  Test3-5vsb

 

What really let this product down was the ripple suppression. The 5V and 3.3V rails had just over double the maximum allowed ripple throughout most of the testing. The 12V rail fared slightly better, but its ripple was also too high during Test 3. The 5vsb rail had very low ripple during Test 1 (with no load – an unrealistic scenario in a PC). During Test 2, it was right at 50mV (the maximum limit), and was just above that during Test 3. The −12V rail did stay in spec, but only just during Test 3.

Poor ripple suppression will increase the stress on the motherboard’s capacitors, and shorten their lifespan. Like in the El-Cheapo Power Supply Roundup, all 10 points will be coming off the final score, because I couldn’t pull anything with the ripple even close to being in spec.

Pages: 1 2 3 4