The 2014 El-Cheapo Power Supply Roundup

Aywun A1-3000

First Look

Aywun power supplies are no strangers to my load tester. At worst, they can be total junk, and at best, they have been just average. This particular model is available for around $30, so it might not be quite as bad as the A1-1000, but maybe not as good as the A1-550S. Let’s find out for sure.

The label makes this power supply out to be another 12V heavy unit with two 12V rails. As we will see later, however, that is not true. It is a single rail unit with a more powerful 5V rail.

Test Results

Test 1 (76.69W Load)

Rail	Load	Voltage	Ripple
12V	2.44A	12.33V	14.8mV
5V	5.15A	5.15V	11.0mV
3.3V	4.9A	3.33V	19.6mV
−12V	0.1A	−11.77V	11.6mV
5Vsb	0.51A	5.12V	12.4mV
AC Power	102.1W
Efficiency	75.11%
Power Factor	0.6
Intake Temp	24°C
Exhaust Temp	28°C

.
Test 2 (105.63W Load)

Rail	Load	Voltage	Ripple
12V	4.82A	12.24V	14.2mV
5V	5.16A	5.16V	10.8mV
3.3V	4.88A	3.32V	19.2mV
−12V	0.1A	−11.87V	12.0mV
5Vsb	0.51A	5.12V	13.4mV
AC Power	135.9W
Efficiency	77.72%
Power Factor	0.63
Intake Temp	25°C
Exhaust Temp	29°C

.

Test 3 (160.83W Load)

Rail	Load	Voltage	Ripple
12V	9.41A	12.1V	15.8mV
5V	5.2A	5.2V	12.0mV
3.3V	4.87A	3.31V	18.0mV
−12V	0.1A	−12.05V	14.2mV
5Vsb	0.51A	5.11V	14.6mV
AC Power	203.5W
Efficiency	79.03%
Power Factor	0.64
Intake Temp	25°C
Exhaust Temp	32°C

.

Test 4 (206.48W Load)

Rail	Load	Voltage	Ripple
12V	9.5A	12.22V	19.4mV
5V	10.14A	5.07V	13.0mV
3.3V	9.94A	3.28V	17.4mV
−12V	0.1A	−12.30V	90.0mV
5Vsb	1.01A	5.06V	19.6mV
AC Power	260.3W
Efficiency	79.32%
Power Factor	0.64
Intake Temp	26°C
Exhaust Temp	35°C

.
Test 5 (258.97W Load)

Rail	Load	Voltage	Ripple
12V	13.93A	12.07V	23.0mV
5V	10.2A	5.1V	13.6mV
3.3V	9.91A	3.27V	17.8mV
−12V	0.1A	−12.54V	19.2mV
5Vsb	1.01A	5.05V	20.4mV
AC Power	332.2W
Efficiency	77.96%
Power Factor	0.64
Intake Temp	27°C
Exhaust Temp	38°C

.
Test 6 (307.55W Load)

Rail	Load	Voltage	Ripple
12V	18.13A	11.93V	34.2mV
5V	9.88A	4.96V	12.6mV
3.3V	10.26A	3.32V	21.0mV
−12V	0.11A	−12.77V	22.0mV
5Vsb	1.01A	5.03V	23.6mV
AC Power	407.5W
Efficiency	75.47%
Power Factor	0.64
Intake Temp	27°C
Exhaust Temp	42°C

.
The 12V rail started off at 12.33V and had dropped to 11.93V by Test 6. This gives us 0.33V (2.75%) worst-case regulation and a 0.4V (3.33%) drop. The 5V rail had maximum and minimum values of 5.2V and 4.96V respectively, giving us 0.2V (4%) regulation and 0.24V (4.8%) or variation. The 3.3V rail had maximum and minimum values of 3.33V and 3.27V respectively, which equates to 0.03V (0.91%) regulation and 0.6V (1.82%) variation. On average, this gives us 2.55% regulation and a 3.32% variation. This result is passable, but not excellent. The 5V rail was what really let it down, as it got close to the maximum limit of ATX specifications.

The efficiency came very close to achieving 80%, but didn’t quite make it. The power factor results were slightly better than some other power supplies in this roundup, but still not great. By Test 6, the exhaust temperature was 15°C warmer than the intake, which is hot for 300W. In fact, it’s warmer than the other three units I have already tested. 300W was all I could pull from this unit. I attempted to increase the load to 350W for another test, but the power supply exploded after about a minute.

Rail	Test 5 (258.97W)	Test 6 (307.55W)
12V
5V
3.3V
−12V
5Vsb

.
The ripple was quite well suppressed on this power supply, with all of the rails staying well below half of the maximum allowed ripple at all times.

Disassembly

The input filtering consists of two X caps, two common-mode chokes and four Y caps, which is enough components. Instead of a bridge rectifier, we have 3A diodes. The primary capacitors are rated at 680µF and look to be from Nippon Chemi-Con (KMG series to be exact). However, we are sure that these parts are fake; the snap-in KMG series is long obsolete, and real 105°C Chemi-Cons have brown sleeves (with the exceptions of LZA, LXZ and LXY which are dark blue, older KZE which were green, and KZJ and some TMZ which were black with gold print). The switching transistors are Jilin Sino Micro 3DD13009K BJTs rated at 12A. Such parts are not generally capable of delivering more than about 350W.

The 12V rectifier is a YG906C2R, which is a Fast Recovery Rectifier rated at 20A. Considering that the 12V label claimed 23A of 12V capacity, a higher rated part should have been used. The other two rails use ESAD83-004R Rectifiers, which are 30A Schottky rectifiers and ample for what the label claims the 5V and 3.3V rails to be capable of. All of the secondary side capacitors are branded BH – a small obscure manufacturer.

Again, this power supply blew the switching transistors when it failed. Not only is there visible damage to the washers and transistors, but it appears that they got hot enough to start de-soldering themselves!

The fan is branded BaoDiKai, and it has also very little lubricant in the bearing. It is wired directly to 12V and was quite audible throughout the testing. The heatsinks are fairly thick, but they do not have a lot of surface to air contact. There was a small marking near the front of the PCB on the primary side saying that Keerda is the OEM. Keerda power supplies are often sold by Logisys in the USA, and do not have a reputation for quality products.

Specifications and Conclusions

Real Wattage	300W
OEM	Keerda Electronics
PFC	None
Price	$30 AUD
ATX Connector type	20+4 pin
Worst-case voltage regulation (12v, 5v, 3.3v)	2.8%, 4.0%, 2.6%
Worst-case ripple (12v, 5v, 3.3v)	34.2mV, 12.6mV, 21.0mV
Worst-case efficiency	75.47%
Input filtering	Adequate
CPU Connector	ATX12V (4 pin)
PCIe Connectors	1x 6 pin
Molex (Peripheral) Connectors	5
FDD Power connectors	1
SATA Power connectors	2

.

Pros: Good ripple suppression

Cons: Can’t deliver labelled rating (−2), Low quality capacitors (−2), Mediocre voltage regulation (−1), Low quality fan (−1), Inefficient (−1), Loud (−1), 5V-heavy (−1)

Score: 1/10

< Previous Page | Next Page >

Pages: 1 2 3 4 5 6 7 8 9 10 11