The 2014 El-Cheapo Power Supply Roundup

Ultraview 750W

First Look

Ultraview is another brand of power supply which has recently appeared on eBay. Buy it now prices are only $25 from some sellers, but I was able to snag this one for just under $20 as the starting bid. Let’s see if we can get some better results out of it (or a better fireworks display) than our last eBay special unit.

Like on the Numan, the label incorrectly claims this to be a dual rail unit. It is really a single rail. The label itself does not appear to be very professionally made – it does not have the full gloss finish like on all other power supplies in this roundup, and the printing is fairly poor quality.

Test Results

Test 1 (75.13W Load)

Rail	Load	Voltage	Ripple
12V	2.4A	12.25V	42.8mV
5V	5.05A	5.05V	16.6mV
3.3V	4.9A	3.33V	12.6mV
−12V	0.1A	−11.66V	35.4mV
5Vsb	0.5A	5.04V	27.0mV
AC Power	99.1W
Efficiency	75.81%
Power Factor	0.64
Intake Temp	32°C
Exhaust Temp	34°C

.
Test 2 (103.18W Load)

Rail	Load	Voltage	Ripple
12V	4.76A	12.12V	49.2mV
5V	5.05A	5.05V	17.4mV
3.3V	4.9A	3.33V	13.0mV
−12V	0.1A	−11.76V	44.8mV
5Vsb	0.5A	5.03V	30.4mV
AC Power	132.3W
Efficiency	77.99%
Power Factor	0.63
Intake Temp	32°C
Exhaust Temp	35°C

.
Test 3 (154.95W Load)

Rail	Load	Voltage	Ripple
12V	9.22A	11.87V	73.0mV
5V	5.06A	5.06V	19.4mV
3.3V	4.88A	3.32V	16.0mV
−12V	0.1A	−11.93V	61.8mV
5Vsb	0.5A	5.01V	40.6mV
AC Power	196.3W
Efficiency	78.94%
Power Factor	0.64
Intake Temp	33°C
Exhaust Temp	37°C

.
Test 4 (201.11W Load)

Rail	Load	Voltage	Ripple
12V	9.39A	12.07V	92.6mV
5V	9.86A	4.93V	25.2mV
3.3V	10.0A	3.3V	22.6mV
−12V	0.1A	−12.31V	77.8mV
5Vsb	0.99A	4.95V	53.6mV
AC Power	251.6W
Efficiency	79.93%
Power Factor	0.63
Intake Temp	33°C
Exhaust Temp	39°C

.
Test 5 (249.97W Load)

Rail	Load	Voltage	Ripple
12V	13.68A	11.84V	114.2mV
5V	9.9A	4.95V	27.4mV
3.3V	9.97A	3.29V	25.6mV
−12V	0.1A	−12.55V	93.4mV
5Vsb	0.99A	4.94V	61.2mV
AC Power	321.9W
Efficiency	77.65%
Power Factor	0.61
Intake Temp	33°C
Exhaust Temp	41°C

.
Test 6 (291.79W Load)

Rail	Load	Voltage	Ripple
12V	17.57A	11.61V	135.2mV
5V	9.9A	4.95V	30.2mV
3.3V	9.94A	3.28V	28.8mV
−12V	0.11A	−12.82V	111.4mV
5Vsb	0.98A	4.91V	70.2mV
AC Power	387.3W
Efficiency	75.34%
Power Factor	0.63
Intake Temp	33°C
Exhaust Temp	43°C

.
The 12V rail started at 12.25V and dropped to 11.61V in Test 6, giving us worst-case regulation of 0.39V (3.25%) and a drop of 0.64V (5.33%). The 5V rail started at 5.05V and dropped to 4.95V, which gives us 0.05V (1%) regulation and a drop of 0.1V (2%). The 3.3V rail started at 3.33V and dropped to 3.28V, giving us 0.03V (0.91%) regulation, and a drop of 0.05V (1.51%). The 5V and 3.3V rails results are pretty good, but the 12V rail’s result is poor. Only the Honli ATX-680 could manage worse regulation than this.

As for the efficiency, I didn’t quite see an 80% reading, but we got so close in test 4 that I’ll give the PSU the benefit of the doubt, since my equipment doesn’t offer the absolute best of accuracy. The exhaust temperature was 2°C warmer to start with and 10°C warmer in Test 6, which is fairly normal for this power level. The power supply could not even deliver half its rating. I attempted to increase the load to about 350W for a seventh test, but the power supply would shut down, so there will be no fireworks from this power supply, unfortunately.

Rail	Test 5 (249.97W)	Test 6 (291.79W)
12V
5V
3.3V
−12V
5Vsb

.

The ripple suppression was unacceptable at high loads. At 200W load, the 5V_SB rail was just slightly above the maximum allowed limit. At 300W load, the 5V_SB was over by 20mV and the 12V rail was also out of spec. The 5V and 3.3V rails were just slightly above half the maximum.

Disassembly

The input filtering consists of two X capacitors, two common-mode chokes and two Y capacitors, which is enough components. Instead of a bridge rectifier, the unit uses 2A bridge diodes. The two primary capacitors are branded Zhifa. They are labelled as 470µF, but one tested 295µF and the other read 297µF – indicating that they are over-rated. The switching transistors are JD Semiconductors 13009A parts, which are rated for 12A. If the results of other units in this roundup are anything to go by, they would probably have blown at 350W load if the OPP hadn’t stepped in. The 5V_SB uses a 2-transistor circuit, with a JD BU3150 MOSFET rated for 3A as the main switcher. The critical capacitor is made by ChengX – not one of the high quality brands.

The 12V rectifier is a MOSPEC F20C20C – a Fast Recovery Rectifier rated for 20A. The other rails use S20C40C Schottky rectifiers, which are also rated for 20A. This is just barely enough for the 3.3V rail, but the other two rails are claimed to be capable of more than 20A, so these parts are insufficient. Some of the secondary capacitors are supplied by H.Q. and others are supplied by Zhifa – neither of which are high quality brands. There are also no PI filtering coils , which explains why the ripple was high.

There are also some noteworthy problems with the soldering. The first thing that caught my eye was that some solder had been spilled on the component side of the PCB – joining two resistors. Luckily, the two pins which it touches are also joined under the PCB, so it didn’t stop the unit from working, but the manufacturer really should be a bit more careful with their soldering. The other problem is that there is not enough solder holding one of the wires down. The wire in question is a 5V wire leading to a molex and SATA connector. If it came out, it would stop the connected drive(s) from working.

The fan is branded Wam, and it is not temperature controlled. It was audible throughout the testing, but not disturbingly loud. It doesn’t have a removable plug under the sticker, so I can’t check the lubricant, but I have seen several WAM fans get noisy and fail, so we can be fairly sure that the lubricant is insufficient. The heatsinks are on the thin side, but could be much worse.

Specifications and Conclusions

Real Wattage	150W
OEM	Unknown
PFC	None
Price	$25 AUD
ATX Connector type	20+4 pin
Worst-case voltage regulation (12v, 5v, 3.3v)	3.25%, 1.0%, 0.9%
Worst-case ripple (12v, 5v, 3.3v)	135.2mV, 30.2mV, 28.8mV
Worst-case efficiency	75.34%
Input filtering	Adequate
CPU Connector	ATX/EPS12V (4+4pin)
PCI-E Connectors	1x 6-pin
Molex (Peripheral) Connectors	2
FDD Power connectors	1
SATA Power connectors	4

.

Pros: Working OPP

Cons: Can’t even deliver half of labelled rating (−3), Low quality capacitors (−2), Ripple out of spec at high loads (−2), Mediocre voltage regulation (−1), Low quality fan (−1), Problems with soldering (−1)

Score: 0/10

< Previous Page | Next Page >

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