Articles: Cases/PSU

Bookmark and Share

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 ]

The junior model’s cross-load diagrams look good except that the +12V sags 3-4% in the bottom right of the diagram (when the load on the +12V rail is high). The voltages approach the 5% deflection only when there is a great misbalance of overall load towards the +5V rail (the top left of the diagram) which is an unrealistic situation for a modern PC.

The diagrams of the 420W model look much alike to the junior model, but the PSU doesn’t make it to the promised 145W on the +5V and +3.3V rails combined because the +5V voltage sinks below the permissible limit. This is not a serious problem, though. There just cannot be a high load on that rail in a modern PC. So if you want to find a fault with this PSU, you should instead gripe about the +12V voltage sagging 3-4% under load.

The oscillograms of the output voltages look somewhat odd. The +3.3V rail, very calm regularly, is full of short high spikes here. This is usually the consequence of using low-quality capacitors with a high equivalent series resistance that do not cope with high-frequency noise well enough. So, the PSU just barely meets the requirements of the standard in terms of output voltage ripple due to those spikes.

The same thing can be observed about the 420W model. Added to it, there appear spikes on the +5V rail, too. The output voltage ripple on the +12V rail is far within the limit (120 millivolts).

Both PSUs come with 120x120x25mm A1225L12S fans from Hong Sheng Technology with a rated speed of 1800rpm.

The next test in my program is to find the correlation between the fan speed and the PSU temperature. The load is steadily increasing from 50W to the allowable maximum, the PSU working for 20-30 minutes under each intermediary load. Alas, the A-360 passed this test, but the A-420 failed after about 3 minutes of operation under full load:

Quite a large portion of the PSU was burned out, including the PWM-controller and most of its accompanying components up to resistors. The chip itself and a couple of neighboring transistors exploded into pieces.

The 360W model not only worked well under full load but also was rather quiet at that. Its fan speed was lower than 900rpm at minimum load and barely reached 1300rpm at max load.

The senior model is somewhat noisier. The fans being identical, I guess there are different ratings of components in the speed controllers. As I wrote above, the senior model failed at full load and the diagram ends at 350W.

The efficiency is rather low by today’s standards, barely reaching 79%. The power factor is 0.65 on average, which is typical of PSUs without power factor correction.

So, the Ascot A-360 and A-420 are products of moderate quality. The former will make a good PSU for midrange PCs, providing stable voltages at a low level of noise, but you shouldn’t pay extra for the latter. The A-420 doesn’t have better parameters than the A-360 and cannot actually deliver the declared maximum output power. Thus, it has only one advantage – two SATA connectors instead of one in the A-360. By the way, this is the most serious problem of these two PSUs from a user’s point of view. The manufacturers should think about equipping their PSUs with at least three or four SATA power plugs placed on two separate cables. This won’t even make the PSU more expensive to make because the PSU may come with fewer Molex connectors that are quickly becoming unnecessary these days.

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 ]


Comments currently: 12
Discussion started: 12/15/15 07:45:13 PM
Latest comment: 12/21/15 11:37:15 AM

View comments

Add your Comment