SPCR

NeilBlanchard wrote:Let's hope this is an incentive for makers to take up the challenge.

A QUIBBLE ABOUT HEAT

My only quibble with the 80 Plus program is that the qualifying tests are conducted at normal room temperature, typically 25°C. (Thermal conditions during each test are not disclosed in the test reports for public consumption.) This is fine at 20% load, but by 50% load, and certainly at 100% load, the temperature of the actual working environment for any PSU rated higher than about 300W will invariably be hotter than the typical room. Not only does a PSU have to deal with its own internal heat generated during the AC/DC conversion, but also the heat of the components in the computer system it is powering.

As some of the approved models in Bronze are rated for very high output (750W to 1200W), at 50% or 100%, the heat they're generating is prodigious. Even at 82% efficiency, the heat produced with 1000W output comes to 219W, which will invariably cause an increase in the operating temperature of the PSU and its immediate environment inside a computer. Why does this matter? The efficiency and capacity of electronic components such as capacitors decreases as temperature rises. At high loads with real applications inside a real chassis, a PSU gets hot and its efficiency tends to decline. By conducting the qualifying tests in unrealistically cool conditions, many of the higher power models that fare well in the 100% load test get off easy. A more thermally rigorous test would see fewer high power units achieve such high efficiency numbers.

Adding thermal requirements and conditions to 80 Plus could complicate testing. It may also make it more difficult to achieve 80 Plus qualification. Then, there's the question of what to do about the nearly 500 models that 80 Plus has already qualified. Still, from the perspective of realistic testing, the absence of thermal requirements in 80 Plus today seems a deficiency that should be addressed.

oberbimbo wrote:I may be missing something but even when testing the PSU open bench at 25° it would have to dissipate its own heat, no?

oberbimbo wrote:So the point would be reduced to the fact that the PSU will, in general, have higher temperature at the intake than open bench testing provides it. (This is not insurmountable, AFAIK there are a few cases that provide outside air to the PSU)

FartingBob wrote:Considering the only product that is rated at silver is a prototype and even the best today only scrape through the bronze standard, i think it'll be a looong time before we see a sensibly priced commercial grade Gold standard PSU.
But thats a good thing, it'll encourage companies to continue improving efficiency, and i expect the first gold standard PSU will be a big seller around here.

alleycat8675309 wrote:There appears to be one Silver rated PS:
http://80plus.org/manu/psu_80plus/psu_join.php

Aris wrote:I dont see 90% happening anytime soon, unless 1 of 2 things happen:

1. Its running on 220v

mcv wrote:So as long as they aim at the European market, there's no problem. Apart perhaps from the fact that 80plus is a North American initiative...

90Plus to become new power supply standard in next 3-5 years, says Delta executive Kung-Tien Huang, Taipei; Steve Shen, DIGITIMES [Friday 18 April 2008]

Parameters  Load@ 80 Plus Bronze Silver Gold  90 Plus
             20%    80%    82%    85%    87%    90%
Efficiency   50%    80%    85%    88%    90%    90%
            100%    80%    82%    85%    87%    90%

The first point that we discussed with him the famous question 90Plus qu'arborait Glacialtech label. Mr Wickes has confirmed that it was false . . .

Even at 82% efficiency, the heat produced with 1000W output comes to 219W,