We call our new efficiency measurements program ETA, deriving from the Greek letter “η”, which is widely used to represent efficiency.

Contrary to existing methodologies we plan to apply close to 1600 different load combinations in the DUT (Device Under Test), which with the interpolation of results, can provide up to 25,000 measurement points, while current methodologies only take three to four measurements. The overall efficiency will be the average of all measurements, which cover the PSU’s entire operational range. This way it is impossible for a manufacturer to tune its products in order to meet some specified load levels, since in essence, we take under consideration the efficiency levels under a greater number of different load combinations, evenly spread throughout the full load range. Besides efficiency we also take voltage, ripple, power factor, noise and temperature measurements. For the moment we will use the power factor and noise results, while in the future we can use some of the remaining factors (or all of them) should we want to enrich our methodology. Furthermore, the huge load of data that our methodology provides allows us to easily modify our efficiency certification program, in case this is needed. Finally, we start our tests at close to 30 °C with the PSU inside a hot-box, which simulates a case environment. During the end of the test the ambient temperature inside the box reaches up to 32-34 °C so it is close to real life conditions.  

Vampire power (power consumption with no load on the 5VSB rail) is of high importance, since all this amount of energy goes wasted and most PC system’s aren’t kept in operation 24/7, meaning that for a significant part of the day the PSUs just consume energy without doing anything useful. We will evaluate each PSU by following closely the EN 50564:2011 and IEC 62301 measurement guidelines. In case the DUT doesn’t meet our standards, it will be automatically downgraded to the next lower efficiency certification level. 

Besides all above ETA will also take into account the overall efficiency of the 5VSB rail. We will measure efficiency on this rail per 0.05 A steps up to its max current output and the average of all measurements will be the final efficiency result. We expect all PSUs to deliver over 70% overall efficiency output on this rail, with this threshold set even higher for units that fall into the top categories of the ETA program. 

In order to make impossible for anyone to fake one of our badges we plan to use a plain but highly effective method. Each of the badges that we provide will address only a specific PSU model so it cannot be used by another one, even if this is based on the same platform but it is sold under a different brand. Users will be able to check if the badge is legit by scanning the QR code that comes along with it, or by visiting the short URL that we will also provide. The landing page of both the QR code and the short URL will be our database entry with the test results of the corresponding PSU model. So, if the user has purchased the X PSU and its performance badge leads to the Y unit, then he/she will know that the unit carries a fake badge and we should be informed immediately about this, in order to take action. We created a special email ( for this purpose, in order to make easier the reporting of fake badges.
Table (1)
Efficiency  Levels (115V Input)Efficiency  (1) (2)PF  (3)5VSB Efficiency  (1) (3) (4)Vampire Power  (3)
A+≥94% & <97% overall efficiency≥0.985 overall PF>79%<0.10W
A≥91% & <94% overall efficiency≥0.98 overall PF>77%<0.15W
B≥88% & <91% overall efficiency≥0.97 overall PF>75%<0.20W
C≥85% & <88% overall efficiency≥0.96 overall PF>73%<0.23W
D≥82% & <85% overall efficiency≥0.95 overall PF>71%<0.25W

(1) In case the PSU is close enough to the limits (0.1%), we will ask its manufacturer if it is willing to send a second sample for re-evaluation. If the second sample registers a significant performance difference compared to the first one (>0.5%), we will ask for a third one, to verify the test results. Moreover, in case the difference with the upper limit is <= 0.05% (e.g. 91.95%), we will round the number to the second digit and the PSU will be taxed to the higher category. 

(2) Besides 115V testing, we also conduct a number of 230V load tests in order to make sure that the under evaluation PSUs meet all requirements mentioned in the (EU) No 617/2013 regulation:

(a) 85% efficiency at 50% of rated output power;
(b) 82% efficiency at 20% and 100% of rated output power;
(c) power factor >= 0.9 at 100% of rated output power.

(3) If the PSU fails in any of those requirements, it will lose one class and will drop to the lower one, despite of its efficiency performance. The same applies for the requirements listed in Table (2) except for the first test (2013 ErP Lot 6 requirement), where we allow for a 5% deviation (40-45% range) due to the very low applied load and the 0.55 A load test where we also allow for the same 5% deviation (70-75% range). In any case vampire power must be less than 0.25 W, even with 230 V input. 

(4) The 5VSB rail should also achieve the required efficiency levels listed in Table (2), to meet various requirements including the CEC and ErP Lot 6/3 ones. Those requirements are also listed and recommended by the Power Supply Design Guide for Desktop Platform Form Factors, Revision 1.4 (April 2017). [Update: 5 May 2017]
  Table (2) 
5VSB LoadEfficiency Target (110V & 230V)Notes
3 Ampere or Full Load75%Recommended
1.50 Ampere75%Required - ASM for CEC & ErP Lot 3 2014
1.00 Ampere75%Recommended
0.55 Ampere75%Required - ASM for CEC & ErP Lot 3 2014
90 Milliampere45%Required - ErP Lot 6 2010
45 Milliampere45%Required - ErP Lot 6 2013