| Measuring and publishing UPS efficiency – a dark art? |
| Thursday, 29 January 2009 00:00 | |||
|
Chloride says that some companies have left an air of mystery on their UPS efficiency figures. This article shows what to look for in the stats A quick desktop survey of published energy efficiency data for UPS machines will result in a wide variation of figures, with as much as three percentage points of difference between the lowest and the highest in any given segment of the market. However this apparent disparity is difficult to explain when one looks deeper into the base technology and discovers that the internal power topology of competing machines are similar despite the difference in published efficiency. This article explores the issue and reaches the conclusion that, in the absence of any independent testing and certification authority, it is far too easy for OEMs to resort to one-up-man-ship in the dark art of the data sheet game.Why has efficiency advanced up the agenda?
In many parts of Europe, the cost of electrical energy has climbed dramatically. At the same time the cost of IT hardware has been steadily declining to the point where, for the first time, over the typical three year life cycle, the energy consumption cost actually exceeds the initial purchase price. If we consider higher capacity UPS the difference in Total Cost of Ownership (TCO) based upon a 3% delta in efficiency can lead to a pay back of the entire UPS hardware in less than 3 years but only if it’s realised in practice! We will now review how one, entirely white brand, machine could have two, very different, published efficiency figures: The most interesting feature of this type of table is that we immediately see that for any given machine we have three efficiency figures that could be published: • Guaranteed 92.5% How the client, or the OEM, views these alternatives is a very subjective matter. The most important conclusion is that until mandatory International Standards are introduced, the efficiency can be measured and interpreted in a variety of ways and each request for the data should be treated with due care and attention. For example it is not true to say that a UPS runs all its life at minimum input voltage and maximum inlet temperature and so the guaranteed figure is far too pessimistic but it is vitally important to size the UPS cooling plant to cope with that disaster scenario. Reputable manufacturers have traditionally published the Guaranteed figure, the lowest of the three, and the nominal figure should be taken into account when TCO calculations are carried out - generally a full 1% higher. It is not recommended to use any best case figures for TCO or plant- room heat capacity planning. Is that the end of the story? No, we have two issues to consider before ending: Hardly any UPS systems run at 100% load and those that are in N+1 or 2(N+1) redundant configurations rarely exceed 30% load per module. We saw that three different efficiency figures can, with some stretching of the facts, be applied to one machine. In fact for most manufacturers there are four efficiency figures, not three since most series on-line (double conversion) machines can run in economy mode with the load on the bypass and the inverter ready to have the load transferred if the mains power deteriorates. In this mode the machine efficiency is typically 97/98% and is useful for loads that require continuous voltage but not computer grade fidelity. It also counters the off line or line interactive efficiency claims by correctly pointing out that efficiency must be higher when the level of protection is lower. However, if the load can run on a line interactive type of power system then ‘up to’ is perfectly permissible. Therefore, for the client, the term caveat emptor applies to data sheets that state efficiency up to 98% without further qualification or explanation.
|
