Uninterruptible power supplies
The regular maintenance of uninterruptible power supplies (UPS) is vital because interruptions can really impede progress. Without that ingredient of regular inspection and, where necessary, replacement of parts, investment in a UPS system is a waste of money, writes PETER ALLPORT.
The importance of regular inspections of uninterruptible power supplies (UPS) equipment for mission-critical applications cannot be overstated. Mains power failure incidents are unpredictable and often it can be as long as a year between disruptions, and so it’s vital to be assured that the UPS is functioning correctly. In short, without check-ups on the state of health (SOH) of batteries and electronics, you may well find that when the mains power is interrupted, the UPS won’t take over the task of supplying essential power to the critical functions in your organisation.
THE ARCHITECTURE OF A UPS
The diagram in Figure 1 is helpful in comprehending the basics of operation. No specialist knowledge is required and it is beneficial in understanding the important areas of regular inspection, testing and, where indicated, component replacement. The diagram illustrates a UPS typical of larger installations, technically described as an ‘in-line’ system.
Normally, mains electricity supply is alternating current (AC), as shown graphically at the input to the UPS, under mains availability conditions. The sine wave nature of voltage and current (AC as supplied by the electricity network) has to be mimicked by the UPS when mains power fails.
The power for your facility, in the event of an interruption to the mains power, will come from a battery bank, as shown in the diagram. The charge for the battery bank is supplied during times when power is available from the network. Batteries store direct current (DC) energy and require DC energy to charge them. The diagram shows a ‘converter’ block, basically changing the AC supplied by the network to DC for battery charge replenishment (see Figure 1).
The last section of importance is the block labelled ‘inverter’, where DC from the battery bank is changed (inverted) back to AC. Something has emerged that you may not have been aware of: the in-line system illustrated is operating as a UPS, even when mains power is available. In other words, the batteries are being continually charged (trickle-charged) and the inverter is always supplying AC to your critical loads.
The great advantage of the in-line architecture is that there is no switchover time in the event of a network failure, when, during the period, even if very short, there is no power for the critical loads. Apart from the danger of loss of data in IT operations, there is also the risk of damage to essential services motors.
WHAT CAN AND DOES GO WRONG (BATTERIES)
Now that the structure of an in-line system is clear, you would think that if there was a failure of the UPS you’d be aware of it the moment it occurs. But that is not the case – you will, in all likelihood, be blissfully unaware that you have lost your critical power back-up. Figure 2 shows the diagram from the first figure overlaid with an ‘automatic bypass circuit’ (shown in red). This bypass is automatically activated if there is a failure in the UPS, switching the critical load directly to the network mains (see Figure 2).
In Image 1, the Salvador Dali-like melting batteries tell a story. This photo, taken of a UPS in a hospital, makes the point of blissful ignorance glaringly obvious. Fortunately, there had been no interruptions of mains power when a service team was called to do an inspection of the system.
Batteries are an obvious area of concern and further down there is a brief description of typical tests conducted as to their SOH. There are other critically important areas as well.
In both Figures 1 and 2, the capacitor bank is shown ahead of the battery bank. Capacitors are essentially very short-term electric charge-holding components, and very similar to batteries. For example, they spill electrolyte when they get hot, causing a lot of damage. The function of capacitors is to ensure smooth trickle-charging of the batteries during normal operation of the UPS. In Image 2 the breakdown of capacitors is shown. Electrolyte has spilled, creating a fire hazard.
The converter (converting AC to DC) and the inverter (converting DC to AC) are examples of power electronics. Both use up power, can get very hot and can fail as a result. This is particularly the case for the inverter, which is basically a group of very fast operating switches – they can typically perform 4000 or more on-off operations per second. These electronic switches use significant amounts of power and consequently require cooling.
To achieve this, the individual switches (transistors of a special type; a popular one being the ‘insulated gate bi-polar transistor’ [IGBT]), are mounted to finned metal sections using special electrically-insulating, thermally-transmitting washers and silicone grease and a torque wrench to seat them, so that there is maximum contact area for the heat to flow from the transistor to the heat sink. Fans blow air over the metal heat sink sections so that the electronic switches operate in a thermally stable environment.
There’s an old saying ‘familiarity breeds contempt’, and this is very true for batteries. The common battery in UPS systems is a lead-acid cell, which doesn’t require regular inspection or topping up on the level of the acid-water electrolyte. This is the valve-regulated lead-acid (VRLA) battery. It is the usual battery for motor vehicles, where it is treated with benign neglect until a roadside replacement takes place. This is no way to treat such a critically important part of a UPS, however, as it’s required to keep things going for periods as long as half an hour. Battery banks in UPS systems comprise paralleled strings (a string is simply a number of batteries daisy-chained in series to provide the required voltage). The strings are bundled together (paralleled) so that the battery bank can deliver not only the required voltage but also the power.
The VRLA batteries require frequent testing and, by the time 80 percent of capacity has been used, this should be at yearly intervals. Part of a battery maintenance practice can include amp-hour data and, in any event, the measurement of open circuit voltage and battery impedance. This latter parameter measures the resistance to current flow through the battery, and the resistance increases over time. A build-up in the value of the impedance parameter is an excellent predictor of battery failure. The combination of the open circuit voltage test and impedance provides the decision level for battery replacement. Record keeping is therefore essential and an integral part of a maintenance program.
In the event of a failure in the UPS, no matter what its cause may be, the automatic bypass circuit has to supply power to the critical load. In effect, the UPS is taken offline, but in the process it automatically switches the bypass circuit to ‘on’, thus maintaining power for the critical loads. This particular test is conducted in consultation with the client because it involves a short disruption of supply.
RELYING ON YOUR UPS SYSTEM
Irrespective of the make of your UPS system, they all need maintaining. However, the age of your UPS does make a difference, for the older the UPS the more important is regular inspection and service. The absence of power failures notwithstanding – in fact, even more so – makes a regular, thorough examination essential. ●
Peter Allport, national service manager, Power Parameters Pty Ltd.
This article also appears in the October/November issue of FM magazine.