In some respects, a load bank test of a data center is much like a road test of an automobile. Suppose you are in the market for a car, and you meet with a dealer or a private party to check out what they have to offer. You start it up, look under the hood, and suddenly you tell the hopeful seller, “I’ll take it!” But is that likely? Not really. Who would buy a car without taking it for a test drive? Surely you want to know what it can do, how it performs on the road. Is it reliable? You would probably even want to take it out on the highway. But many data centers test generators without putting them under load. It would help to put it through the paces to see if it can reliably support the critical load when required. A load bank test can do all that for you.
The general consensus among experts is that load bank testing of data center generators should be done at least annually. The National Fire Protection Association stipulates that routine maintenance, inspection, and operational testing of emergency generators should be completed by properly trained personnel. Those requirements are particularly stringent for facilities such as health care centers. The governmental requirements for generator testing may vary from state to state or country to country.
Some kind of reliability testing is recommended at least weekly.
Some kind of reliability testing is recommended at least weekly. That may involve just starting up the generator and letting it run for a while. This gives you the chance to check such things as oil pressure, give it a listen, and record measurements to ensure the generator is working properly. But just starting it up (like the analogy of the car buyer) is not sufficient. It can even make things worse. Experts warn that if a generator is used at less than 30% capacity, it could result in unburnt fuel as well as the build-up of carbon in the exhaust — a condition known as “wet stacking”. Generator load bank testing — like taking an old car that’s been idle out on the highway — can take care of that. Ensuring the reliability of the generator under critical load is one reason for a full load bank test.
There are good reasons for routine load bank testing. A video from Power Control Limited says that load bank testing:
The folks at Authorized Services of New England (ASNE) tell us that load bank testing “allows the engine to reach this full operating temperature and ‘burns out’ this accumulation of un-burnt fuel”. They also write that it is critical to the reliability of a standby generator. “Generator load bank testing is an essential part of good maintenance practices. Performing yearly load bank tests helps to extend the generator’s engine life, and ensure reliable operation during a utility power failure.” They say the unit will run cleaner and more efficiently.
The best reason to conduct regular load bank testing is to prevent downtime. In our blog post about proactive IT maintenance, we reminded readers that “an ounce of prevention is worth a pound of cure”, and that, like the Boy Scouts say, it’s always good to “be prepared”. Generator load bank testing is another procedure to add to your already robust preventive maintenance plan.
This is a tough one. On the one hand, you don’t want to risk it by putting all your data center equipment on a standby generator. What if it fails? Is there another standby? On the other hand, what better way to see whether the generator could stand the load of the full data center in a complete power loss? And what better way to test the reliability of the automatic transfer switch (ATS)? Plus, using the actual load would save the money required to rent a portable load bank tester.
Load bank testers are meant to evaluate whether the generator meets the design specifications indicated by the manufacturer.
But these testers are professionally made for that specific purpose. Load bank testers are meant to evaluate whether the generator meets the design specifications indicated by the manufacturer. Whether the generator is sufficient for the data center is a matter to be determined by those who installed them. The question is whether the generator is still able to perform at the level of its own design.
As for the testing of the ATS, that should be done as well. But that should not preclude the use of load bank testers. One alternative to the portable load bank tester rental is to install a permanent one onsite.
Verifying power capacity, reliability, and redundancy are all part of the power management of a data center. The load testing of generators is only one facet of that discipline. Of course, there are other things to consider. How reliable are the uninterruptible power supply (UPS) units in the data center? Those may be used in addition to the standby generator. Is there any testing procedure for them? Are there written maintenance procedures for the mechanical and electrical parts of the generator? Are those physically inspected regularly? What are the repair procedures when defects are found? There’s a lot to deal with in data center power, but in summary of this article, let us say: Don’t forget to load test your data center generator at least once a year.
On Thursday night August 20th, 2015, a blast occurred in the basement of 811 West Wilshire Blvd. in downtown Los Angeles taking out an on-site power station leaving 12 buildings in the area without utility power for much of Friday, as reported by ABC News and other sources. The outage significantly affected major network provider Level 3 […]
When it comes to IT infrastructure, there’s not much room for error. In fact, businesses have come to expect their data technology environment to be completely reliable. As we wrote in “The Essence of Uptime“, “The standard for network uptime is 99.999% availability.” Data center reliability is essential to the success of the enterprise now […]
After unabashedly extolling the virtues of redundancy in a recent article , you may be wondering why we would follow up with another post questioning whether sometimes too much (redundancy) was just too much. Credit fellow staffers for the suggestion that we revisit the issue. The problem was clearly a part of our initial research, and it deserves […]
IT systems go down for a lot of reasons. Some downtime causes are obvious, while others take some time to understand. And still others are just plain comical. In this article we’ll have a look at different approaches to assigning blame for outages, and we’ll offer a short list of our own. The concept of downtime applies […]