– Michael Petrino, Vice President of PTS Data Center Solutions (www.PTSdcs.com), says:
Battery related issues can be one of the biggest challenges for an organization due to the fact that most issues are not realized until an event occurs. Commonly learned in a post event recovery is that fact that the battery bank can only support a few minutes of runtime as compared to the expected runtime of the system.
Regular battery maintenance is the most common approach at the time of UPS or quarterly preventative maintenance. Even with maintenance it is possible to have a weak battery bank pass a short battery discharge event, followed by the batteries failing during an extended outage. We advise to add advanced battery monitoring systems with alarm notification. This improves the users view from the typical string voltage as provided by the UPS to the information for each individual cell. Isolating and replacing bad cells will extend the life of the battery string as well as delay the need to incur the costly expense of the entire battery string replacement.
Proper maintenance of the data center electrical power path. Properly maintaining all of the electrical connections on the power path can be a challenge.
The standard procedure of IR (infrared) thermal scanning on regular maintenance intervals is the best approach to get an accurate view of the critical connections. For this to be most beneficial this must be done while these connections are supporting a power load, when possible we advise to add load banks to further stress connections and bring them to future load conditions.
For the most critical connections we advise to install permanently mounted IR (Infrared) scanners and thermal sensors. When scanning a connection 2, 4, or 12 times per year is not enough these sensors can provide real-time data to the status of these connections. The benefit is that once these sensors are installed until there is an issue there is no need to open the cabinet, suit up in the arc flash suit, and take a momentary scan. When the sensor reports an issue it would be necessary to schedule a time to address the problem.
Voltage Regulation. This is a challenging issue facing IT professionals working in any size data center. You know you have voltage issues when a UPS generates an error such as “bypass not available” or “bypass out of tolerance”.
This is a very common issue and can be challenging on multiple levels. It is a typical problem during the summer on very hot days when the utility is struggling to meet the demand on the grid, due to the increased usage of air conditioning systems. As the grid becomes stressed, customers will see sags where the voltage will drop, sometimes in a steadily decreasing fashion which can be very stressful to watch and worry as you grow concerned about what if my UPS no longer accepts utility and goes onto battery but my generator does not come on. This is a real issue, when UPS tolerances are tighter or narrower than a generator for example a UPS most often will accept a voltage range of +/- 10% while a generator ATS will accept a typical range of +/- 15%.
Some things that you should understand going into this. First the most common three phase voltage in the US is 480V from the utility. In some areas the utility company will provide 460V. Here starts your challenge. If the default ‘normal’ setting on your UPS and ATS is 480V, but your normal utility voltage is actually 460V your are already in a 4% under voltage situation under normal operation. Causing you to be very close to the bypass alarms mentioned above. There are remedies for this, by working with your engineer and eventually support vendors to alter the equipment. I caution an advise working with the engineer who understands the transformer and the possibility that there may be adjustments or ‘taps’ on the generator that could be manipulated. Further if the transformer is not adjusted or manipulated it is possible to make adjustments on most UPS systems and ATS that are less than 5 years old.