A client needed to address power and cooling issues with their 5,000 square foot server room. The facility was the single most critical piece of real estate for the global media company. An outage at the facility would mean damage to the brand and complete and instant loss of revenues until services could be restored.
While FEA did not design the original power and cooling systems, having been working with the client since 1995, we were aware of the data center’s power and cooling shortcomings. Working with the facilities department, we presented the issues to upper management in early 2017 which included:
Misapplication of the air conditioning system and inability to cool more than 50% of the UPS capacity.
Single points of failure in the power distribution system.
Non-maintainable equipment (switchgear, circuit breakers, etc.).
Not meeting the company’s critical infrastructure design criteria.
FEA was asked to study four options and provide a schematic design. The options included:
Upgrade the existing facility in place.
Renovate an existing space on campus.
Build a new facility.
The technical advantages, implementation requirements, cost, schedule and risk for each option were determined and evaluated. Ultimately, the team recommended upgrading the existing data center in place because that option met their technical requirements with an acceptable budget and schedule. It was also the least disruptive to their operations both near and long term. The biggest challenge was to design and implement the upgrades in an active data center without taking a shutdown.
Upgrading the existing chiller plant was not possible without taking a shutdown, and a shutdown was not an option. Since no work could be performed on the existing chiller plant, FEA designed a new 350 ton water-cooled chiller plant and CRAH galleries. The new chiller was piped to back up the existing plant and the CRAHs replaced the existing data center cooling system that was inefficient and ineffective.
The biggest challenge was removing portions of the old air handlers and ductwork system to make way for the new, while maintaining cooling to the operating data center. To accomplish this the design included three phases where approximately one third of the old system was removed at a time. Temporary cooling was installed and commissioned before each phases’ demolition. Close coordination was required between the facilities department, trade contractors and FEA engineers to plan, test and execute each step of the project.
The existing electrical systems included an A and B substation, however the way the cooling plant was powered, if either substation was down for a scheduled or unscheduled outage, all cooling was lost. The only way to address this issue was to install new substations to refeed the old and provide the capacity to properly checkerboard the new and existing cooling systems. The new substation allowed us to design in critical component maintainability and automatic restoration features as well. The primary feeders to the building were also reconfigured to provide redundant feeds. This allowed the client to meet their mission critical design standards and switch loads to free up circuits for preventative maintenance. The benefit of the new power distribution and cooling plant was immediately realized before the project was completed.
FEA served as the prime professional providing MEP-FP engineering. We lead a multi-disciplinary team including the architecture, and civil and structural engineers. FEA was engaged from programming, design development, final design, construction administration and commissioning. This was one of the most challenging projects we have ever taken on from a design, coordination and implementation aspect, due to the existing conditions and criticality of the environment.
The project has recently reached the substantially complete milestone and has already payed dividends. The new power and cooling infrastructure permitted the team to perform electrical shutdowns so new, redundant feeds could be installed in de-energized panels, without subjecting workers to arc flash safety hazards. This resiliency will serve the data center well in the future, allowing them to perform scheduled maintenance. Prior to this upgrade they were living on borrowed time. Sooner or later an equipment failure would take them down. Now they can be confident the new system will be able to handle equipment failures while increasing the overall cooling effectiveness and efficiency.