Cooling for data centres: what the system has to deliver – and where efficiency comes from

The cooling system of a data centre runs continuously and is at the same time one of the largest items in operating costs. How far availability and efficiency can be reconciled is decided early on: with the chilled water temperature, the redundancy concept, the control strategy and the question of how much cooling capacity can be delivered without a compressor.

Why cooling in a data centre is a special case

In many industrial applications the cooling load varies considerably – with shift patterns, with batches, over the course of the day. In a data centre the heat load is largely constant: virtually all of the electrical power drawn by the IT equipment is released again as heat and has to be dissipated. That shifts the priorities of the design.

Stability before peak output. Maximum cooling capacity says little about how well a system performs in data centre operation. What matters more is how tightly it holds the supply air temperature over long periods. The decisive factor is the permissible operating conditions of the IT equipment – temperature and humidity within the limits set by manufacturers and operators. Fluctuations within that window are generally uncritical; it becomes critical when those limits are exceeded or when the control system cannot follow the change in load.

Redundancy across the entire chain. That includes chillers, pumps, hydraulics, controls and power supply. A redundant chiller is of little use if the distribution system is built as a single circuit.

Serviceability during operation. A data centre cannot be shut down to replace a pump. Isolation valves, accessibility and a workable changeover concept therefore belong in the design from the outset.

Server room or data centre: two different tasks

“Server room cooling” and “data centre cooling” are often used interchangeably. Technically they are two different classes.

A Server room cooling in a company typically lies in the single-digit to low double-digit kilowatt range. Direct expansion split units or compact chillers are usually used there; redundancy is provided by a second, independent unit.

A Data centre begins where power density, redundancy requirements and continuous operation justify a dedicated refrigeration system: central refrigeration, a chilled water network, defined load zones and a higher-level control system. From this size onwards, the design is about the structure of the system rather than the choice of equipment. That is also where the differences in energy consumption lie.

Considering refrigeration and air conditioning separately

In tenders, “data centre air conditioning” and “data centre cooling” often appear in the same breath. For the design it pays to separate them:

• The Refrigeration supplies chilled water – chillers, dry coolers, pumps, hydraulics.
• The Air conditioning delivers that cooling to the hardware – computer room air conditioners, air routing, hot and cold aisle containment, air volumes.

Both sides have to be matched to each other. Efficient refrigeration loses its advantage if the air routing mixes cold and warm air, because the return temperature then no longer matches the design. When assessing existing installations, experience shows this interface to be the most rewarding place to start: it can often be improved without touching the refrigeration plant itself.

Free cooling and the chilled water temperature

The greatest influence on operating costs comes from free cooling – the mode in which outside air dissipates the heat and the compressor is fully or partially shut down.

A key parameter here is the chilled water temperature. Higher flow temperatures generally increase the usable potential for free cooling. How many free cooling hours actually accumulate over a year, however, depends on the system as a whole – on the design of the dry coolers, on the hydraulics, on the control system and on the climatic conditions at the site. A chilled water temperature set unnecessarily low restricts this potential across the entire service life of the plant and can only be corrected later at considerable expense.

The transition between operating modes is equally important. A control system that merely switches between free cooling and compressor operation leaves the mixed range unused. A gradual transition that adapts both modes to the current IT load is the better approach.

The heat removed can be put to further use through Heat recovery instead of releasing it entirely into the environment. This does require a consumer and a temperature level that matches their requirements – both belong in the concept at an early stage.

Natural refrigerants in the data centre

The European F-Gas Regulation is progressively restricting synthetic refrigerants with a high global warming potential (GWP). For a refrigeration system intended to run for 15 years or more, this affects the economics over its lifetime: top-up quantities, availability in the event of servicing and the residual value of the plant.

Propane (R290) has established itself in this sector. It has a GWP of 0.02, good thermodynamic properties and achieves high efficiency levels even at the higher flow temperatures required for free cooling and heat recovery. Propane is flammable and classified in safety group A3; limited fill quantities and a suitable safety concept are therefore essential. In practice, this is often resolved by locating the plant room outdoors.

You will find an overview of CO₂, ammonia and propane with their key figures and fields of application on our page on natural refrigerants.

What the Energy Efficiency Act means for refrigeration

For data centres in Germany, the Energy Efficiency Act (EnEfG) sets requirements for energy efficiency. The measure used for this is PUE (power usage effectiveness), that is the ratio of total energy demand to the energy demand of the IT equipment. There are also requirements regarding the use of waste heat.

Which values, deadlines and reporting obligations apply in an individual case depends among other things on the connected load, on the date of commissioning and on the version of the legal requirements in force at the time. For a specific project this should be checked at the start of the design phase, because it determines the requirements for metering and for the system concept.

This has a number of consequences for refrigeration, irrespective of the individual limit value. The share of cooling in total electricity consumption becomes a figure that has to be demonstrated – which requires meters, sensors and a metering concept planned in from the outset. Chilled water temperature and free cooling concept are major factors in determining what values a site can achieve at all, and they are fixed during the design. And heat recovery moves beyond a purely commercial consideration.

From practice: NorthC Datacenters, Munich

For the NorthC Datacenters GmbH KSI Kältetechnik implemented the first expansion stage of the central cooling system at the Munich site – as an energy refurbishment during ongoing data centre operation, without interrupting the existing IT infrastructure.

The key facts:

• Refrigerant: R290 (propane)
• Cooling capacity: 1.0 MW
• Redundancy: 100%, all pumps designed redundantly
• Operating mode: dynamic switching between mechanical cooling and free cooling, controlled by the current IT load
• Heat recovery integrated into the cooling plant
• Control system: Siemens central control system with visualisation and secure remote access
• Construction: plant room built entirely outdoors, in parallel with ongoing operation
• Planning: entirely in 3D, owing to the limited roof area

Two points shaped the execution. The plant room was built entirely outdoors and in parallel with ongoing operation, so that the data centre continued to run throughout the construction period. And given the limited roof area, complete 3D planning was the prerequisite for coordinating all components reliably.

The system is designed so that further expansion stages can be connected without intervening in the basic structure. Full details of the project can be found in the NorthC Datacenters reference project.

How to recognise a sound design

Anyone assessing a cooling concept for a data centre – whether new build or refurbishment – will get a long way with these questions:

1. What chilled water temperature has been specified, and how is it justified? Higher flow temperatures increase the potential for free cooling but call for a correspondingly dimensioned air side.
2. How many free cooling hours are calculated for the site, and on what data basis? The assumptions should be disclosed.
3. Is the redundancy continuous? Across chillers, pumps, hydraulics and controls.
4. Can maintenance be carried out during operation? Isolation valves, accessibility, changeover concept.
5. How does the system behave at part load? Data centres rarely run at their design point.
6. Will the refrigerant last over the planned service life of the plant?
7. What metering is planned? Without meters on the refrigeration side, it will later be impossible either to optimise or to demonstrate anything.

Where these questions are answered soundly, the design is usually sound in energy terms as well. If they remain open, experience shows that this only becomes apparent during operation – and corrections are then costly.

How KSI Kältetechnik can help

KSI designs, builds and maintains industrial refrigeration systems using natural refrigerants – from energy planning and concept development through revitalisation of existing systems to technical support with 24/7 standby.

You can find a summary of what this means in practice for data centres on our website Cooling technology for data centres.

For data centres this covers load zone analysis and efficiency assessment of existing installations, the design of chilled water and free cooling concepts, refurbishment during ongoing operation and continuous monitoring of operating conditions. An overview of the requirements in this sector is given on our page Sectors: data centre.