High Relative Humidity in Data Centres: The Final Solutions5 min read

Learn how relative humidity impacts data centers, and discover actionable solutions to optimize RH levels and protect critical infrastructure. Read more now!

Ever wonder how something as simple as air humidity can impact millions of dollars’ worth of equipment? In data centres, controlling relative humidity (RH) isn’t just about comfort—it’s critical for performance and safety.

What is Relative Humidity and Why Does it Matter?

Relative humidity (RH) is a crucial environmental factor in data centers, influencing the overall air quality and the risk of damage to sensitive electronic equipment. RH measures the amount of water vapor in the air relative to the maximum amount the air can hold at a specific temperature. The formula for RH is:

\( RH = \left( \frac{e}{e_s} \right) \times 100 \)

Where:

  • e = actual vapor pressure (in Pa or mbar)
  • es​ = saturation vapor pressure (maximum amount of water vapor the air can hold at a given temperature)

This means that RH is dependent on both temperature and the actual water content in the air. Warmer air can hold more moisture, so a drop in temperature without changing the moisture content will increase RH.

To better understand how RH impacts a data center, let’s calculate the absolute amount of water in the air based on the room size, temperature, and RH.

Step 1: Saturation Vapor Pressure

The saturation vapor pressure ese_ses​ (in Pa) at a given temperature can be calculated using the Clausius-Clapeyron equation. For practical purposes, we can use a simplified empirical formula for temperatures between 0°C and 50°C:

\( e_s = 6.11 \times 10^{ \left( \frac{7.5 \cdot T}{237.3 + T} \right) } \)

Where T = temperature in °C

Step 2: Actual Vapor Pressure

Once you have es​, you can calculate the actual vapor pressure e using the RH:

\( e = \frac{RH \times e_s}{100} \)

Step 3: Amount of Water Vapor in a Room

Now, let’s determine the total amount of water vapor in the air. The absolute humidity AH (in g/m³) is related to the actual vapor pressure eee through the following equation:

\( AH = \frac{216.7 \times 15.91}{25 + 273.15} = 12.29 \, \text{g/m³} \)

Step 4: Total Water Vapor in the Room

To calculate the total amount of water vapor in the room, multiply the absolute humidity by the volume of the room (in cubic meters):

\( \text{Total Water} = AH \times V \)

Where V = volume of the room in m³ (room length × width × height)

Example Calculation:

Let’s assume we have a data center room that’s 10 meters long, 5 meters wide, and 3 meters high, at a temperature of 25°C, with a relative humidity of 50%.

  • Calculate es​:
\( e_s = 6.11 \times 10^{ \left( \frac{7.5 \cdot 25}{237.3 + 25} \right) } = 31.82 \, \text{hPa} \)

  • Calculate e:
\( e = \frac{50 \times 31.82}{100} = 15.91 \, \text{hPa} \)

  • Calculate Absolute Humidity AH:
\( AH = \frac{216.7 \times 15.91}{25 + 273.15} = 12.29 \, \text{g/m³} \)

  • Calculate Room Volume V:
\( V = 10 \times 5 \times 3 = 150 \, \text{m³} \)

  • Total Water Vapor:
\( \text{Total Water} = 12.29 \times 150 = 1,843.5 \, \text{g} \)

Thus, the total amount of water vapor in the room under these conditions is approximately 1.84 kg.

Why Does This Matter?

In a data center environment, managing the total water content in the air is essential to avoid condensation or excessive dryness, both of which can damage sensitive equipment. As RH changes with temperature, even a slight fluctuation can lead to a significant change in water content, increasing the risk of condensation on equipment if the temperature drops. This is why RH is carefully regulated, typically kept between 45-55%.

Having the ability to calculate the exact water content in a room can inform decisions on installing dehumidifiers or adjusting the HVAC system to prevent dangerous humidity levels.

humidity in data center

Optimal RH typically falls between 45-55%, minimizing electrostatic discharge (ESD) risks (which spike at RH <30%) and condensation (which rises when RH >60%).

The Impact of High Humidity on Data Centers

Too high a relative humidity can wreak havoc in a data center. When RH exceeds 60%, condensation can form on server components, which can lead to corrosion or short circuits. In the worst cases, downtime from humidity-induced failures can cost data centers an average of $8,851 per minute, according to a Ponemon Institute study.

In data center SLAs (Service Level Agreements), strict environmental conditions are often enforced, where typical ranges for RH fall between 40-60%. Ensuring compliance is critical, as deviations can lead to costly breaches of contractual obligations and equipment failure.


Technical Solutions for Managing Humidity in Data Centers

Effectively managing RH in a data center involves a combination of strategies, leveraging equipment like Air Handling Units (AHUs), CRAC units, and dehumidifiers.

1. Air Handling Unit (AHU):

AHUs play a crucial role by conditioning the fresh air that enters the data center. To lower RH, an AHU can lower the temperature of incoming air. By bringing the air closer to its dew point, the AHU removes moisture from the air before it enters the data center.

Relative Humidity

It’s also beneficial to implement controls that stop AHU operation when absolute humidity exceeds a set threshold, preventing excess moisture from being introduced.

2. CRAC (Computer Room Air Conditioner) Units:

A CRAC (Computer Room Air Conditioning) unit can dehumidify a data center by controlling the room’s temperature and humidity levels. It operates by circulating air through cooling coils, which lowers the temperature of the air and condenses moisture, effectively removing humidity. This process ensures the air remains at optimal levels for the equipment, preventing overheating and related issues.

3. Installing Dehumidifiers:

For extreme cases where AHUs and CRAC units cannot manage RH effectively, dehumidifiers offer a direct approach to reducing RH. The size and power of the dehumidifier must be carefully calculated based on room volume and current RH levels. For example, a 30-liter dehumidifier might handle a small room (up to 300 cubic meters), while larger data halls could require dehumidifiers capable of removing 200 liters per day or more, depending on external conditions.

Stefano Meroli
Stefano Meroli

CERN scientist, history lover.
PhD in Nuclear Physics and counting.

Articles: 18

Leave a Reply

Your email address will not be published. Required fields are marked *