ORCID Identifier(s)

0000-0003-3523-9284

Graduation Semester and Year

2018

Language

English

Document Type

Thesis

Degree Name

Master of Science in Mechanical Engineering

Department

Mechanical and Aerospace Engineering

First Advisor

Dereje Agonafer

Abstract

The energy used by Information Technology Equipment (ITE) and the supporting data center equipment keeps rising as data center expansion continues worldwide. To contain the rising operation costs, data center administrators are resorting to cost cutting measures by not adhering to the ambient data center environment standards. In many cases, free cooling techniques are adopted to partially or completely relieve the usage of refrigeration units, which adds to huge cost savings but also allows for the increased risk of particulate and gaseous contaminations. The ASHRAE TC9.9 subcommittee, on Mission Critical Facilities, Data Centers, Technology Spaces, and Electronic Equipment, has provided the temperature-humidity envelope primarily targeting data centers utilizing free air cooling or air-side economizers, to manage the contamination risk. According to ASHRAE Standard (2009b), to manage the contamination risk, it was recommended to have the air entering the data center to be continuously filtered with MERV 11 or preferably MERV 13 filters to achieve ISO Class 8 cleanliness standards. The HVAC systems in the data center operate with high supply flow rates to maintain the server inlet temperatures within the ASHRAE recommended limits and is continuously changing the air in the server space. This air change causes for different levels of air speed, temperature difference and contaminant concentration at different regions of the white space. The spread of contamination from such a turbulent airflow leads to unanticipated stagnant regions or areas of dense accumulation of contaminants in the cabinet aisles, thereby increasing the risk of failure due to airborne particulate contamination. This paper uses 6SigmaRoom, a data center CFD (Computational Fluid Dynamics) tool, to study the airflow pattern and flow path of sub-micron contaminants in a high-density data center with both hot/cold aisle containment configurations. In particular, it addresses the behavior of particulate (dust) contaminants which are characterized by their spherical shape and of diameters 0.05, 0.1 and 1 micron. The sub-micron particles have little slip relative to the air, which allows for these particles to be considered as completely airborne. For heavier particles of size 1 micron, the properties of dense carbon particle are considered for simulating contamination in the white space. The data obtained from this study can be helpful in predicting the nature of contaminants and their extent of contamination when utilizing air-side economizers. Further, identifying the region of deposition of contaminants can help data center operators to implement design strategies or simply reinforce the regions which have a potential to fail from particulate contamination. MERV filters with higher efficiency ratings can be used to solve the contamination problem, but this leads to added installation and operation costs .With the help of results from this study, data centers utilizing air-side economizers can tightly control the risk of contamination and also save on operational costs from allowing the IT equipment to operate outside the recommended and allowable ASHRAE envelopes.

Keywords

Air-side economizer, Contamination, Data center, 6SigmaRoom, Cold-aisle configuration.

Disciplines

Aerospace Engineering | Engineering | Mechanical Engineering

Comments

Degree granted by The University of Texas at Arlington

Download

Share

COinS