Research Papers

Determination of the Lumped-Capacitance Parameters of Air-Cooled Servers Through Air Temperature Measurements

[+] Author and Article Information
Hamza Salih Erden

Department of Mechanical and
Aerospace Engineering,
Syracuse University,
Syracuse, NY 13244
e-mail: herden@syr.edu

H. Ezzat Khalifa

Fellow ASME
Department of Mechanical and
Aerospace Engineering,
Syracuse University,
Syracuse, NY 13244
e-mail: hekhalif@syr.edu

Roger R. Schmidt

Fellow ASME
IBM Corporation,
Armonk, New York 10504
e-mail: c28rrs@us.ibm.com

1Corresponding author.

Contributed by the Electronic and Photonic Packaging Division of ASME for publication in the JOURNAL OF ELECTRONIC PACKAGING. Manuscript received September 17, 2013; final manuscript received February 25, 2014; published online May 5, 2014. Assoc. Editor: Siddharth Bhopte.

J. Electron. Packag 136(3), 031005 (May 05, 2014) (9 pages) Paper No: EP-13-1107; doi: 10.1115/1.4027092 History: Received September 17, 2013; Revised February 25, 2014

Computer servers can be represented by lumped thermal capacitances for the purpose of simulating server and data center transient thermal response to changes in operating conditions or equipment failures. Two parameters are needed to characterize the transient behavior of a lumped-capacitance server: its thermal capacitance and its thermal conductance, heat transfer effectiveness, or time constant. To avoid the laborious task of obtaining these parameters from measurements or estimations of the thermal characteristics of internal components of the server, a method is proposed to derive these parameters from external measurements that can be easily obtained without performing an “autopsy” on the server. In this paper, we present the mathematical formulation underlying the proposed method and describe how the parameters are to be obtained from external air-temperature measurements using the mathematical model. We then present validation test cases using experimental data from server shut-down and inlet-temperature ramp tests. The experimentally obtained parameters are implemented into a computational fluid dynamics (CFD) case study of server shutdown in which the transient server exit air temperature is computed from the lumped-capacitance parameters via a user-defined function. The results thus obtained are in excellent agreement with the experimental data.

Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Fig. 1

Example of model results for ramp-up of server inlet temperature

Grahic Jump Location
Fig. 2

Comparison of model predictions with the experimental data of VanGilder et al. [15]

Grahic Jump Location
Fig. 3

Extraction of server time constant from the experimental data of VanGilder et al. [15]

Grahic Jump Location
Fig. 4

Chassis air temperature difference during shutdown experiment

Grahic Jump Location
Fig. 5

Variation of chassis air temperature difference due to a jump in inlet temperature: (a) inlet temperature, and (b) chassis temperature difference

Grahic Jump Location
Fig. 6

Normalized temperature difference to estimate chassis time constant

Grahic Jump Location
Fig. 7

Estimated chassis time constants as a function of air flow rate

Grahic Jump Location
Fig. 8

Comparison of computed and experimental server exit temperatures

Grahic Jump Location
Fig. 9

Comparison of computed and experimental chassis air temperature differences




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In