0
Research Papers

Optimized Thermoelectric Module-Heat Sink Assemblies for Precision Temperature Control

[+] Author and Article Information
Rui Zhang, Marc Hodes, David A. Brooks, Vincent P. Manno

Department of Mechanical Engineering,  Tufts University, Medford, MA 02155

The latter two assumptions imply that TEM operation in refrigeration mode is necessary at some, if not all, permissible operating conditions.

These expressions are necessary to determine the permissible range of H and H2 and Hpe defined in Sec. 3.

The benefit of increasing φAsub is considered below.

The ordinate of (most of) the plots below is $W·t$ rather than $W·TEM$. Otherwise, negative values of $W·TEM$ arising when generation mode is encountered may not be displayed when a log scale is used.

Figure 5 is identical to Fig. 4 but contains further information.

When (T∞ ,max  − Tcp ) is sufficiently large, H1 and H2 are the smaller and larger roots, respectively, for which $W·TEM$(H,Φ(H),T∞ ,min ) = $W·TEM$(H,Φ(H),T∞ ,max ) when Ru-∞  = 0 as per Fig. 6.

In Region III, a TEM operates in generation mode and then cooling mode for sufficiently high H. In part of the generation mode region and all of the cooling mode region TEM power increases with H (see inset plot in Fig. 4), but this is over a rather small range of H and not of practical importance.

See Hodes [2] for further discussion on TEM operating modes, albeit using not an entirely consistent set of definitions.

The number of the thermocouples (N) affects the operating voltage and spreading/constriction resistances in the substrates; however, none of these parameters are required to perform the optimization.

TEM power ($W·TEM$) rather than total power ($W·t$) is plotted to emphasize the range of Ru-∞ for which the TEM operates in generation mode.

J. Electron. Packag 134(2), 021007 (Jun 11, 2012) (10 pages) doi:10.1115/1.4005905 History: Received August 24, 2011; Revised December 05, 2011; Published June 11, 2012; Online June 11, 2012

Abstract

Robust precision temperature control of heat-dissipating photonics components is achieved by mounting them on thermoelectric modules (TEMs), which are in turn mounted on heat sinks. However, the power consumption of such TEMs is high. Indeed, it may exceed that of the component. This problem is exacerbated when the ambient temperature and/or component heat load vary as is normally the case. In the usual packaging configuration, a TEM is mounted on an air-cooled heat sink of specified thermal resistance. However, heat sinks of negligible thermal resistance minimize TEM power for sufficiently high ambient temperatures and/or heat loads. Conversely, a relatively high thermal resistance heat sink minimizes TEM power for sufficiently low ambient temperatures and heat loads. In the problem considered, total footprint of thermoelectric material in a TEM, thermoelectric material properties, component operating temperature, relevant component-side thermal resistances, and ambient temperature range are prescribed. Moreover, the minimum and maximum rates of heat dissipation by the component are zero and a prescribed value, respectively. Provided is an algorithm to compute the combination of the height of the pellets in a TEM and the thermal resistance of the heat sink attached to it, which minimizes the maximum sum of the component and TEM powers for permissible operating conditions. It is further shown that the maximum value of this sum asymptotically decreases as the total footprint of thermoelectric material in a TEM increases. Implementation of the algorithm maximizes the fraction of the power budget in an optoelectronics circuit pack available for other uses. Use of the algorithm is demonstrated through an example for a typical set of conditions.

<>

Figures

Figure 1

Cutaway view of a single-stage TEM [3]

Figure 2

Schematic of a TEM embedded in a thermal resistance network [8]

Figure 3

Ru-∞ ,max as a function of H at T∞ ,min and T∞ ,max when qcp  = qcp ,max

Figure 4

.Wt as a function of H at T∞ ,min and T∞ ,max when Ru-∞  = 0 and Ru-∞ ,max

Figure 5

.Wt and.Wt ,ml as a function of H at T∞ ,min and T∞ ,max when (T∞ ,max –Tcp ) is small when Ru-∞  = 0 and Ru-∞ ,max

Figure 6

.Wt and.Wt ,ml as a function of H at T∞ ,min and T∞ ,max when (T∞ ,max  − Tcp ) is large when Ru-∞  = 0 and Ru-∞ ,max

Figure 7

.Wt as a function of H at T∞ ,max when qcp  = 0 and qcp ,max when Ru-∞  = 0 and Ru-∞ ,max

Figure 8

.Wt as a function of H at T∞ ,min when qcp  = 0 and qcp ,max when Ru-∞  = 0 and Ru-∞ ,max

Figure 9

Delineation of TEM operating modes as a function ofH and Ru-∞ when T∞  = T∞ ,min  < Tcp

Figure 10

Delineation of impermissible and Peltier cooling mode regions as a function of H and Ru-∞ when T∞  = T∞ ,max  > Tcp

Figure 11

.WÂ TEM as a function of Ru-∞ (Tcp  = 55 °C)

Figure 12

qcp ,TEM,max as a function of H at T∞,min and T∞,max when Ru-∞  = 0

Figure 13

Ru-∞,max as a function of H at T∞ ,min andT∞ ,max

Figure 14

. Wt andÂ  . Wt ,ml as a function of H at Tcp  = 55 °C when Ru-∞  = 0 and Ru-∞ ,max

Figure 15

. Wt Â  as a function of Ru-∞ at Hmg at T∞ ,min andT∞ ,max

Figure 16

. Wt ,mg as a function of (φAsub )1/2

Discussions

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 Proceedings Articles
Related eBook Content
Topic Collections