Donald C. Price

Performance of Pin Fin Cast Aluminum Coldwalls, Part 1: Friction Factor Correlations

At low to moderate Reynolds numbers, cast pin e n coldwalls provide the best performance and lowest unit cost when castings are economically viable. An extensive literature search revealed that, although signie cant research has been done for relatively short pin e ns (L/d 10 3 , the friction factor is independent of pin length and Reynolds number.

Performance of Pin Fin Cast Aluminum Coldwalls, Part 2: Colburn j-Factor Correlations

An experimental program was conducted over a wide range of geometric cone gurations and Reynolds number with 44 cast, staggered, pin e n test sections producing heat transfer data for streamwise spacing ratios S/d from 1.83 to 3.21, transverse spacing ratios T/d from 2.00 to 6.41, and pin lengths L/d from 1.88 to 7.25. Empirical correlations are presented for the Colburn j factor as a function of the spacing ratios, pin length, and Reynolds number. Analysis of the experimental heat transfer results, as in the case of the isothermal tests for friction factor, indicates that the e ow transitions from laminar to turbulent e ow at Red » 10 3 with unique correlations in each regime. For laminar e ow, the Colburn j factor depends on the three-dimensional cone guration and Reynolds number Red. In the turbulent regime, the Colburn j factor depends primarily on T/d, L/d, and Reynolds number Red. Nomenclature Abase = test section bottom wall area, lw i NpinApin;b, m 2 AHT = Abase C Atop C Apin D 2(lw)C Npin o.davL id 2 =2), m 2 Amin = test section e ow area, wL i Nt Aproe le, m 2 Apin = total pin surface area, Npin odavL, m 2 Apin;b = base area of each pin, m 2 Aproe le = (d Cdtip)L=2DdavL, m 2 Atop = test section top wall area, lw i NpinApin;b, m 2 cp = specie c heat, kJ/ (kg¢K) D = hydraulic diameter, 4 Aminl=AHT, m d = pin base diameter, m dav = average pin diameter, 0.5 (d Cdtip/, m dtip = pin tip diameter, m h = heat transfer coefe cient, q=AHT.Tw;av iTb;av/, W/(m 2 K) j = Colburn j factor, St Pr 2=3 , Nu=Re Pr 1=3 k = thermal conductivity, W/m K L = pin length (see Fig. 1), m L=d = dimensionless pin length l = test section e ow length, m Npin = total number of test section pins Nt = pin count in the maximum transverse row Nu = Nusselt number based on pin diameter, hd=k NuD = Nusselt number based on hydraulic diameter, hD=k Pr = Prandtl number, cpπ=k Q = volumetric e ow rate, m 3 /s q = heat transfer rate, W

Thermal Design of an Airborne Computer Chassis With Air-Cooled, Cast Pin Fin Coldwalls

This paper documents the thermal design process required to provide effective thermal management for an airborne computer, consisting of 24 modules (two P/S modules and 22 PWB modules), which are edge-cooled to two cast, pin fin coldwalls. The computer chassis is mounted in an electronics pod mounted underneath the centerline of an aircraft. The pod consists of several electronics bays and a self-contained, air-cycle, environmental control system (ECS). The computer chassis is mounted in the forward bay, and the ECS is mounted in the rear bay of the electronics pod. The ECS is an air-cycle refrigeration system, which operates on captured ram air directed by an inlet/diffuser to an expansion turbine. This turbine produces low-pressure, chilled air, which is then directed through an air-to-liquid, load heat exchanger to produce chilled liquid. The chilled liquid is piped through small liquid lines to the forward bay of the pod, where the air-cooled computer chassis is located. The chilled liquid is converted back to chilled air in an air-to-liquid heat exchanger. The chilled air is supplied to the forward bay volume and is drawn through the computer chassis coldwalls by a fan integral to the computer chassis. The temperature of the chilled air, produced in this manner, becomes a strong function of the altitude and flight speed of the aircraft