February/March_AMP

A D V A N C E D M A T E R I A L S & P R O C E S S E S | F E B R U A R Y / M A R C H 2 0 1 7

6 4

FEATURE

of 27 kW, is capable of heating the coupon surface to 1600°F,

achieving a maximum surface heat flux of 10.8 MW/m

Data collection begins when the coupon is at its max-

imum temperature and stable film boiling is established at

the desired quenchant flow rate and temperature. Collec-

tion continues as the heaters reduce the surface tempera-

ture of the coupon at a controlled rate of 20°F/minute. This

slow cooling approach provides a well-defined relation-

ship between surface temperature and surface heat flux

during pseudo-steady-state conditions. Figure 2 shows the

vapor-liquid interface on the test coupon during film boil-

ing at a relatively low flow rate. The test procedure is then

repeated for each flow velocity, quenchant temperature,

and surface orientation of interest.

The raw data collected are temperature values ob-

tained from thermocouples inserted below the surface of

the copper coupon. The values are used to derive surface

temperature and heat flux rate. The procedure may seem

similar to the inversing processes typically used in quench

trials, although there are a few key differences. First, all data

are collected in near steady-state conditions, eliminating the

complexities inherent in transient cooling processes. Sec-

ond, two thermocouples are used at each surface location,

placed at different depths, allowing for advanced extrapola-

tion techniques to the surface of the coupon

[1]

. And finally,

because surface heat flux data are linked to surface tem-

perature, quenchant temperature, quenchant velocity, and

surface orientation, they can be applied to any surface under

the same conditions.

QUENCH TESTING, SIMULATION

AND VALIDATION

To validate themethod, quenching datawere collected

on a generic turbine disk shape made of Inconel 718. Nearly

two dozen thermocouples were inserted into the part at crit-

ical locations. Most of the thermocouples were placed 0.100

in. from the surface, while five were buried deeper into the

part. Near the rim of the disk, the near-surface thermocou-

ple was repeated at 60° intervals to assess the circumferen-

tial uniformity of the quenching operation.

In quench tests performed at the Wyman-Gordon

research facility in Houston, the disk was soaked at 1800°F,

then quenched in Houghton 3420 quenchant under both

still and agitated conditions. Temperature data were col-

lected at 10 Hz during both the heating and quenching

cycles.

The quenching operation was simulated with

ANSYS-Fluent version 16, using a 0.1 second time step. For

the still oil quenching simulation, radial symmetry of the

part allowed for a 2D model, and 17,557 cells were used to

represent both the disk and quench tank in cross section.

Simulating a quench of 1500 seconds required about five

hours of processing time on a 3.5-GHz Linux workstation

with eight processors.

Boiling functions constructed from the flow boiling

database were incorporated into a Fluent User Defined

Function (UDF) that adjusted the surface heat transfer

coefficients (HTCs) on an iterative basis for each surface

Fig. 2 —

Filmboiling on side surface at low velocity and high temperature. Flow is from left to right. Vapor film is thick and uneven, and

vapor bubbles are being shed.