Lynn Gear

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/283720017
Local Film Boiling and Its Impact on Distortion of Spur Gears During Batch
Quenching
Article · January 2012
CITATIONS READS
8 1,491
4 authors:
Nikolai I. Kobasko Michael A. Aronov

IQ Technologies Inc., Cleveland, Ohio, USA and Intensive Technologies Ltd, Kyiv, U… IQ Technologies, Inc.
491 PUBLICATIONS 1,841 CITATIONS 94 PUBLICATIONS 338 CITATIONS
SEE PROFILE SEE PROFILE
Lynn Ferguson Zhichao Li

DANTE Solutions, Inc. DANTE Solutions, Inc.
89 PUBLICATIONS 616 CITATIONS 67 PUBLICATIONS 353 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Modifying intensive quenching (IQ - 2 and IQ - 3) processes based on recent achievements of science, View project
Agueous Quenchants View project
All content following this page was uploaded by Nikolai I. Kobasko on 12 November 2015.
The user has requested enhancement of the downloaded file.

Materials Performance and Characterization, Vol. 1, No. 1
Paper ID MPC104533
www.astm.org
N. I. Kobasko,1 M. A. Aronov,2 B. L. Ferguson,3 and Z. Li3
Local Film Boiling and Its Impact on

Distortion of Spur Gears During Batch
Quenching
REFERENCE: Kobasko, N. I., Aronov, M. A., Ferguson, B. L., and Li, Z., “Local Film Boiling and Its Impact
on Distortion of Spur Gears During Batch Quenching,” Materials Performance and Characterization, Vol. 1,
No. 1, 2012, pp. 1–15, doi:10.1520/MPC104533. ISSN 2165-3992.
ABSTRACT: The paper discusses results of computer simulation connected with the double distortion during
batch quenching of spur gears caused by a local film boiling between teeth. A carburized gear, outside diameter
2.5 in., was intensively quenched in conditions that provided heat transfer coefficient (HTC) equal to 25 000
Wm2K1. In some places between teeth local film boiling took place where HTC was 800 Wm2K1. Com-
puter simulation showed that maximum displacement is observed between teeth where local film boiling took
place. The authors came to the conclusion that increasing critical heat flux densities and elimination of local
film boiling can result in decreasing distortion of spur gear. That is true for different sizes of gear during their
quenching when using the second type of intensive quenching process (IQ-2) technique (a two or three-step
quenching process). It is underlined that critical heat flux densities have a great effect on distortion during batch
quenching. The authors also came to the conclusion that a small amount of special additives can decrease signif-
icantly distortion during quenching of gears. That is why a global database on cooling capacity of quenchants
should be available which must contain critical heat flux densities of different kinds of quenchants.
KEYWORDS: computer simulation, displacement, critical heat flux density, local film boiling, additives,
spur gear, IQ-2 process, optimization
Introduction
The paper discusses double distortion during batch quenching of gears caused by local film boiling
between two teeth. Double distortion means that the top of each tooth moves in opposite direction
when between two teeth a local film boiling takes place. Computer simulation showed that such
behavior of teeth generates two times larger distortion. The present work is the second part of the
authors’ investigations. The first part was discussed at the World Scientific and Engineering Acad-
emy and Society (WSEAS) Conference held on August 20–22, 2010 in Taipei, Taiwan [1]. On the
basis of experiments, computer fluid dynamics (CFD) modeling, and finite elements method
(FEM) calculations of residual stress distribution in gears, the authors came to the conclusion that
distortion and variations in dimensional change can be significantly decreased by combining water
flow velocity with special chemical additives, which increase the first critical heat flux density [2,3].
During batch quenching of large gears in an agitated, anticorrosion water-salt solution of optimal
concentration, distortion variation is small and dimensional change is repeatable from tooth to
tooth. During intensive quenching of small gears in a high velocity intensive quenching (IQ-3)
Manuscript received November 8, 2011; accepted for publication May 1, 2012; published online October 2012.
1
IQ Technologies, Inc., Akron, OH 44309, USA (Corresponding author), e-mail: nikobasko@yahoo.com.
2
IQ Technologies, Inc., Akron, OH 44309, USA
3
Deformation Control Technology, Inc., Cleveland, OH, 44130, USA
Copyright V
C 2012 by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. 1
Copyright by ASTM Int'l (all rights reserved); Mon Feb 11 14:50:32 EST 2013
Downloaded/printed by
Dr. Nikolai Kobasko (IQ Technologies Inc) pursuant to License Agreement. No further reproductions authorized.
2 MATERIALS PERFORMANCE AND CHARACTERIZATION
system distortion sometimes is not repeatable, especially when local film boiling occurs between
gear teeth. Computer simulation shows that double distortion (twice as much distortion) can occur
when local film boiling takes place between two adjacent teeth. Several measures have been
addressed to decrease distortion variation, namely to optimize the concentration of brine solution.
Investigations of critical heat flux densities and local film boiling can result in decreasing distortion
of steel parts during quenching. Unfortunately, in heat treating industry there are no data on criti-
cal heat flux densities; no standard on their correct measurement and testing quenchants with spe-
cial additives. Computer simulation showed that such data should be available for engineers to
optimize distortion of steel parts during quenching.
IQ-2 Technological Process

An IQ-2 technique (a two or three-step quenching process that initially cools parts under the nu-
cleate boiling mode of heat transfer and then by convection) is often used for quenching of gears.
The first step of the IQ-2 process involves intensive cooling without film boiling until a superficial
layer of the part being quenched contains 50 % martensite. At this point, the intensive cooling pro-
cess is interrupted; steel parts are removed from the quench and cooled in the air. During this pe-
riod of time, the temperature is equalized throughout the part cross-sections and self-tempering of
the newly formed martensite in the superficial layer occurs. Then the parts are moved back to the
quench, and intensive cooling continues until the martensite transformation is completed, or until
parts are cooled completely to room temperature in the air. However, it is very difficult to eliminate
a local film boiling, which can take place between teeth during batch quenching. Usually, local film
boiling is observed when initial heat flux density is equal to the first critical heat flux density qcr1.
The IQ-2 quench method is suitable for a variety of steel products, including automotive and off-
highway equipment components (gears, coil springs, kingpins, torsion bars, bearing products, ball
studs), fasteners of different types, and tool products (punches, dies, die components). For intensive
quenching of steel parts, different IQ-2 systems were designed and built (see Figs. 1 and 2).
Akron Steel Treating (AST) company has one production 6000-gal IQ water tank that is
equipped with an atmosphere Surface Combustion furnace. AST has fully automated the IQ water
tank and modified the load lifting mechanism to reduce the time required for loading=unloading
FIG. 1—Layout of production IQ system installed at Summit Heat Treating Co. (AST) [2].
KOBASKO ET AL. ON DISTORTION DURING BATCH QUENCHING 3
FIG. 2—Production IQ system installed at Euclid Heat Treating Co. [2].
the parts during intensive quench. AST also developed a standard procedure for controlling the
concentration of the sodium nitrite in the IQ water tank, water temperature, and rate of the quen-
chant agitation. A full automation of the IQ water tank and a precise control of the quench param-
eters allow intensive quenching of a variety of steel products in the production IQ system.
AST runs several jobs from three its customers on a continuous basis. These jobs are the
following:
• Gear blanks made of alloy 4140 and 4340 steel for manufacturing rack and pinion gears
• Shafts made of carburized steel
• Stampings made of carburized steel
• Punches made of AISI S5 steel
• Dies made of AISI H13 steel
• Chisels made of AISI 1045 and 4340 steels
• Kingpins, etc.
Euclid Heat Treating Co. bought its first IQ production unit from AFC-Holcroft, Inc. in 2003. It
is a three-chamber, integral quench, batch-type furnace with working dimensions of 91 cm 91
cm 183 cm (36in. 36 in. 72 in.) for implementing the batch IQ process. The new system is
used now for:
• Quenching of big gears
• Different plates made of alloy and plain carbon steels
• Dies made of AISI H13 steel
• Chisels made of AISI 1045 and 4340 steels
• Big forgings, etc.
In many cases the shortcoming of this system is distortion of steel parts during batch quenching.
That is why the goal of our investigation is to find out how local film boiling impacts distortion of
spur gears. Prior to making calculations let us consider the importance of critical heat flux density
values.
FIG. 3—Cylindrical steel parts prepared for batch quenching using IQ-2 technology.
Importance of Critical Heat Flux Densities

The critical heat flux densities qcr1 and qcr2 are inherent properties of any vaporizable quenchant.
The first critical heat flux density (qcr1 ) is the maximum amount of thermal energy coming out of a
unit of surface area needed to create film boiling in the given liquid over a hot surface area. The
more resistant a liquid is to boiling when heat is applied, the higher the liquid’s qcr1 is [2]. The
more resistant a quenchant is to boiling, the more likely it is to quench a part uniformly (with no
film boiling), thus yielding less distortion. Also, with greater resistance to boiling, there is less likeli-
hood of a “slack quench”—a quench that is sufficiently slow to produce spotty or lower than opti-
mum as-quenched hardness for a given steel alloy. On the other hand, the second critical heat flux
density (qcr2 ) is the minimum amount of heat energy necessary to support film boiling over the
given surface area. A scheme of heat transfer modes during quenching of steel parts is shown in
Ref [2].
The following experimental correlation between qcr2 and qcr1 is used [2]
qcr2
0:2 (1)
qcr1
To predict modes of heat transfer, the first critical heat flux density should be compared with the
initial heat flux ðqin Þ at the moment of immersion of steel part into quenchant. The initial heat flux
density ðqin Þ means the heat flux that appears during immersion of steel parts into quenchant. It
depends on configuration and size of steel part, thermal properties of the material, and
TABLE 1—Pool film boiling of water on spheres [4].
Sphere Material Diameter, mm Water Subcooling, DTsub HTC, W=m2K Number of Observation
Stainless steel 19 0 243–252 4
Stainless steel 25.4 0 223–249 7
Stainless steel 25.4 30 417–440 5
Silver 19 0 230–260 6
Silver 19 30 436–466 5
TABLE 2—Forced flow film boiling of water on 19 mm diameter AISI 304 stainless steel sphere [4].
Water Subcooling Water Flow Velocity, HTCmin, Minimum Film-Boiling Number of

DTsub U (m=s) W=m2K Temperature, ðDT Þmin Observation
0 0.02 233–261 114–120 3
0 0.45 460–480 126–127 2
30 0.02 440–457 334–344 2
30 0.45 940–960 328–346 3
50 0.02 540 469–474 2
50 0.45 1070–1100 469–484 4
austenitizing temperature. It could have any of three variants of such comparison: qin > qcr1 ;
qin < qcr1 ; qin qcr1 .
When qin > qcr1 , full film boiling is established around the steel part (for example gear) and
smooth cooling is provided in this case. Distortion of steel part (gear) will be minimum due to
smooth cooling, however surface hardness and depth of hardened layer will be minimum too
because of very slow cooling during smooth film boiling. Only high alloy steels can be quenched
properly by using this method of cooling.
When qin < qcr1 , full film boiling is absent around the steel part (for example gear) and smooth
cooling is provided by transient nucleate boiling process. Distortion of steel part (gear) will be simi-
lar to oil quenching, however surface hardness and depth of hardened layer will be maximum
because of a very high cooling rate during nucleate boiling. In this case high alloy steels can be sub-
stituted with lower alloy steel or plain carbon steel. Size of steel part after intensive quenching
slightly changes due to more complete martensite transformation.
If qin qcr1 , local film boiling is observed on the surface of steel parts. During batch quenching
of cylindrical steel parts (for example chisels), local film boiling can be inside the load and leads to
great distortion of the chisels. To eliminate this effect, heat treaters try to separate chisels far from
each other to provide enough room between chisels that decreases the probability of local film boil-
ing. Such a load, prepared for batch quenching, is shown in Fig. 3.
Dhir and Purohit developed the theory and accurate method for measuring experimentally pool
film boiling of water on spheres [4]. Results of their investigations are provided in Table 1 and
Table 2. It is well known that during quenching of plates film boiling is more stable and the heat
transfer coefficient (HTC) has less value as compared with the quenching of spheres.
On the basis of experimental results obtained, it makes sense to consider HTC during local film
boiling as 800 W=m2K for poor cooling of teeth, which is the reason for large distortion of gears
during batch quenching. Other areas of the gears are cooled intensively where HTC of 25 000
W=m2K is provided. These values are taken into account based on experimental data and CFD
modeling [2,4].
Impact of Local Film Boiling on Distortion and Residual Stresses After

Batch Quenching
Impact of local film boiling on distortion and residual stresses after batch quenching was investi-
gated by DANTE software, which allows receiving also metallurgical data like amount of martensite,
bainite, etc. For computer FEM modeling a gear, shown in Fig. 4, was used. The maximum outer
diameter of the spur gear shown in Fig. 4 is 63.5 mm. The gear has 28 teeth and has a common
configuration that is often used in machine building.
FIG. 4—A spur gear made of AISI 8620 steel.
FIG. 5—Carbon content in spur gear after vacuum carburization (only flank faces and root were carburized).
During the batch quenching, many gears are located near each other and quenched in agitated
cold water or water-salt solution of optimal concentration. As a result of the proximity of gears,
local film boiling as a rule is observed between teeth that are directed perpendicular to the liquid
stream. Agitation of water increases the first critical heat flux density, however it cannot be
increased enough and can be equal to qcr1, which creates conditions for local film boiling. Small
amount of additives, dissolved in the quenchant, can eliminate film boiling and decrease distortion.
Only functional surfaces were carburized, i.e., flank faces and root.
It was established by experiment that distortion depends on thickness of the carburized layer.
This problem will be considered in detail by the authors later when more experimental data are col-
lected and more computer simulations are done. Thickness of the carburized layer and content of
carbon in it affect distortion because of changing specific volume of martensite and current stress
distribution. As one can see from Fig. 5, the content of carbon in the carburized layer was within
0.6 %–0.73 % and in other areas content of carbon was 0.2 %.
As is known, the martensite start temperature MS depends on content of carbon in steel. The
higher the carbon in steel is, the lesser is the martensite start temperature MS. For example, for
AISI 8620 steel the martensite start temperature is about 405 C (760 F), see Fig. 6. For AISI 8660
steel the martensite start temperature drops to 260 C (500 F), see Fig. 7. For high carbon steels the
martensite start temperature could be equal to the boiling point of liquid.
For a given content distribution in the gear (see Fig. 5), computer simulation was performed
with existing poor cooling between some teeth. The red areas of the gear (see Fig. 8) show poor
cooling caused by local film boiling, which differs from transient nucleate boiling by almost 31
times.
Each red area in Fig. 5 shows poor cooling where during local film boiling heat transfer coeffi-
cient (HTC) is 800 Wm2K1 (see Table 1 and 2). Other areas of gear are cooled intensively where
average effective HTC is equal to 25 000 Wm2K1. Average effective HTC is significantly less as
compared with real HTC because heat flux density is divided by Tsf Tm.
Here Tsf Tm Tsf TS; TS is saturation temperature (boiling point) [6]. The real HTC is cal-
culated as q=(Tsf TS) [6].
For calculating distortion and residual stresses a sector of a gear was considered (see Fig. 9). At
each time and space step, the calculation results were compared with the continuous cooling
FIG. 6—Isothermal transformation diagram for AISI 8620 steel austenitized at 900 C (1650 F). Composition: 0.18 C; 0.79
Mn; 0.52 Ni; 0.56 Cr; 0.19 Mo. Grain size 9 to 10 [5].
FIG. 7—Isothermal transformation diagram for AISI 8660 steel austenitized at 845 C (1555 F).Composition: 0.59 C; 0.89 Mn;
0.53 Ni; 0.64 Cr; 0.22 Mo. Grain size 8 [5].
transformation (CCT) diagram of the supercooled austenite to choose appropriate thermal and
mechanical properties depending on temperature. For boundary conditions average effective HTC
of 25 000 W=m2K for boiling process and 800 W=m2K for local film boiling process were used.
The process of calculation was within the austenitizing temperature and room temperature.
Such an approach can be explained by the following. After a certain intensity of quench heat
extraction, the thick part of a gear cannot give up its heat any faster than the rate of heat conduc-
tion through the part. This is why one cannot quench “too fast” during the intensive portion of the
quench. Once the part surface layer has reached the temperature of the quenchant, conduction
within the part sets a natural limit on the rate of cooling in the subsurface layers and the core of
the part. Because conduction is also a very rapid and very uniform form of heat removal, intensive
quenching is able to reach the ultimate goal of any quench—the most uniformly rapid removal of
heat that yields the less part distortion.
FIG. 8—Red areas of gear show poor cooling caused by local film boiling.
FIG. 9—The area of a gear for meshing and FEM calculations the stresses and distortion.
When quenching large gears in agitated water or water-salt solutions, convection can prevail.
The criterion, which provides direct convection, is [2,3]
2ð#0 #I Þ
Bi ¼ (2)
#I þ #uh
The value #I is calculated from Eq 3

1 2kð#0 #I Þ 0:3
#I ¼ (3)
b R
75k0 ðq0 q00 Þ0:5 g 0:5

b¼ (4)
00 Þ0:7 Pr0:2
r0:5 ðq00 r W
where:
k is thermal conductivity of steel (W=mK)
R is radius or half thickness of plate shaped steel parts
#0 ¼ T 0 T S ;
#I ¼ T I T S ;
#uh ¼ TS Tm ;
T0 is initial temperature,
TI is initial temperature of the surface at the beginning of nucleate boiling,
Tm is temperature of a quenchant (liquid),
Pr ¼ a is Prandtl criterion (dimensionless number),
is kinematic viscosity (m2=s),
k0 is thermal conductivity of the liquid (W=mK),
r is surface tension (N=m),
g is acceleration due to the gravitational force (9.8 m=s2),
q0 is liquid density (kg=m3),
q00 is vapor density (kg=m3),

r* is latent heat of evaporation (J=kg), and
W 00 is vapor bubble growth rate (m=s).
As one can see from Fig. 10, a more complete martensite transformation occurs in non-
carburized AISI 8620 steel, where 99 % martensite is formed. In the carburized layer, where content
of carbon is 0.60 %–0.73 %, only 90 %–96 % of martensite is formed.
As a result of local film boiling and a low HTC, the transformation of austenite to martensite is
delayed in the area where film boiling takes place. The transformation starts first on the left side of
tooth 1 and on the right side of tooth 2, as shown in Fig. 10. Because a specific volume of the mar-
tensite is greater than that of austenite by 4 %, the surface layers during quenching expand, causing
the movement of tooth 1 to the right and tooth 2 to the left. This tooth movement generates resid-
ual compressive stresses at the root area and near it (see Fig. 11) resulting in double distortion
between them. Note that in Fig. 11, S22 means axial residual stresses. The hoop stresses are desig-
nated as S33, and radial as S11 [2].
The distribution of minimum and maximum principal stresses in spur gear after batch quench-
ing and double distortion are presented in Fig. 12 and 13. Similar to S22 stresses, the larger princi-
pal compressive residual stresses are observed between tooth 1 and tooth 2 (see Fig. 12 and 13).
In Fig. 14 the total displacement between two teeth where local film boiling took place is
presented.
On the basis of investigations concerning the effect of local film boiling on distortion of gears,
our main recommendations are as follows:
To introduce widely an intensive quenching IQ-2 technology into the heat-treating practice, first
of all, special additives should be developed to increase the first critical heat flux density that can
prevent local film boiling effectively and thus decrease distortion of steel parts.
It is important to apply IQ-2 processes in captive heat treating shops where mass production
exists that can be completely automated and carefully controlled.
FIG. 10—The martensite volume fraction in the spur gear in carburized and non-carburized areas after batch quenching. The
local film boiling takes place between tooth 1 and tooth 2.
FIG. 11—Distribution of S22 stress in spur gear after batch quenching.
Special attention should be paid to designing and manufacturing of IQ equipment that mini-
mizes or completely eliminates local film boiling.
Discussion
As a rule, when performing computer simulation, investigators do not pay attention to local film
boiling and take into account only the value of HTC or heat flux, which are characteristics of cool-
ing intensity of the process. Nobody paid serious attention to local film boiling because it is
FIG. 12—Distribution of max principal stresses in spur gear after batch quenching.
FIG. 13—Distribution of minimum principal stresses in spur gear after batch quenching.
connected with critical heat flux densities, which are unknown. At the same time, local film boiling
generates large distortions because there are big differences between HTCs on the boundary line of
local film boiling–transient nucleate boiling process. As known, HTCs can differ by an order of 10.
The transient nucleate boiling process provides immediate martensite transformation and local
film boiling delays martensite transformation. As a result, on the boundary line of local film
boiling–nucleate boiling big distortions occur due to larger specific volume of martensite (4%).
From the computer simulation it follows that distortion is minimum when local film boiling is
absent. One can find new ways of decreasing distortion of gears and different kinds of steel parts in
FIG. 14—Total displacement between tooth 1 and tooth 2 where film boiling took place.
TABLE 3—Time required for the surface of steel spheres of different sizes to cool to different temperatures when quenched
from 875 C (1,605 F) in 5 % NaOH-water solution at 20 C and moving at 3 ft/s (0.914 m/s), according to French [7].
Time, s

Diameter, mm 700 C 600 C 500 C 400 C 300 C 250 C 200 C 150 C
6.35 0.025 0.030 0.033 0.040 0.06 0.10 0.21 1.05

0.025 0.040 0.050 0.063 0.12 0.23 0.42 0.67
0.030 0.040 0.043 0.050 0.09 0.13 0.23 0.36
0.027 0.037 0.043 0.051 0.09 0.15 0.29 0.69
12.7 0.033 0.040 0.050 0.053 0.07 0.11 0.15 0.43
0.035 0.038 0.046 0.060 0.09 0.13 0.22 0.49
0.032 0.050 0.073 0.090 0.11 0.14 0.32 0.92
0.016 0.043 0.050 0.083 0.17 0.24 0.35 0.65
0.020 0.040 0.060 0.077 0.10 0.15 0.26 0.53
0.028 0.042 0.058 0.071 0.11 0.15 0.26 0.60
25.4 0.035 0.040 0.045 0.060 0.08 0.10 0.15 0.40
0.050 0.050 0.080 0.083 0.11 0.19 0.40 1.20
0.028 0.040 0.045 0.064 0.14 0.21 0.34 0.71
0.020 0.020 0.050 0.086 0.19 0.32 0.32 0.99
0.033 0.042 0.055 0.074 0.13 0.21 0.35 0.82
general from the results of the computer simulation. When performing intensive quenching, three
methods should be taken into account:
1. Use small amounts of additives, which destroy completely local film boiling.
2. Use an optimal concentration of water-salt solution, which provides maximal critical heat flux
density.
3. Increase the boiling point of a liquid to delay martensite transformation during nucleate boiling
process.
These measures could decrease distortion of steel parts during batch quenching.
According to the IQ-2 process, full film boiling and local film boiling should be absent com-
pletely. In this case surface temperature of a steel part drops very rapidly to the boiling point of liq-
uid quenchant, slightly exceeding it, and maintains at this level for a rather long time [7,8].
French discovered that surface temperature within the austenitizing temperature and boiling
point for different sizes of probes is almost the same (see Table 3).
Kobasko [3] established that surface temperature during quenching of steels can be adjusted by
pressure or concentration of salts, which increase the boiling point of liquid (see Fig. 15).
This means that the transformation from austenite to martensite during nucleate boiling process
can be delayed, especially when quenching high carbon steels.
From Fig. 6, Fig. 7, and Fig. 15 it follows that during quenching of carburized steel parts mar-
tensite transformation begins in the inner layers first, where content of carbon is low and martens-
ite start temperature is high. This situation can change the residual stress distribution in quenched
steel parts. In order to take this fact into account, real heat transfer coefficients should be used dur-
ing computer simulations [7,8]. At present time, a global database is being developed that takes
into account effective heat transfer coefficients (HTC) and real HTC. Effective HTC generate less
quenching distortion. In spite of that, it is shown that local film boiling is a major disaster for dis-
tortion after batch quenching.
FIG. 15—CCT diagram for AISI W1 steel (a) and adjustment of surface temperature of a cylindrical specimen 20 mm in diam-
eter made of stainless steel (AISI 304) by changing pressure (b) [2]. i is normal pressue 0.1 MPa; ii is pressure 0.4 MPa; iii is
pressure 0.7 MPa.
The real HTC depends on the overheat of a boundary layer, DT ¼ TSf TS and does not
depend on the underheat, DT ¼ TSf Tm . The point is that the formation of nucleating centers
depends on overheating of a boundary layer determined by the equation
2rTS
Rcr ffi ; (5)
r q00 DT
where:
Rcr is a critical size of a bubble that can grow and function,
r is surface tension (N=m),
TS is saturation temperature (R),
r is latent heat of vapor formation (J=kg),
q00 is vapor density (kg=m3), and
DT ¼ TSf TS .
Active nucleating centers are the basic carriers of heat that remove heat from a surface and transfer
it to a cold bath.
The effective HTC is evaluated as
q
h¼ (6)
Tsf Tm
and real HTC as

q
h¼ : (7)
Tsf TS
The real HTC could be up to ten times larger as compared with the effective HTC. Historically, the
effective HTC are widely used in the heat treating industry for recipe development. It has been
shown by Kobasko [7,8] that effective HTC can be used successfully for cooling time and cooling
rate evaluation at the core of steel parts. During computer simulation the effective HTC generates
less distortion due to less gradient of the temperature at the surface of steel parts during quenching.
The next step in our investigation is a more precise calculation of gear distortion by using real and
effective HTC [8–10].
Summary
1. Local film boiling occurs when qin qcr1 and it is the reason for significant distortion of steel
parts, especially during batch quenching with many steel parts loaded.
2. The double distortion during quenching is observed when local film boiling takes place between
two teeth. The left-hand tooth moves to the right and the right-hand tooth moves to the left.
3. There is a need to develop special additives that can increase the first critical heat flux density to
eliminate local film boiling.
4. In the paper the role of local film boiling on changing the displacement of teeth on a spur gear is
shown. It is established that significant distortion of gear teeth takes place where local film boiling
exists.
5. For computer simulation average HTC were used (for local film boiling 800 Wm2K1 and for
nucleate boiling process effective HTC 25j000 Wm2K1). When using real HTC, distortion of
gears could be even greater.
6. A database for cooling capacity of quenchants is needed, which will help engineers to minimize
distortion of steel parts during quenching.
References
[1] Kobasko, N. I., Aronov, M. A., Powell, J. A., Ferguson, B. L., and Dobryvechir, V. V., “Critical
Heat Flux Densities and Their Impact on Distortion of Steel Parts During Quenching,” New
Aspects of Fluid Mechanics, Heat Transfer and Environment, N. Mastorakis, V. Mladenov, and
Z. Bojkovic (Eds.), WSEAS Press, Athens, 2010, pp. 338–344.
[2] Kobasko, N. I., Aronov, M. A., Powell, J. A., and Totten, G. E., Intensive Quenching Systems:
Engineering and Design, ASTM International, West Conshohocken, PA, 2010.
[3] Kobasko, N. I., “Intensive Steel Quenching Methods,” Quenching Theory and Technology, Sec-
ond Edition, B. Liščić, H. M. Tensi, L. C. F. Canale, and G. E. Totten (Eds.), CRC Press, Boca
Raton, FL, 2010, pp. 509–568.
[4] Dhir, V. K. and Purohit, G. P., “Subcooled Film Boiling Heat Transfer From Spheres,” Nuclear
Engineering and Design, North-Holland Publishing Company, Amsterdam, 1978, pp. 49–66.
[5] Heat Treater’s Guide, Practices and Procedures for Iron and Steels, 2nd ed., H. Chandler (Ed.),
ASM International, Materials Park, OH, 1995.
[6] Kobasko, N. I., Steel Quenching in Liquid Media Under Pressure, Naukova Dumka, Kiev,
Ukraine, 1980, pp. 1–206.
[7] French, H. J., The Quenching of Steels, American Society for Steel Treatment, Cleveland, Ohio,
1930.
[8] Kobasko, N. I., “Discussion of the Problem on Designing the Global Database for Different
Kinds of Quenchants,” Recent Advances in Fluid Mechanics, Heat & Mass Transfer and Biol-
ogy, A. Zemlliak and N. Mastorakis (Eds.), WSEAS Press, Athens, 2011, pp. 117–125.
[9] Liščić, B. and Filetin, T., “Global Database of Cooling Intensities of Liquid Quenchants,” Pro-
ceedings of the European Conference on Heat Treatment 2011, “Quality in Heat Treatment”,
Wels, Austria, 2011, pp. 40–49.
[10] Quenching Theory and Technology, 2nd ed., B. Liščić, H. M. Tensi, L.C.F. Canale, and G. E.
Totten (Eds.), CRC Press, Boca Raton, FL, 2010.
View publication stats

Lynn Gear

Uploaded by

Copyright:

Available Formats

Lynn Gear

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lynn Gear

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Article · January 2012

Nikolai I. Kobasko Michael A. Aronov

SEE PROFILE SEE PROFILE

Lynn Ferguson Zhichao Li

SEE PROFILE SEE PROFILE

Agueous Quenchants View project

The user has requested enhancement of the downloaded file.

N. I. Kobasko,1 M. A. Aronov,2 B. L. Ferguson,3 and Z. Li3

Local Film Boiling and Its Impact on

IQ-2 Technological Process

FIG. 2—Production IQ system installed at Euclid Heat Treating Co. [2].

Importance of Critical Heat Flux Densities

TABLE 1—Pool ﬁlm boiling of water on spheres [4].

Water Subcooling Water Flow Velocity, HTCmin, Minimum Film-Boiling Number of

Impact of Local Film Boiling on Distortion and Residual Stresses After

FIG. 4—A spur gear made of AISI 8620 steel.

The value #I is calculated from Eq 3

75k0 ðq0 q00 Þ0:5 g 0:5

q00 is vapor density (kg=m3),

FIG. 11—Distribution of S22 stress in spur gear after batch quenching.

6.35 0.025 0.030 0.033 0.040 0.06 0.10 0.21 1.05

and real HTC as

You might also like