Mathematical Modeling and

M. C. Leu
Experimental Verification of
P. Meng Stationary Waterjet Cleaning
E. S. Geskin Process
L. Tismeneskiy The use of stationary waterjet for the removal of coating material from the substrate
is investigated analytically and experimentally. In the analysis, the cleaning width
Department of Mechanical Engineering,
as a function of standoff distance, water pressure, and nozzle radius is derived by
New Jersey Institute of Technology, considering the structure of waterjet and the cleaning mechanism. Also derived are
Newark, New Jersey 07102 the relations of the optimal cleaning standoff distance and maximum cleaning width
to the critical cleaning standoff distance, and how the water pressure and nozzle
radius affect this critical standoff distance. These derived analytical relations are
verified with experimental results.

1 Introduction with the standoff distance until the maximum cleaning rate is
reached at a certain standoff distance, after that the cleaning
Pure waterjets have been used for removal of various coatings
rate reduces with increase in the standoff distance (Galecki and
or deposits from the substrates as well as cutting of many mate-
Vickers, 1982; Xu and Summers, 1994; Geskin et al., 1995;
rials. Waterjet cutting normally involves the penetration of a
Wu and Kim, 1995). The optimal standoff distance is related
solid by a continuous jet. Waterjet cleaning, on the other hand,
to the coating and substrate materials and the operating parame-
is often achieved by an erosion process involving discrete drop-
ters such as nozzle (orifice) size, water pressure, jet travel
lets. While the past investigations on waterjet cutting included
speed, etc. Although useful knowledge has been generated from
both analyses and experiments (e.g., Hashish and Duplessis,
the various experimental studies in the past, there exists no
1978 and 1979; Chao et al., 1995), present knowledge on wa-
theoretical model that is capable of explaining the observed
terjet cleaning has come mainly from experiments (Erdmann-
experimental results.
Jesnitzer et al., 1978 and 1980; Labus, 1982; Louis and Schi-
korr, 1982; Haferkamp et al., 1984; Blickwedel et al., 1988; This paper describes an analytical and experimental study of
Conn, 1992; Singh et al., 1992; Leu et al., 1994; Geskin et al, cleaning by stationary waterjets. A mathematical equation is
1995; Meng et al., 1996). Because of the lack of mathematical developed to express the cleaning width as a function of standoff
models, the present understanding of parameter effects in wa- distance, water pressure, nozzle radius and other system param-
terjet cleaning and hence the ability to optimize the cleaning eters based on the waterjet structure and a material erosion
process is quite limited. model. In the erosion model, removal of material occurs when
the equivalent dynamic stress generated by the water droplet
The continuous water flow gradually becomes a stream of flow exceeds the endurance limit of the coating material under
water droplets after it comes out of an nozzle due to a continuing fluctuating stress. The maximum cleaning width is shown to
interaction between the water and its surrounding air. The water exist at a certain standoff distance which has a certain ratio
droplets impinging on the target coating create impact forces. with the critical cleaning standoff distance. The mathematical
The mechanism of material removal is highly complex. Adler relations derived are verified experimentally.
(1979) reported that material erosion by the droplets consists
of four damage modes. These are direct deformation, stress
wave propagation, lateral outflow jetting, and hydraulic penetra- 2 Structure of Waterjet
tion. One or more damage modes may exist in a particular It is generally recognized that there exist three waterjet re-
erosion process. The first two modes are responsible for initia- gions: the initial, main, and final regions, as illustrated in Fig.
tion of cracks. The effect of direct deformation is not obvious 1. In the initial region, which is close to the nozzle exit, the
if there exist few low velocity impacts. The propagation of stress instability of the tangential surface separation in the continuous
waves caused by the impact forces is generally responsible for flow stream causes eddies. The eddies bring about an exchange
crack initiation in the erosion process, as reported by many of matter between the water and air. The surrounding air me-
researchers (Springer, 1976; Louis and Schikorr, 1982; Ramulu dium is entrained into the water stream and breaks up the contin-
et al., 1991; Li et al., 1992; Kang et al., 1993; Watson, 1993). uous water stream into water particles due to an intensive trans-
Lateral outflow jetting and hydraulic penetration cause the ex- verse transfer of mass, momentum, heat, and constituents. Inside
tension, enlargement and propagation of existing cracks. In the the jet there is a wedge-like region known as the potential core,
erosion of coating material, the adhesion between the coating which is surrounded by a mixing layer. The velocity inside the
and the substrate may also need to be considered. core is equal to the jet velocity at the nozzle exit. The waterjet
Experimental observations have shown that there exists an in the initial region is considered to be a continuing flow having
optimal standoff distance at which the volume of material re- very little air inside the jet. At the end of this region, the effect
moval is the greatest (Louis and Schikorr, 1982; Kang et al., of air dynamics and continuous interaction of the water with
1993). It has also been shown that the rate of cleaning increases the surrounding air medium results in the breakup of the waterjet
stream into droplets. This begins the main region. In this region,
the mixing of the waterjet stream with air medium continues to
Contributed by the Manufacturing Engineering Division for publication in the the full extent, and the jet stream is disintegrated into droplets
JOURNAL OF MANUFACTURING SCIENCE AND ENGINEERING. Manuscript received
Feb. 1996; revised Jan. 1997. Associate Technical Editor: K. Rajurkar.
continuously due to the entrained air. The smaller the distance

Journal of Manufacturing Science and Engineering AUGUST 1998, Vol. 120 / 571
Copyright © 1998 by ASME

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Initial
Region

Final Region

Potential Core

Water Droplet Zone

Fig. 1 Structure of waterjet in air

to the center line of the waterjet stream, the bigger the water
droplet size, and the more concentrated the droplet flow. This
results in a gradual expansion of the cross section and reduction
in the velocity and pressure of the waterjet. Between the droplet
zone and the surrounding air, there is a mist zone consisting of
very fine droplets. The droplets at the boundary of the droplet
zone and the mist zone can be considered to have zero velocity.
The final region is a diffusion region in which the waterjet is
totally broken up into small droplets. Figure 2 provides a photo
showing a typical waterjet. Detailed discussions on the structure
of waterjet in air can be found in (Yanaida and Ohashi, 1978;
Yanaida and Ohashi, 1980; Yanaida and Ohashi, 1982; Zou et
al., 1985).
From the investigation of Yanaida and Ohashi (1978) and
Zou et al. (1985), the radius of the jet in the droplet zone, R,
relates to the distance from the nozzle exit, x, as follows (refer
to Fig. 3 for the parameters):
R = Cx (1)
where C is the spreading coefficient. Its value was experimen-
tally observed by Yanaida and Ohashi (1978) to be about Fig. 2 Photo of typical waterjet in air
0.03 in the main region and increased to about 0.06 in the
diffusion region. Although C may be a function of water
pressure and nozzle radius, this dependence relationship is
recognized to be complex and it is not available from the
= (l-£''5)3 (2)
literature. We will assume C to be independent of water pres-
sure and nozzle radius in the latter part of our mathematical
derivation, and will show that despite this simplification, the where m is the mass flow rate of water droplets per unit area
numerical results from our analytical model agree well with at some point of consideration in the flow field, mm is the mass
experimental results. flow rate of water droplets per unit area at the center of the
According to Erastov's experiment (Abramovich, 1963), the same cross section, and £ is a dimensionless parameter defined
mass flow rate in a waterjet has the following relationship by

Nomenclature
C = jet spreading coefficient r0 = radius of nozzle w„, = maximum cleaning width
k = flow resistance coefficient of water R = radius of the jet cross section x = standoff distance
system S = endurance limit of the coating ma- xc = critical cleaning standoff distance
m = water droplet flow rate at any point terial xm = optimal standoff distance
inside the jet U = velocity of waterjet p = density of water
mm = water droplet flow rate at a point U0 = velocity of waterjet at the nozzle \ = stress coefficient
on the jet center line exit £ = dimensionless parameter defined by
P = water pressure from the intensifier w = cleaning width £ = rlR
or pump wc = cleaning width at the critical stand- tjj - sound speed in water
r = distance of any point of consider- off distance
ation from the jet center line

572 / Vol. 120, AUGUST 1998 Transactions of the ASME

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Coating Workpiece

Impact Pressure Distribution ^

Nozzle y Mist Zone ^ _ \ _
:::::i::
^=== ^ *nr
Uo / ^

Droplet Zone /

Cleaning Width (w)

Standoff Distance

Fig. 3 Schematic of waterjet cleaning

the cleaning width is not as wide as the jet width, and an optimal
£= (3)
cleaning width exists in the main waterjet region. The variation
R
of cleaning width as a function of standoff distance is governed
where r is the distance of the point of consideration from the by two factors. One is the jet structure. As the waterjet propa-
jet center line (see Fig. 3). rh is equal to pU multiplied by the gates with continuing air entrainment, the jet width grows lin-
volumetric ratio of water in the water-air mixture, where p is early as the standoff distance increases. The other is the impact
the water density and U is the flow speed. pressure. The impact pressure generated by water droplets de-
During the waterjet spreading, the total mass flow rate in creases with increase in the standoff distance. There exists a
each cross section is equal to the total mass flow rate at the exit critical standoff distance at which the coating can not be re-
of the nozzle. Therefore, the following relation holds: moved at all, due to the impact pressure which has become too
small. The distribution of impact pressure has the shape shown
mairn = Z7r I mrdr = 2rrmm — rdr (4) in Fig. 3. The impact pressure is the strongest at the center of
Jo Jo m,„ each jet cross section and decreases to zero at the jet edge.
where m0 is the mass flow rate per unit area at the nozzle exit Beside standoff distance the impact pressure is also a function
and r0 is the radius of the nozzle. of water pressure and nozzle radius. Due to the jet structure
Substituting Eqs. (2) and (3) into Eq. (4) we obtain and impact pressure distribution, the maximum cleaning width
occurs somewhere between the nozzle exit and the critical clean-
ing standoff distance. At the critical standoff distance, the jet
mm = 5.62m0 (5) loses its capability to create an impact pressure high enough to
perform any cleaning at all (Meng et al, 1996).
When a liquid droplet hits a solid surface, the sudden deceler-
3 Analysis of Cleaning With Stationary Waterjets ation of the droplet generates an impact pressure which induces
Cleaning by stationary waterjets involves in coating material a stress wave at the liquid-solid interface. By making some
removal by a waterjet at a fixed nozzle position and orientation simple assumptions, Springer (1976) showed that when the
(without traveling). Experimental observations have shown that droplet velocity is perpendicular to the surface of impingement,

POLYMER ADDER

CLEANING
RWgON/OFF SURFACE
WVALVE
=^t£
BOOSTER PUMP

HYDRAULIC PUMP

CONTROL PANEL
H; O

CATCHER 'yywjvwv

Fig. 4 Schematic of experimental setup

Journal of Manufacturing Science and Engineering AUGUST 1998, Vol. 120 / 573

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 High Pressure Water Based on fatigue theorems established for bending and tor-
sion in solid mechanics, Thiruvengadam (1967) and Springer
(1976) investigated the repetitive impacts of multiple liquid
droplets on a solid surface. Material removal occurs due to
fatigue at a certain number of stress cycles when the equivalent
dynamic stress is between the ultimate strength and the endur-
ance limit of the material. Material removal does not occur at
all if the equivalent dynamic stress is smaller than the material's
endurance limit.
We apply the above general results to coating material re-
moval by water droplets. In stationary waterjet cleaning, the
number of stress cycles is theoretically infinite. Since the impact
stress is compressive, an appropriate equivalent dynamic stress
is the amplitude of the fluctuating stress, which is one half of
the impact stress (Shigley, 1977). Coating material is removed
when the equivalent dynamic stress is higher than or equal to
the endurance limit of the coating material. Mathematically,
this translates to
\mifj a S (6)
where S is the endurance limit of the coating material, m is the
mass flow rate of liquid droplets per unit area as before, and X
is a stress coefficient depending on the droplet size, coating
thickness, and properties of the liquid, coating, and substrate
material. Note that we have only considered failure of the coat-
ing material, and have not considered failure at the coating-
substrate interface because of the lack of knowledge on the
bonding strength.
Fig. 5 Schematic of cleaning head
3.1 Critical Cleaning Standoff Distance. Critical clean-
ing standoff distance is the shortest distance from nozzle exit
the stress on the surface is equal to the well-known water ham- at which the waterjet is unable to remove the coating material.
mer pressure pUifi, where p is the liquid density, U is the droplet Theoretically, cleaning at the critical cleaning standoff distance
velocity, and tp is the sound speed in liquid. Although the real happens at a single point (with zero cleaning radius) which is
droplet impact situation differs from his assumptions, this rela- the center of a certain cross section. Thus Xrnmip = S at the
tion provides a good characterization of the average impact critical standoff distance. By the use of Eqs. (1), (5) and (6)
stress. For the case of a droplet impinging on the surface of a together with the relationship m0 = pU0, it can be shown that
coating, he also showed that the impact stress on the coating the critical cleaning standoff distance, xc, relates to the cleaning
surface is equal to pUip multiplied by a coefficient which is a parameters as follows:
function of the droplet size, coating thickness, densities of the
liquid, coating and substrate, as well as speeds of sound in the (7)
liquid, coating and substrate. The value of this coefficient is 1 5.62Xpi//
when the coating and substrate have the same material. The waterjet velocity, U0, at the nozzle exit is related to

• Experimental, P=311 MPa

• Experimental, P = 2 4 2 MPa
Analytical
Experimental. P=1 72 MPa
11 - Analytical
• Experimental, P=1 0 4 MPa
10 Analytical
*^-——• —-\t
y
^s^7-*\^ s

*~~* H *"^*' ^-^ ~""~ "-v. - ^ v

E »-^**'
6 7 X * s
• \
*
x;
6 • * •••-, \ \ v
•o ' '"••-. V \ \
5 \ \ \
\. \ \
c 4 \ \ \ •
ID S
* \ '- \
O \\ \•? \ \
2
\ \ \
1
*
0 I * ' ,

0 1 0.2 0.3 0.4 0.5 0.6

Standoff Distance (m)

Fig. 6 Cleaning width vs. standoff distance for epoxy-based paint removal with sapphire
nozzle no. 14 at four different pressures

574 / Vol. 120, AUGUST 1998 Transactions of the ASME

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 • Experimental, P = 3 1 1 MPa

• Experimental, P = 242 MPa

Analytical
Experimental,P = 1 72 MPa
Analytical
• Experimental, P=1 0 4 MPa
9 • Analytical

^ 7
E

3 5

en
c
4 \ • \ \
1 3 \ \ V
o 2

0
0 1 0.2 0.3 0.4 0.5
Standoff Distance (m)

Fig. 7 Cleaning width vs. standoff distance for epoxy-baaed paint removal with sapphire
nozzle no. 12 at four different pressures

the water pressure, P, generated by the pump or intensifier as U0r20

follows: (10)
\2CxJ CV 5.62\pil>

Ua (8) where x is the standoff distance. The above equation can be

obtained by using Eqs. (1), (2), (5), and (6) and letting r =
where k is a coefficient which accounts for flow resistance in w/2. Substituting Eq. (7) into Eq. (10) results in
the waterjet system and is usually around 0.96 ~ 0.99. The (2/3)-
critical cleaning standoff distance can thus be expressed as w = 2Cx 1 - - (ID
25
—(?)"(?)<">•
xc = z . 8 / | — T[^)(Ppf- (9)
As a check, the cleaning width at the critical cleaning standoff
distance is 0, i.e. w - wc = 0, when x = xc, from the above
3.2 Cleaning Width vs. Standoff Distance. If the coat- equation.
ing surface is placed somewhere between the nozzle exit and The maximum cleaning width w = wm can be obtained by
the critical cleaning standoff distance, the cleaning width w can letting dwidx - 0. Thus, by differenting Eq. (11) with respect
be shown to satisfy the following equation to x and letting x = xm, we obtain

. Experimental, P = 3 1 1 MPa
Analytical
Experimental, P = 2 4 2 MPa
* Analytical
ft Experimental, P=1 72 MPa
Analytical
Experimental, P=1 04 MPa
11 * Analytical
10
9

7
6
5
4
3}
o 3h
2
1
0
0.1 0.2 0.3 0.4 0.5 0.6
Standoff Distance (m)

Fig. 8 Cleaning width vs. standoff distance for oil-based paint removal with sapphire
nozzle no. 14 at four different pressures

Journal of Manufacturing Science and Engineering AUGUST 1998, Vol. 120 / 575

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Experimental, P=31 1 MPo
Analytical
• Experimental, P = 2 4 2 MPa
Analytical
Experimental, P=1 72 MPa
. ,
* Analytical
. Experimental, P = l 04 MPa
M Analytical
- •
a-
s
^ - w ' V ~~~ ""~-< . ~'~ \

ing Widt h (mm

^ i

6 . •
\ *
5 - '•. \ \ \

\ '\ \
4 - , •
o
<D
3 \ \.
o 2
\\ \
\
1 \
0
0.1 0.2 0.3 0.4 0.5 0.6
Standoff Distance (m)

Fig. 9 Cleaning width vs. standoff distance for oil-based paint removal with sapphire
nozzle no. 12 at four different pressures

of two cleaning processes are the same, i.e. the two values of
= 0.576 (12) \rhm\\i are the same. By the use of Eqs. (7) and (8), we have

Substituting Eq. (12) back into Eq. (11) results in XifrVS, \2P2 'r2 02
(14)
CiXr
wm = 0.9l2Cxm = 0.525Cxc (13)
If the critical cleaning standoff distance xc is known, the optimal where Pi, P2, r 0] , r02, Ci, C2, A-i, X-2, xcl and xc2 represent
standoff distance xm can be calculated using Eq. (12), and the the water pressures, nozzle radii, spreading coefficients, stress
maximum cleaning width wm can be calculated from Eq. (13). coefficients, and critical standoff distances of the two cleaning
processes.
3.3 Effects of Water Pressure and Nozzle Radius on Strictly speaking, C and X. are functions of water pressure
Critical Cleaning Standoff Distance. To investigate the ef- and nozzle radius. Since these functions are highly complex
fects of water pressure and nozzle radius on the critical cleaning and unknown, we will assume them to be independent of P and
standoff distance, we start by noting that the maximum equiva- r in predicting the dependence of critical standoff distance on
lent dynamic stresses at the critical cleaning standoff distances water pressure and nozzle radius. We will show that the pre-

• Experimental Nozzle No.1 4

Analytic al
" Experimental Nozzle No.12
0.6 * Experimental Nozzle No,10
Analytical
» Experimental Nozzle No.7
Analytical
0.5

0.4

0.3

0.2

0.1
20 70 120 170 220 270 320
Water Pressure (MPa)

Fig. 10 Critical cleaning standoff distance vs. water pressure for epoxy-based paint
removal with four different sapphire nozzles

576 / Vol. 120, AUGUST 1998 Transactions of the ASME

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 • Experimental, Nozzle No.l 4

• Experimental, Nozzle No.l 2

1
Experimental, Nozzle No. 10
1? 0.6 Analytical
» Experimental, Nozzle No.7
<D - - ••• Analytical
U
c
D
% 0.5
b

~o
a 0,4 -
(/)
CD

'c
§ 0.3
o . - - - ' -v---'"""""
"5
u

5 0-2
20 70 120 170 220 270 320
Water Pressure (MPa)

Fig. 11 Critical cleaning standoff distance vs. water pressure for oil-based paint removal
with four different sapphire nozzles

dieted dependence relations agree fairly well with experimental X,:2 r

02
observations even with these assumptions. By assuming C and (17)
>oi
\ to be independent of P and r, Eq. (14) can be rewritten as
x
Therefore, we have shown that the critical standoff distance is
<:2 _ rp2 / P2
(15) linearly proportional to the nozzle radius and is proportional to
XcA l"tt\ \P\ the one-fourth power of water source pressure. It should be
noted that these relations are only "approximate" because of
With the consideration of the same nozzle radii, i.e. r0, = r02, the assumptions on C and \ .
Eq. (15) becomes
0.25
XC2
v

(16)
4 Experimental Verification
Xd Cleaning experiments are carried out with an Ingersoll-Rand
waterjet system in order to verify the derived analytical rela-
Also, with the same water pressures, i.e. P2 = P\, Eq. (15) tions. The cleaning head in the waterjet system is mounted on
becomes a 5-axis gantry robot. The movement is controlled by an Allen

• Experimental, P=311 MPa

^ 0-7
• Experimental, P = 242 MPa

- *
— —-—
Experimental, P = 1 72 MPa
Analytical
C
D » Experimental. P = 1 04 MPa
Analytical
.» 0.5 J/^^ " ^^
0

1 0.4
c
0

™ 0-3 -
en

§ 0.2
0

.1 °-1
S
0.0
1 4 6 8 10 12 14 16
Sapphire Nozzle Number

Fig. 12 Critical cleaning standoff distance vs. sapphire nozzle number for epoxy-based
paint removal at four different water pressures

Journal of Manufacturing Science and Engineering AUGUST 1998, Vol. 120 / 577

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 . Experimental, P = 311 MPa
Analytical
•? 0.7 - . Experimental, P = 2 4 2 MPa
Analytical
'--'
<D Experimental, P = 1 72 MPa
o 0.6 * Analytical
v Experimental, P=1 04 MPa ^s' ^
o
c **
Analytical
tn
O 0.5
. .^ ^i . - * • ' '

s * ^ ••' - ^

*o .^-'"*"
0.4
^ • ^ i . . - '
s^. ^ ."-'
CO ^ •• ^
en 0.3 s * ^ ^ .---" "
c ^ c \ ^ *• '
c s ^ ^ ^ >*
-"' -
lea 0.2
o -^ r >- ,-^ -^
D
(J 0.1 ^^>"^
^ ^
."t^
o 0.0 i 1 . I . I , I
, i

2 4 6 8 10 12 1 4 1 6
Sapphire Nozzle Number

Fig. 13 Critical cleaning standoff distance vs. sapphire nozzle number for oil-based paint
removal at four different water pressures

Bradley 8200 series CNC controller. The water is pressurized (11). Comparison between experimental and analytical results
by an intensifier using a hydraulically driven, double acting, on cleaning width vs. standoff distance with the same pressures
reciprocating plunger pump and then carried through a stainless and nozzles in the removal of oil-based paint is shown in Figs.
steel pipe to the cleaning head. Figure 4 shows the schematic 8 and 9. Figures 10 and 11 show the variation of critical cleaning
of the experimental setup. A flat sapphire nozzle is inserted into standoff distance vs. water pressure for the removal of these
a cleaning head for generating the waterjet. The cleaning head two coating materials with four different nozzles. The experi-
is shown in Fig. 5. mental data are compared with the analytical curves calculated
In all of the cleaning tests, the waterjet stream is perpendicu- using Eq. (16), again with the measured critical standoff dis-
lar to the surface of coating. A number of tries of the standoff tance with nozzle no. 14 at the water pressure of 311 MPa.
distance are made to find the critical cleaning standoff distance, Figures 12 and 13 show the experimental results of critical
above which the surface cannot be cleaned. After that, a series cleaning standoff distance vs. nozzle number for different pres-
of tests begins. The variation in cleaning width is obtained with sures in the two coating materials. Also shown are the analytical
systematic increase in the standoff distance. For each standoff curves calculated using Eq. (17) with the measurement of only
distance, the corresponding cleaning width is measured and one critical cleaning standoff distance (nozzle no. 14 and 311
recorded. There are two combinations of the coating and sub- MPa). In all of these figures, the analytical results agree fairly
strate: a yellow epoxy-based paint on a steel substrate and an well with the experimental results.
oil-based paint on a steel substrate. The oil-based paint made
by Paris Paint & Varnish is coated by hand brushing on the
surface of the AISI1018 steel substrate, while the epoxy-based 5 Conclusion
paint made by Krylon is coated by spraying the paint on the A mathematical model of cleaning using stationary waterjets
surface of the steel substrate. The measured thicknesses of the has been established by applying the theoretical structure of
oil-based and epoxy-based paints are 0.098 and 0.082 mm, re- waterjet in air and considering that cleaning occurs when the
spectively. The water pressures used in the experiments include equivalent dynamic stress due to water droplets is greater than
104, 172, 242, and 311 MPa (i.e. 15,000, 25,000, 35,000 and or equal to the endurance limit of the coating material. The
45,000 psi). Sapphire nozzle nos. 14, 12, 10 and 7 made by model relates the cleaning width to the standoff distance, water
Ingersoll-Rand are used, which correspond to the nozzle diame- pressure, and nozzle radius. Based on this model, the maximum
ters of 0.014, 0.012, 0.010 and 0.007 inches, respectively. The cleaning width is shown to be linearly proportional to the critical
cleaning width is obtained by measuring the diameter of the cleaning standoff distance, and optimal cleaning is shown to
cleaning spot using a caliper. Three measurements are taken occur at 0.576 times the critical standoff distance. By assuming
and the average of them is used as the measured cleaning width. that the spreading and stress coefficients (C and \ ) are indepen-
In same cases the cleaning spot looks a little like an ellipse due dent of water pressure and nozzle radius, the model also predicts
to nozzle imperfection and wornout. For such cases the mean that the critical cleaning standoff distance is linearly propor-
of the major axis length and minor axis length of the ellipse is tional to the nozzle radius and is proportional to the one-fourth
used as the cleaning width. power of the water pressure. The quantitative relations have
The experimental results on cleaning width vs. standoff dis- been verified by waterjet cleaning experiments with various
tance under various water pressures are shown in Figs. 6 and water pressures, nozzle radii, coating materials, and standoff
7, respectively, for nozzle nos. 14 and 12 in the removal of distances.
epoxy-based paint from the steel substrate. Also shown in these The derivation of equations presented in this paper is of
figures are plots of analytical curves using Eq. (11) and the practical importance to waterjet cleaning. Without the analytical
critical cleaning standoff distance measured with nozzle no. equations, it would be necessary to perform many tests in order
14 at the water pressure of 311 MPa. The waterjet spreading to obtain the optimal standoff distance for the maximum clean-
coefficient used in the calculation is 0.0335, which is obtained ing width for each cleaning situation (i.e. each set of water
from regression of the experimentally obtained data using Eq. pressure and nozzle radius). The derived mathematical model

578 / Vol. 120, AUGUST 1998 Transactions of the ASME

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 makes it possible to determine the optimal standoff distances Haferkamp, H„ Louis, H., and Schikorr, W., "Jet-Cleaning Investigation on
Polymeric Model Layers," Proceedings of 7th International Symposium on Jet
for various cleaning situations (i.e. different water pressures and Cutting Technology, BHRA, Cranfield, England, June 26-28, 1984, pp. 119-
nozzle radii) by measuring only one critical cleaning standoff 134.
distance for each combination of coating and substrate materials Hashish, M., and duPlessis, M. P., "Theoretical and Experimental Investigation
(with constant coating thickness). of Continuous Jet Penetration of Solid," ASME JOURNAL OF ENGINEERING FOR
INDUSTRY, Vol. 100, Feb. 1978, pp. 88-94.
We have only considered cleaning with stationary waterjets Hashish, M., and duPlessis, M. P., "Prediction Equations Relating High Veloc-
here. In most of real cleaning work, however, the waterjet is ity Jet Cutting Performance to Standoff Distance and Multipasses," ASME JOUR-
moving and not stationary. The effect of traveling would require NAL OF ENGINEERING FOR INDUSTRY, Vol. 101, Aug. 1979, pp. 311-318.
a modification of the mathematical model of stationary waterjet Kang, S-W., Reitter, T„ and Carlson, G„ "Target Response to the Impact of
High-Velocity Non-Abrasive Waterjet," Proceedings of 7th American Waterjet
cleaning derived in this paper. We will discuss cleaning with Conference, Seattle, Vol. 1, Aug. 28-31, 1993, pp. 71-86.
moving waterjets in a separate paper. Leu, M. C , Geskin, E. S., Meng, P., Tismenetskiy, L., and Uschitsky, M.,
"Waterjet In-Situ Cleaning," Report to IAB Meeting of the Emission Reduction
Research Center, The Pennsylvania State University, Nov. 29-30, 1994.
6 Acknowledgment Li, H. Y., Geskin, E. S., and Gordon, E. I., "Investigation of the Pure Waterjet-
Workpiece Interaction," Proceedings of 11th International Symposium on Jet
The research is supported by the Emission Reduction Re- Cutting Technology, BHRA, Dordrecht, Netherlands, July 1992, pp. 3-15.
search Center (an NSF Industry /University Cooperative Re- Louis, H., and Schikorr, W., "Fundamental Aspects in Cleaning with High
Speed Water Jets," Proceedings of 6th International Symposium on Jet Cutting
search Center) and by the Center for Manufacturing Systems Technology, BHRA, Guildford, England, Apr. 6 - 8 , 1982, pp. 217-228.
(a New Jersey Advanced Technology Center) at NJIT. We also Meng, P., Leu, M. C , Geskin, E. S„ and Tismenetskiy, L„ "Cleaning with
like to express our special thanks to Mr. Yu Quan for various High-Pressure Directed Waterjets," Proceedings of Japan-USA Symposium on
help in the drawing. Flexible Automation, Boston, MA, July 8-10 1996, pp. 1131-1138.
Ramulu, M., Yeh, H., Wong, K. P., and Raju, S. P., "Photoelastic Investigation
of Jet Piercing Process," Proceedings of 6th US Waterjet Conference, Houston,
TX, Aug. 24-27, 1991, pp. 1-16.
References Shigley, J. E., Mechanical Engineering Design, McGraw-Hill Co, 1977.
Abramovich, G. N„ The Theory of Turbulent Jets, MIT Press, Cambridge, MA, Singh, P. J., Munoz, J., and Chen, W. L., "Ultra-High Pressure Waterjet Re-
1963. moval of Thermal Spray Coatings," Proceedings of 11th International Symposium
Adler, W. F., ' 'The Mechanisms of Liquid Impact, Treatise on Materials Sci- on Jet Cutting Technology, BHRA, Dordrecht, Netherlands, July 1992, pp. 4 6 1 -
ence and Technology," Vol. 16, Erosion, Carolyn M. Preece, ed., Academic 480.
Press, 1979. Springer, G. S., Erosion by Liquid Impact, Scripta Publishing Co., Washington,
DC, 1976.
Chao, J. F., Zhou, G., Leu, M. C , and Geskin, E. S„ "Characteristics of Abra-
Thiruvengadam, A.,' 'The Concept of Erosion Strength," Erosion by Cavitation
sive Waterjet Generated Surfaces and Effects of Cutting Parameters and Structure
or Impingement, ASTM STP408, 1967.
Vibration," ASME JOURNAL OF ENGINEERING FOR INDUSTRY, Vol. 117, Nov.
Watson, J. D., "Thermal Spray Removal with Ultrahigh-Velocity Waterjets,"
1995, pp. 516-525.
Proceedings of 7th American Waterjet Conference, Seattle, WA, Aug. 28-31,
Conn, A. F., "Waterjet Cleaning for In-Factory Application," Proceedings of 1993, pp. 583-598.
11th International Symposium on Jet Cutting Technology, BHRA, Dordrecht,
Wu, S. S., and Kim, T. J., "An Application Study of Plain Waterjet Process
Netherlands, July 1992, pp. 443-449. for Coating Removal," Proceedings of 8th American Waterjet Conference, Hous-
Erdmann-Jesnitzer, F., Louis, H., and Wiedemeier, J., "Material Behavior, ton, TX, Vol. 2, Aug. 26-29, 1995, pp. 779-792.
Material Stressing, Principle Aspects in the Application of High Speed Water Xu, J., and Summers, D. A., "Experimental Evaluation of the Performance of
Jets," Proceedings of 4th International Symposium on Jet Cutting Technology, Fan Jet Systems," Proceedings of 12th International Symposium on Jet Cutting
BHRA, Canterbury, England, April 12-14, 1978, pp. 29-44. Technology, Rouen, France, Oct. 25-27, 1994, pp. 37-46.
Erdmann-Jesnitzer, F., Louis, H., and Wiedemeier, J., "The Action of High Yanaida, K., ' 'Flow Characteristics of Water Jets,'' Proceedings of 2nd Interna-
Speed Water Jets on Materials, Measurement Methods and Their Practical Appli- tional Symposium on Jet Cutting Technology, BHRA Fluid Engineering, Bedford,
cation: A Critical Review," Proceedings of 5th International Symposium on Jet England, 1974, Paper A2.
Cutting Technology, BHRA, Hanover, F. R. Germany, June 2 - 4 , 1980, pp. 7 5 - Yanaida, K., and Ohashi, A., "Flow Characteristics of Water Jets in Air,"
85. Proceedings of 4th International Symposium on Jet Cutting Technology, BHRA,
Galecki, G., and Vickers, G., "The Development of Ice Blasting for Surface Canterbury, England, vol. 1, April 12-14, 1978, pp. 39-53.
Cleaning," Proceedings of 6th International Symposium on Jet Cutting Technol- Yanaida, K., and Ohashi, A., "Flow Characteristics of Water Jets in Air,"
ogy, Guildford, England, April 6 - 8 , 1982, pp." 59-80. Proceedings of 5th International Symposium on Jet Cutting Technology, BHRA,
Geskin, E. S., Leu, M. C., Meng, P., Tismcnetskiy, L., and Uschitsky, M., Hanover, F. R. Germany, June 2 - 4 , 1980, pp. 33-44.
"Waterjet In-Situ Equipment Cleaning," Progress Report to the Emission Reduc- Zou, C-S., Dang, L., Duan, X., and Cheng, D. Z., "Investigation on Anatomy
tion Research Center, New Jersey Institute of Technology, Newark, NJ 07102, of Continuous Waterjet for Updating Jet Performance," Proceedings of 3rd Amer-
April 13, 1995. ican Waterjet Conference, Pittsburgh, May 21-23, 1985, pp. 160-178.

Journal of Manufacturing Science and Engineering AUGUST 1998, Vol. 120 / 579

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use