A benchmark experiment on gas cavitation
Received: 3 June 2014 / Revised: 1 July 2014 / Accepted: 4 July 2014 / Published online: 20 July 2014
Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Cavitation research is often a matter of exper-
iments conducted in complex machinery. There, it is
extremely difﬁcult to look into one of the most important
issues of cavitation which is nucleation. This work inves-
tigates gas cavitation under well-deﬁned ﬂow conditions.
Nuclei are placed in wall bound cavities and are exposed to
a radial gap ﬂow featuring independent pressure and shear
stress. A reciprocating bubble generation is achieved.
Bubble frequency and size are evaluated which turn out to
depend on pressure and wall shear stress. The experiment
lends itself to systematic research in cavitation.
The formation of gas bubbles from the liquid phase can be
either vapor cavitation or gas cavitation (also named
pseudo cavitation) of dissolved gas. Both processes may
have similar appearance and may, in fact, happen simul-
taneously. A necessary condition for cavitation to happen is
a change of the liquid pressure or temperature such that the
liquid or the dissolved gas becomes supersaturated with
respect to a saturated equilibrium state. In most cases, these
changes are brought forward by interaction of the ﬂow with
walls like ﬂow past a hydrofoil or ﬂow through a journal
Another necessary condition to get the phase change
actually going is nucleation of some kind. Jones et al.
(1999) provide a comprehensive review on this subject.
The most prominent nucleation sites are micro bubbles,
spots of high vorticity or wall bound cavities holding small
amounts of gas. The latter ones, which were ﬁrstly postu-
lated by Harvey et al. (1944), are subject of this work. The
existence of wall cavities in real ﬂow systems, for example,
crevices or patches of surface corrosion, is beyond dispute.
It is a good deal more challenging to accept that the cavity
gas, acting as a nucleus, should form one bubble after the
other without being depleted.
We think that the study of this process is fundamental to
the understanding of the inception of gas bubble formation
in ﬂows, for gas as well as vapor cavitation. In experiments
carried out in realistic environments like at hydrofoils [for
impressive pictures see Guennoun et al. (2003)] or in
throttle devices, the appearance of bubbles is readily
detected, yet the inception of bubble formation remains
unresolved. Therefore, we suggest a bench mark experi-
ment which makes a point of accessibility and reproduc-
ibility rather than proximity to real machinery. The
experiment deals with gas cavitation in oil. A choice has
been made for a suitable ﬂow in a gap. A simple Bernoulli
ﬂow which provides control of the pressure is not satisfying
because bubble formation, taking place at a wall, is a
transport phenomenon in terms of phase change and
momentum. Therefore, the ﬂow should offer pressure
control and variable wall shear stress. This concept has
been suggested in good time by Nu
cker (1932) and Fro
(1943) without follow-ups.
2 Experimental set-up and procedure
A radial gap ﬂow was realized as sketched in Figs. 1 and 2.
Oil ﬂows from the outer edge into the gap of width
F. Peters Á R. Honza (&)
Fluid Mechanics, Ruhr University Bochum,
tsstraße 150, 44780 Bochum, Germany
e-mail: email@example.com; firstname.lastname@example.org
Exp Fluids (2014) 55:1786