The value quoted above is sometimes stated as the maximum head rise
possible at a check valve on closure. This is only the case as far as the
initial head rise is concerned, as observations have revealed consider-
ably greater head rise after initial closure as described in section
20.10.2. At the upstream valve face, the corresponding head drop
will be:
H
v
¼aV
o
=g
20.10.2 Cavitation upstream of the valve and resulting peak
pressures
At the upstream valve face there is a fall in head which in the absence of
other influences will be aV
o
=g. Since pressure head H
o
at the valve
just prior to closure may be relatively low, it is quite likely that minimum
pressure at the upstream face of the valve calculated as H
o
aV
o
=g will
be less than i:l: h
vap
, where i:l: is invert level and h
vap
is the vapour
pressure head. Should minimum head reach the vapour pressure
head, a vapour-filled cavity will start to develop so that:
V
o
þ g=aH
o
¼ Jþ¼V þ g=aði:l: h
vap
Þ
Taking the cavity to be distributed across the entire pipe cross-section,
then velocity at the front of the cavity will be:
V ¼ V
o
þ g=aH
o
g=aði:l: h
vap
Þð20:11Þ
Flow in the pipe branch upstream of the valve will decelerate and
reverse becoming positive.
The cavity will then collapse with an attendant shock pressure rise
against the upstream valve face. By this time, pressure wave effects in
the pipe branch downstream of the check valve may well have dissipated.
In his investigations of 50 mm swing check valve behaviour, Worster
(1959) observed that in almost all cases studied, after initial seating of
the valve, the door repeatedly opened by a very small amount and then
shut due to the action of pressure waves in the pipe branch upstream of
the valve. The differential transient pressure across the valve door was
sufficient to reopen the valve. In more severe tests when appreciable
noise was produced by initial closure, the valve remained shut for a
longer period and then reopened sometimes by as much as 128 from
its seat. Following closure, further periods of reopening and closure
occurred. Worster attributed this behaviour to high surge pressures
after cavity closure, producing a thrust against the valve door causing
it to reopen. Forward velocity in the entire upstream water column as
397
Check valve dynamics