Varian 990 dCLD II, Руководство по эксплуатации

Страница: 151 / 162

Model 990 dCLD II Component Leak Detector
C-5
C.5.4 Facts About Leak Rates
Visualizing Leaks in
Everyday Terms 10
−
5
std cc/sec: approximately 1 cc/day
10
−
7
std cc/sec: approximately 3 cc/year
Audible or Visual Detection by Observer
a. Bubbles rising
in water
10
−
4
std cc/sec or larger
b. Audible Leaks 10
−
1
std cc/sec or larger
Sizes of Leaks in
Man-Made Joints Studies indicate that almost all leaks at joints are about
5 x 10
−
7
std cc/sec (about 1 cc/month) or larger. This is true of
ceramic-to-metal, plastic-to-metal seals, welded, soldered and
brazed joints. Some long-path leaks are slightly smaller. Diffusion
of helium through glass can be as high as 10
−
8
std cc/sec per
square centimeter of surface area.
Variation in Leak Sizes Leaks unintentionally built-in at joints during manufacture can vary
from hour to hour and day to day. Breathing on a 10
−
6
std cc/sec
leak provides enough moisture to close it temporarily; perhaps for
several days. Atmospheric particles can close a leak of this size.
Never depend on an
accidentally made leak to remain constant.
Manufacturing standard leaks for calibration purposes requires spe-
cial techniques.
C.6 Leak Detection Methods
Most leak detection methods depend on the use of a tracer gas passing through the leak and
being detected on the other side (for example, visual detection of air bubbles in water).
The mass spectrometer leak detector operates with helium as a tracer and is widely used
because it combines high sensitivity with production testing capability. There are three
basic methods in common use.