Introduction
Colormetric Leak Detection, the primary topic of this report, is a
relatively new, reliable, economical method of Leak Detection. Such
Leak Detection is an outgrowth of industries need for better methods
of leakage testing. Before launching into the specific qualities of
this method, I would like to briefly comment on the total Leak
Detection field.
The Alaskan Pipeline is a dramatic illustration of the need for more
attention to the field of leakage testing. Even the extensive flaw
testing performed for the Alaskan Pipeline was not enough, leaks
still developed. The Alaskan Pipeline is not the only project that
has suffered, sometimes at great expense from failure to understand
the important role of Leak Detection. As a result of such leakage
problems as experienced by the Pipeline, quality control experts are
demanding that leak testing as well as flaw testing, be in their
control.
Leak Testing can be as refined as flaw testing. Each company must
require of their supplier, quality leak testing products. Since leak
detection, in the past was assigned a minor role, it often remains a
branch of alchemy to many people. Often companies are not sure how
to derive a leakage standard for their product which adequately
balances the need for legal liability protection with the demands
for economic production.
The rush to know more has brought a mixture of information on the
market, sometimes to the point of confusion. Much of the confusion
could be eliminated by a clear set of terminology, some of which I
will set forth here.
In its simplest form, all leakage rates are expressed in mass loss
per unit of time, flowing across a pressure barrier. The variety of
ways of expressing this leakage is large and a uniform means of
communication is necessary. It is easy to see how confusing this can
be when leakage is defined in an unfamiliar standard. For instance,
I think of leakage in terms of standard cubic centimeters per
second. However does a standard cubic centimeter look like? If I say
to you that a leakage rate of 10-4 std. Cc/sec. Is the equivalent of
the loss of 1 cc. of air over 3 hours, this sounds like a large
leakage rate. However, if I say that it is the equivalent of losing
a pound of freon over 27 years, you will probably think the leakage
quite small.
Some of the equivalents to standard cc. per seconds are illustrated
here:
EQUIVALENT LEAK SIZE ( to Std. cc/sec)
Std.cc/sec= One cubic centimeter of gas flow per second at 14.7 psi
of pressure and a temperature of 77o.
|
Std cc/sec |
Time for lb
of
freon to leak |
Torr
Liters/sec |
Time for
cc to leak |
Bubble Time
in Water |
| 10-1 |
10 Days |
7.6x10-2 |
10 seconds |
1.3 seconds |
| 10-2 |
3 Months |
7.6x10-3 |
100 seconds |
13.3 seconds |
| 10-3 |
2.7 Years |
7.6x10-4 |
16.67 minutes
|
14.5 minutes |
| 10-4 |
27 Years |
7.6x10-5 |
2.78 hours |
24 minutes |
| 10-5 |
270 Years |
7.6x10-6 |
27.8 hours |
4 hours |
| 10-6 |
2,700 Years |
7.6x10-7 |
11.57 days |
|
| 10-7 |
27,000 Years |
7.6x10-8 |
3.86 months |
|
| 10-8 |
270,000 Years |
7.6x10-9 |
3.22 years |
|
| 10-9 |
2,700,000 Years |
7.6x10-10
|
32 years |
|
| 10-10 |
27,000,000 Years |
7.6x10-11
|
320 years |
|
Equal in importance to an understanding and uniformity of Leak
Detection standards is an understanding of the leakage testing
products. A multitude of Leak Detection and leakage measurement
devices exist. Some devices are pneumatic. Some are electronic. Some
are chemical. Each device has advantages and disadvantages. To use
any method effectively, a careful analysis of the device is
necessary.
The device I will attempt to analyze today is one form of chemical
leakage testing. This device, "Colormetric Developer System", is a
two component leak locator, consisting of a tracer-activator and a
developer-indicator.
The first component, the tracer, can be either gaseous or liquid;
such as ammonia, chlorine, hydrogen chloride, carbon dioxide,
hydrogen sulfide, nitrogen dioxide, hydrazine, water and liquid
hydrocarbons. Obviously, in some cases, adding tracers is
unnecessary since the tracer is inherent in the product being
tested. The most important characteristic of a tracer is its ability
to rapidly penetrate through a leak.
The second component, the developer-indicator, must have the ability
to react with the tracer and to indicate colormetrically minute
leakage of the tracer. The developer should enhance the visibility
of the leakage indication by drawing the tracer to it, by
absorption. This process should cause the indication to spread.
Also, a good developer creates a sharp color contract between its
reacted and unreacted appearance. Ideally, the color reaction should
be the compliment of the original developer color. The developer
should be such that it is easy to remove by rinsing or wiping. And
obviously, the developer should not be harmful to the material being
tested.
How Colormetric Systems Work
To locate a system leak, a thin coating of the indicator developer
should be applied to the outer surface of the object being tested.
The application can be made by spraying or dipping the test object.
Or, the developer can be incorporated into a dry paper which can be
taped on to the test object. In most cases, only a small area of the
test object need developer coating such as seams, welds, couplings
and joints; these areas are often the susceptible areas.
Next, the tracer-activator should be introduced into the test object
at a predetermined concentration and pressure. The tracer-activator
must be given a specific soak time so that it may effuse through any
leaks and thus react with the developer.
For any given Colormetric System, sensitivity is affected by four
factors - the physical properties of the tracer, the differential
pressure across the leak, the concentration of the tracer and the
soak time. These four factors, which I will outline, make
significant differences in the test results and they need to be
accurately calculated.
1). Physical Properties of the Tracer. The viscosity and molecular
weight of the tracer determine its ability to rapidly diffuse
throughout the test object and transverse a leak. Usually, the
tracer is a combination of two gases; one is the activator, one is
inert. For maximum effectiveness, the molecular weights of the
activator and inert constituent should be similar.
Graph #1 shows the relationship between viscosity of the tracer
and its ability to diffuse. This graph illustrates that the lower
the viscosity of the tracer, the more rapid will be the rate of
diffusion. Likewise, low molecular weight increases the rate of
diffusion.
2). Differential Pressure Across the Leak. Generally, the higher the
pressure across the leak, the more sensitive the test.
Graph #2 shows the relationship of Colormetric sensitivity as a
function of pressure and soak time. The curve itself represents a
spot of 1mm diameter produced by the colormetric interaction between
the tracer activator and indicator developer. As the pressure
differential is increased, the time required for a visible
Colormetric reaction decreases. In many systems, as this graph
indicates, at a certain point, further increases in pressure will
only marginally reduce the development time.
Leak Rate 1x10-4 std cc/sec
Tracer Activator: 1% NH3 in 99% N2
Indicator Developer: Pinpoint ADP-119
3). Tracer Concentration. Changes in the concentration of the
tracer-activator will vary the Colormetric reaction time with the
developer. By increasing the concentration of the tracer-activator,
Colormetric development time decreases.
Graph #3 shows the relationship between tracer-activator
concentration and development time (in seconds) to produce a 1mm
spot on ADP-119.
4). Soak-Time. Soak time
is the time allowed for the tracer to diffuse through the system,
transverse the leak and react with the developer. Soak time must be
long enough to allow sufficient tracer to transverse the leak. The
smaller the leak, the longer soak time required.
By increasing the pressure differential, tracer concentration and
soak time, the maximum sensitivity for any given Colormetric System
can be achieved. However, the smallest leak which can be found with
a Colormetric System is limited by the diameter of the tracer
molecule. Obviously, if the leak size is smaller than the size of
the tracer molecule, the tracer cannot transverse the leak and cause
a reaction. For practical purposes, the theoretical limit of
sensitivity is reached when the leak size is one decade larger than
the molecular size of the tracer.
Colormetrics In Use
Having given an overview of Colormetric Systems, I will describe one
system is use in the marine industry - The Colormetric System now
being used to test the tightness of integrated LNG tanks. This is an
ammonia-tracer system, which is being used because of the relatively
high sensitivity, which can be achieved with ammonia as a tracer.
Ammonia, has a molecule small enough to transverse leaks with a
minimum diameter of 10-3 microns. It has low molecular weight and
low viscosity. Because of these physical characteristics, ammonia
Colormetric Systems can be satisfactorily used for leakage rates as
small as 10-7 std. cc/second. Chart #2 compares the physical
characteristics of ammonia with two of the most used Non-Colormetric
tracers - Freon and Helium. As can be seen from the chart, the
physical properties of ammonia make it an excellent tracer.
| Chart #2 Tracer
Gas Comparison |
| |
Viscosity in Centipoise |
Molecular Weight |
Molecule Diameter in Microns
(M) |
| Ammonia |
.0094 |
17 |
3x10-4 M |
| Freon |
.0118 |
121 |
6.96x10-4 M |
| Helium |
.0178 |
4 |
2.18 x 10-4 M |
The formulas used to precisely calculate are described in the ASTM
Specification E1066
Standard Method for Ammonia Colorimetric Leak Testing.
Refer to Section 11.1.2.
Application of Colormetrics
Once the variables are calculated, the application of the ammonia
Colormetric System is fairly straightforward. The recommended
process for testing pipes and containment tanks with an ammonia
system is:
1). Clean the inside and outside of the tank thoroughly.
2). Dry the outside wall of the container with a hot air gun.
3). Coat the tank seams and other points of suspected leakage with a
thin layer of developer.
4). If the atmosphere is humid, dry the developer with a hot air
gun.
5). When the developer is dry, in some instances a vacuum should be
drawn on the inside of the tank to reduce the moisture and provide
for an even tracer mixture in the tank.
6). Meter in the ammonia-nitrogen mixture until atmospheric pressure
has been achieved. Large leaks will develop immediately.
7). If no leaks are apparent, slowly increase the pressure of the
tracer mixture until test pressure is achieved. Maintain vigilance
for the development of large leaks.
8). At the end of the soak time, mark any leakage indications,
evacuate the tracer, and rinse with nitrogen.
The ammonia tracer process just described is one of many Colormetric
Development Systems. Each Colormetric System though different in
specifics, has several common features. One of the common features
is that system sensitivity can be adjusted on a formula basis to
meet any given leakage standard. Once the calculation is done,
unlike electronics, no further calculation or adjustment is
required.
Colormetric Systems are easy to operate. They do not require highly
trained personnel slowly investigating each seam. Color reactions
can usually be seen quickly in a wide visible range. Since personnel
time and initial system cost is low, Colormetric Leak Detection is
relatively inexpensive. Colormetric Systems are operator
independent. Colors readouts are clear and repeating eliminating
decisions based on personal opinion.
Because of these features, colormetric systems have assumed a major
role in the rapidly advancing field of leak location.
|
