|

Sensistor Product Webpage
Sensistor FAQ
Sensistor Product Brochure
HYDROGEN AS
TRACER GAS FOR LEAK DETECTION
M. Block 1
1 Sensistor Technologies GmbH;
Muehlheim am Main, Germany
Abstract:
Helium is the most commonly used
tracer gas for leak detection.
However, it has
a number of limitations in many industrial applications. The
molecular weight
of hydrogen is only half that of helium, and hydrogen has a much
higher molecular
velocity. As a result, hydrogen diffuses rapidly inside test objects,
and dissipates
far more quickly than helium. These properties, and the low price of
hydrogen, make
it a very attractive alternative.
There are two
main reasons why hydrogen is not already more commonly used than
helium as a
tracer gas: a) the perception that hydrogen is too dangerous and b) the
availability of
suitable hydrogen detectors.
The flammability
problem is easily solved by using a readily available standard mix of
5% hydrogen and
95% nitrogen, which is classified as non-flammable.
The other
obstacle was overcome by the introduction of a new type of hydrogen
detector, based
on microelectronic hydrogen sensors. These have high sensitivity
and selectivity
to hydrogen. Moreover, they are cost-effective and sufficiently robust
for industrial
applications, allowing leaks of 5x10 -7
atm cc/s to be detected using the
5% hydrogen mix.
The use of hydrogen as a tracer gas is now increasing rapidly in
the automotive,
chemical, power generation, aerospace and telecommunications
industries,
among others.
Introduction:
There are many different
technologies and methods currently in use
for leak
testing. Water baths and foaming agents, for example, are widely
employed
in industrial
leak location, and are capable of detecting leaks with leak rates down
to
10 -3
atm cc/s. However, test objects
need to be cleaned or dried after testing, and
both methods are
unsuitable for test objects where it is important to avoid exposure
to moisture.
Leak testing
often involves the measurement of pressure changes. In industrial
conditions, this
method, too, is capable of detecting leak rates of approximately 10 -3
atm cc/s for
small-volume test objects. With larger volumes, however, sensitivity
decreases
significantly. The accuracy of measurements is also affected by
temperature
fluctuations and elastic deformation of test objects. These variables
must therefore
be taken into account when determining leak rates. If leak testing by
means of
pressure measurement reveals leakage in excess of permissible levels,
the
next step is to
locate the leak.
The tracer-gas
method can be used for both leak location and leak testing. It is also
ideal for
locating smaller leaks. Test objects remain dry, and measurements are
unaffected by
temperature changes and elasticity. Various tracer gases are currently
used for this
purpose. Helium and hydrogen are most common for a wide range of
industrial
applications. Other gases, such as sulfur hexaflouride (SF 6)
or carbon
dioxide (CO 2),
are employed for highly specialized applications.
Results:
Hydrogen has a number of properties
that make it particularly well suited
to use as a
tracer gas. It has a very low viscosity and the background concentration
of hydrogen in
ambient air is relatively low.
Comparison of
the physical properties of hydrogen and helium with those of air:
Hydrogen Helium
Air
Dynamic
viscosity 8.7 µPas 19.4 µPas 18.3 µPas
Background
concentration in air 0.5 ppm 5 ppm 100 %
Mol. mass 2
g/mol 4 g/mol 29 g/mol
Assuming laminar
flow (generally the case for leakage rates up to 10 -5
atm cc/s), the
amount of tracer
gas that escapes through a leak during a defined unit of time
depends on its
dynamic viscosity. The background concentration of a tracer gas in
ambient air
affects the detection limit.
A standard
mixture of 5% hydrogen (H 2)
and 95% nitrogen (N2)
is generally
employed in leak
testing. Certified as non-flammable to ISO 10156, this gas presents
no safety risks.
It is primarily deployed as a shielding gas for welding, soldering, and
brazing, and is
usually provided by industrial gas suppliers in 200 bar (2.900 psi) and
300 bar (4.350
psi) pressurized gas cylinders. The mixture costs a fraction of helium,
the
price of which
has recently risen sharply due to increased demand. Moreover, as
global resources
are finite, helium is expected to become even more expensive. By
contrast, the
cost of hydrogen, which is increasingly seen as a source of energy, is
steadily
falling.
Unlike helium,
hydrogen is a renewable resource, which is an important
consideration
for environmental audits under ISO 14001. Furthermore, hydrogen is
non-toxic and
non-corrosive.
The physical
properties of hydrogen also offer a number of practical benefits.
Hydrogen
diffuses much more rapidly than helium, quickly achieving uniform
concentration
within the test object. Evacuation of the test object is not generally
required before
testing. Moreover, there is far less risk of contamination. In addition,
hydrogen does
not adhere to surfaces, and can be flushed out much more rapidly.
The above
factors make hydrogen the ideal tracer gas.
Discussion:
We have seen that hydrogen is
highly suitable as a tracer gas.
However, the
hydrogen method requires a testing device with corresponding range
and selectivity
(i.e. exclusive to one gas).
The testing
device presented here is based on a semiconductor sensor similar in
design to a
field-effect transistor. Hydrogen molecules adhere to the surface of the
sensor, where
they dissociate into hydrogen ions (protons). These protons can then
penetrate the
lattice structure of the sensor, where they cause a change in an
electrical
field, triggering a signal that passes along a signal-processing chain
to be
displayed by the
testing device. The sensor responds selectively to hydrogen. The
detection limit
is based on the natural background concentration of hydrogen in air
(0.5 ppm).
To detect a
leak, the sensor, which is integrated into the probe, is placed directly
over the leak.
This means that no pumps are required to suck gas into the device.
The sensor is
linked to the testing device by a single cable. As a result, no filters
or
other
maintenance-intensive components are needed. A further benefit is that
this
direct
measurement method has very rapid response times, irrespective of the
cable
length, and very
short recovery times. The sensitivity of the device can be adjusted
at the touch of
a button across a range from 1 to 4,000, depending on the task
involved.
The
easy-to-operate, rugged devices are ideal for industrial applications,
and can be
quickly and
easily calibrated using a reference leak or calibration gas. During
calibration, the
current status of the sensor is displayed. A battery- operated version
allows several
hours of uninterrupted mobile use.
Conclusions:
The hydrogen method is suited to
applications where a flow of tracer
gas is emitted
from the test object. Differential pressure can range from zero to
several thousand
bar. Even with absolute differential pressure of zero bar, partial
pressure
differential is sufficient to cause hydrogen gas exchange via the leak.
The device is
primarily employed to perform leak location or integral leak testing by
means of the
hood method. Leak location
This type of detection is also
commonly
known as
"sniffing" - which is not strictly accurate here, since tracer gas does
not
have to be
sucked into the device. With forming gas (95/5), the detection limit is
5x10 -7
atm cc/s. In the hydrogen industry,
testing is sometimes performed with
higher hydrogen
concentrations - and corresponding safety measures - or even using
pure hydrogen.
Here, the detection limit drops in proportion to the concentration,
enabling lower
leakage rates to be detected.
Example
applications:
* Leak location
on objects of complex shapes.
If a test object
(e.g. engine, transmission or axle) has been found to be leaking, the
leaks need to be
located as quickly and accurately as possible to enable repairs. The
properties of
the tracer gas outlined above and the variable sensitivity of the
testing
device enable
rapid, straightforward, and precise location of all sizes of leaks.
Results
are unaffected
by background concentrations of hydrogen. Since tracer gas does not
adhere to
surfaces, the effectiveness of repairs can be conclusively verified by
subsequent
testing. * High background levels of helium.
Components
previously tested using helium are often installed in cooling and
airconditioning
systems. As a
result, the background concentration of helium at the
assembly site is
too high for accurate subsequent measurement. * Rapid leak
location for
vacuum systems.
In vacuum
systems where there are gross leaks or considerable moisture, leak
location can be
extremely time-consuming or impossible. Creating a sufficient
vacuum is a
difficult and prolonged process, which the hydrogen method accelerates
significantly.
Leak location can begin after the test object has been flushed with
forming gas.
Experience shows that the majority of leaks lie within the range of 10 -2
to 10 -5
atm cc/s, i.e. within the device
measuring range. Due to the properties of the
tracer gas
described above, results are unaffected by the background concentration
of hydrogen. *
Mobile applications In large, complex systems of the kind
encountered in
plant engineering, and the chemical and aerospace industries, leak
detection is
often extremely time-consuming. Where testing equipment is
insufficiently
mobile, and there are background effects due to the tracer gas, the
process can take
even longer. The hydrogen leak detection device described above is
available in a
battery-powered version for applications of this type. * Hydrogen
systems Systems
that run on or contain hydrogen, such as electrolysers, reformers
or even fuel
cells, can be tested during operation. This solution is particularly
attractive for
regular testing during ongoing operations, where there is a need to
avoid system
downtime.
Hood method
(integral testing)
As in leak
location, the test object is filled with tracer gas. The increase in the
hydrogen
concentration is measured in a known volume created around the test
object. The
duration of the test depends on the permissible leakage rate and on the
specific volume
involved. Depending on the application, the known volume is created
using either a
flexible hood or a solid enclosure.
Example
applications
* Systems
engineering With heat exchangers and valves, for example, it is
important
to check for
leaks at the interfaces between individual areas, as well as for
external
leaks of the
test object. In many cases, overpressure testing involves
significantly
less time and effort than vacuum methods. Independent studies verify
that results of
equal quality are achieved using both methods. * Volume production
In the
automotive industry and refrigeration systems manufacture, components
with
low permissible
leakage rates are often tested in a vacuum chamber. As a result,
tests performed
using this method often involve high capital expenditure and running
costs. In many
cases, leak testing can be performed in a simple test chamber at
atmospheric
pressure, considerably reducing investment and running costs. *
Research and
development In R&D, it is often necessary to determine the leakage
rate of a
variety of test objects with a minimum of effort. Measurement using a
flexible hood is
particularly suitable. As described above, the tracer gas rapidly
diffuses
uniformly within the hood and does not adhere to surfaces, allowing
repeated
measurements to be quickly and easily performed. * Car manufacturing
The tightness
requirements for fuel systems in cars are becoming ever more
stringent. With
conventional fuels, it is increasingly common to measure the total
emissions of the
stationary vehicle (zero evap). For hydrogen-powered vehicles, a
simple test of
the overall leakage rate of the entire vehicle is required.
Measurements are
performed using a hood. * Permeability Testing of permeability is
a specialized
application. Forming gas is applied to one side of a material sample,
and the increase
in the hydrogen concentration over time is measured on the
opposite side.
This procedure is used for packaging materials and membranes.
|