Source:
http://www.bronkhorst.com
The footprint of mass flow and pressure sensors can be substantially
reduced using MST/MEMS
DR. J.C. LoTTERS
Equipment manufacturers are looking for compact
solutions to monitor or control the gas flow or pressure
within their systems. As these systems are getting smaller
and smaller, the incorporated flow and/or pressure
control units need to be miniaturized.
For example, the dimensions of a conventional gas
chromatograph (GC) are approximately 1 x 1 x 1 m. The
latest generation of (portable) GCs may have dimensions
down to circa 25 x 25 x 25 cm. In a GC, usually 3 to 6
mass flow and/or pressure controllers are applied.
Conventional mass flow controllers (MFCs) and pressure
controllers (EPCs) have dimensions of circa 2.5 (w) x 7.5
(l) x 12.5 (h) cm, making it impossible for manufacturers
to further decrease the dimensions of their equipment.
Moreover, the use of individual instruments introduces
potential leak points.
Until now, conventional mass flow and pressure sensors
and controllers have needed a footprint of 1.5” (ca. 40 mm), as for instance specified in the NeSSI™ [1] system.
Due to micro system technology (MST/MEMS),
Bronkhorst High-Tech has been able to halve this dimension
to 0.75” (ca. 20 mm), enabling the realization of ultra
compact flow and/or pressure measurement and control
instruments and systems.
In this article, a new generation of instruments is presented
that meets the size and other requirements
imposed by compact equipment manufacturers.
Furthermore, due to their modular construction, compact
manifold solutions can be built, thereby reducing
potential leak paths and ensuring space efficiency.
Flow sensor structure and basic
operating principle
The actual flow sensor chip consists of a thin square
membrane with two heaters and two temperature sensors
(T-sensors) on its surface. The
membrane is suspended in a
frame around it and can withstand
200 psi (15 bar).
The measurement principle is
as follows: the two heaters are
heated to a temperature ΔT over
the medium temperature by providing
them with constant power.
From each T-sensor, the hot side
is located on the membrane, and
the cold side is on the frame
around the membrane. The
temperature difference between
the hot sides of the upstream and
downstream T-sensors is proportional
to, and a measurement
of, flow.
|
|
At zero flow, the heaters are
powered in such a way that the
differential output voltage of both T-sensors equals zero. The resulting
temperature distribution on the
membrane is shown in Figure 1 (the
green line, indicated with Φm = 0).
When a flow occurs, the temperature
of the upstream T-sensor decreases,
and the temperature of the downstream
T-sensor increases. The
resulting temperature difference ΔT
[°C] is the measurement for the flow,
as shown in Figure 1 (the blue line,
indicated with Φm > 0). |
Figure 1. Basic operating principle of flow sensor chip
Figure 2. Exploded view of the flow sensor
(left) and control valve (right) module |
Unfortunately, the transfer function
between mass flow Φm [kg/s]
and resulting temperature difference
ΔT [°C] is not a simple expression.
The temperature difference ΔT is
dependent on the dimensions and
physical properties of the membrane
material and on the physical properties
of the gas that is applied.
The exact relation between mass
flow Fm [kg/s] and resulting temperature
difference ΔT [°C] has been
determined using finite-element
method (FEM) simulations. Using
this exact relation, the behavior of
the flow sensor for different gases
can be predicted. |
Micro fluidic modules
The flow sensor chip is mounted
in a modular housing with a 20 mm footprint. Additional modules with
this footprint, necessary for the realization
of miniature flow pressure
control systems, are:
- Flow sensor; typical flow ranges
from 20 mln/min through 2000
mln/min (full scale values)
- Pressure sensor; typical
pressure ranges of the order
of 100 psi
- Control valve
- Electronic circuitry
- Three-way valve
- Shut-off valve
- Filter
- Mixing chamber
Both individual instruments and
customer specific designs can easily
be configured with these modular
“building blocks”, as shown in
Figure 3.
Due to these small-sized modules,
we have manufactured mass
flow and pressure controllers (as
shown in figures 3a and 3b) with the
dimensions 20 (w) x 40 (l) x 60 (h)
mm which, to our knowledge, are
the smallest manufactured instruments.
Figure 3. (a) IQ+Flow mass flow controller; (b) IQ+Flow pressure controller; (c) customized
manifold flow pressure controller based on IQ+Flow modules (chip flow sensor, chip pressure
sensor, control valves, electronic circuitry, three-way valve)
Experiment
A sample group of mass flow
controllers utilizing the chip flow
sensor, control valve and electronic
circuitry modules were subjected to
the same test (see Figure 3a).
The output signal of the instruments
was measured for flow ranges varying
between 20 and 2000 mln/min.
nitrogen, hydrogen, helium and argon. Furthermore, the dynamic
behavior of each of the instruments
was measured by performing various
set point variations. The resulting
mass flow sensor response was
measured by a digital oscilloscope.
The measured output signals as a
function of the mass flow are shown
in Figure 4a. The measured dynamic
behavior of the mass flow controller
is shown in Figure 4b.
The measured curves, as displayed
in Figure 4a, correspond well
with the theoretically expected values
as obtained with the FEM simulations.
The measured response
times, of which a typical example is
shown in figure 4b, are all within the
value of t98% = 0.5 s.
Example of an analytical
application
A miniature flow pressure control
unit can be applied at the injector of
a GC, as shown in Figure 5. The carrier
gas flow is controlled by the
MFC and fed through the injector to
the analytical column and thus to the
detector. It is vital that the carrier
gas flow is highly stable under all circumstances,
since the analytical
result is very much dependent upon
the stability of the flow velocity
through the analytical column.
The pressure within the injector
is kept constant by the EPC. When a
liquid sample is injected it immediately
evaporates due to the high temperature
in the injector and the
resultant pressure pulse is directly
vented by the EPC. It is very important that the pressure in the injector
is stable under all circumstances
since the analytical result is again
very much dependent on the
stability of the flow velocity through
the analytical column (which is
affected by pressure variations on
the column).
Therefore, stability and fast
response are important features for
both the MFC and the EPC.
Moreover, both controllers should
be independent of each other’s
behavior.
The system as described above
has been realized with the miniature
flow pressure control modules, built
both as individual instruments (as
shown in Figure 5b) and as a
customized manifold (as shown in
Figure 3c). Both versions of the
miniature flow pressure control unit
showed excellent performance.
Figure 4. Measurement results obtained with an IQ+Flow MFC, adjusted for N2: 500 mln/min º100 % (a) response to different gases: N2, H2, He and
Ar; (b) response to change in set point (gas: N2)
Conclusions
In this article, a new generation of
micro fluidic flow pressure control
modules has been presented that are
capable of meeting the special
needs—and more importantly, the
size constraints of analytical and
other markets. Due to the use of
micro system technology (MST), we
are able to halve the size of functional
modules thereby enabling the realization
Figure 5. (a) Schematic and (b) realised miniature flow pressure control unit at the injector side of a GC
References:
1. New Sampling/Sensor Initiative;
www.cpac.washington.edu/NeSSI
Dr. J. C. Lötters is Manager
Research & Development at
Bronkhorst High-Tech B.V.,
Nijverheidsstraat 1a, 7261 AK
Ruurlo, The Netherlands. He can
be reached at +31-573-458800 or
j.c.lotters@bronkhorst.com. |
