The present invention relates to a mass flow controller (MFC) calibration setup.
More particularly, the present invention relates to the field of solar photovoltaics.
This setup shall be used to calibrate mass flow controller (MFC) used for the deposition
of thins films of semiconductors for solar cell application.
BACKGROUND OF THE INVENTION
Deposition of good quality thin films of semiconductors is a requirement for producing
high efficiency solar cells. Thin film deposition is carried out using atmospheric pressure
chemical vapor deposition (APCVD) and Plasma Enhanced Chemical Vapor Deposition
(PECVD) techniques. The flow of process gases needs to be controlled with great
precision. Generally, the flows are in the range of a few cc per minute to a few litre per
minute. For maintaining accurate flow rates of process gases, MFCs are used which
work on the principle of measuring the change in the temperature of incoming and
outgoing gas mass. Generally the calibration shifts after a continuous use and this
effects the critical properties of the deposited films. Recalibration is necessary and it is
a tedious process as the MFC needs to be sent to OEM which is expensive and time
consuming. Replacement is also not easy as it may be very expensive. Therefore a new
simple technique has been attempted where MFC can be calibrated very quickly and
accurately without removing it from the gas manifold.
PRIOR ART
Thorough patent search has been conducted on the net and the related patent does not
reveal any similar design. The patents in general, deal with mass flow controller (MFC)
as a whole and/or techniques to improve the performance of the MFC. Some of the
patents talk about self calibrating MFCs and method to control the flow but no patent
deals with the technique to calibrate MFC with non-expensive and simple principle of
downward displacement of water.

JP2018010696A (2018-01-18) titled “The mass flow controller” deals with real time
calibration during its operation using two flow meters, control valve and controlling
signal. It is different from present innovation as we are using one time calibration that
to using non-expensive items like plastic tube, water etc.
JP6249024B2 (2017-12-20) titled “A plurality of kinds of fluids and method for improved
performance over the mass flow controller” corresponds to a system and method for
improving the control of a flow of a variety of fluid types. The method includes selecting
a process gas type for the process gas that will be controlled and obtaining molecular
mass information for the selected processed gas type. It is completely different from
the present work.
EP3105647A4 (2017-09-13) titled “system for and method of providing pressure
insensitive self verifying mass flow controller” gives details of an MFC comprising: a
pressure-based flow meter, a thermal-based flow meter, a control valve, and a system
controller. The pressure-based flow meter and thermal-based flow meter each measure
flow rate of mass through the mass flow controller. This is also similar to the real time
calibration and different from the present work.
US20030236643A1 (2003-12-25) titled “Apparatus and method for calibration of mass
flow controller” deals with a calibrator having a variable-flow fluid source, a receptacle
of known volume, and a pressure differentiator. The variable-flow fluid source supplies
gas at varying rates to the mass flow sensor being calibrated and at proportional rates
to a receptacle of known volume. A pressure differentiator computes the time derivative
of gas flow into the receptacle of known volume and, from that, the actual flow into the
receptacle. The purpose here is somewhat similar but the process of calibration of the
present work is unlike that of present work.

OBJECTS OF THE INVENTION
The object of the invention is to design a simple and inexpensive setup for calibrating
mass flow controller (MFC) without removing from the gas manifold.
SUMMARY OF THE INVENTION
Mass flow controllers (MFCs) are the very important component of a thin film deposition
system which is used for the deposition of semiconducting films for use in solar cell
production. The quality of the film depends a lot on the accuracy of the flow of process
gases. These MFC are very expensive and generally these are used for flowing process
gases which are in general toxic and pyrophoric such as silane, diborane, phosphine
etc. Due to the hazardous nature and very high purity of gases, removal of the MFC for
repair or calibration is quite cumbersome. It has also been felt and found that these
MFC do require recalibration after every three to four months of operation. In view of
these problems, a new simple and inexpensive technique has been developed where
these MFCs can be calibrated without removing from the gas manifold very quickly
using one of the first principle of physics i.e. downward displacement of water by gases.
BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
Fig. 1 – MFC calibration setup.
Fig. 2 – Calibration chart for Silane MFC.
BRIEF DESCRIPTION OF THE INVENTION
For the calibration of MFCs a simple inexpensive setup has been developed using the
concept of downwards displacement of water. The setup is portable and can be taken
close to the MFC which are generally installed in gas manifold and their removal from
the gas manifold is quite cumbersome. Using nitrogen gas at input of the MFC, it can be

calibrated by measuring the actual flow of gas as recorded by the displacement method
using a stop watch. The accuracy is very good as large amount of water is displaced for
long time (~ a few min to 60 min). As the operation is very simple and inexpensive, the
set up can be easily developed in any laboratory and can be used by any person
without the expert knowledge of physics.
DETAILED DESCRIPTION OF THE INVENTION
Mass flow controllers (MFCs) are the very important component of a thin film deposition
system which is used for the deposition of semiconducting films for use in solar cell
production. The quality of the films depends a lot on the accuracy of the flow of process
gases. These MFCs are very expensive and generally are used for flowing process gases
which are in general toxic and pyrophoric such as silane, diborane, phosphine etc.Due
to the hazardous natures of gases and very high purity, removal of the MFC for repair
or calibration is quite cumbersome. It has also been felt and found that these MFC do
require recalibration after every three to four months of operation. In view of these
problems, a new simple technique has been developed where these MFC can be
calibrated very quickly using one of the first principle of physics of downward
displacement of water by gases and that too without its removal from the gas manifold.
In this set up a large plastic tank (A) of about 50 litres capacity is placed on a table and
is filled with clean water (C). A beehive shelf (B) is placed on the bottom of the tank
and its height is selected such that when one liter graduated cylinder (D) is placed on
top of the beehive shelf, all the graduations of the cylinder are easily visible to operator.
The arrangement is shown in figure 1. Another very important part of the setup is the
three way stop cock which connects the inlet tube of the beehive shelf and the output
of MFC. This is an especially designed stop cock which keeps one opening of the cock
always open either to beehive shelf or the other port. This feature does not allow the
pressure to be built-up when input gas is connected to the MFC. The time of flow is
recorded using a stop watch (G).

OPERATION
The mass flow controller (MFC) to be calibrated is connected to the nitrogen gas supply
at the input and the output is connected to the tube which is going to three way stop
cock. The power supply and readout of the MFC is put ON and left for more than 30
minutes for warming up. The measuring cylinder is filled with water to the brim and
inverted inside the water tank and brought carefully to be placed on top of the beehive
shelf. Care is taken so that no air bubbles are permitted to be into the measuring
cylinder. This is ensured for accurate measurement. Before placing the measuring
cylinder on the beehive shelf, it is confirmed that the three way stop cock is in a
position where the gas will not go to beehive shelf and will bubble out of the cock itself.
The gas is allowed to pass through the MFC and starts bubbling out of the three way
stop cock into the water and not into the inlet tube of beehive shelf. The stop watch is
kept ready and the three way stop cock is turned to allow gas passage into the beehive
shelf. Simultaneously the stop watch is also started. The gas bubbles into the
measuring cylinder and the water starts receding. Keeping a close watch on the lower
meniscus of the water in the measuring cylinder, the stop watch is stopped when the
meniscus reaches zero mark on the measuring cylinder. Now we know that how much
time is taken to pass one 1000 cc of gas. The experiment is repeated by changing the
flow rates at MFC at the digital display and a chart is prepared for at least six flow rates
as determined by this set up and compared with the reading displayed at the MFC. If
the chart shows a linear behavior, the flow in under control and MFC is controlling the
gas properly. A calibration table is prepared w.r.t. actual flow and the displayed value at
MFC.

EXPERIMENTAL DATA FOR ACTUAL CALIBRATION OF A SILANE MFC
A Silane MFC of 0-100 Standard Cubic Centimeter per Minute (sccm) has been
calibrated. The conversion factor (K)for nitrogen to silane is taken as 0.6 from the table
of standards.

A graphical representation is made in the form of calibration chart for Silane MFC which
can be seen in Fig.2
It can be easily seen that the calibration of MFC has drifted and error is more for the
low flow rates.
ACCURACY
Generally it is felt that if the calibration done with simple techniques it may not be very
accurate. This is not so in the present case. The same is explained below with the help
of an example.
In general the MFCs are in the range 0-100 sccm. Let us select a value of 50 sccm. This
means that for displacing a volume of 1000 cc (one fill of one lit jar), it will take 20
minutes. Using an ordinary digital stop watch, one can easily manage very accurate
measurement of time correct to ms. Let us imagine that one has introduced an error of
0.5 seconds in time measurement and about 10 cc in volume measurement. This results
in an inaccuracy of about 2%. In actual practice it has been observed that these MFCs
drift from their rated value to more than 15 to 20 % in a period of about few months.
Researchers working in R&D laboratories do find it very difficult to get these MFCs
recalibrated due non availability of inexpensive simple setup. This set up can solve their
problems very easily.

WE CLAIM
1. A mass flow controller (MFC) calibration setup comprising:
a large plastic tank (A) of about 50 litres capacity containing clean water (C);
a beehive shelf (B) placed on the bottom of the said tank (A) with its height so
selected such that when a one litre capacity graduated cylinder (D) is placed on top
of the beehive shelf (B), all the graduations of the cylinder (D) are easily visible to
the operator;
a three way stop cock (E) connecting the inlet tube (H) of the beehive shelf (B) and
the output end (K) of mass flow controller (F);
an inlet (J) is provided at the MFC (F) for entry of nitrogen into the controller,
wherein the three way stop cock (E) so designed which keeps one opening of the
cock (E) always open either to beehive shelf (B) or the other port ( ) in order to
allow the pressure to be built up when input (J) gas is connected to the MFC, the
setup using a stop watch (G) for recording the time of flow of gas during calibration.
2. The method for using the mass flow controller (MFC) calibration setup as claimed in
claim 1, comprising :
- filling the plastic tank (A) of 50 litre capacity with clean water (C);
- filling the one litre capacity graduated measuring cylinder (D) with filled in water
upto the brim and placing the filled in cylinder (D) in inverted position inside the
water tank (A) carefully on top of beehive shelf (B);
- connecting the MFC (F) to be calibrated to nitrogen gas supply at the input (J)
and output (K) and finally to the tube inlet (H) of beehive shelf (B) via three
way stop cock (E) ensuring that no air bubble are permitted to be into the
measuring cylinder (D) in the initial stage;
- ensuring before placing the measuring cylinder (D) over the beehive shelf (B)
that the three way cock (E) is in a position where the gas will not go to beehive
shelf (B) and also will bubble out of the cock (E) itself;

- putting ON the power supply and read out of the MFC (F) and left for 30 minutes
for warming up;
- allowing the nitrogen gas to pass through the MFC (F) and bubbling out of the
three way stop cock (E) into the water, but not into the inlet tube (H) of beehive
shelf (B);
- the stop watch (G) being kept ready, the three way stop cock (E) is turned on for
allowing the gas passage into the beehive shelf (B) when the stop watch (G)
starts taking reading;
- allowing the gas bubbling into the measuring cylinder (D) occupying at the top
surface while at the same time water level inside the measuring cylinder (D)
receding;
- keeping a close watch on the lower meniscus of the water in the measuring
cylinder (D), the stop watch (G) is stopped at that period when the lower
meniscus reaches zero-mark;
- writing down the time taken for passing 1000 c.c gas;
- repeating the experiment by changing the flow rates at MFC (F) at the digital
display for at least six different flow rates as determined by the setup;
- preparing a chart by comparing with the reading displayed at the MFC (F),
- preparing a calibration table w.r.t. actual flow and the displayed value at MFC
(F) wherein a linear behavior of chart shows the flow is under control and the
MFC (F) is controlling the gas supply.