CONTROL OF A GASEOUS ENVIRONMENT IN A WAFER LOADING
CHAMBER
CROSS REFERENCE TO RELATED APPLICATIONS
The present application claims the benefit of and priority from commonly assigned U.S. Provisional Patent Applications Serial Nos. 60/396,536, entitled Thermal Processing System, and filed July 15, 2002, and 60/428,526, entitled Thermal Processing System and Method for Using the Same, and filed November 22, 2002, both of which are incorporated herein by reference in their entirety.
BACKGROUND OF THE INVENTION
1. Technical Field
This invention relates generally to semiconductor manufacturing equipment, and more specifically to methods and apparatuses for controlling the gaseous environment in a wafer loading chamber.
2. Description of Related Art
Furnaces are commonly used in a wide variety of industries, including in the manufacture of integrated circuits or semiconductor devices from semiconductor substrates or wafers. Thermal processing of semiconductor wafers include, for example, heat treating, annealing, diffusion or driving of dopant material, deposition or growth of layers of material (including chemical vapor deposition), and etching or removal of material from the substrate. These processes often call for the wafer to be heated to a temperature as high as 250 to 1200 degrees Celsius before and during the process. Moreover, these processes typically require that the wafer be maintained at a uniform temperature throughout the process, despite fluctuations in the temperature of the process gas or the rate at which it is introduced into the process chamber inside the furnace.
When manufacturing semiconductor wafers, the presence of oxygen during crucial manufacturing steps may render a wafer useless. Specifically, when the wafer is thermally processed in the furnace, free oxygen or water vapor on or near the wafer's surface may bond with silicon in the wafer to form an oxide, such as silicon dioxide. Oxides are nonconductive films. Should such a reaction occur on parts of the wafer where electrical conductivity is necessary, the wafer's functionality may be completely destroyed. Accordingly, even trace amounts of oxygen (on the order often parts per million) may disrupt the semiconductor manufacturing process. Oxygen in unwanted manufacturing areas may lead to production delays and excessive costs.
Further, oxygen must be limited or eliminated in areas other than the furnace. In staging or handling areas, such as the wafer boat handling unit (BHU), oxygen concentration must be minimized in order to prevent the gas from migrating into the furnace with the wafers. When a wafer carrier (or "boat") is pre-heated prior to entering the furnace, oxidation may begin even before the wafers are moved into the process chamber. The movement of the boat or wafer carrier into the furnace is colloquially referred to as a "boat push." Prior art systems have minimized oxygen in staging areas by introducing nitrogen in large quantities into a sealed chamber. Typically, nitrogen is introduced through a manually-set mass flow controller, which maintains a steady nitrogen flow into the chamber. While pressurized nitrogen is pumped into the chamber, the ambient gas inside the chamber is evacuated through a purge or bypass valve. When the nitrogen sufficiently dilutes the oxygen inside the chamber (as measured by an oxygen sensor), the purge valve is closed.
Sometimes, however, oxygen may migrate into the chamber through cracks or other flow paths. This can cause the oxygen level to rise beyond the maximum limits. In such a case, the bypass valve is again opened until the oxygen concentration drops. Thus, the bypass valve may endlessly cycle in an attempt to maintain the proper oxygen density.
Additionally, because the mass flow controller is manually set, it constantly introduces nitrogen into the BHU. This causes two problems. First, nitrogen is relatively expensive. The constant gas flow increases manufacturing costs. Second, since the nitrogen is pressurized, the chamber pressure may slowly increase when the purge valve is closed.
Further, such a system may control oxygen density, but exerts little to no control over the chamber pressure. Overpressurizing the BHU may cause panel distortion, which accelerates oxygen migration. In other words, high pressure yields a greater oxygen concentration, and requires further valve cycling.
Generally, prior art solutions have ignored pressure control in this situation because only a single control variable (nitrogen flow) is present in the system. Most conventional systems cannot use a single variable to control two outputs, namely pressure and oxygen density. Accordingly, pressure control is ignored in order to minimize the chance of oxidizing a wafer.
Accordingly, there is a need for an apparatus and method to overcome the aforementioned problems.
BRIEF SUMMARY OF THE INVENTION
Generally, one embodiment of the present invention takes the form of a gaseous flow control system. The flow control system monitors and controls gaseous levels and gas flow into and out of an enclosed space, such as a boat handling unit (BHU) in a semiconductor manufacturing environment. In one embodiment, the control system minimizes oxygen density within the boat handling unit, although other embodiments may monitor and minimize the concentration of other gases.
A first sensor monitors a first gas density within the chamber, while a second sensor monitors the chamber pressure. The sensors may be discrete or integrated. Each sensor transmits its monitoring data to a controller. The controller may manipulate a flow controller in order to adjust either oxygen density or the chamber pressure. Generally, the flow controller takes the form of a mass flow controller, which adjusts nitrogen flow into the chamber. By opening the flow controller, more nitrogen enters the boat handling unit, thus minimizing oxygen concentration inside the BHU. However, since the nitrogen is pressurized, this also increases the BHU internal pressure. In one embodiment, the mass flow controller may assume a variety of positions between fully open and fully closed.
Similarly, the controller may also open or close a purge valve. The purge valve controls evacuation of the gases from inside the BHU. When the purge valve is opened, a low-pressure vent drains gas from the BHU interior. When the valve is closed, no escape route exists for such gas. Accordingly, opening the purge valve reduces pressure inside the BHU, while keeping it closed increases pressure. Further, because oxygen levels inside the boat handling unit remain relatively static (with the exception of some oxygen migration through cracks or spaces in the BHU walls), flushing the gas mixture inside the BHU through the purge valve may rapidly reduce the oxygen density within the BHU. This is especially true when the evacuated gas is replaced by nitrogen flowing through the mass flow controller. The present invention also contemplates a method for regulating an operational parameter of a chamber. Generally, the method comprises the steps of monitoring a first gas density, monitoring a chamber pressure, calculating a proportional contribution based on the first gas density, determining whether the chamber pressure is within a normal pressure range, in the event the chamber pressure is not within the normal pressure range, calculating an integral contribution based on the first gas density, and adjusting the operational parameter based on the proportional contribution and integral contribution. Operational parameters may include chamber pressure and first gas density.
More specifically, the method may be implemented by a proportional/integral derivative (PID) controller, which in turn is implemented via software. Alternate embodiments may use different types of controllers (for example, fuzzy logic), or may use a hardware implementation. The PID controller regulates nitrogen flow into the chamber, and may adjust either chamber pressure or oxygen density through its operation. Generally, the proportional term of the controller is applied to oxygen density, while the integral term of the controller may be applied to either oxygen density or chamber pressure. Typically, the integral term applied to oxygen density only if the pressure is within acceptable levels. The proportional and integral terms are used to calculate a proportional and integral contribution, respectively. In turn, the proportional and integral constants may be used to adjust nitrogen flow into the chamber. If the BHU pressure exceeds a maximum value, oxygen density is ignored in favor of applying the integral term to reducing pressure.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 displays an embodiment of the present invention. Fig. 2 displays an exemplary operating environment for an embodiment of the present invention.
Fig. 3 is a table displaying exemplary operating parameters for the embodiment of Figs. 1 and 2.
Fig. 4 displays a flowchart detailing the operation of the embodiment of Figs. 1 and 2.
DETAILED DESCRIPTION OF THE INVENTION
1. General Overview
Generally, one embodiment 100 of the present invention takes the form of a gaseous flow control system, as shown in Fig. 1. The embodiment 100 monitors and controls gaseous levels and gas flow into and out of the chamber 120, or another enclosed space. The embodiment 100 may monitor levels and control flow of the same type of gas, or may monitor gas levels of a first gas and control flow of a second gas. The present embodiment monitors oxygen (O2) levels and controls nitrogen (N2) gas flow levels.
Operation of the present embodiment 100, including both monitoring and flow control functions, is generally carried out by a controller 110. The controller 110 is operationally connected to one or more sensors 130, which monitor both pressure and gas levels within the chamber 120. Typically, the sensor(s) 130 electrically connect to the controller 110. Alternate embodiments may use different connections, such as a pressure or other mechanical connection. Gas and pressure data is relayed from the sensor 130 to the controller 110. The controller 110 is also operationally connected to a flow controller 165. The controller 110 may open, close or throttle the flow controller 165, thus altering the flow rate of a gas into the chamber 120 through an inlet port 140. The flow controller 165 is placed in a gas pipe or line 160 connecting a gas source 150 to the chamber 120.
Additionally, the controller 110 may be operationally connected to a purge or outlet valve 170 (referred to herein as a "purge valve"). Typically, the controller and purge vale are electrically connected, so that the controller 110 may transmit electrical control signals to the purge valve 170. The purge valve 170 is placed inline along a pipe 190 leading from an outlet 180 in the chamber 120 to a vent 195 or other low pressure area. When the purge valve 170 is opened, gas escapes from the chamber 120, through the pipe 190, and ultimately to the vent 195. As with the flow controller 165, the controller 110 may open or close the purge valve 170 as necessary to ensure proper gas levels inside the chamber 120. In the present embodiment 100, the valve has only two states. In alternate embodiments an adjustable, or throttling, valve may be used as a purge valve. In the present embodiment 100, the controller 110 takes the form of a proportional/integral/derivative (PID) software controller. The operation of a PID controller is well-known to those skilled in the art. The present embodiment's software controller 110 assigns the proportional term of the PLD algorithm to gas flow through the flow controller 165. Unlike a standard PID controller, the software controller 110 includes two integral terms, rather than just one. The first integral term adjusts the pressure of the chamber 120 (i.e., the "pressure variable") while the second integral term is assigned to the chamber's monitored gas level. The software controller 110 neither associates a variable with nor calculates a derivative term. Alternate embodiments may assign a variable to the derivative term, such as the rate of change of the pressure in the chamber 120, a gas level in the chamber, and so forth. Alternate embodiments may also assign the proportional (P) and/or derivative (D) terms to the pressure variable.
Although the present embodiment 100 includes a software controller 110, altemate embodiments may employ a variety of PLD controls. For example, a PID controller may be implemented as a hardware solution, such as a programmable logic controller (PLC), a custom-manufactured control board, and so on. Further, alternate embodiments may employ different control schemes other than PID logic, such as pure proportional (or error) control or a fuzzy logic implementation. Returning to the discussion of the present embodiment 100, the software implementing the controller 110 is typically resident on a computer located at a monitoring station or in a control room. The computer may be of any type known to those skilled in the art, including minicomputers, microcomputers, personal or desktop computers, UNIX stations, SUN stations, network servers, and so forth. Still with reference to Fig. 1, in the present embodiment 100, the sensor 130 is a combined oxygen level and pressure sensor. Of course, alternate embodiments may use two discrete sensors, one to detect gas levels and one to detect chamber 120 pressure. Generally, oxygen levels are detected and expressed in parts per million (ppm), while pressure is detected and expressed in Torr. A Torr is a unit of pressure equal to 1/760* of an atmosphere, or approximately the pressure exerted by a one millimeter column of mercury. As previously mentioned, the sensor 130 is electrically connected to the controller 110.
The flow controller 165 of the present embodiment 100 is typically a mass flow controller, although other embodiments may employ different types of flow controllers, such as a pressure flow controller, a volume flow controller, or a flow rate controller. The mass flow controller 165 receives commands from the PID controller 110 and accordingly adjusts the mass flow rate of gas into the chamber 120. In the present embodiment 100, the flow controller 165 may be fully open, fully closed, or throttled between the two states. Unlike the purge valve 170, discussed above, the flow controller 165 may variably adjust gas flow. Alternate embodiments may use one or more two-state (i.e., open and closed) mass flow controllers.
Thus, the embodiment 100 generally monitors two variables - oxygen level and pressure within the chamber 120. Similarly, the embodiment 100 may change both variables only through a single input, namely, gas flow through the mass flow controller 165 into the chamber 120. (It should be noted that the term "chamber" as used in this document refers generally to any substantially sealed area or enclosure rather than specifically to a room.) By varying gas flow through the mass flow controller 165, the embodiment 100 may alter both the chamber 120 pressure and gas density detected by the sensor 130. In the present embodiment 100, nitrogen is fed through the mass flow controller 165. Generally speaking, the more nitrogen introduced into the chamber 120, the lower the oxygen density becomes. When the embodiment 100 initially begins operation, the chamber 120 contains relatively high oxygen levels. This is detected by the oxygen sensor 130, which in turn relays the data to the controller 110. In response, the PID controller 110 opens the mass flow controller 165, introducing nitrogen into the system. Feeding nitrogen into the chamber 120 positively pressurizes the chamber, thus evacuating gases already in the chamber through the purge valve 170. Since nitrogen is continually fed into the chamber 120 while the gas mix is evacuated, with time the nitrogen concentration inside the chamber increases while the oxygen concentration decreases. Nitrogen flow is maintained by applying a calculated proportional term of the PID software controller 110 to the nitrogen flow. Once the sensor 130 detects that the oxygen concentration has reached a nominal level, the controller 110 closes the purge valve 170.
Once proper oxygen levels have been achieved inside the chamber 120, the integral term of the PID software controller 110 maintains the nitrogen flow through the mass flow controller 165 necessary to stabilize the gas mix inside the chamber. This may be accomplished in one of two fashions depending on the type of mass flow controller 165 used. First, the software controller 110 may "flutter" the mass flow controller 165, oscillating between a completely open and a completely closed position. Generally, this is used only with a non-variable MFC. Alternately, the PID controller 110 may throttle the mass flow controller 165 as necessary to ensure constant flow. Ideally at this point the chamber 120 pressure should be approximately 2 - 4 Torr above atmospheric pressure. By slightly positively pressurizing the chamber 120 migration of oxygen and other gases from the atmosphere in the chamber exterior is eliminated. However, if the chamber 120 is too greatly overpressurized, chamber walls may begin to distort and bow. When the chamber walls distort, stress is placed on seals between the walls and/or around inflow and outflow regions of the chamber. Such stress causes the seals to more rapidly degrade, and may eventually lead to the formation of unintended airflow paths into and out of the chamber 120. In order to avoid overpressurization, the sensor 130 continuously monitors the chamber 120 pressure. If the sensor 130 detects the chamber 120 is overpressurized, it relays this data through the PID controller 110. The controller 110, in turn, may lower the chamber 120 pressure, first by reducing gas flow through the mass flow controller 165, and second by opening the purge valve 170. Although the chamber's oxygen content is monitored during depressurization it is ignored until the chamber 120 pressure returns to an acceptable state. That is, even if the depressurization process results in oxygen concentration exceeding the maximum of acceptable value as determined by the PID controller 110, nitrogen flow through the mass flow controller 165 is not increased until proper pressure is attained. Effectively, the embodiment 100 regards an overpressurized state as overriding an over-oxygenated state. Alternate embodiments may reverse this classification.
2. Operating Environment
Fig. 2 displays an exemplary operating environment for an embodiment 200 of the present invention. Specifically, the operating environment is a boat handling unit (BHU) chamber 220 of a semiconductor manufacturing environment. The boat handling unit 220 generally accepts and stores semiconductor wafers prior to processing the wafers in a semiconductor furnace 205. The term wafer is used broadly herein to indicate any substrate containing a plurality of integrated circuits, one or more flat panel displays, and the like. Generally, wafers are loaded into the BHU 220 from a front-opening unified pod (FOUP) 201, or wafer cassette. Wafers are taken from the FOUP 201 by a robotic arm 202 and inserted into a wafer carrier 203.
The carrier 203 is raised through the BHU 220 and into the fumace 205 by an elevator 204 located beneath the carrier. Wafers placed in the furnace 205 by this boat push are then processed according to methods well known to those skilled in the art. For example, vapor deposition is used to place a film on the wafer surface without causing a chemical reaction with the underlying wafer layer.
Prior to loading the carrier 203 into the furnace 205, the carrier 203 and associated wafers may be pre-heated to speed up processing. Should oxygen remain on the wafer surface during deposition in the furnace 205 and/or pre-heating, a dielectric film may be formed on the wafer. This ruins some or all of the wafer's functionality. Oxygen can cling to the wafer while it is stored in the BHU chamber 220. Accordingly, prior to processing, the oxygen content of the BHU 220 should be minimized. The embodiment 200 operates in much the same way previously described in order to reduce the oxygen (or ambient atmosphere) concentration within the BHU 220. The sensor 230 monitors both BHU chamber 220 pressure and oxygen concentrations. This data is relayed to the software PID controller 210, which in turn adjusts nitrogen flow through the mass flow controller 265 into the BHU 220. The PID controller 210 may also open or close the purge valve 270 to either evacuate oxygen from the chamber or to avoid overpressurization.
Additionally, the PID controller 210 is connected to an elevator 204 control system (not shown). The elevator 204 will not push the wafer carrier 203 into the furnace 205 until the oxygen level inside the BHU chamber 220 falls below a maximum threshold. This threshold may be adjusted by changing a setpoint within the PID software controller 210.
3. Operational Parameters In the embodiment 200 discussed with respect to Fig. 2, operation of the controller 210 is controlled by one or more user settable values. In the embodiment of Fig. 2, the user settable values include: 1) a desired oxygen level inside the BHU 220; 2) a warning oxygen level inside the BHU; 3) a maximum allowable oxygen level inside the BHU chamber 220; 4) a minimum pressure inside the BHU; 5) a maximum allowable pressure inside the BHU; and 6) a safety pressure inside the BHU. Each of these variables will be described in turn. Typically, these variables are implemented as software, but alternative embodiments may employ hardware implementations for each variable.
The "desired oxygen level" variable indicates the level of oxygen in the BHU chamber 220 the controller 210 ideally maintains by adjusting the positions of the MFC 265 and purge valve 270. In the present embodiment 200, this variable is set to ten parts per million (ppm).
The "warning oxygen level" variable is set to an oxygen concentration that, while still acceptable for the BHU chamber 220, is higher than the desired oxygen level. In the present embodiment 200, this variable is set to twenty ppm. The "maximum oxygen level" variable indicates the maximum allowable oxygen concentration in the BHU 220 during a boat push. In the present embodiment 200, this value is nominally set to thirty parts per million.
The next user-settable variable, in the present embodiment 200, is the "minimum chamber pressure" variable. The value of this variable indicates the bottom of the standard or normal pressure range maintained by the controller 210 inside the BHU chamber 220. Still with respect to the present embodiment, the minimum pressure variable is nominally set to two Torr above atmospheric pressure. Another user-settable variable is the "maximum chamber pressure" variable, the value of which represents the upper bound of the standard or normal range of acceptable pressure inside the BHU chamber 220. In the present embodiment 200, this variable is set to four Torr above atmospheric pressure.
The final user-selectable variable in the present embodiment 200 is the "safety chamber pressure" variable. Generally, this variable dictates a safety threshold for the pressure in the BHU chamber 220. The controller 210 operates to keep pressure below the safety chamber pressure variable value. In the present embodiment, this value is set to nine Torr above atmosphere.
Generally, the present embodiment 200, through the controller 210, employs these variables to control both the oxygen and pressure levels inside the BHU 220. The oxygen and pressure levels are generally sampled by the sensor 230. An example of the embodiment's operation will be given for illustrative purposes.
Beginning the example, when the oxygen concentration exceeds the value of the maximum oxygen level variable, the controller 210 opens the purge valve 270 to allow greater flow of nitrogen through the system. This, in turn, aids in rapidly reducing oxygen concentration in the BHU 220. In some embodiments, the purge valve may be opened when the oxygen concentration equals the maximum oxygen level.
Similarly, if the oxygen level drops below the warning oxygen level, the purge valve 270 is closed by the controller 210, since the oxygen level is approaching a desired level, and any pathways permitting oxygen to migrate back into the BHU 220 should generally be minimized. Thus, the purge valve 270 is opened if the oxygen density in the BHU 220 exceeds the maximum oxygen level and closed if the oxygen concentration is less than the warning oxygen level.
Continuing the example, if the BHU 220 oxygen density is less than or equal to the desired oxygen level (i.e., the value of the desired oxygen variable) and the BHU pressure is between the value of the minimum chamber pressure and maximum chamber pressure variables, operating conditions for the BHU 220 are generally in a typical, or normal, range. Accordingly, the controller 210 closes purge valve 270 (if the purge valve is open), and maintains a gas flow rate through the MFC 265 sufficient to maintain the chamber pressure in the normal range, as well as oxygen density at or below the value of the desired oxygen variable. Typically, the MFC 265 remains partially open in order to replace any nitrogen lost through cracks of other transient pathways from the BHU 220 interior to the outside. As yet another part of the example, if the BHU 220 pressure exceeds the maximum chamber pressure variable, then the calculation of the flow rate through the MFC 265 (as performed by the controller 210) will emphasize reducing the nitrogen flow in order to bring BHU pressure back down into the normal range, even if the oxygen level is higher than desired. In other words, the controller 210 emphasizes maintaining a desired pressure over a desired oxygen level. Similarly, if the chamber 220 pressure drops below the minimum BHU pressure variable value, then controller 210 will increase gas flow through the MFC 265 in order to bring the chamber pressure back up into the normal range, even if this adversely affects the oxygen concentration in the chamber. As a final part of the example, if the BHU 220 pressure ever exceeds the value of the safety chamber pressure variable, then the controller 210 closes the MFC 265, and opens the purge valve 270. Accordingly, with no gas inflow and the purge valve open, excess pressure should generally bleed off quickly. In the present embodiment 200, such a condition rarely (if ever) occurs during normal operation. Nonetheless, such functionality is provided for the controller 210 in order to ensure system safety.
It should be understood that the particular variables mentioned above, as well as the particular values given, are exemplary only. Alternative embodiments may employ different variables, or may use different values for such variables. Further, alternative embodiments may vary the values of one or more variables based on environmental or other changing conditions. Accordingly, the above information is provided by way of example and not limitation.
4. Operation of the Embodiment Figure 3 is a table 300 displaying an exemplary set of constants used by the proportional/integral/derivative controller 110 of the embodiment 100 of Fig. 1. (These constants may be used by the embodiment 200 of Fig. 2, as well.) It should be understood that these constants may vary in alternative embodiments, and may be zero or non-zero values, as necessary. Generally, a zero value for any given variable suppresses the operation of that portion (i.e., proportional, integral, or derivative) of the controller 210.
Generally, the table 300 is divided into three rows and six columns. The uppermost row represents the values for constants when the chamber 120 pressure is below the minimum chamber pressure variable. The middle row depicts constant values when the chamber pressure is in between the minimum and maximum chamber pressure variables. The bottommost, and final, row represents the values for constants when the chamber 120 pressure exceeds the maximum chamber pressure variable. Generally, the columns represent different constants. For example, the "O2-
P" column represents the proportional oxygen band, discussed below with respect to step 435 of Fig. 4. The "O2-I" column represents an oxygen integral constant, discussed below with respect to step 450 of Fig. 4. The "O2-D" column represents an oxygen derivative constant, which may be used by the controller 110 in some embodiments to adjust gas flow through the MFC 165. The "Press-P" column represents a proportional pressure constant. As with the oxygen derivative constant, this constant may be used by an embodiment of the present invention to control gas flow through the MFC 165, and correspondingly control oxygen density and overall pressure within the chamber 120. The "Press-I" column represents a pressure integral constant, which may be used for similar control calculations by the controller 110. Finally, the "Press-D" column represents a derivative pressure constant, which may also be used by the controller 110 to calculate a gas flow through the MFC.
The values shown in Fig. 3's table 300 are exemplary for the present embodiment 100. The zero values in all three rows of the "O2-D", "Press-P", and "Press-D" columns indicate that these particular constants are not employed by the embodiment. Alternative embodiments, however, may make use of such constants.
Figure 4 displays a flowchart detailing the general operation of the embodiment 100 shown in Fig. 1. This flowchart generally applies to the embodiment shown in Fig. 2, as well. The steps shown in Fig. 4 are typically executed by a software or hardware system logic. Such a system logic is often (though not always) implemented as part of the proportional/integral/derivative controller 110.
The operational process begins in step 400, where the PID controller 110 opens the purge valve 170, and nitrogen begins flowing through MFC 165.
Once step 400 is completed, the embodiment 100 executes a loop comprising steps 405 to 465. This is typically executed on a periodic basis for a set period of time or after a certain time period has passed, for example in one embodiment the loop is executed about every four seconds. Alternative embodiments include executing the loop in response to a trigger, such as a change in an environmental variable. Exemplary triggers may include a variable being exceed or a user-initiated command. In yet another alternative embodiment this loop may be executed only once, or it may executed a set number of times.
First, in step 405, the embodiment 100 determines if the chamber 120 pressure exceeds the value of the safety chamber pressure variable, mentioned above. If so, the embodiment executes step 410. In step 410, the controller 110 closes the MFC 165 and opens the purge valve 170. After completing step 410, the embodiment 100 executes step 465, in which it waits for the next iteration of the loop. If the embodiment 100 determines the chamber 120 pressure is below the safety chamber pressure variable's set value, then step 415 is executed. In step 415, it is determined if the oxygen concentration exceeds the maximum oxygen variable's value. If so, then step 420 is executed and the purge valve 170 is opened (if it is not already open). After step 420, the embodiment executes step 435, as detailed below.
If the oxygen density does not exceed the maximum oxygen level variable in step 415, then step 425 is executed. In this step, the embodiment 100 determines whether the oxygen concentration in the chamber 120 is less than the warning oxygen level variable's value. If the oxygen concentration is above this value, step 435 is accessed. Otherwise, step 430 is executed. In step 430, the controller 110 closes the purge valve 170, if it is open.
In step 435, the embodiment 100 determines a "proportional contribution" for the gas flow through the MFC 165. This proportional contribution is based on the current chamber 120 oxygen concentration. The proportional contribution generally equals the difference between the current oxygen level and the desired oxygen level, divided by the size of an "oxygen proportional band." The oxygen proportional band is a user-changeable parameter that defines a range in which the proportional term of the controller 110 responds in a proportional manner.
Mathematically expressed, the proportional contribution may be determined in the following manner:
Pc = (O2C - O2D )/ O2P; where
Pc = proportional contribution;
O2C = current chamber 120 oxygen concentration;
O2D = desired chamber 120 oxygen concentration (i.e., the "desired oxygen level" variable discussed above); and O2P = oxygen proportional band (in the present embodiment 100, shown in column "O2-P" above).
As an example, the proportional band in an embodiment of the present invention may be 1000 ppm (as opposed to the 5000 shown in Fig. 3), while a desired oxygen level variable might be 10 ppm. Accordingly, a current chamber oxygen concentration of 110 ppm would result in a proportional contribution of 10% of the MFC maximum flow rate:
(110 - 10)/1000 = 10% of MFC flow capacity After step 435 is completed, the embodiment 100 determines in step 445 if the current chamber 120 pressure is between the minimum chamber pressure and maximum chamber pressure variables. If the pressure is within this normal range, then step 450 is executed.
In step 450, the embodiment 100 calculates an "integral contribution." The integral contribution is used by the controller 110 to at least partially control gas flow through the MFC 165. Generally, the integral contribution represents a gas flow into the chamber 120, expressed here as a percentage of the maximum gas flow through the MFC 165. The integral and proportional contributions are generally used to maintain operating parameters within the chamber 120. "Operating parameters" here generally refer to a chamber 120 oxygen density and a chamber pressure. The proportional and integral contributions together assist in determining a gas flow rate through the MFC 165, as explained in more detail in step 460.
Continuing with the discussion of step 450, the embodiment computes the integral contribution by taking the difference between the current chamber 120 oxygen concentration and the desired oxygen concentration in the chamber, then multiplying it by an oxygen integral constant. This formula may be mathematically expressed as follows:
Ic = (O2C - O2D )* O2I; where
Ic = integral contribution; O2C = current chamber 120 oxygen concentration;
O2D = desired chamber 120 oxygen concentration (i.e., the "desired oxygen level" variable discussed above); and
O2I= oxygen integral constant (in the present embodiment 100, shown in Fig. 3 in the "O2-I" column). As an example, if the current chamber 120 oxygen level is 110 ppm and an embodiment employs an oxygen integral constant of 0.01 and a desired oxygen level variable of 10 ppm, the integral contribution is:
(110 - 10) * 0.01 = 1% of MFC flow capacity
If the embodiment 100 determines in step 445 that the chamber 120 pressure is outside the normal range, then step 440 is executed. In step 440, the embodiment calculates the integral contribution to be added for this loop iteration based on the chamber pressure (as opposed to basing the integral contribution on oxygen density, as in step 450). In step 440, the embodiment 100 calculates the difference between the current chamber 120 pressure and the minimum pressure (if the current chamber pressure is below the minimum pressure variable) or maximum pressure (if the current chamber pressure is above the maximum pressure), and multiplies this difference by a pressure integral constant. The resulting number is the integral contribution. Mathematically expressed, the formula for the integral constant, when the current chamber pressure is less than the minimum chamber pressure variable: Ic = (Cvar - Ccurr )* PI; where
Ic = integral contribution;
Cvar= chamber pressure variable exceeded (either the minimum or maximum chamber pressure variables discussed in Section 3., above); Ccuπ-= curcent chamber 120 pressure; and
PI= pressure integral constant (in the present embodiment 100, shown in Fig. 3 in the "Press-I" column).
For example, in an embodiment with a current chamber 120 pressure of 1.9 Torr, a minimum chamber pressure variable of 2 Torr, and a pressure integral constant of 0.1, the integral contribution is:
(2 - 1.9) * 0.1 = 1% of MFC flow capacity
Another example may be helpful. In an embodiment having a maximum chamber pressure variable set to 4 Torr, a current chamber 120 pressure of 4.2 Ton, and a pressure integral constant of .1, the integral proportion would be: (4.0 - 4.2) * 0.1 = -2% of MFC flow capacity.
Once the integral contribution is determined in either step 440 or step 450, step 455 is executed. In this step, the embodiment 100 adds the integral contribution calculated to an "integral sum." The integral sum is the sum of all integral contributions calculated in previous loop iterations. Accordingly, the integral sum integrates the integral contributions over successive loop iterations. Using the above example where the integral contribution was -2%, and assuming a previous integral sum of 65%, the new integral sum would be 63%. Generally, a rise in the integral sum represents an increase in gas flow through the MFC 165, while a decrease in the integral sum represents a reduction in the flow through the MFC.
After step 455 is completed, step 460 is executed. In step 460, the embodiment 100 determines the final flow rate (or "operating percentage") for the MFC for the particular loop iteration. The flow controller 165, in turn, implements the operating percentage by adjusting the position of the MFC. Generally, the MFC 165 flow rate is calculated by adding the proportional term calculated in step 435 to the integral sum calculated in step 455. Continuing the example above, the proportional term was set to 10% in step 435, and the integral sum set to 63% in step 455. Accordingly, the new MFC 165 operating percentage is 73% of its maximum possible flow. Thus, the flow controller 165 sets the MFC 73% open.
After step 460 is completed, step 465 is executed. In this step, the embodiment 100 waits for a timer to expire to initiate the next loop cycle.
In an alternative embodiment in which the loop is executed in response to a trigger, such as a change in either the oxygen density or pressure inside the chamber 120, once the trigger is detected, the loop begins again at step 405. Such changes typically occur, due to the controller's 110 adjustment of the positioning of the MFC 165
5. Conclusion As will be recognized by those skilled in the art from the foregoing description of sample embodiments of the invention, numerous variations on the described embodiments may be made without departing from the spirit and scope of the invention. For example, a gaseous control system may use different setpoints, flow controller, implementations of a PID/PI controller, and so on. Further, while the present invention has been described in the context of specific embodiments and processes, such descriptions are by way of example and not limitation. Accordingly, the proper scope of the present invention is specified by the following claims and not by the preceding examples.