scispace - formally typeset
Search or ask a question

Showing papers on "Process variable published in 1996"


Journal ArticleDOI
TL;DR: Statistical process control methods for monitoring processes with multivariate measurements in both the product quality variable space and the process variable space are considered.
Abstract: Statistical process control methods for monitoring processes with multivariate measurements in both the product quality variable space and the process variable space are considered. Traditional multivariate control charts based on X2 and T2 statistics ..

644 citations


Journal ArticleDOI
TL;DR: In this paper, the effects of restraining force on the springback phenomena when stamping channels from aluminum sheet are investigated and a strategy to control the binder force during the forming operation in order to reduce springback and simultaneously avoid tearing failure is described.
Abstract: Springback in a forming process is due to the elastic deformation of the part during unloading This manufacturing defect can be accounted for through proper tooling design or through proper design and control of the magnitude and history of restraining force Using finite element analyses of the process : (1) the effects of restraining force on the springback phenomena when stamping channels from aluminum sheet are investigated ; (2) a strategy to control the binder force during the forming operation in order to reduce springback and simultaneously avoid tearing failure is described ; and (3) a binder force control strategy which provides robustness in the presence of process parameter uncertainty is implemented The process history and controller designed using finite element analyses is then experimentally verified : excellent agreement between simulation and experiments is obtained A binder force history, which leads to a significant reduction in the amount of springback incurred by the formed part without reaching critical stretching conditions, was proposed Although an optimal forming history was found, in order to ensure that part shape error remained minimized even in the event of variations in processing parameters such as friction, a closed-loop control algorithm was developed whereby the binder force is altered during the process in order to provide a robust, repeatable stretching history Experiments were performed using a double-action servo-controlled process and were found to produce the desired results demonstrating both the accuracy of the numerical simulation and the success of the proposed active-binder force control method to obtain net shape

75 citations


Patent
19 Dec 1996
TL;DR: In this article, a processor for processing a time domain reflectometry (TDR) signal having a plurality of reflection pulses to generate a valid output result corresponding to a process variable in a vessel and a method for a process measurement signal.
Abstract: The present invention provides a processor apparatus for processing a time domain reflectometry (TDR) signal having a plurality of reflection pulses to generate a valid output result corresponding to a process variable in a vessel and a method for a process measurement signal. One method includes the use of at least two different techniques for detecting a valid reflection pulse generated by the process variable to calculate an independant result using each of the at least two techniques, and applying a wheighted factor to the independant results from each of the at least two different techniques to provide weighted results. Another method includes establishing an initial boundary signal before the process variable is located in the vessel, storing the initial boundary signal, detecting the TDR signal, determining a baseline signal by subtracting the initial boundary signal from the TDR signal, determining a maximum value of the reflection pulse and calculating an output result based on the maximum value.

53 citations


Patent
09 Aug 1996
TL;DR: In this article, an auto-tuning circuit is proposed to tune the control function to the process model based on a process model, which can be implemented with hardware or software in low power and low memory applications.
Abstract: A process control system (100) controls a process (110) through a control output signal (u(t)) based on a set point (YSET) and a measured process variable (y(t)). The process control system (100) includes a control circuit having a set point input, a process variable input and a control output. The control circuit generates the control output signal (u(t)) on the control output as a function of the set point (YSET) received on the set point input and the measured process variable (y(t)) received on the process variable input. An auto-tuning circuit (106) excites the process (110), estimates a process model based on a rising dead time, a rising rate-of-change, a falling dead time and a falling rate-of-change in the measured process variable (y(t)) and then tunes the control function to the process (110) based on the process model. The auto-tuning circuit (106) obtains robust results, but is computationally simple such that the circuit can be implemented with hardware or software in low-power and low-memory applications, such as in field-mounted control units.

53 citations


Proceedings ArticleDOI
06 May 1996
TL;DR: In this paper, the influence of component and mold design decisions as well as process parameters selection on the process energy efficiency through a framework that integrates a 3D solid modeling environment, numerical analysis of the filling and post-filling behavior, and an energy-based process model.
Abstract: A significant part of the development of electronics products is the design of injection-molded thermoplastic parts ranging from chassis panels and enclosures to connectors. While much effort has been spent on analyzing the end-of-life issues associated with thermoplastic parts, such as disassembly, separation, and recycling, a primary driver for design-for-environment of plastic parts is the energy expended in the forming process. Traditional part design and process parameter selection, which is performed with the goal of minimizing process cycle time, often overlooks environmental issues such as energy efficiency. This paper analyzes the influence of component and mold design decisions as well as process parameters selection on the process energy efficiency through a framework that integrates a 3D solid modeling environment, numerical analysis of the filling and post-filling behavior, and an energy-based process model. The process model that is presented describes the modes of energy utilization in injection molding processes as they relate to design-for-environment concerns. Case studies are performed on a center-gated plastic disk, and a small plastic enclosure to illustrate the use of this framework.

39 citations


Patent
25 Jun 1996
TL;DR: In this paper, an abrasive waterjet cutting system is monitored and controlled by measuring at least one process parameter and using that measurement as a comparison with a running average of the process parameter as it may change over time.
Abstract: An abrasive waterjet cutting system is monitored and controlled by measuring at least one process parameter and using that measurement as a comparison with a running average of the process parameter as it may change over time. Optionally, the rate of change of the process parameter may also be used to signal an emergency condition, and extreme outer limits of the process parameter may be used to define unacceptable operating conditions.

33 citations


Patent
07 Jun 1996
TL;DR: In this article, a method for refining lignocellulose-containing material into pulp in a disc refiner comprises preheating the material to a temperature greater than the glass transition temperature of lignin in the material, and holding this temperature for under one minute.
Abstract: A method for refining lignocellulose-containing material into pulp in a disc refiner comprises preheating the material to a temperature greater than the glass transition temperature of lignin in the material, and holding this temperature for under one minute. The heated material is then subject to high speed refining in a disc refiner to produce pulp. The resulting pulp may then be subject to secondary refining steps to produce paper quality pulp. The preheat retention time is preferably in the range of 5-30 seconds, and can be controlled as a process variable to optimize energy savings, pulp strength, and optical qualities. High quality pulp can be obtained with preheat at high temperature and low retention time, followed by primary refining at disc speed of at least 2300 rpm.

25 citations


Journal ArticleDOI
TL;DR: In this article, an advanced multivariable in-line process control system, which combines traditional statistical process control (SPC) with feedback control, has been applied to the CVD tungsten process on an Applied Materials reactor.
Abstract: An advanced multivariable in-line process control system, which combines traditional statistical process control (SPC) with feedback control, has been applied to the CVD tungsten process on an Applied Materials reactor. The goal of the model-based controller is to compensate for shifts in the process and maintain the wafer-state responses on target. The controller employs measurements made on test wafers to track the process behavior. This is accomplished by using model-based SPC, which compares the measurements with predictions obtained from process models. The process models relate the equipment settings to the wafer-state responses of interest. For CVD tungsten, a physically-based modeling approach was employed based on the reaction rate for the H/sub 2/ reduction of WF/sub 6/. The Arrhenius relationship for the kinetic model was linearized so that empirical modeling techniques could be applied. Statistically valid models were derived for deposition rate, film stress, and bulk resistivity using stepwise least-squares regression. On detecting a statistically significant shift in the process, the controller calculates adjustments to the settings to bring the process responses back on target. To achieve this, two additional test wafers are processed at slightly different settings than the current recipe. This local experiment allows the models to be updated to reflect the current process state. The model updates are expressed as multiplicative or additive changes in the process inputs and a change in the model constant. This approach for adaptive control also provides a diagnostic capability regarding the cause of the process shift. The adapted models are used by an optimizer to compute new settings to bring the responses back to target. The optimizer is capable of incrementally entering controllables into the strategy, reflecting the degree to which the engineer desires to manipulate each setting. The capability of the controller to compensate for induced shifts in the CVD tungsten process is demonstrated. Targets for film bulk resistivity and deposition rate were maintained while satisfying constraints on film stress and WF/sub 6/ conversion efficiency. The ability of the controller to update process models during routine operation is also investigated. The tuned process models better predict the process behavior over time compared to the untuned models and lead to improved process capability.

25 citations


Journal ArticleDOI
TL;DR: In this paper, a computer model of modular structure complete with an integrated input parameter database for high speed milling based on theoretical considerations is presented, and the results obtained from these experiments correlate well with the process parameters predicted by the computer model.

23 citations


Patent
Anoop K. Mathur1, Tariq Samad1
11 Jul 1996
TL;DR: In this article, a neural network is trained using a process model to approximate a function which relates process input and output data to process parameter values, and the network can be used as a system identification tool.
Abstract: The method of making the tool, for process system identification that is based on the general purpose learning capabilities of neural networks. The method can be used for a wide variety of system identification problems with little or no analytic effort. A neural network is trained using a process model to approximate a function which relates process input and output data to process parameter values. Once trained, the network can be used as a system identification tool. In principle, this approach can be used for linear or nonlinear processes, for open or closed loop identification, and for identifying any or all process parameters.

20 citations


Patent
31 Dec 1996
TL;DR: In this paper, a membrane system controller and control method for maximizing retentate product output and incorporating compressor load detection apparatus and process parameter instrumentation for feeding essential monitoring information to a control unit is presented.
Abstract: A membrane system controller and control method for maximizing retentate product output and incorporating compressor load detection apparatus and process parameter instrumentation for feeding essential monitoring information to a control unit. Product output maximization is accomplished by increasing the membrane operating temperature during periods when compressor capacity is underutilized.

Patent
03 May 1996
TL;DR: In this paper, a method of controlling an output variable of a process having long dead times and delay times, for example for controlling the interior temperature of an absorber refrigerator, and also relates to an apparatus for carrying out such a method.
Abstract: The present invention relates to a method of controlling an output variable of a process having long dead times and delay times, for example for controlling the interior temperature of an absorber refrigerator, and also relates to an apparatus for carrying out such a method. The output variable of the process, such as the interior temperature, is sampled and compared with a desired value in order to determine a desired value deviation. In dependence on the determined desired value deviation, a manipulated variable, such as the power supplied to the absorber, is calculated in such a manner that the output variable of the process is brought near to the desired value. After the lapse of a waiting time, a value of the manipulated variable is newly determined. During the waiting time, the manipulated variable is not altered. The waiting time is calculated on the basis of the difference between the value of the manipulated variable and an earlier value of the manipulated variable. A controller for carrying out the method comprises a memory for an earlier calculated manipulated variable and a timer which is designed to receive the earlier calculated value of the manipulated variable and the current value of the manipulated variable, to calculate a waiting time from these and, after the lapse of this waiting time, to start a new calculation of a value for the manipulated variable.

Patent
Bernd-Markus Pfeiffer1
16 Dec 1996
TL;DR: In this article, a method and device for controlling a self-regulating, delayed process, in particular a temperature control system, in which, after a first setpoint step change, the manipulated variable is set to its maximum value and an IT1 model of the process is identified on the basis of the step response is calculated.
Abstract: A method and device for controlling a self-regulating, delayed process, in particular a temperature control system, in which, after a first setpoint step change, the manipulated variable is set to its maximum value and an IT1 model of the process is identified on the basis of the step response. The time at which the manipulated variable must be reset for time-optimum control is calculated on the basis of this model. An additional, more precise model is determined if the rate of change of the controlled variable drops below a specific limit after the shut-off time. The parameters of a linear controller are set as a function of the more precise model, after which the controller controls the process in a non-linear mode. When additional setpoint step changes occur, a time for switching to a steady-state manipulated variable is calculated on the basis of the more precise model in order to achieve time-optimal control. The described control device adjusts to the process as early as after the first setpoint step change without requiring any knowledge about the process. The method and device may be used with adaptive controllers.

Journal ArticleDOI
TL;DR: In this paper, a review of hydrodynamic, heat-transfer studies and temperature change models related to convection-heated canned foods is made, based on the developed sensitivity functions, a parametrical analysis is performed and the significance of the process parameters and the basic thermo-bacteriological, thermo physical models is established.

Patent
14 Aug 1996
TL;DR: In this article, a neural network process control system is configured with five neural process models A1-5U, each of which has four inputs for signals from different process parameter sensor A6-9U outputs.
Abstract: The neural network process control system is configured with five neural process models A1-5U, each of which has four inputs for signals from different process parameter sensor A6-9U outputs. Each model has an output fed to a common weighting stage A10U that provides an output to a regulator A11U coupled to the process. The regulator can be a fuzzy controller. The process neural models are trained with data sets and the weighting module determines a standard variation value relative to an average value. This is compared with a defined limit and if exceeded an alarm A12U is activated.

Patent
20 Mar 1996
TL;DR: In this paper, a process and apparatus for controlling and/or regulating a solids-liquid separation process, such as in a dewatering or drying device, includes at least two photodetectors, each associated with a light source, a detector to detect light of the associated light source reflected from the surface of the solids liquid mass, a controller generating an output signal as a function of the detected light and converting and processing the output signals of the photoderectors into correcting variables for controlling or regulating at least one process parameter of the separation process.
Abstract: A process and apparatus for controlling and/or regulating a solids-liquid separation process, such as in a dewatering or drying device includes at least two photodetectors, each associated with a light source, a detector to detect light of the associated light source reflected from the surface of the solids-liquid mass, a controller generating an output signal as a function of the detected light and converting and processing the output signals of the photodetectors into correcting variables for controlling and/or regulating at least one process parameter of the separation process. At least one of the photodetectors is associated with at least one second photodetector for detecting light reflected into the photodetector at an angle, preferably an angle of 30° to the main direction of reflection. The output signals of the photodetectors are advantageously processed into correcting variables for the step-wise adjustment of the at least one process parameter to optimize the separation process.

Patent
25 Jun 1996
TL;DR: In this article, the compatibility of a process parameter is inspected based on the check data received from the higher control device, and the result is informed to the higher controller by a host computer and a block controller.
Abstract: PURPOSE: To make it possible to reliably accomplish the assembling and setting treatment of a process parameter by a method wherein the compatibility of the process parameter is inspected based on the check data received from the higher control device, and the result is informed to the higher control device. CONSTITUTION: A host computer 1 and a block controller 2, as the higher control device, and a communication circuit 5, with which a control device 4 and the higher control devices 1 and 2 are connected, are provided and a control panel 6, a main control part 7 and recipe inspection part 8 are provided on the control device 4. The recipe inspection means 8 inspects whether the process parameter, which is going to be assembled into the process recipe, is compatible by comparing the process recipe and the recipe check data sent from the higher control devices 1 and 2. As a result, check data can be transmitted by the process parameter by reforming it to one having no hindrance, and accordingly, the process parameter, sent from the higher control device, can be incorporated into the process recipe quickly.

Journal ArticleDOI
TL;DR: A method which relies on an empirical response surface and an iterative computational technique to minimize process variance is presented and the results show that the method works regardless of the process variable being optimized, i.e. temperature, oxide thickness, etc.
Abstract: Many RTP reactors used in production presently employ single point temperature measurement along with multiple lamp zone control. Optimum zone settings in these systems are known to vary depending on the radiative properties of the wafer and the process. Therefore, tuning the system to meet specifications for a new process is achieved by an operator analyzing metrology results and drawing on experience with the system to determine proper zone settings. It would be desirable to have a technique which can be used to optimize the lamp zone settings in an automatic manner that would be the same for every user. Furthermore, such a technique should be both simple to use and robust. In this work a method is presented which relies on an empirical response surface and an interative computational technique to minimize process variance. The results show that the method works regardless of the process variable being optimized, i.e. temperature, oxide thickness, etc., and that the method relies on no user input.

Proceedings ArticleDOI
02 Sep 1996
TL;DR: BABAS, the automatic simulation system capable of very efficient process control sweeping, statistical variation analysis and output data processing, which has functions for producing input data files automatically and has an easy-to-use GUI has been developed.
Abstract: For the realization of today's miniaturization and low voltage drive of ULSIs, it has become much more important to understand and predict the effect of process variations on device performance. Predicting the quantitative effect of every process variation on device characteristics has the following advantages: in designing, required performance range can be appropriately specified, and in manufacturing, when some device characteristic is out of specification, processes to control can be feed-forwarded for proper production. In order to realize this by utilizing simulations, it would be crucial for the simulation system to have a user interface with ease of operation and an efficient data processing ability based on the actual process parameter distribution of the manufacturing equipment. There are some simulation tools which enable the above tasks. However, they have difficulty in handling mask shape variations and in utilizing actual process distributions which do not follow the Gaussian distribution. We have developed BABAS, the automatic simulation system capable of very efficient process control sweeping, statistical variation analysis and output data processing, which has functions for producing input data files automatically and has an easy-to-use GUI. Furthermore, BABAS has succeeded in reducing CPU time drastically by utilizing moment expansion method and enabled Monte Carlo simulations based on real process parameter variations.

Journal Article
TL;DR: In this paper, the influence of process parameter settings on the microstructure of the gas atomized Ni-B-Si powder was found out, and the interaction between the apex angle and the temperature of the melt shows that the choice of angle is most important when the temperature is low.
Abstract: Background. The structure of an engineering material has great influence on its properties and must therefore be under control to give the desired result. The high cooling rates (10 2 -10 5 K/s) connected with gas atomized powder produce a fine dendritic microstructure. The aim of the present study was to find out the influence of process parameter settings on the microstructure of the gas atomized Ni-B-Si powder. Methods. The parameters studied were gas pressure, superheat, melt nozzle diameter and apex angle. Design of experiments has been used to investigate the complex relationship between process parameters and microstructure. This method allows the detection of interaction between 2 parameters and minimizes the number of experimental runs. Results. Increasing the apex angle results in a finer microstructure of the powder, and increasing the superheat gives coarser microstructure and larger standard deviation of secondary dendrite arm spacing, SDAS. Furthermore, the interaction between the apex angle and the temperature of the melt shows that the choice of angle is most important when the temperature is low. Conclusion. The results shows that the superheat and the apex angle play a key role in the formation of the microstructure of the powder.

Patent
16 Jan 1996
TL;DR: In this paper, an electropneumatic regulator is connected to the air supply pipe, operating pressure control valves 57a, 57b adjusting the supply pressure of steam through a three-way solenoid valve 2 and steplessly operated according to a change of an electric signal.
Abstract: PURPOSE:To suppress the foaming pressure at the time of heating to a necessary min. degree and to efficiently use necessary energy while shortening a molding cycle by making the supply pressure of steam in a heating process variable. CONSTITUTION:An electropneumatic regulator 3 is connected to the air supply pipe 1 operating pressure control valves 57a, 57b adjusting the supply pressure of steam through a three-way solenoid valve 2 and steplessly operated according to a change of an electric signal to make operation air pressure variable to steplessly regulate the pressure control valves 57a, 57b. The adjustment of the supply pressure of steam is performed by detecting foaming pressure changing with the passage of time by a foaming pressure detector 67 to take out the obtained data as an electric signal (e) and controlling the electropneumatic regulator 3 by a control unit 4 so as to make the supply pressure of steam optimum. By this constitution, foaming pressure can be set to necessary min. pressure and the conservation of energy can be achieved by the reduction of the use amt. of steam.

Patent
01 Oct 1996
TL;DR: In this paper, a method combining fuzzy logic and PID complex procedure control is used to generate control output for controlling one process, which comprises of calculating one input goal control value and process variable generated from the above process, calculating one control offset; according to fuzzy logic, by the above control offset generating one auxiliary goal value so as to supply modification to the above goal controller value.
Abstract: A method of combining fuzzy logic and PID complex procedure control, which is used to generate control output for controlling one process, comprises of: (1) according to one input goal control value and process variable generated from the above process, calculating one control offset; (2) according to fuzzy logic, by the above control offset generating one auxiliary goal value so as to supply modification to the above goal control value; (3) adding the above goal control value and the above auxiliary goal value, generating one virtual goal value; and (4) according to the above virtual goal value and the above process variable, by one PID controller generating control output suitable to the above process.

Journal ArticleDOI
TL;DR: ExSENSEL is the successor to TRANSELEX, which is a single expert system for selection of transducers in the area of temperature, pressure, and flow measurement.
Abstract: A novel expert-system-based method for selection of sensors for process variables is presented. EXSENSEL (expert system based sensor selection) deals with 12 process variables (gas chromatography, conductivity, humidity, level, moisture, nuclear radiation, oxygen content, pH, pollution, reaction products, strain, viscosity) and currently has 94 rules in its knowledge base. Despite its large knowledge base, users have to answer a set of only a few questions regarding a particular process variable, which is selected from a menu of 12 variables. A general description of the chosen process variable can be viewed before invoking the rules. Once a sensor has been selected, a brief write up on that particular sensor is also available on user's request. EXSENSEL is the successor to TRANSELEX, which is a single expert system for selection of transducers in the area of temperature, pressure, and flow measurement.

Patent
16 Feb 1996
TL;DR: In this article, a linear process function is proposed to describe the properties of a semiconductor device with linear process functions from the process parameters in a mfg. process for the device, extracting one group of optimal parameters which fulfil the envisaged specification, with the use of the process functions.
Abstract: The semiconductor device mfr. involves describing one property of the semiconductor device with linear process functions from the process parameters in a mfg. process for the device, extracting one group of optimal parameters which fulfil the envisaged specification, with the use of the process functions, and mfg. the semiconductor device by the mfg. process with the extracted process parameter. The step for describing the property of the semiconductor device, incorporates the step of establishing the process functions based on experimental and/or simulated values, and the process functions can then be linearly combined. The process parameters can includes gate oxide film thickness, channel width, side wall width, ion implantation energy for source or drain regions, and heat treatment time.

01 Jan 1996
TL;DR: In this article, an advanced multivariable in-line process control system, which combines traditional statistical process control (SPC) with feedback control, has been applied to the CVD tungsten process on an Applied Materials reactor.
Abstract: Abszi-act- An advanced multivariable in-line process control system, which combines traditional statistical process control (SPC) with feedback control, has been applied to the CVD tungsten process on an Applied Materials reactor. The goal of the model-based controller is to compensate for shifts in the process and maintain the wafer-state responses on target. The controller employs measurements made on test wafers to track the process behavior. This is accomplished by using modelbased SPC, which compares the measurements with predictions obtained from process models. The process models relate the equipment settings to the wafer-state responses of interest. For CVD tungsten, a physically-based modeling approach was employed based on the reaction rate for the Ha reduction of WFs. The Arrhenius relationship for the kinetic model was linearized so that empirical modeling techniques could be applied. Statistically valid models were derived for deposition rate, film stress, and bulk resistivity using stepwise least-squares regression. On detecting a statistically significant shift in the process, the controller calculates adjustments to the settings to bring the process responses back on target. To achieve this, two additional test wafers are processed at slightly different settings than the current recipe. This local experiment allows the models to be updated to reflect the current process state. The model updates are expressed as multiplicative or additive changes in the process inputs and a change in the model constant. This approach for adaptive control also provides a diagnostic capability regarding the cause of the process shift. The adapted models are used by an optimizer to compute new settings to bring the responses back to target. The optimizer is capable of incrementally entering controllables into the strategy, reflecting the degree to which the engineer desires to manipulate each setting. The capability of the controller to compensate for induced shifts in the CVD tungsten process is demonstrated. Targets for film bulk resistivity and deposition rate were maintained while satisfying constraints on film stress and WF6 conversion efficiency. The ability of the controller to update process models during routine operation is also investigated. The tuned process models better predict the process behavior over time compared to the untuned models and lead to improved process capability.

Proceedings ArticleDOI
11 Dec 1996
TL;DR: The cause of ill-conditioned process problems is investigated by focusing on poor PID settings and Optimization calculation techniques are used to identify the closed-loop system and the small gain theorem is used to verify the model.
Abstract: An ill-conditioned process is one in which there exists equipment with unmeasurable variables between the input and output data to be modeled. The identification of closed-loop systems is investigated using actual closed-loop data. Optimization calculation techniques are used to identify the closed-loop system, and the small gain theorem is used to verify the model. The targeted control loop is the temperature control loop of a TW303 distillation tower. The only measurable process variable which greatly affects the TW303 bottom temperature is the reboiler flow rate. Hence, input and output variables of the heat exchanger cannot be measured. The temperature and reboiler flow rate data has a periodic oscillation of 12 minutes. Even though the periodic oscillation is caused by the equipment, the controllers must be able to control any type of disturbance. We investigate the cause of these problems by focusing on poor PID settings.

Proceedings ArticleDOI
13 Sep 1996
TL;DR: In this paper, an in-situ "dark field" particle sensor was used to monitor process defect levels during tungsten deposition, and the results were used to define the CVD Tungsten process operating window in terms of in situ defectivity.
Abstract: Upstream control and in-situ defect monitoring have been successfully implemented to track and improve particle performance for CVD tungsten films. An in-situ 'dark field' particle sensor was used to monitor process defect levels during tungsten deposition. The sensor was mounted directly below the CVD tungsten deposition chamber on the foreline. Information from the in-situ particle sensor and the CVD system process parameter analog output were routed to a stand alone computer where the incoming signals were recorded. In-situ particle and process parameter signals were monitored using the standard production recipe for tungsten deposition. From this data the internal variables (process parameters) affecting in-situ particle levels were determined. The internal factors were varied over a selected operating range and the in-situ particle levels quantified for each factor. These results were used to define the CVD tungsten process operating window in terms of in-situ defectivity. Additionally, other external sources were identified as contributors to in-situ defectivity.© (1996) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Journal ArticleDOI
TL;DR: In this paper, a low-order model of the milling process is estimated with force and speed signals, and the control system adapts to process parameter changes based on the signals of position sensors.

Journal ArticleDOI
TL;DR: In this paper, the relationship between the manipulated variable and the tank temperature is non-linear, due to variations in operating conditions and process behaviour affecting controller performance within the stability limitations of the feedback loop.