Patent Application
20120259575
1. Field of the Invention
The disclosed embodiments relate to modeling and monitoring device performance degradation due to various failure mechanisms and, more particularly, to an integrated circuit chip incorporating a test circuit that allows for on-chip stress testing to either model device performance degradation or to monitor (i.e., track) device performance degradation.
2. Description of the Related Art
Various mechanisms associated with different classes of devices (e.g., gate oxide integrity (GOI), negative bias temperature instability (NBTI), positive bias temperature instability (PBTI), and hot carrier injection (HCI) associated with metal oxide semiconductor field effect transistors (MOSFETS); reverse bias secondary breakdown (RBSB) associated with bipolar junction transistors (BJTs); etc.) can cause the performance of such devices to degrade over time. Typically, in order to predict how an integrated circuit chip will perform over time, all devices available in a given semiconductor technology node associated with the chip or at least all devices contained in the design chip are subjected to accelerated voltage and/or temperature stress tests in a laboratory environment and/or in a test system environment. Then, based on the results of the accelerated stress testing, performance degradation models and useful life predictions for the devices individually and for the chip as a whole are generated. Unfortunately, existing techniques for generating performance degradation models and useful life predictions require multiple experiments on a statistical sample of hardware and, thus, can be costly and time-consuming. Furthermore, techniques are currently not available for monitoring (i.e., tracking) device performance in the field (i.e., for monitoring the performance of devices on an integrated circuit chip incorporated into a product) in order to allow the performance degradation models to be updated in real-time. Therefore, there is a need in the art for an on-chip test circuit that can be used to stress test devices in a laboratory or test system environment in order to model performance degradation as a function of various class-specific failure mechanisms and that can also be used to stress test devices in the field in order to monitor (i.e., track) device performance degradation as a function of the same failure mechanisms.
In view of the foregoing, disclosed herein are embodiments of an integrated circuit chip that incorporates a test circuit having multiple logic blocks. Each logic block is a matrix of individually selectable, physically different, test devices in a specific class of devices. An embedded processor is operatively connected to the logic blocks, a sensor system, and temperature and voltage regulation systems for the chip. The embedded processor ensures that specific stress conditions are selectively applied, by the temperature and voltage regulation systems, to the test devices and further controls selective testing, by the sensor system, of the test devices to determine the impact of the applied stress conditions. In a laboratory or test system environment, accelerated stress conditions are selectively applied to the test devices and the testing results are used to model device performance degradation due to class-specific failure mechanisms. In the field (i.e., when incorporated into a product), stress conditions are selectively applied to the test devices so as to mimic the stress conditions impacting active devices in use on the same chip and the testing results are used to indirectly monitor (i.e., indirectly track) performance degradation of the active devices due to class-specific failure mechanisms.
More particularly, one embodiment of an integrated circuit chip circuit can comprise a test circuit for use in a laboratory environment. This test circuit can comprise multiple logic blocks (i.e., logic clouds). Each one of the multiple logic blocks can be associated with a specific class of devices and can comprise a matrix of individually selectable, physically different, test devices in the specific class. In this case, the test devices can comprise, for example, duplicates of the devices contained in a given functional circuit design or duplicates of devices in a given semiconductor technology node. The test circuit can further comprise a sensor system and an embedded processor operatively connected to the multiple logic blocks and the sensor system. The embedded processor can control selective stressing (e.g., accelerated stressing) of the test devices. Specifically, the embedded processor can cause a specific test device in the matrix of a specific logic block to be subjected to specific stress conditions (e.g., a specific voltage as applied by a discrete voltage regulator and/or a specific temperature as applied by a heat source, such as an oven). The embedded processor can also control selective testing of the test devices to determine the impact of the applied stress conditions. Specifically, the embedded processor can cause the sensor system to determine an actual value of a specific electrical characteristic exhibited by the specific test device following application of the specific stress conditions. Testing results can then be used to model device performance degradation due to various class-specific failure mechanisms.
Another embodiment of an integrated circuit chip can be incorporated into a test system and can comprise at least one functional circuit as well as a test circuit. The functional circuit(s) can comprise a plurality of active devices. As in the previously described embodiment, the test circuit can comprise multiple logic blocks (i.e., logic clouds). Each one of the multiple logic blocks can be associated with a specific class of devices and can comprise a matrix of individually selectable, physically different, test devices in the specific class. However, in this case, the test devices can comprise duplicates of the active devices contained in the functional circuit(s). The test circuit can further comprise a sensor system and an embedded processor operatively connected to the multiple logic blocks and the sensor system. The embedded processor can control selective stressing (e.g., accelerated stressing) of the test devices. Specifically, the embedded processor can cause a specific test device in the matrix of a specific logic block to be subjected to specific stress conditions (e.g., a specific voltage as applied by the voltage regulation system of the test system and/or a specific temperature as achieved by regulating processing in the test system). The embedded processor can also control selective testing of the test devices to determine the impact of the applied stress conditions. Specifically, the embedded processor can cause the sensor system to determine an actual value of a specific electrical characteristic exhibited by the specific test device following application of the specific stress conditions. Testing results can then be used to model device performance degradation due to various class-specific failure mechanisms. It should be noted that this embodiment is only made practicable if the embedded processor is remote access service (RAS) enabled so as to allow for remote communication with embedded processor (e.g., to update embedded processor programming for selective stressing and/testing, to initiate selective stressing and/or testing on demand, to receive testing results, etc.).
Yet another embodiment of an integrated circuit chip can be incorporated into a product in use and can comprise at least one functional circuit as well as a test circuit. The functional circuit(s) can comprise a plurality of active devices in use in the product. As in the previously described embodiments, the test circuit can comprise multiple logic blocks (i.e., logic clouds). Each one of the multiple logic blocks can be associated with a specific class of devices and can comprise a matrix of individually selectable, physically different, test devices in the specific class. In this case, the test devices can comprise duplicates of the active devices contained in the functional circuit(s). The test circuit can further comprise a sensor system and an embedded processor operatively connected to the multiple logic blocks and the sensor system. The embedded processor can control selective stressing of the test devices. Specifically, the embedded processor can cause a specific test device in the matrix of a specific logic block to be subjected to specific stress conditions (e.g., a specific voltage as applied by the voltage regulation system of the product and/or a specific temperature as achieved by regulating processing by the product). However, in this case, the specific stress conditions to which the test device is subjected will approximate the in-use stress conditions imparted on a corresponding active device (i.e., an active device for which the test device is a duplicate). The embedded processor can also control selective testing of the test devices to determine the impact of the applied stress conditions. Specifically, the embedded processor can cause the sensor system to determine an actual value of a specific electrical characteristic exhibited by the specific test device following application of the specific stress conditions. Since the test devices and active device are subjected to essentially the same stress conditions, the testing results for the test devices can be used to indirectly monitor performance degradation of the active devices due to various class-specific failure mechanisms. It should be noted that this embodiment is also only made practicable if the embedded processor is remote access service (RAS) enabled so as to allow for remote communication with embedded processor (e.g., to update embedded processor programming for selective stressing and testing, to initiate selective stressing and/or testing on demand, to receive testing results, etc.).
The embodiments disclosed herein will be better understood from the following detailed description with reference to the drawings, which are not necessarily drawn to scale and in which:
The disclosed embodiments and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description.
As mentioned above, various mechanisms associated with different classes of devices can cause the performance of such devices to degrade over time. Typically, in order to predict how an integrated circuit chip will perform over time, all devices available in a given semiconductor technology node associated with the chip or at least all devices contained in the design of the chip are subjected to accelerated voltage and/or temperature stress tests in a laboratory environment and/or in a test system environment. Then, based on the results of the accelerated stress testing, performance degradation models and useful life predictions for the devices individually and for the chip as a whole are generated. Unfortunately, existing techniques for generating performance degradation models and useful life predictions require multiple experiments on a statistical sample of hardware and, thus, can be costly and time-consuming. Furthermore, techniques are currently not available for monitoring (i.e., tracking) device performance in the field (i.e., for monitoring the performance of devices on an integrated circuit chip incorporated into a product) in order to allow the performance degradation models to be updated in real-time.
In view of the foregoing, disclosed herein are embodiments of an integrated circuit chip that incorporates a test circuit having multiple logic blocks (i.e., logic clouds). Each logic block is a matrix of individually selectable, physically different, test devices in a specific class of devices. An embedded processor is operatively connected to the logic blocks, a sensor system, and temperature and voltage regulation systems for the chip. The embedded processor ensures that specific stress conditions are selectively applied, by the temperature and voltage regulation systems, to the test devices and further controls selective testing, by the sensor system, of the test devices to determine the impact of the applied stress conditions. In a laboratory or test system environment, accelerated stress conditions are selectively applied to the test devices and the testing results are used to model device performance degradation due to class-specific failure mechanisms. In the field (i.e., when incorporated into a product), stress conditions are selectively applied to the test devices so as to mimic the stress conditions impacting active devices in use on the same chip and the testing results are used to indirectly monitor (i.e., indirectly track) performance degradation of the active devices due to class-specific failure mechanisms.
More particularly, referring to
Referring to
Specifically, the physical difference between each test device 202 in a given matrix 201 can be, for example, a difference in the dimension (i.e., height, thickness, etc.) of a component of the test device, a difference in the geometry (i.e., shape) of a component of the test device, a difference in a material used in a component of the test device, a difference in a dopant or dopant concentration used in a component of the test device, etc. As a result, the test devices 202 in a given matrix 201 can exhibit different electrical characteristics (e.g., different voltage values, current values, resistance values, etc) and may further be subject to different performance degradation rates as a function of different failure mechanisms. For example, in a given matrix 201 comprising all MOSFET-class test devices 202, each MOSFET-class test device can have a different channel length, a different channel width, a different gate dielectric thickness, a different gate dielectric material, and/or a different gate conductor material than other MOSFET-class test devices in the same matrix. Furthermore, as a result of these physical difference(s), the MOSFET-class test devices in the same matrix may exhibit different electrical characteristics (e.g., different threshold voltages (Vts), different drain currents (Id), etc.) and may further be subject to different performance degradation rates as a function of different failure mechanisms (e.g., as a function of gate oxide integrity (GOI), negative bias temperature instability (NBTI), positive bias temperature instability (PBTI), or hot carrier injection (HCI).)
Finally, each of the test devices 202 within each matrix 201 in each logic block 102 can be individually selectable for stressing and testing purposes. That is, each test device 201 can be individually activated for stressing (i.e., for application of a specific voltage and/or a specific temperature for a predetermined period of time and/or in a predetermined pattern) and for testing (i.e., for determining the impact of applied stress conditions on the electrical characteristics of the test device) by means of one or more multiplexors 204 that provide for matrix column and row selection.
Referring to
Those skilled in the art will recognize that the specific stress conditions applied to specific test device will vary depending upon the class of device under test as well as the failure mechanism at issue. For example, for MOSFET-class test devices, stressing to evaluate for gate oxide integrity (GOI) can include stressing at multiple temperatures and voltages for varying timer periods; stressing to evaluate negative bias temperature instability (NBTI) or positive bias temperature instability (PBTI) can include stressing at a stress temperature ranging between −40° C. and 140° C. (preferably 125° C.) and at multiple voltages for varying time periods; stressing to evaluate for hot carrier injection (HCI) can include stressing at a stress temperature ranging between −40° C. and 140° C. (preferably 30° C.) and at multiple gate/drain voltage combinations for varying time periods; and stressing to evaluate BJT-class test devices for reverse bias secondary breakdown (RBSB) can comprise stressing at a constant voltage at a high temperature (preferably 140° C.) for varying time periods.
The embedded processor 110 can also control selective testing of the test devices 202 to determine the impact of the applied stress conditions. That is, the embedded processor 110 can cause (i.e., can be adapted to cause, configured to cause, programmed to cause, etc.) the sensor system 103 to determine an actual value of a specific electrical characteristic exhibited by the specific test device 202 during testing following application of the specific stress conditions. To accomplish this, the multiple logic blocks 102 can each be connected to a corresponding data-in register 104 (i.e., inbound wear-out isolation register (WIR)) and a corresponding data-out register 105 (i.e., an outbound WIR). The embedded processor 110 can first cause data-in 106 to be received by a specific logic block 102 containing the specific test device 202. That is, the embedded processor 110 can transmit a first enable signal 108 to the corresponding data-in register 104 of the specific logic block 102 so that the data-in 106 is released to that specific logic block 102 for processing. The embedded processor 110 can then cause, via multiplexor 204, the data-in 106 to be processed by the specific test device 202. During processing of the data-in 106, the embedded processor 110 can also cause the sensor system 103 to take any measurements required to determine the actual value of the specific electrical characteristic(s) at issue. Then, upon completion of processing, the embedded processor 110 can transmit a second enable signal 109 to the corresponding data-out register 105 of the given logic block 102 so that data-out 107 is released, e.g., for storing in an on-chip data storage 115, for additional processing by the embedded processor 110, etc.
It should be noted that the specific electrical characteristic(s), for which the actual value is determined by the sensor system 103, will vary depending upon the specific class of device under test as well as the specific failure mechanism under test. That is, the sensor system 103 can have the ability to determine (i.e., can be adapted to determine, configured to determine, programmed to determine, etc.) different electrical characteristics associated with the test devices 202 in the logic blocks 102, as directed by the embedded processor 110 and depending upon the specific class of device under test as well as the specific failure mechanism under test. Additionally, since each matrix 201 is configured so that the test devices 202 contained therein are individually selectable for stressing and for testing purposes, the sensor system 103 can determine (i.e., can be adapted to determine, configured to determine, etc.) an electrical characteristic at issue for one or more test devices in one or more of the matrices at a time, while other test devices in the same or different matrices are being subjected to stress conditions.
For example, the embedded processor 110 can initiate testing of a specific MOSFET-class test device 202 in a matrix 201 of a specific logic block 102 to evaluate that specific test device for negative bias temperature instability (NBTI), positive bias temperature instability (PBTI), and/or hot carrier injection (HCI) by causing (i.e., directing) the sensor system 103 to determine a drain current (Id)-gate voltage (Vg) value for that test device. The embedded processor 110 can also test that specific MOSFET-class test device to evaluate for gate oxide integrity (GOI) by causing (i.e., directing) the sensor system 103 to determine a gate current (Ig) value. It should be noted that other electrical characteristics, which can be used to evaluate for specific failure mechanisms and which can be determined by the sensor system 103 for such MOSFET-class test devices, can include, but are not limited to, saturated threshold voltage (Vtsat), linear threshold voltage (Vtlin), off current (Ioff), saturated drain current (Idsat), linear drain current (Idlin), etc.
The embedded processor 110 can similarly initiate testing (i.e., measuring) of other classes of devices in other logic blocks for associated failure mechanisms. For example, the embedded processor 110 can initiate testing of a bipolar junction transistor (BJT)-class test device for reverse bias secondary breakdown (RBSB) by causing (i.e., directing) the sensor system 103 to determine a collector-base current (Icb) value, an emitter-base current (leb) value, etc. for that BJT-class test device.
The various electrical characteristics at issue can be determined by the sensor system 103 either directly (i.e., by taking a direct measurement) or indirectly (i.e., by measuring one or more device parameters and extracting the electrical characteristic based on the measurement(s) taken), if necessary. Sensor systems that determine the electrical characteristics, directly or indirectly, of different classes of devices under test are well-known in the art and, thus, are omitted from this specification in order to allow the reader to focus on the salient aspect of the disclosed embodiments.
It should be noted that the bus 203, multiplexor 204 and driving circuits, described above and illustrated in
More specifically, referring to
The specific stress conditions selectively applied to a test device 202 can comprise, for example, accelerated stress conditions. Such accelerated stress conditions can be applied for a predetermined period of time and/or in a predetermined pattern and can comprise a specific voltage that is higher than the target operational voltage of the device and/or a specific temperature that is higher than target operational temperature. In the laboratory environment, the voltage regulation system 111 for the chip 100 can comprise a discrete voltage regulator (e.g., a stand-alone voltage regulator within the laboratory), which is electrically connected to the logic blocks 102 and the test devices 202 contained therein via at least one bus 203. The embedded processor 110 can cause (i.e., can be adapted to cause, configured to cause, programmed to cause, etc.) this voltage regulator 111 to selectively apply a specific voltage (e.g. via the bus 203) to a specific one of the logic blocks 102 and, more particularly, to at least one specific test device 202 in the specific logic block 102 (e.g., by means of a multiplexor 204) for a predetermined period of time and/or in a predetermined pattern. Additionally, the temperature regulation system 112 for the chip 100 can comprise a discrete heat source (i.e., a standalone heat source) positioned adjacent to the integrated circuit chip 100. This heat source 112 can comprise, for example, an oven that contains and globally heats the integrated circuit chip 100. Alternatively, this heat source 112 can comprise a heating unit that applies directed heat to specific regions of the integrated circuit chip 100. In either case, the embedded processor 110 can cause (i.e., can be adapted to cause, configured to cause, programmed to cause, etc.) the heat source 112 to heat a specific logic block 102 and, more particularly, a specific test device 202 contained therein to a specific temperature for a predetermined period of time and/or in a predetermined pattern.
The embedded processor 110 can further control selective testing of the test devices 202 to determine the impact of the selectively applied stress conditions to one or more electrical characteristics of the specific test device 202. This can be accomplished, as discussed in detail above, by causing data-in 106 to be received by the specific logic block 102 containing the specific test device 202, by causing that specific test device 202 to process the data-in 106, and by causing the sensor system 103 to determine the actual value(s) for the specific electrical characteristic(s) at issue during processing of the data-in. The testing results (i.e., the results of accelerated stress testing) can then be used to model device performance degradation due to various class-specific failure mechanisms and, ultimately, to make device lifetime predictions. That is, the testing results can be used to model performance degradation of the various devices that are contained in either the given functional circuit design (i.e., the target circuit design) or in the given semiconductor technology node and also to make useful life predictions.
The embedded processor 110 can further be in communication with a controller 450 (i.e., a control unit or testing unit). This controller 450 can (i.e., can be adapted to, configured to, programmed to, etc.) install and update, as necessary, the embedded processor programming for selective stressing and testing (i.e., the software or computer program instructions to be executed by the embedded processor in order to perform selective stressing and testing of the test devices), initiate on-demand stressing and/or testing, access testing results for further processing (e.g., to generate and update, as necessary, device performance degradation models), decide to terminate and/or modify stressing/test conditions based on testing results, etc. In the laboratory environment, the embedded processor 110 can be directly wired to the controller 450. Alternatively, the embedded processor 110 can be remote access service (RAS) enabled so as to allow remote communication between the controller 450 and the embedded processor 110. That is, the embedded processor 110 can be configured with the required framework and interface 105 to allow for remote communication therewith. RAS enabled embedded processors are well-known in the art and, thus, the details of the framework and interface required is omitted from this specification in order to allow the reader to focus on the salient aspects of the disclosed embodiments.
Referring to
As in the laboratory environment, the specific stress conditions selectively applied to a test device 202 can comprise, for example, accelerated stress conditions. Such accelerated stress conditions can be applied for a predetermined period of time and/or in a predetermined stressing pattern and can comprise a specific voltage that is higher than the target operational voltage of the device and/or a specific temperature that is higher than target operational temperature. However, since the integrated circuit chip 100 is incorporated into a test system 500, the voltage regulation system 111 for the chip 100 can comprise the voltage regulation system for the entire test system 500. This voltage regulation system 111 can be electrically connected to the logic blocks 102 on the integrated circuit chip 100 and the test devices 202 contained therein via at least one bus 203. The embedded processor 110 can cause (i.e., can be adapted to cause, configured to cause, programmed to cause, etc.) this voltage regulation system 111 to selectively apply a specific voltage (e.g. via the bus 203) to a specific logic block 102 and, more particularly, to a specific test device 202 in that specific logic block 102 (e.g., by means of a multiplexor 204) for a predetermined period of time and/or in a predetermined stressing pattern. Additionally, the temperature regulation system 112 for the chip 100 can comprise a processing control system for one or more additional processors within the test system 500. The embedded processor 110 can regulate the amount of processing performed by the additional processor(s), via the processing control system 512, so as to heat a specific logic block 102 and, more particularly, a specific test device 202 contained therein to a specific temperature for a predetermined period of time and/or in a predetermined pattern. Techniques for using processing control systems to regulate chip temperature are well-known in the art and, thus, the details of such techniques are omitted from this specification in order to allow the reader to focus on the salient aspects of the disclosed embodiments.
The embedded processor 110 can further control selective testing of the test devices 202 to determine the impact of the selectively applied stress conditions to one or more electrical characteristics of the specific test device 202. This can be accomplished, as discussed in detail above, by causing data-in 106 to be received by the specific logic block 102 containing the specific test device 202, by causing the specific test device 202 to process the data-in 106, and by causing the sensor system 103 to determine the actual value(s) for the specific electrical characteristic(s) at issue during processing of the data-in. The testing results (i.e., the results of accelerated stress testing) can then be used to model device performance degradation due to various class-specific failure mechanisms and, ultimately, to make device lifetime predictions. That is, the testing results can be used to model performance degradation of the various devices that are contained in the functional circuit(s) 501 and to make useful life predictions.
As in the laboratory environment, the embedded processor 110 can further be in communication with a controller 550 (i.e., a control unit or testing unit). This controller 550 can (i.e., can be adapted to, configured to, programmed to, etc.) install and update, as necessary, the embedded processor programming for selective stressing and testing (i.e., the software or computer program instructions to be executed by the embedded processor in order to perform selective stressing and testing of the test devices), initiate on-demand stressing and/or testing, access testing results for further processing (e.g., to generate and update, as necessary, device performance degradation models), decide to terminate and/or modify stressing/test conditions based on testing results, etc. However, since the integrated circuit chip 100 is incorporated into a test system 500, this embodiment is only made practicable if the embedded processor 110 is remote access service (RAS) enabled so as to allow remote communication between the controller 550 and the embedded processor 110. That is, in this embodiment, the embedded processor 110 must be configured with the required framework and interface 105 to allow for remote communication therewith. As mentioned above, RAS enabled embedded processors are well-known in the art and, thus, the details of the framework and interface required is omitted from this specification in order to allow the reader to focus on the salient aspects of the disclosed embodiments.
Referring to
In this case, during selective stressing, the specific stress conditions (i.e., specific voltage and/or temperature conditions) applied to a specific test device 202 can approximate (the in-use stress conditions imparted on a corresponding active device. That is, the specific stress conditions applied to the specific test device 202 can be essentially the same as or mimic the stress conditions imparted on an active device 602, which is in use in the product 600 and for which the specific test device is a duplicate. To accomplish this, the integrated circuit chip 100 can further comprise a monitoring system 610 that monitors (i.e., is adapted to monitor, configured to monitor, etc.) the use conditions (e.g., operating times, temperatures, voltages, etc.) of the active devices 602. Systems that monitor use conditions of devices are well-known in the art and, thus, the details of such systems are omitted from this specification in order to allow the reader to focus on the salient aspects of the disclosed embodiments. The monitoring system 610 can further communicate (i.e., be adapted to communicate, configured to communicate, etc.) the determined use conditions to the embedded processor 110 and the embedded processor 110 can subject (i.e., can be adapted to subject, configured to subject, programmed to subject, etc.) the test devices 202 to essentially the same conditions as their corresponding active devices 602 in functional circuit(s) 601. Thus, performance degradation of the test devices 202 due to various class-specific failure mechanisms will be indicative of performance degradation of the corresponding active devices 602 due to the same failure mechanisms.
Since the integrated circuit chip 100 is incorporated into a product 600, the voltage regulation system 111 can comprise the voltage regulation system for that product 600. This voltage regulation system 111 can be electrically connected to the logic blocks 102 on the integrated circuit chip 100 and the test devices 202 contained therein via at least one bus 203. The embedded processor 110 can cause (i.e., can be adapted to cause, configured to cause, programmed to cause, etc.) this voltage regulation system to selectively apply a specific voltage (e.g. via the bus 203) to a specific logic block 102 and, more particularly, to a specific test device 202 in that specific logic block 102 (e.g., by means of a multiplexor 204) for a predetermined period of time and/or in a predetermined pattern. Additionally, the temperature regulation system 112 for the chip 100 can comprise a processing control system 612 for one or more additional processors within the product 600. The embedded processor 110 can regulate the amount of processing performed by the additional processor(s), via the processing control system 612, so as to heat a specific logic block 102 and, more particularly, a specific test device 202 contained therein to a specific temperature for a predetermined period of time and/or in a predetermined pattern. As mentioned above, techniques for using processing control systems to regulate chip temperature are well-known in the art and, thus, the details of such techniques are omitted from this specification in order to allow the reader to focus on the salient aspects of the disclosed embodiments.
As in previously described embodiments, the embedded processor 110 can further control selective testing of the test devices 202 to determine the impact of the selectively applied stress conditions to one or more electrical characteristics of the specific test device 202. This can be accomplished, as discussed in detail above, by causing data-in 106 to be received by the specific logic block 102 containing the specific test device 202, by causing the specific test device 202 to process the data-in 106, and by causing the sensor system 103 to determine the actual value(s) for the specific electrical characteristic(s) at issue during processing of the data-in. Since, as mentioned above, the test devices 202 are subjected to essentially the same use conditions as the active devices 602 in the functional circuit(s) 601, performance degradation of the test devices 202 due to various class-specific failure mechanisms will be indicative of performance degradation of the active devices 602 due to the same failure mechanisms.
As in the laboratory and test system environments, the embedded processor 110 can further be in communication with a controller 650 (i.e., a control unit or testing unit). This controller 650 can (i.e., can be adapted to, configured to, programmed to, etc.) install and update, as necessary, the embedded processor programming for selective stressing and testing (i.e., the software or computer program instructions to be executed by the embedded processor in order to perform selective stressing and testing of the test devices), initiate on-demand stressing and/or testing, access testing results for further processing (e.g., to generate and update, as necessary, device performance degradation models), decide to terminate and/or modify stressing/test conditions based on testing results, etc. However, since the integrated circuit chip 100 is incorporated into a product 600, this embodiment is only made practicable if the embedded processor 110 is remote access service (RAS) enabled so as to allow remote communication between the controller 550 and the embedded processor 110. That is, in this embodiment, the embedded processor 110 must be configured with the required framework and interface 105 to allow for remote communication therewith. As mentioned above, RAS enabled embedded processors are well-known in the art and, thus, the details of the framework and interface required is omitted from this specification in order to allow the reader to focus on the salient aspects of the disclosed embodiments.
It should be noted that the results of in the field stress testing can be used not to generate initial performance degradation models, but rather to diagnose active device failures, to monitor (i.e., track) performance degradation of the active devices in use over time and to update previously generated performance degradation models and useful life predictions. Specifically, the embedded processor 110 or, alternatively, the controller 450 can compare (i.e., be adapted to compare, configured to compare, programmed to compare, etc.) the actual value for a specific electrical characteristic, as determined by the sensor system 103, with a threshold value for that specific electrical characteristic to determine whether or not the device has failed and/or estimate how close to failure the device may be based on existing models and threshold values. The embedded processor 110 or, alternatively, the controller 650 can also compare (i.e., be adapted to compare, configured to compare, programmed to compare, etc.) the actual value for the specific electrical characteristic, as determined by the sensor system 103, with a predicted value for that specific electrical characteristic as indicated by a previously generated performance degradation model for the device. Then, based on the determined difference between the actual and predicted values, the embedded processor 110 or controller 650 can update the performance degradation model for the device and the predicted useful life. Furthermore, repeating the stressing and testing processes allows a user to monitor (i.e., track) the rate of performance degradation of the active devices over time for comparison with the model.
It should be understood that the embodiments as described above are used in the fabrication of integrated circuit chips. The resulting integrated circuit chips can be distributed by the fabricator in raw wafer form (that is, as a single wafer that has multiple unpackaged chips), as a bare die, or in a packaged form. In the latter case the chip is mounted in a single chip package (such as a plastic carrier, with leads that are affixed to a motherboard or other higher level carrier) or in a multichip package (such as a ceramic carrier that has either or both surface interconnections or buried interconnections). In any case the chip is then integrated with other chips, discrete circuit elements, and/or other signal processing devices as part of either (a) an intermediate product, such as a motherboard, or (b) an end product. The end product can be any product that includes integrated circuit chips, ranging from toys and other low-end applications to advanced computer products having a display, a keyboard or other input device, and a central processor.
It should further be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should further be understood that the terms “comprises” “comprising”, “includes” and/or “including”, as used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. Additionally, it should be understood that the corresponding structures, materials, acts, and equivalents of all means or step plus function elements in the claims below are intended to include any structure, material, or act for performing the function in combination with other claimed elements as specifically claimed. The description of the disclosed embodiments has been presented for purposes of illustration and description, but is not intended to be exhaustive or limiting in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the disclosed embodiments.
Therefore, disclosed above are embodiments of an integrated circuit chip that incorporates a test circuit having multiple logic blocks. Each logic block is a matrix of individually selectable, physically different, test devices in a specific class of devices. An embedded processor is operatively connected to the logic blocks, a sensor system, and temperature and voltage regulation systems for the chip. The embedded processor ensures that specific stress conditions are selectively applied, by the temperature and voltage regulation systems, to the test devices and further controls selective testing, by the sensor system, of the test devices to determine the impact of the applied stress conditions. In a laboratory or test system environment, accelerated stress conditions are selectively applied to the test devices and the testing results are used to model device performance degradation due to class-specific failure mechanisms. Such a test circuit reduces the number of experiments required for target design qualification (i.e., for validating, debugging and testing designs) and, thereby reduces the time and costs associated with such qualification. In the field (i.e., when incorporate into a product), stress conditions are selectively applied to the test devices so as to mimic the stress conditions impacting active devices in use on the same chip and the testing results are used to indirectly monitor (i.e., indirectly track) performance degradation of the active devices due to class-specific failure mechanisms over time. Such a test circuit allows pro-active measures to be taken in order to meet a client needs.
