David Johnson
(The author works at Cypress Semiconductor)
Light-emitting diodes ( LEDs ) are commonly used in a variety of modern electronic designs and systems to provide image display and status inspection. As the board becomes more complex, vendors are looking for more testing capabilities to test each component. This will allow the Boundary Scan Test (BST) to evolve to include a full-featured Build-in Self Test (BIST) technology. The challenge for the industry is that there are still some components that have limited automatic test capabilities and must rely on manual testing or visual inspection to detect fault conditions. These LEDs can be used to create $1 cost components for use in line cards or systems that cost $10,000. Therefore, the quality of the components can directly affect customer satisfaction and the overall product quality. .
LED test bottleneck
LED faults can usually be divided into two broad categories. First, when an LED is inserted into a front panel display or other small space, electronic failure and short circuit can occur due to component failure, board assembly problems, and mechanical assembly damage. The second is when the adjacent LEDs are intensified to the same state, their color and brightness will be inconsistent. Using BST to verify the continuity of the line, the first type of failure can be found in other components. However, in stand-alone LEDs, the built-in BST function is not a viable option, and the manufacturing engineer has to visually inspect the LEDs. The visual inspection process is performed by the technician to observe the LED status, and then return all components are normal. This repetitive work is prone to human error, so the test is more complicated to ensure that the technician is not distracted. This test process is not an efficient and value-added approach for testers and practitioners.
(Figure 1) General LED lighting circuit schematic Before looking for an alternative to LED testing, it is necessary to examine the operating procedures of the LED under normal system operation. Basically, when a voltage is applied to put the diode into a biased state, the LED will illuminate as shown in (Fig. 1) (a). Since the diode response is a non-linear mode, a current limiting resistor is typically added to the circuit to ensure that the upper limit of the drive pin is not exceeded. In general, a single-color LED is usually controlled by one driver and connected to one pin, and the other pin is grounded (GND). This design allows one pin to control one LED. To display multiple colors, simply add LEDs of other colors and connect their inputs to other pins.
Simple LED test solution
Figure 1 (b) shows a two-color LED configuration using a common ground and connecting different drivers. One of the things that can happen when combining two or more different color LEDs in the same space is that if the two LEDs are simultaneously energized, the color mixture they exhibit depends on the color displayed by the two LEDs. Another method of constructing a dual-turn bi-color LED is to connect the two configured LEDs in a head-to-tail manner, and the illumination state of either side depends on the other LED entering the bias state. If you need to display the color mixture, the situation becomes more complicated, the system must switch between the two intensified states at a faster speed, so that the naked eye will not see the flicker at the time of switching, and the color mixture will be in a single color state, such as Figure 1 (c). In the following discussion, this two-color LED will be introduced because it represents the most complex conditions and covers other types of construction methods.
(Figure 2) The operation of the LED in the two-color LED circuit connected to the comparator (Figure 2). If pin 1 and pin 2 input the same voltage (usually Vcc or GND), no current will be generated, but all the circuits The point will measure the same voltage. When the voltages of the two pins are different, the two-color LED will be biased, and the voltage at the measuring point will become a diode drop (usually 0.7 volts), which will be higher or lower than the voltage of pin 2. If the voltage at this point can be measured, the state of the LED can be determined, and an automated test mechanism covering each LED can be developed; and this can be used to identify most of the fault conditions of the LED during manufacturing and testing. At the most basic level, if each LED is connected to a comparator and the appropriate set point is selected as the input source for the comparator, the LED can be tested. In this test, the tester is in a completely passive state, so the LED controller must place the LED in a suitable electronic state.
In addition, since the ideal situation is that the LED is tested under various conditions (displayed different colors or turned off after being activated), the set point of the comparator can preferably be adjusted. But this requires more components, and board developers must also increase power consumption. The main disadvantage of this method is that the number of components is too high, because each LED needs its own dedicated comparator, or some form of multi-tasking mechanism to improve the coverage of LEDs, but on the other hand has to reduce the number of components. In addition, you must face complex operations that control all set points to ensure that the correct values ​​are checked at the appropriate set points.
A slightly more integrated solution uses a variety of A/D analog-to-digital converters, sampling all test points through a multitasking mechanism, and merging the results into a format of processing elements. This information can be used to analyze the measured voltage values ​​and is appropriate for the existing LED configuration under test. While this approach reduces the number of components, multiple components are still required to perform the job and process the data retrieved from the test points. A more integrated system uses a microcontroller with multiple analog functions. This approach integrates A/D sampling and processing functions into a single component.
The actual test flow is similar to other board tests, where assembly (board and chassis) errors can occur. A single test mode can only verify that an excited LED is not sufficient to ensure proper operation, so the industry must develop a complete test solution. The following shows an example of a two-color LED:
(1) All LEDs are off (two pins connected to Vcc) – the ground is detected as a short circuit condition;
(2) All LEDs are off (two pins connected to the ground) – the Vcc terminal is detected as a short circuit condition;
(3) All LEDs are on (color one) – a fault is detected in the color-circuit channel;
(4) All LEDs are on (color 2) – a fault is detected in the color two circuit channel;
(5) All LEDs are turned off (the adjacent LED lines are switched between the Vcc terminal and the ground terminal) - a short circuit condition is detected between the LED lines;
(6) Repeat step (5) to detect the channel in the opposite direction.
By completing these six steps, it is determined that all of the LED functions are correct and that there are no fault conditions in the original board assembly or the mechanical components of the front panel. As a result, the reliance on manual visual LED functionality can be significantly reduced, enabling manufacturing engineers to test LED functionality at any stage of the manufacturing process. In addition, the designer must additionally consider the coordinated operation between the control and drive LED components and the test components. There must be a handshaking mechanism between the components to ensure that the test component knows the "expected" state of the current LED. The voltage value at point 1 represents the failure of passing the test and test, as shown in Figure 2. (Figure 3) shows a programmable test circuit configuration designed to monitor 26 LEDs and interface with the entire system via an I2C interface. This design allows the system to adjust all set points and specify which components to test next. Use the I2C interface to allow external systems to analyze the results of any test and the performance of each LED.
(Figure 3) Programmable LED Test Circuit Configuration Another way to add a microcontroller to a design is to integrate control and test functions into the same component. The commonly used link åŸ expander can be used to support the LED control functions in the design, and the analog functions in the components can be used to perform test work at the same time. This integration simplifies the design effort because the designer only has to issue a test command that allows the microcontroller to take control of all the programs and automatically switch to normal system operation after completion. Since the components must support control and test functions, additional pin and software complexity will result in more hardware requirements; on the other hand, it will also reduce the load on the LED processor during system manufacturing and normal system operation. .
Conclusion
In summary, many LEDs are currently tested visually, and it is easy to encounter human error. Therefore, the industry has developed many alternative test methods. These methods not only improve the reliability of the test, but also bring the benefits of some automated tests to replace the manual test process. Not only does this reduce costs, but it also allows LEDs to be tested at any stage of the assembly process and even becomes part of the normal startup process. None of these features are available with existing LED test procedures.
USB Power Adapter,Ce Usb Power Adapter,Portable USB Power Adapter
US Power Cord,Europe Power Cord Co., Ltd. , http://www.chpowercord.com