CA2364564A1

CA2364564A1 - Short-circuit locator for printed circuit boards using a visible camera and an ir camera

Info

Publication number: CA2364564A1
Application number: CA002364564A
Authority: CA
Inventors: Marc Pastor
Original assignee: Individual
Current assignee: Individual
Priority date: 2001-12-07
Filing date: 2001-12-07
Publication date: 2003-06-07

Abstract

A Printed Circuit Board (PCB), bare or populated with electronic components, may present some short-circuits occurring during the manufacturing or the assembly process. By taking different images of the entire PCB, at specific times, with an Infrared (IR) camera, before and after the short-circuit current is applied, and by processing them, it is possible to detect the warm pixels of the short-circuit fault on the PCB. The IR Focal Plan Array (FPA) of the camera, which is sensitive in the IR band, is able to sense and to detect a signal due to the heat radiation of the short-circuit area.
Once this extra heat radiation has been detected for some identified pixels, these pixels are reported to the Visible image captured with a Visible camera in order to pin point the exact location of the short-circuit fault. Then, the fault can be corrected by the usual repair methods.

Description

BACKGROUND OF THE INVENTION.
The present invention relates to inspection or test apparatus for electronic Printed Circuit Board (PCB), and more particularly to detection and location of specific defects occurring during the manufacturing of the bare PCB itself or during the assembly process of the populated PCB
(populated PCB means bare PCB with the electronic components installed and already soldered on it).
On a Printed Circuit Board, where all the electronic components (Integrated Circuits, resistors, capacitors, connectors,... ) are already installed and soldered, it is easy to know the presence of some short-circuit, between the power supplies conductors or between two signal conductors, but it is impossible to locate it in order to remove this fault to get again a functional PCB. Simple devices or apparatus exist already to detect the presence of a short-circuit but none of them is able to pin point precisely their location on a Printed Circuit Board (bare or populated).

2 Other current techniques don't allow to find the location because the short may be anywhere all along the path which present a low or close-to-zero resistance.
The first aspect related to this invention is the detection of the short-circuit area. The approach is based on the thermal properties of a typical short-circuit. The short-circuit thermal signature matches the exact and precise location of an hardware short-circuit (due to excess of solder, solder bridge, manufacturing or assembly process weakness,...) on the PCB and, as a consequence, the opportunity to repair the PCB later on by removing the short-circuit fault, once the exact precise area has been identified and pin pointed on the PCB. This aspect of the invention is based on the fact that a short-circuit has a certain resistance value, greater than the PCB conductor traces carrying the same current. As a consequence of the short-circuit current passing through, this small resistance dissipates more heat that the normal path carrying the current through the PCB
traces. The short-circuit creates a local temperature elevation (Joule's effect) which is detectable by an IR camera sensitive to Infrared radiation.
The second aspect related to this invention is the display of the precise location of the previously detected short-circuit area on the PCB. It is known that an IR
image is practically uniform because the PCB is at a constant and stable temperature. Only the warm spots of the short-circuit thermal signature will be detected over an uniform background because the other area of the PCB stays at the same temperature, From this fact, it is impossible, for an operator, to locate the warm pixels on the PCB in relation with the electronic components mounted on it. To solve this problem, a Visible image is captured by a conventional video camera with a medium to high resolution to give a clear image of the PCB. By knowing the relationship, or the correspondence, between any IR pixel of the IR image and any Visible pixel of the Visible image, it is easy to report, on the Visible image, the warm or hot pixels found on the IR image.
The third aspect related to this invention is the calculation of the previously mentioned relationship, between any IR pixel of the IR image and any Visible pixel of the Visible image. This process, called a spatial calibration, will calibrate spatially the IR pixel coordinates of the IR image and the Visible pixel coordinates of the Visible image. This cross-reference between pixels is essential and mandatory in order to properly overlay the identified warm or hot IR pixels over the Visible image, in order to pin point precisely to the operator the faulty area with the surrounding electronic components. This method is based on the fact that, during a simple calibration process, the two camera (IR and Visible) will see the same scene (a square or a rectangle called a Field-Of View mask) with their respective resolution. By computing the spatial relationship between an IR
pixel and a Visible pixel, it is easy to convert any IR pixel coordinates in Visible pixels coordinates and, then, to display these identified Visible pixels on the Visible image to show to the operator the faulty area on a clear visible PCB image.

SUMMARY OF THE INVENTION.
The present invention is directed to electro-optic image capturing system for obtaining images of a bare or populated PCB. The system includes a base plate support to install the PCB
under test and a single wall light protective enclosure having the internal side covered with an absorbing (high emissivity) paint or coating. This enclosure has to absorb all the internal Infrared radiation and to isolate the imaging system from the environment light noise and variation (Visible and IR).
As a first part, the image capturing system consists of a Visible camera with associated optics. This Charge Coupled Device (CCD) Visible camera detects any change in Visible light intensity which gives information about the light level and color information of the different electronic components on the PCB. A typical resolution is 768 x 494 pixels for a common CCD
technology.
As a second part, the image capturing system consists of an IR camera with associated optics. This IR camera detects any change in IR radiation intensity which gives information about the level of measured temperature or heat of the PCB. A typical resolution is 320 x 240 pixels for a common microbolometer technology.
As noticed, a Visible camera (768 x 494) has more spatial resolution than an IR camera (320 x 240). As a general statement, all the captured images are stored and processed in the Computer Device and the results displayed on the monitor or any output peripheral.
Reference to FIG. 1, a typical spectral response of a Visible CCD Focal Plan Array (FPA) shows the response in the Visible band (between 400 and 800 nanometers). This sensitivity permits to capture a Visible image of the PCB which will be used to display a clear picture of the PCB.
Reference to FIG. 1, a typical spectral response of a IR Focal Plan Array (FPA) shows the response in the IR band (between 2 and 14 micrometers). This sensitivity enables to capture an IR
image of the PCB which will be used to detect the warm or hot spots on the PCB.
Reference to FIG. 1, a typical IR emission radiation signature of a short-circuit in the IR
band. The large overlap of the spectral IR emission of the short-circuit and the IR camera spectral response permits the detection of a small amount of heat, due to the Joule's effect, of the short-circuit presenting a non-zero resistance value.

The sensitivity of the IR camera in this IR band is sufficient enough to generate a signal higher when an area on a Printed Circuit Board area is warmer or hotter due to the presence of this extra heat. This short-circuit defect has a small resistance value which generates enough detectable heat in the IR band when the current is passing through it.
First, an IR image is captured before applying the short-circuit current in order to get an IR
Reference image. Reference to FIG. 2A , this is the illustration of a typical histogram (distribution of pixel amplitude of the entire PCB) of a PCB image captured at room temperature before the short-circuit current is applied. This image is called the Reference image.
Next, just before the short-current time duration is finished, a second IR
image called Heat image is captured, containing information of the warmer pixels due the sensitivity of the IR camera in the IR band. Reference to FIG. 2B, this is the illustration of the histogram of the same previous PCB but the IR image has been captured just before the pulse current is gone.
This image is called the Heat image. The warm area of the short-circuit location will emit radiation in the IR band, which are detectable by the IR camera. Some pixels will have a higher value ( black colored).
Then, Reference to FIG. 2C, the histogram of the resulting image made by subtracting the IR Reference image from the IR Heat image, pixel by pixel, and keeping the non-zero value pixels.
These are the pixels of interest. The zero-pixels value are not shown because they are not carrying any heat information (means there is no change in temperature between the Reference image and the Heat image for these pixels which are at the same temperature before and after the short-circuit current has been applied). FIG. 2C shows also the detection threshold used to increase the noise immunity.

BRIEF DESCRIPTION OF THE DRAWINGS.
FIG. 1 is an illustration of the detector sensitivity of both camera and short-circuit IR heat radiation, to show their respective bandwidth.
FIG. 2A is an example of an histogram for a IR Reference image, before applying the short-circuit current.
FIG. 2B is the histogram for the IR Heat image, just before the short-circuit current pulse is gone, showing some warm or hot pixels with higher values.
FIG. 2C is the histogram for the IR Difference image, which is the difference between the IR Heat image and the IR Reference image, described in FIG. 2B and FIG. 2A
respectively, showing the pixels which are detected as warmer, due to the short-circuit radiation in the IR band. The zero-value pixels are not represented because they don't have any interest;
FIG. 3 is a bloc diagram of the apparatus showing the hardware components of the system.
FIG. 4 shows details for the limiting current resistors and the connecting relays.

DESCRIPTION OF THE INVENTION.
Reference is now made to FIG. 4 of the drawing wherein a Printed Circuit Board 2 (PCB) to be tested is located on the base plate support 1, inside a protective light enclosure 4 to minimize the environmental light variation and external temperature perturbations.
As a initial set-up, connect the DC or AC Power Supply 18 leads at the appropriate location to the PCB, (normally a connector 22 with one or more contacts). Then set, on the Computer Device 11, the power supply pulse duration programmable from 50 milliseconds up to 2,000 milliseconds (increment of 50 milliseconds). This pulse duration is the time the connecting relays 19 will be closed to enable the current to flow through the PCB.
To protect the PCB 2 and to avoid the Power supply 18 to drop its own voltage or to be damaged, it is necessary to put a limiting resistor Rp in series to limit the short-circuit current flowing through the PCB by using the simple formula:
Vs Imax -Rp If the operator wants to limit the short-circuit current to Imax (in Amperes), and if the Power supply voltage is Vs (in Volts), the limiting resistor value (in Ohms) will be:
Vs Rp - ( Rp in Ohms ) Imax The resistor can be connected or installed in the resistors box 20 and FIG. 5 shows a typical implementation of such limiting current resistors.

The short-circuit test operation can start now described in the following steps and all the hardware components are referenced to FIG. 4.
As a first step, it is necessary to calibrate the two camera by using a so-called Field-Of View (FOV) mask which is essentially a square or rectangle of reflective metal (aluminum, copper,...) which can be seen or detected by both Visible and IR camera. As an initial action the operator has to adjust the adjustable optical focus of each camera to get the best contrast and to install the FOV
mask 3 on the support base plate 1 and, then, capture the same FOV mask image by both camera.
To capture the Visible FOV mask as a Visible image, the Computer Device 11 will send an electrical pulse signal through the Digital Output Port Card 14 to enable the lamp driver 6 to illuminate the FOV mask with the lamp 5 on. This Visible FOV mask image is stored in the PC
memory. Because this FOV mask is reflective to the Visible light generated by the lamp, the Visible FOV mask image is clear and easy to process to extract the four corners coordinates of the square or rectangle mask in Visible pixels coordinates.
To capture the IR FOV mask as an IR image, the Computer Device 11 will send an electrical pulse signal through the Digital Output Port Card 14 to enable the lamp driver 6 to illuminate the FOV mask with the lamp 5 on. This lamp 5 will create a sufficient heat emission to reflect on the FOV metal mask to get a contrasted IR image. This IR FOV mask image is stored in the PC memory. This contrasted IR image is easy to process to extract the four corners coordinates of the square or rectangle mask in IR pixels coordinates.
Because the two camera have captured the same spatial FOV mask, it is possible to calculate the relationship between the pixels coordinates of the two camera.
FIG. 2A shows an example of Visible image and IR image for each camera with their respective FPA size (768 x 494 pixels for the Visible camera and 320 x 240 pixels for the IR
camera).
The FOV mask corners coordinates on the IR image, in IR pixels, found by searching the four corner pixels are:
left top corner:(a,b) right top corner:(a,d) left bottom (c.b) corner:

right bottom (c,d) corner:

The FOV mask corners coordinates on the Visible image, in Visible pixels, found by searching the four corner pixels:
left top corner: (A,B) right top corner: (A,D) left bottom corner: (C,B) right bottom corner: (C,D) It is now necessary to compute the spatial relationship between any IR pixel (x,y) and a Visible pixel (X,Y) in order to report any warm or hot identified pixel of the IR image to the Visible image as an overlay.
Suppose the coordinates of a warm or hot pixel found on the IR image is (i,j), the corresponding pixel (I,J) on the Visible image will be computed by using simple calculations:
D-B
J = B + ( j - b) horizontal coordinate in Visible pixels d-b C - A
I = A + ( i - a) vertical coordinate in Visible pixels c-a As a second step, it is necessary to capture a Visible image of the PCB.
Remove the FOV
mask 3 and install the PCB 2 under test on the base plate 1. The Computer Device 11 will send an electrical pulse signal through the Digital Output Port Card 14 to enable the lamp driver 6 to illuminate the PCB with the lamp 5 on. During the time the lamp 5 illuminates the PCB, the Computer Device 11 will capture a video Visible image by using the Visible Frame Grabber card 12 and store this image in the Computer Device memory. When this image capture is done, the lamp 5 is turned off. The Visible camera Frame Grabber converts the Visible Video signal 9, coming from the Visible camera 8 and associated optics, in a digital bit map image. This Visible image will be used later on to show physically where is the location of the short-circuit by overlaying the warm or hot pixels.
As a third step, it is necessary to capture an IR image to be used as an IR
Reference image with no short-circuit current applied. The Computer Device 11 will capture an IR Reference image by using the IR camera Frame Grabber card 13 and store this image in the Computer Device memory. This IR image is the IR image of the PCB at a uniform temperature (room temperature) before any short-circuit current is applied.
As a fourth step, it is necessary to capture an IR image to be used as a Heat image with the short-circuit current applied. The Computer Device 11, through the Digital Output card 14, will send a Command signal 17 to activate the Connecting relays 19. By closing the contacts of the relays, the current coming from the Power Supply 18 will pass through the Limiting current resistors 20 and feed the PCB through the wires 21 and the connector 22. After the programmed time duration, the Connecting relays 19 are deactivated, current flowing through the PCB 2 is then stopped, and an IR Heat image is captured by the camera and store in Computer Device memory.
FIG. 5 show a typical implementation of such connecting relays 19.
As a fifth step, it is necessary to identify the IR pixels which are warmer due to the short-circuit current. The Computer Device 11 will compute the IR Difference image, which is the subtraction between the IR Heat image and the IR Reference image, on a pixel-by-pixel basis, to extract the pixels which are warmer than the IR Reference image. These warmer pixels come from the heat activity generated by the current passing through the short-circuit path which gives a local heat increase. The sensitivity of the IR camera in the IR wavelength bandwidth permits the detection of a such extra amount of Infrared radiation.
FIG. 3C illustrates the IR Difference image with warm detected pixels.
As a sixth step, it is necessary to select the warm or hot pixel and to report them on the Visible PCB image. These identified pixels in IR pixels coordinates on the IR
image will be converted in Visible pixel coordinates on the Visible image by using the simple formula explained and calculated above, in the first step section. The monitor 16 will display the warm pixels, now converted in Visible pixel coordinates, by overlaying these pixels, in a different color, on the Visible PCB image captured initially, in the second step section, as an image of the PCB. This final image ( Visible PCB image with the addition of warm pixels converted in Visible pixel coordinates) is called the Visible Result image.
An operator adjustable programmable threshold enables the operator to adjust the display of these identified warm pixels on the monitor 16. The Computer Device is operated by the operator with the keyboard and the mouse devices 15.
These identified warm pixels, on the Visible Result image, pin point the location of the electrical short-circuit fault with all the surrounding electronic components of the PCB.

Claims

CLAIMS:

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. An apparatus comprising:
-an IR camera with its associate optics, -a Visible camera with its associate optics, -a protective surrounding enclosure to absorb internal and external IR
radiation acting as a IR radiation shield, -a lamp emitting visible light and heat radiation, -a Field-Of View (FOV) mask used for the spatial calibration between the two camera in order to find the relationship between any IR pixel coordinates of the IR
image and any Visible pixel coordinates of the Visible image, -a set of connecting relays, -a set of limiting current resistors, -a computer device including a Frame Grabber per camera and Digital Output card and keyboard/mouse peripherals, -an application software to capture and to store and to process digital images and to display the results on a monitor, or on any peripheral, and to activate digital output pulses to switch ON an OFF the connecting relays and the lamp.
-an application software for spatial calibration for the two camera, in order to get the spatial relationship between the IR pixels coordinates and the Visible pixels coordinates.

2. An IR camera, as defined in claim 1, in which the Focal Plan Array (FPA) is sensitive to IR radiation between 2 to 14 micrometers

3. A Visible camera, as defined in claim 1, in which the Focal Plan Array (FPA) is sensitive to Visible light radiation between 400 to 800 nanometers.

4. A single wall protective enclosure, as defined in claim 1, with the internal side covered with an absorbing (high emissivity) paint or coating or made with an absorbing material in the IR
band, to isolate the imaging system from external Visible and IR perturbation or variation.

5. A lamp, as defined in claim 1, to illuminate the FOV mask and the PCB under test during the image captures for both Visible camera and IR camera.

6. A Field-Of View (FOV) mask, as defined in claim l, square or rectangle, made of metal or reflective material to reflect Visible light and IR radiation in order to capture a contrasted image in the Visible band and in the IR band.

7. A set of connecting relays, as defined in claim 1, to enable the current to flow through the PCB under test.

8. A set of selectable limiting current resistors, as defined in claim 1, to limit the Power Supply current to a maximum value in order to protect the Power Supply and the PCB under test.

9. A computer device, as defined in claim 1, to capture and to process video frames from the Visible camera and IR camera and to display the result image and to provide electrical signals to the connecting relays and to the lamp driver.

10. An application software, as referenced to claim 1, to capture the frames coming from the IR camera at specific times in direct relation with the switching ON and OFF
of the connecting relays, to average consecutive frames to get a better quality image, to transfer the frames into the computer device memory.

11. An application software, as referenced to claim 10, to process the frames according to the software application rules or algorithms and, more precisely, to subtract the initial image (IR
Reference image) from the short-circuit image (IR Heat image) to extract only the IR variations in the pixel value in order to get the IR Difference image.

12. An application software, as referenced to claim 11, to adjust the decision threshold for the pixels of the IR Difference image to identify the hot and warm pixels with the largest IR
increments in order to detect more precisely the warmer spots, directly related to the short-circuit fault.

13. An application software, as referenced to claim 9, to display, as an overlay, the warm pixels on the Visible image to pin point physically the short-circuit spatial location to get the Visible Result image.

14. An application software, as referenced to claim 13, to store or to print the Visible Result image in order to give the short-circuit fault spatial location to the repair technician.

15. An application software, as referenced to claim 10, with a user programmable time duration feature for the short-circuit power supply pulse which has to be applied to the PCB under test.

16. An application software, as referenced to claim 15, to give the command signals to switch ON and OFF the connecting relays, at a controlled starting time and for a programmable time duration.

17. An application software, as referenced to claim 1, to capture FOV mask images by both Visible camera and IR camera by using the lamp as a light source and heat source respectively.

18. An application software, as referenced to claim 17, to extract the four corners coordinates on both Visible FOV image (A,B,C,D) and IR FOV image (a,b,c,d).

19. A simple algorithm, as referenced to claim 18, to calculate the relationship between any IR pixel (x,y) and any Visible pixel (X,Y) in order to report any warm or hot pixel identified on the IR Difference image to the Visible image as an overlay.
Suppose the FOV mask coordinates on the IR image, in IR pixels, are:
left top corner: (a,b) right top corner: (a,d) left bottom corner: (c.b) right bottom corner: (c,d) Suppose the FOV mask coordinates on the Visible image, in Visible pixels, are:
left top corner: (A,B) right top corner: (A,D) left bottom corner: (C,B) right bottom corner: (C,D) If the warm or hot IR pixel coordinates on the IR image is (i,j), then the corresponding Visible pixel (I,J) coordinates on the Visible image will be computed by using the simple calculations:~