Frequently Asked Questions
What is the highest resolution I can get with an acoustic microscope?
The real question most people want to know is, "what is the smallest detectable defect size with a Sonix™ system?" In reality, this is a complicated question.
There are three different types of resolutions that are relevant to discuss when addressing this question, scan resolution, image resolution and data acquisition resolution. Data acquisition resolution relates to the sampling rate of the analog-to-digital conversation of the ultrasonic data obtained by a Sonix™ instrument. With the latest Sonix™ A/D board, data acquisition resolution is rarely a bottleneck in determining the minimum detectable defect size. To better understand scan resolution picture the scan area as a chessboard, each square is a location where image data is plotted; these locations are referred to as pixels (picture elements). When the transducer visits each pixel location in the image, the ultrasonic signal is processed and a pixel is filled in. If the spacing is very large, then it is possible that the ultrasonic beam never illuminates some objects because they fall between the pixel locations. But if the spacing is very small, then many more scan passes are required to complete the image, and the scan takes longer. The number of data points or image resolution is determined in software by the user.
With Sonix™ WinIC™ software, there is not fixed limit to the number of data points obtainable. However, one concern is it is possible to obtain more data points than is capable of being displayed on the monitor at any one time, so not all data points are displayed on the screen. If the image is zoomed, those extra data points will be displayed. So the issue of the smallest detectable defect size depends on the number of data points acquired and whether the zoom factor and monitor resolution allows those data points to be seen by a human eyeball. The remaining variable deals with the ultrasonic resolution of a Sonix™ system. This resolution is application dependant. The higher the frequency ultrasound, the better the "detectablility", but the poorer the sound penetrates into the sample. If too high a frequency ultrasound is sent into a sample the power of the sound returned would not be adequate to be resolved by the system. What frequency of ultrasound is optimal is determined by the type of material in the sample and the depth of the interface to be imaged. For assistance in determining what the minimum detectable defects size for your applications, contact your local Sonix™ office.
For more detailed explanation please view the presentation of the Minimum Detectable Defect Size.
Is my application a good fit for acoustic micro-imaging?
See our applications library for a good description of what types of applications are best. The best way to find out if your application matches our capabilities is to find out. Contact our Applications Lab or your nearest Sonix™ office to arrange for a demonstration.
What is a C-Scan?
A C-Scan is a map showing the value of a selected ultrasonic measurement at every location. The scan area is divided into discrete positions like a checkerboard. As the transducer visits each position during the scan, the ultrasonic signal is evaluated and a color-coded pixel is plotted on the image. The color may represent the intensity of the reflected signal, the phase of the signal, the time required for the echo to return, etc.
C-Scan Image of a PLCC
Click to Enlarge
What is a B-Scan?
The B-Scan can be visualized as a cross-sectional view from the top to the bottom of the component, along a single scan line, in other words a slice in the X-Z or Y-Z plane. Compare this with the C-Scan view, which represents an X-Y slice at a single depth.
Click to Enlarge
B-Scans are very useful for visualizing different depths at which objects are located. A number of defects in plastic packages are clearly represented in the B-Scan, such as die tilt, voids in the molding compound, popcorn cracks, and internal lead deformation.
A Curtain B-Scan is a multi-page image consisting of individual B-Scans acquired at a number of different locations. Typically, the transducer makes a pass in the scan direction to create a B-Scan image page. It then indexes a small amount in the Step direction, and makes another pass to create a second B-Scan page. This is repeated to cover a user-defined area.
Curtain B-Scan images can contain a very large amount of data compared to a C-Scan. In the C-Scan image each location along the scan line is represented by a single piece of data (e.g. the amplitude or time-of-flight value). In the B-Scan, there is an entire ultrasonic waveform for each location. Because such data files can grow quickly, limit the B-Scan area to the smallest region that will provide the data of interest; keep the B-Scan gate as short as possible; and remember that higher resolution means more data collected, so do not use higher resolution than required.
What type of gates can be used?
The ultrasonic signal includes many echoes, only some of which are of interest for creating a useful image. It is the user's responsibility to select the specific echoes to be used. A number of gates are provided for this purpose. The gate identifies a region or time-slice; the segment of the signal within the gate is analyzed and the result is contributed to the image.
If you are not familiar with the concepts of Front Surface echo, Internal echoes, Delay Lines, and Water Path Multiples, please read the section on the Ultrasonic Signal.
Because there may be different goals for each inspection, and because of the large variety of package types, it is not possible for WinIC to select the echoes of interest automatically. The user must examine the signal, identify what internal structure is responsible for each echo, and place the gates in such a way as to use only the echo of interest for imaging.
Front Surface Follower
As the transducer moves about during the scan, the water path (distance between the transducer and the top of the component) can vary, due to a warped tray, bent pins on the IC, surface imperfections, etc. A variation in the water path will cause the echo of interest to move left or right on the oscilloscope. If the data gate is in a fixed position, the echo may drift in and out of the gate as the scan progresses, rendering the image meaningless.
To compensate for this, a special gate called the Front Surface Follower (FSF) is placed so that no matter where the transducer is positioned, the front surface echo always falls inside the FSF. The point where the signal breaks the FSF threshold is detected automatically, and used as a reference point for the other gates. Using this technique, normal variations in the water path are detected and the data gate is always placed correctly.
When setting the threshold for the FSF, the rule of thumb is to ensure that the first signal to cross the threshold has a clean, sharp edge. The best position for this purpose may be above or below the center line of the scope display. Often the front surface signal begins with a small "pip" followed by a very large, steep rising edge. If the FSF threshold is near the pip, sometimes the pip will break the threshold and other times it will not. This causes the reference time to jitter, and therefore the data gate will jitter.
The Yellow Region is the FSF
Click to Enlarge
Data Gate
The Data Gate has three adjustable parameters: Start, Length and Threshold. Although the Start is a fixed value, it is with respect to a changing reference point (the FSF threshold crossing described above). The Length is set so that the Data Gate includes only the echo(es) of interest. The Threshold can be set so that echoes below a certain strength are ignored. Typically the threshold is set near zero percent.
The Red Region is the Data Gate
Click to Enlarge
Phase and TAMI and B-scan gates can also be used.
What is a Pulse-Echo and Through Transmission image?
Can they be done simultaneous?
Launching ultrasonic energy at the component and analyzing the echoes that return creates the PE (pulse-echo) image. The same transducer is used to send and receive. Since echoes from locations deeper in the component return later, the elapsed time before the return indicates depth. By placing a data gate to select echoes from different depths, the operator can correlate objects in the image to specific depths or layers in the component. For example, a gate can be set to analyze only the echoes from the die attach layer, or between the die and the encapsulant material.
The TT (through-transmission) image detects delaminations at any depth in the device rather than at a specific depth set by the gate. For this reason, it is very useful for an overall accept/reject test. However, TT images are typically not as clearly in focus as Pulse-Echo images.
Pulse-echo and through transmission can now be done simultaneous. Sonix™ PETT feature increases throughput by generating two images in one pass of the part.
What is a Phase Inversion image?
Ultrasonic echoes occur when there is a discontinuity in the path. The discontinuity can be a normal transition from one material to another (e.g. mold compound to die) or because of a defect (e.g. disbond or delamination). The most commonly used indicator is the strength (amplitude) of the return signal.
One of the properties of any material is the velocity of sound. When the ultrasonic impulse crosses a boundary between two materials, the phase of the echo signal depends on the relative speeds of sound in the two materials. If the impulse is traveling from a slower material to a faster material, the phase is unchanged. When it travels from a faster material to a slower material, the phase is inverted.
When the phase of the echo is different from what is expected, it may be indicative of a defect. Specifically, when the normal path is from slower to faster, but the signal is phase-inverted, this indicates an abnormal condition. For example, plastic molding compound is slower than silicon. Therefore, the echo from the top of the die in a plastic package should not be phase-inverted. But if there is a delamination between the plastic and the die, the air gap will cause the signal to invert. It is possible to create an image that uses a special indication for regions of phase inversion.
This signal has normal phase. The negative threshold is broken first.
Click to Enlarge
This signal has inverted phase. The positive threshold is broken first.
The red border around the scope signals phase inversion.
There are many methods to detect phase inversion. A typical reflection begins with a small positive excursion followed by a large negative excursion. If the order is reversed, the signal is inverted. The trick is to quantify what is meant by "large" and "small" excursions, and also to make the algorithm insensitive to overall amplitude variations. Sonix™ uses a phase detection algorithm developed by Texas Instruments, which operates as follows:
- The absolute peak amplitude of the signal is obtained. (This is the largest excursion from zero, regardless of whether it is in the positive or negative direction.)
- Thresholds are established relative to the peak value. Two positive thresholds, typically at 80% and 96% of the absolute peak value. A single negative threshold is set, typically at -60% of the absolute peak.
- The signal in the phase detection gate is evaluated for threshold crossings. If the negative threshold is crossed first, the signal is of normal phase. If one of the positive thresholds is crossed first, the signal is phase-inverted. The image is color coded according to which threshold is crossed. If only the lower positive threshold is crossed, the image is colored yellow. If the upper threshold is crossed the image is colored red.
Since the echo from a delamination is typically stronger than normal, weaker signals are probably not indicative of delamination even if their phase is inverted. For this reason, an additional threshold is used to screen out signals that are too weak. This threshold is referred to as the "Consider for Delam" threshold, and is generally set above 50% of full-screen height.
The default threshold settings were established by Sonix™ over many years of experience and are reliable for the 15MHz V313 transducer. However, the user may wish adjust the thresholds to obtain desired results based on samples with known defects. Special transducers may also require adjustment of the thresholds.
What is a TAMI image?
When setting up an inspection the first time, it is often necessary to experiment with ultrasonic parameters including gate position and focus. To refine the technique can involve a time-consuming and tedious process of scanning, changing gate position, rescanning, comparing results, etc.
Many inspections require separate images representing different depths within the component (e.g. top of the die and die attach, underfill and solder ball, etc.). This can require multiple scans, with the data gate repositioned for each scan.
Sonix's unique Tomographic Acoustic Multiple Imaging (TAMI) system collects multiple images simultaneously in a single scan, using a different data gate for each image. The user examines the images to identify those that best provide the desired results. This information can then be used to set the gate up for a standard single-gate scan for future production use. Or, you can continue to use TAMI as the standard operating mode.
Note that some TAMI images are not as bright as others. This is due to the fact that the individual gates are placed at different locations on the ultrasonic signal. Some parts of the signal have high amplitude, corresponding to a bright image, while other parts have low amplitude, resulting in a dimmer image. Palette adjustments can be used to bring out the details you wish to see.
NOTE: TAMI is an optional feature that is not available on all systems.
What is the latest software release?
As of late 2002, the latest release of WinIC™ is version 3.5.
What is a transducer?
Just as a photographer selects different camera lenses depending on subject matter and light conditions, the operator of an ultrasonic imaging system must select a transducer that is appropriate for the imaging application.
The transducer has an active element (such as a piezoelectric crystal) that generates an ultrasonic impulse when it is energized by the pulser/receiver. The element is mounted in a case with a focusing lens at the tip. The ultrasonic energy does not travel well through air, so a coupling medium (typically water or de-ionized water) is required. Most inspections are done in an immersion tank.
As a general rule, higher frequency transducers are used to image smaller objects with greater clarity, and lower frequency transducers are used when penetrating power is required. Some materials cause more scattering or attenuation of the ultrasonic impulse, which worsen at higher frequencies. In the microelectronics industry, plastic molding compound is one of the worst culprits for signal degradation. Thick plastic packages generally require a lower frequency transducer than thin plastic, ceramic, or bare die applications.
When you start up WinIC you are asked to select an application (thin plastic package, flip chip, ceramic BGA, etc.). A transducer recommendation is included with the parameter file for the application you select.
Some transducers have a fused-quartz "delay line" which aids in focusing the sound beam. When the beam first strikes the end of the delay line, a strong echo is returned, which reverberates inside the transducer. These echoes are visible on the oscilloscope as a series of strong pulses at regular intervals (approximately 4 to 5 microseconds apart). These echoes remain in fixed locations even if you adjust the focus; their intensity changes if you adjust the gain, but not their position in time.
These echoes are not useful for imaging; the echoes from the component to be inspected should typically appear between the first and second delay line echoes. Of course, you must adjust the distance between the transducer and the component in order for these echoes to occur at the correct time.
What upgrades are available for my Sonix™ system?
Contact your local Sonix™ office for more information or fill out our Customer Service Support form and a representative will contact you.
|
