The investigation of a failed chip can be a long and drawn out process. It's not as if the failure analysis engineers blindly take any chip which is brought to them and start performing various tests and procedures on it. The entire operation from start to finish is a structured process which starts off by analyzing the various circumstances under which this particular semiconductor manifested its fault. A detailed analysis of the situation gives semiconductor failure analysis engineers a starting point for determining what has gone wrong. Those who're experienced in the field will be able to make use of the knowledge to cut down on the time required to isolate the possible causes of error. Of course, some defects are far more difficult to identif. For this reason, there may be a need to test the chip in multiple ways before confirming the cause of failure. Accordingly, techniques called "Nondestructive Testing" or NDTs are preferred because they keep the chip intact and open for further testing.

A failure analysis lab makes use of nondestructive testing techniques before they resort to other methods like decapsulation which involves physically opening up the chip to get to its innards. There are a wide range of techniques which can be classified as NDTs for the purposes of failure analysis. These include light emission microscopy, acoustic testing, and micro thermal imaging. They are only broad categories and there are several specific techniques within each whose use is determined by the unique circumstances of each chip.

Emission spectroscopy for example consists of bombarding a chip with either electrons or with x-rays. These powerful beams interact with the material of the semiconductor in different ways. Depending on what they wish to find out, we can either analyze the characteristic x-rays of the material, or capture the Auger spectrum. We can also take extremely fine photographs at the tiniest of scales using an electron microscope.

Acoustic testing probes the chip with high-frequency sound waves while immersed in a liquid to detect structural flaws such as hairline fractures, voids and improper soldering. Once again, depending on the specific requirements we can scan a broad area at a specific death or a tiny portion of the chip.

Microthermal imaging proves to be extremely useful when detecting electrical flaws which give off excess heat. These experiments can be fine-tuned for extreme sensitivity and experienced failure analysis engineers will be able to interpret the readings and arrive at the flaw. These are just a few ways in which a semiconductor failure analysis is conducted through the use of nondestructive testing. They play an important role in improving the safety and reliability of the electronic chips which we take for granted every day.

Given the densities and the miniaturization of electronic components in an integrated circuit today, it is more important than ever for failure analysts to be able to obtain crisp and clear images on tinier scales than ever before. Normally, we use the word "microscope" to denote optical microscopes which we're used to finding in biological labs. But such devices are inherently restricted by their medium – which is light itself. An optical microscope makes use of the visible light spectrum and provides us with images which are magnified to about 2000 times their original size. While this is sufficient for a large number of applications, we are beginning to find it increasingly wanting for the resolution of electronic chip defects. In order to overcome this limitation and to be able to take images with far greater magnification, we need to use an entirely different paradigm – namely electron microscopy.

As we learned in our science classes, electrons have the capability of behaving both like waves and as particles. We can make use of this curious nature in labs and exploit the shorter wavelengths of electrons to obtain images at the tiniest of scales. The electron microscope bombards the specimen with electrons and captures them as they scatter off and interact with the object which is being examined. Depending on the precise technique we wish to use, we can analyze the resulting spectrum in a variety of ways. For example, we have already seen how such a bombardment can result in the generation of Auger electrons which are a secondary effect. Or, we can study what are known as "characteristic x-rays" to obtain more information about the sample. Using such methods, we can obtain very precise information about the composition of the material.

But our interest here is in obtaining an image and in many ways the electron microscope can do a far better job than an optical one. For example, it is much easier to obtain an image with varying levels of depth with electron microscopy. Traditional optical microscopes focus on one layer of the sample whereas objects behind it and ahead of it are out of focus. This limitation is erased with the scanning electron microscope.

Still, obtaining an image is just one part of the long process of failure analysis. Before we actually go ahead with the experiment, a lot of groundwork needs to be done in order to understand which area we need to focus on and why. And this is where the experience of failure analysts comes in. It is thanks to them, that we are able to rely on our electronic systems as much as we do today.

Because an integrated circuit is so sensitive, even the slightest variance in its operating parameters can cause it to malfunction. These malfunctions can take place in a variety of ways. Mechanical stress is one such example. If the integrity of the materials comprising the chip is compromised in any way, it can lead to unpredictable behavior. Sometimes the connections between the components of the circuit and die are compromised and this once again leads to errors. It is the job of failure analysis to identify the cause of the malfunction so that improvements can be made in the manufacturing, storage, and transportation processes to minimize the chances of it happening again. Such is the sensitivity of these semiconductors, that even fractures which are too small for the eye to see can cause the component to fail. A powerful technique which allows us to identify these problems is called Scanning Acoustic Microscopy which makes use of sound waves to probe the electronic circuits.

The principal is very similar to that used in sonar applications. Soundwaves happen to have just the right wavelength which makes them extremely sensitive to anomalies in the materials used to manufacture semiconductor chips. Failure analysis focuses on utilizing these unique properties in order to isolate various faults which routinely appear on electronic packages such as voids and hairline fractures. The complexity is increased by the fact that these chips are layered and the fault can lie at a specific depth. There are different modes of scanning acoustic microscopy which isolate different type of problems.

Sound waves in the air don't possess the necessary qualities which allow them to properly penetrate the semiconductor material and reflect off the faults. For this reason, we place the package in water and propagate the soundwaves through that instead. This allows scanning acoustic microscopy techniques to be far more effective than would otherwise be possible. We can choose to scan semiconductor chips into ways – we can either inspect one entire section at a specific depth below the surface of the chip, or we can focus our attention on the specific area regardless of the chip depth. Failure analysis experts will be able to determine which technique is necessary depending on the condition of the chip and what symptoms it manifests.

This is one of the reasons why an experienced failure analysis team makes a huge difference in the effectiveness of an error detection procedure. Scanning acoustic microscopy is an essential tool in the repertoire of engineers who strive to make the chips we use every day more reliable and long-lasting.which utilizes soundwaves to identify

Anyone who has seen an integrated circuit must have wondered at some point or the other "How does it work?" There are no moving parts, the components are too small to see with the naked eye, no human hand could ever manage the dexterity required to put them together, and the mind boggles at the densities with which they are packed into a limited area. Such individuals will have a proper appreciation of the difficulty therefore, of not only understanding how it works as a whole, but of identifying what has gone wrong with a defective chip. Indeed the ability to isolate errors underlies much of the reliability of modern chips today. Without it, we would not know which manufacturing techniques to improve or how to build better integrated circuits. It's clear from the outset that the task requires innovative and novel methods – ones which have been built specifically for this purpose. Emission microscopy is one such technique used for the failure analysis of semiconductor chips.

How Does Emission Microscopy Work?

There are a large number of techniques in failure analysis which rely on the emission of electromagnetic radiation from a chip in order to find out what has gone wrong. These emissions can be interpreted in a variety of ways depending on the experiment. Typically, each separate technique is used to identify a particular type of fault or error which the failure analysis engineer thinks is the cause of the malfunction. Sometimes, the chip is irradiated with x-rays or electrons in order to stimulate it into emission.

But that isn't always necessary. Even under normal working conditions, every chip emits some radiation. This natural radiation is emitted in the lower half of the spectrum closer towards infrared. Using extremely sensitive equipment, we can pick them up and use it for failure analysis purposes. Merely obtaining the emission microscopy signature of the defective chip however, isn't nearly enough. In the first place we need to have an experienced analyst who is able to make sense of the readings and who already has some inkling of the fault. It also greatly helps to have a normal known functioning chip in order to compare the defective one with. This way, we will be able to quickly identify the variations.

But it's not all smooth sailing from there either. Very often the symptom of an electronic failure manifests itself in a completely different area from that of its cause. This is where the expertise of the person performing the experiment comes into play. It's no wonder that the field of failure analysis is one of the most crucial post manufacturing steps in ensuring that the electronics we use everyday are reliable and trustworthy.

 

Integrated circuit geometries these days are becoming smaller and smaller. The size of these structures is measured in terms of a few dozen nanometers which makes it increasingly difficult to properly analyze the cause of a failure. Electronic circuits are especially difficult to debug because of the complex way in which they interact with each other. It's entirely possible for the cause of the defect to be located in a different area from that of its symptom. Accordingly, IC failure analysis lab techniques are evolving to suit this new paradigm. New phenomena have to be taken advantage of in order to provide the same efficiency in failure analysis detection that the previous methods possessed, only this time at a smaller scale. Examples include the introduction of the Solid Immersion Lens Objective (SIL) which makes it easier to obtain clearer images of exceedingly tiny structures.

There are many other lab techniques which have evolved to manipulate integrated circuits on extremely tiny scales. The categorization of these techniques can be based either on the phenomena which they exploit, or the type of defect which they aim to resolve. Either way, it requires significant engineering expertise and deductive skills to effectively perform a failure analysis on an electronic chip.

Failure Analysis Techniques

Defects in electronic circuits can very often be identified by the fact that they create certain side effects such as a tiny emission of excess heat which can be detected. We can use a variety of techniques for this such as exploiting the properties of liquid crystals whose molecules align themselves in a particular manner when exposed to a temperature gradient. More sophisticated processes involved florescent imaging which can produce a kind of "heat map" for the experienced analyst to decipher.

Yet other techniques involve making use of the properties of light via optical spectrographic methods in order to determine the exact composition of the substances used to manufacture the chip. Even tiny variances and impurities can cause the chip to operate outside its normal parameters and behave in unexpected ways. A great deal of work goes into ascertaining the probable cause of the failure before the device is subjected to an actual test which tends to be time-consuming and expensive. Those who have worked in the field for a long time can make use of their experience and arrive at a more accurate conclusion faster than others.

As IC failure lab techniques continue to evolve, we will see more and more precise methodologies which need to be applied to smaller and smaller chips. The progression of the two industries of chip manufacturing and failure analysis move in tandem and it is this balance which is responsible for making our electronic lives reliable and error-free.

Printed Circuit Boards or PCBs for short are an essential component of almost every electronic device manufactured today. Over the years, they have become more and more complicated as electronic device manufacturers improve their capabilities and continue the trend towards greater miniaturization of electronic components. Coupled with the increased processing speed which is required to implement the burgeoning demands for new features, this has resulted in a massive boom in chip density. Admirable as these achievements are, they have resulted in an increasing number of chip defects which manifest themselves in different ways depending on the situation. For this reason, PCB failure analysis has come into even greater prominence in the past few years. The costs of handling negative publicity as well as implementing callbacks are so primitive that companies are willing to go all the way and perform extensive testing and failure analysis on a chip before it goes into production.

Despite the increased complexity however, the basics and fundamentals of PCB failure analysis remain the same. All defects can be categorized into a few primary slots and all failure analysis techniques can be grouped together based on their underlying principle. A few types of integrated circuit defects are presented below.

PCB Defects

To start with, one needs to ensure that the materials used in the composition of printed circuit boards meets with the stringent requirements necessary for their operation. It's not as simple a matter as ensuring the purity of the substance, rather than making certain that the correct proportion of materials is present. Certain substances when present even in miniscule quantities are responsible for throwing off the operation of an integrated circuit entirely. But failure analysis is not the only reason to determine the precise composition of materials. Places like the European Union have stringent laws prohibiting even tiny amounts of hazardous substances such as cadmium etc. An ROHS certification is necessary for being able to market your product in these countries and for that, failure analysis techniques for detecting tiny amounts of impurities are vital.

It's also important to ensure the integrity of the materials by ascertaining whether or not there are hairline fractures, cracks or voids. These are detected using acoustic techniques similar to those used by sonar applications. Yet other defects revolve around the electronic wiring. These are arguably the most difficult to isolate and fix because the symptoms can vary greatly and do not always show up at the site where the error exists.

Because of all this, the experience of failure analysis engineers is vital when determining whom to go to. The costs of PCB failure analysis pay themselves off many times over by ensuring that you won't have to implement expensive recalls at a later stage.

As machines become more and more complex, it becomes increasingly difficult to find out what's wrong with them when they fail to operate properly. Integrated circuits have taken this to an extreme. They're not merely complex. They're mind-boggling. We've transitioned to a state where we can have literally millions of transistors on a single chip. These chips don't have any moving parts and finding out what is gone wrong with one of them is a herculean task. Even ascertaining what kind of problem isn't easy. It requires experienced engineers who have experience in the field to understand the symptoms and figure out what the issue is. Even so, we need concrete testing in order to confirm or repudiate their suspicions.

The different defects which can plague electronic circuits are categorized by how they are caused with broad classifications, and several sub classifications under each. But this isn't the only way to do this. We can also arrange them based on the methods of detection, though there is a very close correlation between the two. Let's look at a few kinds of defects which regularly appear on integrated circuits.

Integrated Circuit Defects

If we proceed from the bottom up, we find that the first source of flaws in electronic systems deal with the composition of the material with which they are made up. Chips in particular have extremely sensitive operating parameters including the purity of the substances which comprise them. Small percentages of deviations from the optimal composition is enough to cause them to malfunction.

Even if the material has no such impurities, they can still be problems with the integrity. Flaws such as cracks, voids, and delaminations can easily distort the operation of otherwise perfect integrated circuits. Usually we use acoustic technologies to isolate and correct such problems.

Perhaps the most difficult kind of error to detect is one which is electrical in nature. This is because there's no easy way in which we can ascertain where exactly something has gone wrong. The best we can do so far is to isolate the exact spot at which the error manifests itself. But because of the complex nature of this technology, it's entirely possible for the symptom itself to be located somewhere other than where the real problem lies. This is why it is necessary to have experienced failure analysis engineers will be able to spot the problem given their years of experience in the field.

The trend of more and more complex electronic devices shows no signs of stopping. Though some say we are approaching the theoretical limits of compression, there will always be more ways to extract juice out of the hardware. If anything, the role of the failure analysis engineer is set to expand dramatically in the coming years.

Most of us never think about the workings of the electronic devices use everyday. The ubiquitous integrated circuit finds a place in so many appliances across the board, that it's no longer a source of wonder. However, we mustn't forget the true miracle of this important component. For the overwhelming majority of us in our modern world, our lives revolve around the electronic chip. The circuits that make up electronic devices are probably some of the most complex networks known to man. Regular failure analysis of these electronic circuits is crucial to ensuring that they maintain their current levels of reliability. In fact, there's no doubt that the sophisticated levels of testing which are applied to modern electronic chips are largely responsible for their phenomenal success.

Many things can go wrong with an integrated circuit. In this article, we look at the ways and means by which we detect errors in the circuitry. These failures are particularly difficult to diagnose and localize because of the complex interactions by which the errors manifest themselves. The symptom of failure in the electronics might show itself somewhere down the line rather than actual the physical location where it's caused. Because of this, it requires a highly experienced and professional failure analyst to effectively uncover these issues and problems.

Failure Analysis of Electronic Circuits

Infrared thermography is the failure analysis technique which is used for detecting flaws in electronic circuitry. It's based around the principle that incorrectly performing circuits give off an unusual amount of heat. Do to the small currents involved, this extra heat needs to be detected using an extremely sensitive process. For this, we can use either liquid crystals or florescent micro thermal imaging.

It was something like this. A small amount of liquid crystal is spread along the area where the error is suspected to have occurred. It is applied in as thin a layer as possible. We can do this by rotating the chip quickly and allow the centrifugal force to spread it out. Once we have applied the liquid with the desired consistency, we can fire up the circuit. The liquid crystals have been designed in such a way that a certain amount of heat at a particular point will render it opaque. This can be picked up by sensitive instruments and we can isolate the heat producing element.

Florescent imaging, is almost the same except that instead of providing a single spot of heat, we can produce instead a heat map. It's more complicated, but diverges more useful information for the failure analysis of electronic circuits. These are just some of the methods which failure analysis engineers use to isolate faults in integrated circuit circuitry. And it is methods such as this which work under the hood to provide us with the current levels of durability that we have come to expect from our electronics today.

Integrated circuits these days are found everywhere and not just in computers, music players, and tablets. In fact they proliferate the world around us and are found in common everyday devices such as TVs, remotes, fans, and even the seemingly mundane – like toasters. But even though these devices seem very basic to us, they're all based on printed circuit boards or "PCBs" for short. These PCBs can be very complex – so complex in fact that it's likely that no single human being fully understands how one works, down to the very last electronic circuit. Given this level of complexity, it's not surprising that PCB failure analysis is a critical component of the process which enables us to use these devices day in and day out with a high degree of reliability.

There are many failure analysis techniques. Some of them destroy the chip completely. Others preserve it for future analysis. These techniques are called Non Destructive Testing techniques or NDTs for short. We prefer to use NDTs before making use of procedures which destroy the chip. Let's take a look at a few of these procedures which are so instrumental in giving us the electronic devices we use everyday.

Types of PCB Failure Analysis

Nondestructive techniques cover a wide variety of technologies. They range from light emission spectroscopy, to the application of sound waves in order to probe for defects in material. The use case scenario for each varies. For example, light waves are primarily used to isolate defects in the composition of the material. X-rays on the other hand are used to obtain a glimpse into the internal workings of the circuits.

Other procedures involve isolating defective circuits which can lead to overheating. For common users this can manifest itself as a burned remote for example. The rarity of these occurrences is a testament to the thoroughness of the testing which takes place before these devices are allowed to reach us.

When all other methods fail, we might have to open the packages or "decapsulate" them to confirm our suspicions of what has gone wrong. But sometimes it's the only way to be sure. It takes a great deal of experience and skill to correctly analyze a failed chip.

PCB failure analysis experts do their work in the background. They are unheralded and often overlooked. But without them, we would not have nearly the same reliability of electronics that we take for granted every day.

When analyzing the various things which can go wrong with an Integrated circuit, we can classify them in two ways. This classification helps in gaining a more comprehensive overview of the inner workings of a chip and the methods used to analyze it. The first way of classification is by the type of IC defect. For example, the problem can lie with the construction of the chip. Presence of voids, improper soldering and cracks in the material are some of the many ways a chip can be defective. These problems can arise at the time of manufacture or they can occur afterwards. Chips can also fail due to electrical defects. Some of these defects can be caused by the first type of error, but it need not be that way. Electrical defects are difficult to find and isolate due to the mind numbing complexity of the circuitry on an IC.

Next, we can have defects based on the purity or impurity of the material comprising the chip itself. The operating conditions of an integrated circuit are very delicate and the purity of the materials is of paramount importance. A deviation by even a fraction of a percent can skew the operating parameters and lead to an IC defect.

Classification of Methods

Failure analysis methods can also be classified based on the type of techniques used. For example, emission spectroscopy covers a wide range of techniques which can be used to discover a host of facts about the specimen in question. Using spectrograph techniques, we can analyze the composition of a wide range of materials and get truly sensitive measurements. This can even be used to obtain certifications to declare that the materials are free from hazardous substances as per the ROHS directive.

X-Ray and Infrared imaging allows us to view the chip from a variety of angles and perspectives to give us pictures we don't see with the naked eye. For example, we can detect minute hot spots which tell us about deviant electrical activity or look inside the chip to detect its workings.

Finally, we classify techniques by whether they destroy the chip or not. Destructive techniques involve opening up the package to expose its innards and get visual confirmation of a suspected flaw if such a thing is possible. Sometimes it's the only way to expose a flaw when all other methods fail. It's a last resort because we'd rather not destroy the chip.

There are some of the ways we can classify an IC defect depending on the type of error or on the type of method used to detect it.