December 04, 2019 by Sam Holland
ICs (integrated circuits) are ubiquitous in electronic devices: they perform many functions (such as data processing), and consist of countless discrete components integrated on a tiny silicon chip.
ICs may of course fail, however, leading to equipment malfunction: this article explores some prominent causes of IC failures and how to act on them.
ICs fail due to several factors. The most frequent causes of IC failure can be categorised under manufacturing defects, physical damage to the packaging, harsh operating temperature conditions, and electrostatic discharge (ESD).
The next sections show a closer look at each of these problems:
Impurities and contaminants within the bulk silicon, crystal defects, and design issues during IC fabrication contribute to stability issues in the final chip. Small clearances between the die and the substrate (i.e. die-attach failure) can lower the thermal conductivity of the die.
This problem can cause a die to overheat and crack under normal operation, leading to premature failure. Also, tiny cracks at the interface between the die and bond wire, due to excessive bonding pressure (i.e. wire-bond failure), can adversely impact the reliability of ICs.
Cracks and chips on the packaging of integrated circuits (due to high mechanical stress or process temperatures, etc.) can reduce IC reliability and cause failures during operation. Moisture and cleaning chemicals absorbed through cracks can also short internal components, leading to malfunction; and on top of this, IC packaging can be damaged during transit or due to improper handling.
An overheated integrated circuit. Image Credit: Wikimedia Commons.
High operating and ambient temperatures can cause thermal overstress in ICs. An essential metric for ICs is the junction temperature, after all: ideally, each IC’s junction temperature should not exceed 125 to 150°C to avoid excessive internal stress. Indeed, utilising ICs above their maximum-rated voltage and current can cause electrical overstress, which can lead to catastrophic failures. Consider, for example, that cooling system (such as heat sink and cooling fan) failures in consumer electronics, such as laptops, can cause ICs to heat up at a critically fast rate.
Static electricity generated during the fabrication, handling, and storage of ICs can damage the internal circuitry due to the triboelectric effect. ICs are more likely to fail when exposed to ESD. The resulting electric fields induced in the IC may break down oxide layers and junctions within the device and/or increase the current flow—once again causing overheating.
ICs are far from easy to troubleshoot, particularly when they’re already mounted on a PCB. At least, using the following techniques, you can fairly easily determine if an IC is defective and needs to be replaced:
Again, physical damage to ICs can cause them to malfunction of course. Performing a visual inspection of the chip can identify problems such as cracks, chippings, and charring that may have been caused by high mechanical impact or overheating.
Two 24-pin integrated circuits integrated on a printed circuit board. Image Credit: Pixabay.
One of the best ways to ascertain the condition of an IC is to carry out a short-circuit test. You can test for shorts using a digital multimeter in the following steps:
Set the multimeter to its continuity function
Bridge all the pins on one side of the IC and connect it to one of the multimeter’s test probes
Test each of the IC pins with the remaining multimeter test probe: if there’s continuity in more than 50% of the pins, then the IC is likely shorted
To check if an IC overheats:
Turn on the device (or supply power to the inputs of the IC for 10 to 15 seconds)
Touch the surface of the IC with the tip of one finger to check its temperature
Remember that ICs that become very hot after 10 to 15 seconds are likely defective.
An Arduino microcontroller that has two integrated circuits. Image Credit: Pixabay.
Checking the input/output voltages of ICs—namely to see if they deviate from the preset values specified by the manufacturer’s technical document—is perhaps the most accurate way to identify a defective component.
You can use a multimeter to compare the DC input/output voltages of the IC against the values specified in the said technical document. Large disparities in measured and specified voltages indicate a defective component. You can also use an oscilloscope to test ICs more accurately: the basic procedure is to connect the ground test probe of the oscilloscope to a suitable ground point and the other to each of the IC pins. Analysing the output waveform shows whether the IC is good or bad.
Like all manner of circuit components, ICs are prone to fail during use. The said manufacturing defects, physical damage, electrostatic discharge, and less-than-ideal operating conditions are just some of the factors that can significantly impact the reliability of these miniature components.
All in all, the purpose of troubleshooting integrated circuits is often not to repair them (again, fixing ICs is incredibly complex and chips are rather inexpensive to replace); rather, it’s to understand the underlying causes of failures and how they contribute to equipment malfunction.
