ESD Specification Models
Using a series 1.5K-ohm resistor to a 100-pF capacitor. Specification is 2kv with a goal of 4kv. The ramp rate is specified to be 10 nanoseconds or less. Tests for all of these models must be run on a certified ESD tester due to the need for very high-frequency components necessary to guarantee the very fast ramp rates, high-voltage inputs and a safe test environment for the operator. Inductance of the wiring interconnect must be minimized by utilizing short wires, braided 3/8-inch copper or better and appropriate ground planes. In addition, special high-speed, high-voltage capacitors are required for both the 100 pF and 200 pF capacitors.
Using “no series resistor” and 200-pF capacitor, specification is 250 volts with an acceptance of greater than 350 volts and a goal of 700 volts.
Requires a special setup for each package pin type. The “standard” new test may not be accepted. The specification is 1,000 volts with a design goal of 1,500 volts. There are a number of aspects that must be addressed for good ESD protection. Each of the three conditions above addresses a different cause of failure, and each must be considered in order to guarantee a robust design. Even though the cause of failure may vary, many times the failure mechanism within the IC may be the same or very similar. ESD protection is "from a given pin to any other pin.”
ESD CHALLENGES: ESD SPECIFICATION MODELS: HUMAN-BODY MODEL
The Human-Body Model (HBM) has a fast rise time to a very high voltage, but the 1.5K-ohm resistor limits the current to a maximum of 2.67 amperes at 4kV with a decaying RC time constant set by the 1.5K-ohm resistor and the 100-pF capacitor. The type of failure represented by this model is produced by static charge buildup on people handling the integrated circuits in the lab/engineering areas or in the marketing/device transfer areas.
In addition, consumers can produce this type of failure in a dry static-charge buildup environment when they touch an electronic product that does not have appropriate grounding or shielding. Similar ESD damage can occur in automotive environments when, for example, a passenger with a fur coat slides into an automobile in cold, dry conditions and touches a conductor to an electronic component that isn’t adequately protected.
Quantitative investigations in this type environment have shown electrostatic voltages as high as 30,000 volts. Typically the on-chip ESD protection circuitry is either nonexistent or does not perform fast enough to give adequate protection to devices that fail this test. The HBM ESD test is the primary type of acceptance test used for device qualification in the United States.
ESD CHALLENGES: ESD SPECIFICATION MODELS: MACHINE MODEL
The Machine Model (MM) has both a fast rise time and no current limiting resistor, and therefore the peak currents can be as high as 30 amperes for a very short duration depending on the length of wire and the series inductance of the wires used in test hardware for performing this test. In addition, the high-frequency characteristics of the 200-pF capacitor can significantly affect the results of this test. The peak voltage is much lower than the Human-Body Model (HBM) test, but the fast rise time and the peak current can generate failures that aren’t detected by the HBM test.
This type of failure is generally produced by test machines on production floors or by machines that handle electronic components for both test and packaging. The cause of this problem is inadequate grounding of the machines through the use of electrical isolation bushings, anodized aluminum (which produces clean surface finishes but poor electrical conduction) and extensive use of plastics. The Machine Model ESD test is the primary type of acceptance test for device qualification in Japan.
ESD CHALLENGES: ESD SPECIFICATION MODELS: CHARGED-DEVICE MODEL
Like the other two models, the Charged-Device Model (CDM) has fast rise times. It is used to simulate charge buildup that occurs during handling. The type of failure generated is similar to the Human-Body Model (HBM) but with a key difference: The entire device is charged to the high voltage and then discharged to ground. Therefore the ESD energy may travel paths during the discharge time that are different from those identified with either HBM or the Machine Model (MM). As a result, the CDM detects marginal design problems that may not be detected with either MM or HBM. Due to the uniqueness of the CDM test, a special test fixture is required for each package-pin type. CDM is currently proposed but may not be accepted.