Page 146 - NEW Armstrong Book - 2

VSOURCE, ISOURCE, TCASE(MAX), TCASE(MIN), TAMBIENT, ΔTCASE AMBIENT, number of executed cycles (TON/TOFF)
The stress and measurement system created for this type of test is capable of monitoring the source current (IS) at the end of each “on” phase. This current is constantly compared with two reference thresholds that signal, first, a warning if the measured current is contained in an interval between [IMIN_WARNING; IIMAX_WARNING] and, second, a fail if the measured current is contained in an interval between [IMIN_FAIL; IIMAX_FAIL].
A dedicated HWBIN (hardware binning, a fault marker during the testing) is associated with each measurement interval. If the measured current is greater than the second measure- ment threshold, the system is able to deselect the device on the reliability card, preserving it from any catastrophic break- downs. In these conditions, the single device is no longer subject to electrical stimulation and stops undergoing ther- momechanical stress.
In addition to measuring the source current, the system is capable of constantly measuring the temperature on the case of the device in the “on”/”off” phases (TCASE(MAX), TCASE(MIN)) and the ambient temperature reference (TAMBIENT).
As for the high-temperature reverse bias (HTRB)/high- temperature gate bias (HTGB)/RDS(on) mode tests for these families of discrete devices, Eles has developed an auto- mated methodology that allows, in the face of an ambient temperature of 150˚C/175˚C, runtime during the polariza- tion phase of the device:
In addition to a series of IP hardware that acts as a real measure-management unit, a multisite solution has been developed, both for IOL and HTGB/HTRB tests, capable of stressing and measuring devices in different packages whether they are soldered on card-edge (PowerFLAT HV), thus using specific connectors, or in a Hip247 package using the appropriate socket. For these tests, there is a completely automated management with high precision of the quanti- ties to be measured, guaranteeing the controllability, reach- ability, and observability of the device during the various tests and in the various status changes.
These tests play an important role in the reliability of these devices as the failure mechanisms that are able to trigger to highlight a series of defects, such as:
• BVDss and/or Vth degradation (the presence of contaminants in the diffusion process and/or misalignment of the process masks)
• High leakage (IGSS and IDSS) due to wrong diffusion
•High leakage/short-circuit due to passivation problems (creation of silicon oxide [SiO2] used as an insulator in amorphous silicon structures)
Antonio Zaffarami
is chairman of the Eles Board of Directors.
HTRB: HTGB:
VDRAIN = VD, VGATE = 0, TROOM = 175˚C
VDRAIN = 0, VGATE = VG, TROOM = 150˚C
VDRAIN = VD, VGATE = VG, TROOM = typ˚C, IDS = 250 μA
RDS(on) mode:
(with current limitation 1 A)
The IGSS, IDSS, and RDS(on) measurements are performed during the stress cycles, avoiding having to carry them out after a phase of blind stress on specific testbenches that use par- ticular tracers or parameter analyzers. Also, for this type of test, the reliability boards have a parallelism (N) whose limit is fixed by the mechanical dimensions of any sockets/con- nectors on which the devices are housed.
The devices on the reliability card are configured in one of the three ways mentioned above. Once the setup of the polariza- tion parameters of the device is set, the test is started, and the test environment (TAMBIENT) is created and checked with its fully automatic adjustment.
132
ASPENCORE GUIDE TO SILICON CARBIDE