Miniaturization trends in integrated circuit (IC) technology have
caused many testing problems. As bigger packaged ICs with higher pin
counts are more densely packed onto a printed circuit board (PCB),
accessing an IC's pins is harder. No longer are the pins mechanically
accessible to probes or a bed-of-nails fixture. Therefore, determining
which IC or interconnect is faulty is difficult or
... [Show full abstract] impossible. Because
each IC's input pins cannot be controlled, and each IC's output pins
cannot be observed. The boundary scan method was developed with the goal
of improving this controllability and observability problem. A
shift-register is included next to each IC pin so that input and output
values can be serially shifted in and out. This reduces the need to use
probes to control and observe. Also, the output of each IC's scan
register can be connected with the input of another IC's scan register.
This effectively creates one big scan chain per PCB, further reducing
points that must be mechanically probed. The inclusion of a
scan-register on each IC allows: 1) the observation of each IC during
normal operation; 2) the test of interconnects between ICs, and 3) the
isolation of the IC from others so it can test itself. The IEEE Standard
1149.1 Test Access-Port and Boundary Scan defines the test logic for
implementing a boundary scan test architecture. Example circuits were
designed in CMOS. A boundary scan cell is described