Design-for-Testability for Digital Microfluidic Biochips - Digital Microfluidic Biochips - page 140

in table T 3 . The broadcast-addressing method is then applied to replace the
don't-care terms in table T 3 and generate the eventual pin assignment.
Note that the addition of test operations into the bioassay may result in an
increase in the number of control pins, compared to a test-unaware design
using broadcast addressing. As shown in Figure 5.3 (table T 1 ) , before the split-
ting test is added, we can map the two don't-cares in the activation sequence
for E 1 with “10” and map the two don't-cares in the activation sequence for E 4
with “01” to make the two sequences identical. Therefore, the corresponding
electrodes E 1 and E 4 can be connected to a single control pin. As a result, only
three control pins are needed to control the linear array. However, when the
splitting test is added, activation sequences in table T 3 become incompatible.
Therefore, they have to be controlled independently. The linear array now
requires four control pins.
5.2.2 euler-Path-based Functional Test Method for
irregular Chip layouts
The test operations used in the previous testability-aware pin-constrained
design method can be determined using the functional test method in
Chapter 4, Section 4.5. However, this approach requires a rectangular array
structure for the chip under test. As discussed in Section 5.1, to reduce cost
in practical designs, unused electrodes are often removed from the array,
resulting in an irregular chip layout. Irregular layouts also result from the
need for allocating routing tracks under the fluidic layer for connecting the
electrodes to chip pins. In this subsection, we present an Euler-path-based
method for the functional testing of such irregular chip layouts.
For simplicity, we focus on the functional testing of two widely used
microfluidic modules: mixers and splitters. According to the functional test
method described in Section 4.5, a mixing functional test can be reduced to
a droplet-merging test, which actuates a series of three adjacent electrodes to
determine whether two droplets can be merged on them. A split operation can
be viewed as the reverse of droplet merging. Consequently, these two tests can
be combined into a unified splitting and merging test application procedure.
In a splitting and merging test for a single functional unit, a test droplet
is routed to the center electrode of the three-electrode cluster, split, merged,
and finally routed back to a detection site for test readout (Figure 5.4) . To
carry out the mixing and splitting functional test for a chip, this basic
Figure 5.4
The mixing and splitting test for a functional unit.