Biomedical Engineering Reference
In-Depth Information
This efficient wiring plan allows the 181 pins to be wired on a two-layer
PCB. Recall that the direct-addressing method needs 1284 control pins,
which requires a four-layer PCB, thereby increasing the fabrication cost
by a factor of 1.6-2 [72]. Moreover, the 181 pins can be easily incorporated
using standardized 3 mil feature size technology. In contrast, to fit the 1284
pins in the direct-addressing-based design, 2 mil technology, which usu-
ally costs 3-5 times more than 3 mil technology, has to be used. Therefore,
the pin-constrained design achieves a reduction of fabrication cost by a fac-
tor of 5-10×. The reduction is more significant when the wiring-plan design
cost is considered.
In Figure 6.6, every well unit has the same pattern of pin assignment.
This is because the dimension of the unit well is the same as the period
of pin-assignment patterns from the Connect-5 algorithm. This regular pin-
assignment result facilitates the use of an efficient well-loading algorithm,
discussed in Subsection 6.1.2.
6.1.2 Shuttle-Passenger-like Well-loading Algorithm
In this section, we focus on the problem of loading the wells with sample and
reagent droplets on the pin-constrained chip. The goal is to efficiently route
the sample and reagent droplets to their destination wells. Note that, in the
96-well chip design in Figure 6.6, every 6 × 6 well unit has the same pattern
of pin assignment. Therefore, any sequence of manipulations in a single well
unit will cause the same manipulations in all the other well units. Although
this “synchronizing” property leads to reduced freedom of droplet manipu-
lations, it allows the concurrent manipulation of multiple droplets. Based on
this observation, we present a parallel shuttle-passenger-like routing method
for high-throughput well loading.
We illustrate the well-loading algorithm using an example. Figure 6.9
shows a pin-constrained chip that consists of four 6 × 6 well units. A dispens-
ing reservoir is located at the top right corner on the chip. Three droplets, D
1
,
D
2
, and D
3
, are to be dispensed and routed to three destination wells. If the
droplets are placed on the start points as indicated in Figure 6.9, the rout-
ing can be carried out simultaneously by applying the control-pin actuation
sequence 5→2→4→1→3→5→4→3→2→1. The actuation sequence will route
all the droplets (if any) at the upper left corner of the well units to the well
within the same unit, just as synchronized shuttles that carry passengers
from fixed start points to fixed destinations. The shuttles run regularly irre-
spective of whether there is any passenger. To go to a specific destination, a
passenger needs to get to the correct starting point and wait for the shuttle
(pin actuation sequence) for pickup and routing to the destination (well).
Routing of droplets to the starting point can also be carried out using the
shuttle-passenger-like method. As in the example in Figure 6.9, the routing
step can be carried out using the shuttle (pin activation sequence) as shown
in Figure 6.10.
