Browse Prior Art Database

Optimization of High-Current Distributed Nets in Bipolar Masterslices

IP.com Disclosure Number: IPCOM000113017D
Original Publication Date: 1994-Jun-01
Included in the Prior Art Database: 2005-Mar-27
Document File: 2 page(s) / 90K

Publishing Venue

IBM

Related People

Concilio, S: AUTHOR [+2]

Abstract

Disclosed is a method for developing the wiring of a bipolar masterslice so that widely distributed signal lines are auto-wired with standard routing software. Certain high current nets are treated as input/output lines formed as relatively large conductors strategically shaped and pre-placed for the attachment of other nodes through individual wiring.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Optimization of High-Current Distributed Nets in Bipolar Masterslices

      Disclosed is a method for developing the wiring of a bipolar
masterslice so that widely distributed signal lines are auto-wired
with standard routing software.  Certain high current nets are
treated as input/output lines formed as relatively large conductors
strategically shaped and pre-placed for the attachment of other nodes
through individual wiring.

      The Figure shows the physical design process of this method, in
which a high-current net 10 is wired as part of a chip within a
perimeter 12.

      Within a bipolar integrated circuit, the DC currents required
to hold logic levels are typically much higher than such currents
within a CMOS device, where only negligible currents are required.
While this effect is generally not a problem in terms of the drivers
required, since bipolar integrated circuits typically include drivers
capable of driving high currents, the net or wire carrying this
current must also be configured for high current capability.

      For each particular wiring material, such as aluminum, the
current-carrying capacity of a wire is primarily a function of the
cross-sectional area of the wire.  With the thickness of the wire
fixed as a function of the technology used to produce the integrated
circuit, a high-current net must either be designed as a single wire
of sufficient width to carry the required current, or as a number of
parallel wires of a standard width which are together sufficient for
carrying the current.

      The wide single wire solution provides an optimum physical
solution, resulting in a net having the exact width necessary for
carrying the required current.  This method does not use additional
wiring area needlessly and does not increase the capacitance of the
net.  Such capacitance can slow the transmission of a signal.
However, the implementation of this method has traditionally required
an entirely manual placement of the high current net before
auto-wiring the remaining nets.  This process, which is therefore
time-consuming and error-prone, does not lend itself to modificatins
in the implementation of design changes.

      The solution using multiple parallel wires is desirable from a
viewpoint of design efficiency.  The process is completely
auto-wirable, as the requirements for high-current nets a...