Browse Prior Art Database

Active Load with Feedback Controlled Self Terminating Dynamic Pull Down Current for High Speed Bipolar Circuits

IP.com Disclosure Number: IPCOM000109181D
Original Publication Date: 1992-Aug-01
Included in the Prior Art Database: 2005-Mar-23
Document File: 4 page(s) / 135K

Publishing Venue

IBM

Related People

Anderson, CJ: AUTHOR [+3]

Abstract

Emitter-follower with resistor or constant-current pull-down has been widely used in high-speed bipolar circuits due to the simplicity of topology and the emitter-dotting capability. While various active- pull-down schemes have been proposed to reduce the power consumption and improve the pull-down delay of the emitter-follower stage, the lack of emitter-dotting capability in these schemes has limited their applications.

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

Active Load with Feedback Controlled Self Terminating Dynamic Pull Down Current for High Speed Bipolar Circuits

       Emitter-follower with resistor or constant-current
pull-down has been widely used in high-speed bipolar circuits due to
the simplicity of topology and the emitter-dotting capability.  While
various active- pull-down schemes have been proposed to reduce the
power consumption and improve the pull-down delay of the
emitter-follower stage, the lack of emitter-dotting capability in
these schemes has limited their applications.

      Disclosed is an active load with a feedback-controlled
self-terminating dynamic pull-down current to improve the power-delay
of high-speed bipolar circuits.  The active load preserves the
emitter-dotting capability and improves the pull-down delay with
minimum power consumption.  The schematics of the active load is
depicted in Fig. 1 where VCC1 N GND.  When the output node T is at
'High' (0.25 V as shown), Q1 is conducting and node B is at its low
state (-0.35 V as shown).  Q2  is, therefore, at cut-in condition
(barely on with almost no current). When the voltage at node T starts
to drop, the voltage at node B rises, turning Q2 heavily on to
quickly pull down the output node T.  As the output reaches its low
state, Q1 is at cut-in condition with only a very small current,
while Q2 is now conducting.  Hence, at steady-state, one of the
branch is not conducting (either Q1or Q2), thus reducing the power
consumption.

      A preferred scheme is disclosed in Fig. 2.  In this scheme, the
major improvements are:
1.  The resistor RC1 is returned to the output node.  This eliminates
the extra connection to the supply line (VCC1 in Fig. 1), enabling
the active load to be attached directly between the output node T and
the termination voltage VT without any extra connection.
2.  The branch containing RL (in Fig. 1) is eliminated, thus reducing
the power consumption.
3.  In steady-state, there is only one branch of current. When the
o...