Amplifiers – With semiconductor amplifying device – Including differential amplifier
Reexamination Certificate
2002-12-19
2004-05-25
Choe, Henry (Department: 2817)
Amplifiers
With semiconductor amplifying device
Including differential amplifier
C330S254000, C327S359000
Reexamination Certificate
active
06741129
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to operational amplifier (op-amp) slew rate boosting schemes, and more particularly to a technique to boost the slew rate of a fully differential op-amp which has a closed-loop gain that is unity or less.
2. Description of the Prior Art
Fully differential folded cascode op-amps with a closed-loop gain that is unity or less are well known in the prior art. One such op-amp is illustrated in FIG.
1
. The circuit of
FIG. 1
includes input differential pair
20
; transistors
22
-
25
; output buffers
26
and
28
; compensation capacitors
30
and
32
; resistors
34
and
36
; input resistors
44
and
45
(Ri); feedback resistors
46
and
47
(Rf); input nodes IN+ and IN−; tail current source I
tail
; bias voltage V
bias
; common mode feedback V
CMF
; and output nodes OUT+ and OUT−. The output buffers
26
and
28
are typically class AB buffers. The equations, which govern the operation of the op-amp, are also very well known and have been documented by Paul R. Gray and Robert G. Meyer in
Analysis and Design of Analog Integrated Circuits
, pp. 646—onward, John Wiley & Sons, Inc., 3
rd
Edition. It has been shown that if the op-amp is to be made stable, one must control its unity gain crossing frequency. The conventional technique for controlling this unity gain crossing frequency is to use compensation capacitors
30
and
32
that couple nodes
40
and
42
to ground. The frequency where the open loop gain falls to unity for the op-amp is then defined as:
f
unity
=
G
m
2
π
C
comp
(
1
)
where G
m
is the transconductance of the input stage
20
and is determined by the tail current I
tail
shared by the input stage
20
transistors, and C
comp
is the capacitance value for capacitors
30
and
32
. Specifically, G
m
is defined as:
Gm
=
qI
tail
2
kT
(
2
)
The slew rate (maximum rate of change in output voltage for large input signals) for the op-amp is defined as:
Slewrate
=
I
tail
C
comp
(
3
)
where it is well known the compensation capacitance C
comp
must be set to a sufficiently large value in order to make the op-amp stable. Conventional op-amp compensation techniques provide a high small signal open loop voltage gain at a high output impedance at the op-amp gain node. This high output impedance in combination with the compensation capacitor then operates to place the dominant pole. When a conventional op-amp uses a compensation capacitor to achieve stability, all of the input stage tail current will flow into the compensation capacitor in response to a large step input voltage since one end of the compensation capacitor is coupled to the high impedance gain node while the opposite end of the compensation capacitor is coupled to an AC ground. The tail current will charge the compensation capacitor to achieve a slew rate defined by equation (3) above.
If more than two poles exist before the unity gain frequency f
unity
is reached, the op-amp will be unstable. Specifically the compensation capacitors
30
and
32
operate to push the first dominant pole down to a low enough frequency such that the op-amp gain falls below unity before the second pole is reached. Setting the value of compensation capacitance C
comp
to achieve the above desired stability characteristics therefore also establishes the maximum slew rate for the op-amp as set forth in equation (3) above. It is, of course, desirable to have a very high slew rate. In view of the foregoing, it can be appreciated that achieving a very high slew rate requires reducing the value of the compensation capacitance C
comp
. Reducing the value of the compensation capacitance C
comp
however, makes the op-amp less stable since the first dominant pole will then be moved to a higher frequency as discussed herein before. The desired value of the compensation capacitance C
comp
and the desired high slew rate are therefore in direct conflict with one another.
One conventional technique used to address the direct conflict between the desired value of the compensation capacitance C
comp
and the desired high slew rate includes reducing the value of transconductance G
m
associated with the input stage
20
by inserting resistors into the emitter paths of the input stage
20
transistors. A lower value for the compensation capacitance C
comp
can then be used to achieve amplifier stability as seen by equation (1) above, which also then increases the slew rate as seen by equation (3) above. This technique however, is problematic in that the emitter resistors added to the emitter paths of the input stage
20
transistors introduce additional noise that cannot be tolerated in specific applications such as when the op-amp is driving a load that is connected to ground in response to an input signal such as might be used to accommodate ADSL systems.
In view of the foregoing, a need exists for a technique to boost the slew rate of fully differential amplifiers having a closed-loop gain of close to unity or less without introducing additional noise into the system.
SUMMARY OF THE INVENTION
The fully differential amplifier slew rate boosting scheme for use with an amplifier having a closed-loop gain very near unity or less has the first plates of the compensation capacitors conventionally coupled to internal high impedance gain nodes, but has the other plates of the compensation capacitors unconventionally driven with the input signal. The voltages appearing across the compensation capacitors in response to changes in the input signal is significantly less than that achieved using conventional compensation architectures where the other plates of the compensation capacitors are coupled to ground. Since little current is now required to charge the compensation capacitors, the input stage tail current no longer limits the slew rate.
REFERENCES:
patent: 6188282 (2001-02-01), Montalvo
patent: 6304143 (2001-10-01), Corsi et al.
patent: 6404263 (2002-06-01), Wang
patent: 6605994 (2003-08-01), Kim et al.
Corsi Marco
Hellums James R.
Choe Henry
Stewart Alan K.
LandOfFree
Differential amplifier slew rate boosting scheme does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Differential amplifier slew rate boosting scheme, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Differential amplifier slew rate boosting scheme will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3210038