Common mode feedback bias for low voltage opamps

Amplifiers – With semiconductor amplifying device – Including differential amplifier

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C330S253000

Reexamination Certificate

active

06388522

ABSTRACT:

FIELD OF THE INVENTION
This invention generally relates to electronic systems, and more particularly relates to common mode feedback bias for low voltage opamps.
BACKGROUND OF THE INVENTION
FIG. 1
shows a prior art two stage fully differential opamp for low voltage systems. Differential pair transistors M
1
and M
2
with active loads transistors M
3
and M
4
form the first stage, and common source amplifier transistors M
5
and M
6
with active load transistors M
11
and M
12
serve as a class-A output stage driving the output terminals OUTP and OUTM. Miller pole-splitting compensation is shown. Common mode feedback is applied with differential pair transistors M
7
and M
8
, loaded with current mirror transistor M
9
working into the first stage active load devices M
3
and M
4
. The desired level of output common mode voltage is applied at terminal CMDES, and the actual output common mode level is sensed with a pair of matched averaging resistors R
1
and R
2
. The common mode feedback drives the observed level to the desired level with the loop gain. Small capacitors C
1
and C
2
bypassing R
1
and R
2
improve the common mode loop phase margin.
This topology works fairly well, but it is possible that if the opamp is used in a circuit with overall resistive feedback, the common mode loop can settle in a stable operating point which drives both the output terminals to the positive rail. Suppose that in either an initial startup condition or a transient condition both the input terminals INP and INM are close to the positive rail. This will leave the input stage tail current source transistor M
13
with virtually no drain voltage, reducing its output current to very small levels. This will cause transistors M
3
and M
4
to lose control of the common mode output level at the output of the first stage, and second stage devices M
5
and M
6
will be below threshold. This lifts the output voltage close to the positive rail. If the feedback around the opamp is resistive, with no other bias source ensuring proper input terminal common mode biasing, then the input terminals will have their high levels latched, and the opamp will be in a stable but virtually useless state.
In general, this troublesome behavior is caused by the conflict between two common mode feedback loops, one being the intended one through the differential pair of transistors M
7
and M
8
to accurately control the output level, and the other the path through the overall feedback resistors through the input pair of transistors M
1
and M
2
. Usually the high drain resistance of tail source transistor M
13
degenerates the second feedback path, leaving only the intended one to act in the circuit. But if the input voltage is high enough to send transistor M
13
into the ohmic region, its drain resistance falls markedly, increasing the common mode gain through the opamp signal path. Since for a two stage amplifier the common mode gain is positive, this path may take over and latch the opamp state.
Prior art designs have addressed this problem by including two additional devices M
15
and M
16
as shown in FIG.
2
. An appropriate bias voltage is applied to terminal BIAS such that in normal operation, with the input terminals at their desired common mode level, transistors M
15
and M
16
will be held subthreshold by the tail voltage at the sources of transistors M
1
and M
2
. The common mode feedback loop operated normally, and transistors M
15
and M
16
do not contribute any noise into the opamp. However, if the input levels ever rise toward the positive rail, transistors M
15
and M
16
will conduct, maintaining bias current to transistors M
3
and M
4
, and ensuring that the common mode loop will not lose control and latch.
This simple arrangement unfortunately becomes unusable at very low supply voltages. Generally, the lower supply limit for an opamp of this topology will be determined by the supply voltage where tail current transistor M
13
goes into the ohmic region at the prescribed input common mode level. Therefore, operating the opamp at the minimum supply voltage means that there is no value of the BIAS voltage in
FIG. 2
which will leave transistors M
15
and M
16
off, and not degrading performance, yet enabling transistors M
15
and M
16
to catch the condition that transistor M
13
is entering the ohmic region.
SUMMARY OF THE INVENTION
An opamp with common mode feedback bias includes: a first differential pair having first and second inputs; active load devices coupled to the first differential pair; a common mode feedback circuit coupled to the active load devices for controlling the active load devices; a second differential pair having a first input coupled to the first input of the first differential pair and a second input coupled to the second input of the first differential pair; and current drivers having control nodes coupled to the second differential pair and outputs coupled to the active load devices.


REFERENCES:
patent: 4797631 (1989-01-01), Hsu et al.
patent: 6052025 (2000-04-01), Chang et al.
patent: 6140877 (2000-10-01), Forbes
patent: 6265941 (2001-07-01), Lopata

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Common mode feedback bias for low voltage opamps does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Common mode feedback bias for low voltage opamps, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Common mode feedback bias for low voltage opamps will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2831733

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.