Miscellaneous active electrical nonlinear devices – circuits – and – Specific input to output function – By integrating
Reexamination Certificate
2002-01-30
2003-08-19
Cunningham, Terry D. (Department: 2816)
Miscellaneous active electrical nonlinear devices, circuits, and
Specific input to output function
By integrating
C327S345000, C327S336000
Reexamination Certificate
active
06608516
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to electronic circuits and, more particularly, to adjusting the time constant of an integrator.
DESCRIPTION OF THE RELATED ART
Instrumentation or data acquisition (DAQ) systems are often used to obtain measurement data pertaining to physical phenomena (e.g., force, pressure, acceleration, etc.). Such measurement data is useful in laboratory research and testing, process monitoring and control, and control of mechanical or electrical machinery, to name a few examples. A typical analog instrumentation system includes signal conditioning circuitry which performs one or more “signal conditioning” functions upon analog measurement signals. Such signal conditioning functions include amplification, filtering, and direct current (DC) level shifting.
Signal conditioning circuitry often includes filters used to provide anti-aliasing and noise reduction. Filters provide noise reduction by filtering out extraneous signals above and/or below the frequency range of interest. Anti-aliasing filters are often used to attenuate signals input to or output from a sampling system that have frequencies that exceed half the sampling (or Nyquist) frequency in order to prevent distortion that may arise due to aliasing.
In general, signal filters pass signal frequencies within a pass band and attenuate signal frequencies within a stop band outside of the pass band. A “cutoff frequency” or “corner frequency” f
c
defines a boundary between the pass band and the stop band. Common types of signal filters include low pass, high pass, and band pass filters. The pass band of a low pass filter extends from DC (0 Hz) to fc, and the stop band of a low pass filter lies above fc. A high pass filter has a pass band above fc, and a stop band including frequencies below fc. Graphs of ratios of output voltage to input voltage versus frequency for real (i.e., non-ideal) low and high pass filters have finite slopes within the stop bands. Low pass and high pass filters have quality factors or “Qs” which determine the slopes of such graphs within the stop bands.
Band pass filters have a pass band extending between a low corner frequency fl and a high corner frequency fh. Low corner frequency fl defines a boundary between the pass band and a first stop band including frequencies below fl, and high corner frequency fh defines a boundary between the pass band and a second stop band including frequencies above fh. A band pass filter has a center frequency fo representing a geometric mean off fh and fl, a bandwidth bw, and a selectivity or Q, where:
fo={square root over (fh·fl)},
bw=fh−fl
, and
Q
=
fo
bw
Active filters typically use components such as operational amplifiers (op amps), resistors, and capacitors to implement a desired signal filter. Active filters are able to provide signal gains of greater than unity. A state variable filter is a versatile type of active filter circuit that provides low pass, high pass, and band pass outputs simultaneously.
As filtering requirements change from application to application, it is highly desirable to be able to vary the characteristics of active filters (e.g., state variable filters) without having to replace resistors and/or capacitors with like components having different values. In many data acquisition systems, filters have electronically programmable bandwidths and/or corner frequencies, making it unnecessary to physically adjust the filter when reconfiguring the data acquisition system.
FIG. 1
is a circuit diagram of an exemplary prior art state variable filter
10
with programmable characteristics. State variable filter
10
includes two programmable multiplying digital to analog converters (MDACs)
12
A and
12
B that are used to attenuate the input to integrators
16
A and
16
B respectively. Digital values provided to and stored within the MDACs (collectively referred to as MDACs
12
) determine the bandwidth of state variable filter
10
.
Resistors R
1
and R
4
are coupled to op amp OA
1
to form an inverting summing amplifier with a gain of −R
4
/R
1
. The summing amplifier sums the input signal, the amplified band pass signal, and the low pass signal to produce a high pass signal.
The high pass signal output from OA
1
is input to MDAC
12
A. MDAC
12
A attenuates the signal being provided to resistor R
6
. Resistor R
6
and capacitor C
1
are coupled to op amp OA
3
to form inverting integrator
16
A. Inverting integrator
16
A has a time constant equal to the product of resistance value R
6
and feedback capacitance value C
1
. Together, MDAC
12
A and inverting integrator
16
A have a time constant equal to the gain of the MDAC
12
A divided by resistance value R
6
and feedback capacitance value C
1
. Inverting integrator
16
A produces the band pass signal and provides an output signal to OA
2
through R
7
and to the input terminal of MDAC
12
B.
OA
2
is coupled to resistor R
5
and resistor R
7
to form an inverting amplifier. OA
2
amplifies or attenuates the band pass signal output from OA
3
. The adjusted band pass output is provided to the input of OA
1
through resistor R
3
.
MDAC
12
B outputs a signal to inverting integrator
16
B. Inverting integrator
16
B includes resistor R
8
, capacitor C
2
and op amp OA
4
. MDAC
12
B and inverting integrator
16
B form another programmable time constant integrator. Inverting integrator
16
B produces the low pass output at an output terminal. The low pass output is provided through resistor R
2
to the input of OA
1
.
The bandwidth for a state-variable filter is typically set by scaling the time constants of the integrating stages within the filter. The time constant for an integrator is commonly made adjustable by coupling an MDAC to an input of the integrator so that the time constant of the MDAC-integrator combination can be adjusted, as shown in FIG.
1
. The addition of the MDAC to the integrator adds a time constant multiplier G the integrator's time constant. This multiplier G is the gain of the MDAC. Typically, G is less than one.
Two potential problems may arise in the programmable filter design shown in. FIG.
1
. One potential problem is DC offset. For example, if the gain of the MDACs
12
in
FIG. 1
is relatively high, the DC path through the filter will be strong and feedback in the circuit will help to nullify the effects of DC offset. However, if the gain of the MDACs
12
is set very low, the DC path through the filter may be constricted, and the DC offset of the filter may dramatically increase. In addition, the variable gain amplifier may contribute its own DC offset. To make matters worse, the DC offset of the filter may be directly dependent on the filter's bandwidth. Thus, there may be a different DC offset associated with each filter bandwidth setting.
The second potential problem is noise. The time constant of the variable gain integrator is modified by controlling the signal gain going into the integrator. While this changes the time constant of the overall circuit, the integrator itself still has the same time constant independent of any gain setting. The net result is that the inherent noise of the circuit may increase as the filter's bandwidth setting decreases. Unless low noise components are used in the circuit, this may create undesirable noise effects.
SUMMARY OF THE INVENTION
Various embodiments of a system and method for adjusting the time constant of an integrator are disclosed. In one embodiment, a variable time constant integrator includes an amplifier configured to generate an output signal, a capacitor coupled to provide feedback to the amplifier, and a variable gain element coupled to the output of the amplifier and to the capacitor. The variable gain element is configured to provide the product of a gain and the output signal to the capacitor. The variable gain element is also configured to receive an indication of a new value of the gain and to responsively set the gain equal to the new value of the gain. For example, the indication may be a voltage, and a level of the vo
Cunningham Terry D.
Hood Jeffrey C.
Meyertons Hood Kivlin Kowert & Goetzel P.C.
National Instruments Corporation
LandOfFree
Adjustable time constant integrator does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Adjustable time constant integrator, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Adjustable time constant integrator will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3092552