The present invention relates to an electronic circuit as described in the preamble of claim 1, and to a method for realising such an electronic circuit as is described in the preamble of claim 13.
Methods and electronic circuits for realising a desired filter transfer function have been described extensively in the past. Amongst them one can count the classical components-based filter realisations using discrete electronic components as resistors, capacitors and inductors, such as the Chebychev, Butterworth and other filter techniques. Also active filter methods using operational amplifiers and switch-capacitor techniques have been extensively described in literature.
A drawback of the discrete filter realisations is that they are using lumped elements, more in particular lumped inductors. These are bulky and expensive. Although attempts were made in the past years to integrate these devices on silicon, they still remain bulky and consume a lot of integrated circuit area, especially when they have to withstand high voltages and currents.
Active filters can be implemented as part of an integrated circuit. However filters that are to operate at high voltages and currents are generally not realisable by means of active filters since they would also require expensive high voltage technologies.
Another method for realising filters, as used for instance in splitters in telephone circuits, was described in the published PCT patent application WO 03/061121. Therein the low-pass filter of a splitter on a telephone line is realised as an active element in series on each transmission line and preceded on the line side by an inductance type passive element. A drawback of this method is that the active elements in the series path between input and output contribute to a series voltage drop which is undesirable.
It is therefore an object of the present invention to provide an electronic circuit and a method for realising such an electronic circuit of the above known kind, but which is able to operate at high voltages and which does not have a series voltage drop associated with it.
According to the invention this object is realised by the fact that the electronic circuit is further realised as described in the characterising portion of claim 1, by means of the method as described in the characterising portion of claim 13.
In this way, the filter transfer function is almost completely realised by means of an active circuit between the output terminals, whereas the series path between the input and output terminals only contains at least one passive component. This passive component can be a simple resistor or an inductor, or a combination of both or any passive circuit, but these can be selected such as to have only a very small DC series resistance. The active circuit is thus realised as a parallel circuit between the output terminals, and is selected such that the total transfer function of the combination of the series impedance of this at least one passive series component and the parallel active circuit matches the desired transfer function. With active circuit is meant any circuit including at least one active element such as a transistor. Generally speaking, in order to realise any filter function F(s), the active parallel circuit G thereby includes another filter having a filter transfer function H(s). The active circuit G, combined with the previously described series impedance, forms the total F(s).
In a preferred embodiment both series branches between corresponding input and output terminals include such a series impedance as is described in claim 2, whereby both series impedances can be further made substantially identical, as is set out in claim 3.
In order to realise a desired high pass filter function, the active parallel circuit thereby has to include a low pass filter, as described in claim 4, in order to realise a desired bandpass filter function, the parallel circuit also has to include a band-reject filter, as set out in claim 5, in order to realise a desired low pass filter function, the active parallel circuit has to include a high pass filter as set out in claim 6, and in order to realise a desired band-reject filter function, the active parallel circuit has to include a band pass filter as set out in claim 7. Throughout this document we refer to the standard tutorial handbook “Handbook of Filter Synthesis” by A. I. Zverev, J.Wiley and Sons, ISBN 0471 98680 1 with respect to these definitions for particular types of filters.
In a particular embodiment the desired low pass filter function is the low pass filter part of a splitter circuit in telephone transmission lines, as set out in claim 8. A particular characteristic of such a low pass filter circuit in a splitter circuit is that it is bidirectional, meaning that in this case a filter transfer is to be observed from input to output terminals and vice versa, meaning that in such a bidirectional circuit input and output terminals are interchangeable to some respect. Remark that the transfer function in one direction need not be the same as the transfer function in the other direction.
For realising bi-directionality, at least two passive series elements are to be present, surrounding the active circuit as set out in claim 9. Claim 10 states that, for a balanced solution at least four passive series elements are foreseen, which are preferrably equal.
Claim 11 further states that the active circuit can be coupled to the the secondary windings of a transformer, of which the primary windings are coupled between the output terminals. This has as further advantage that the active circuit can be powered at low voltages, by virtue of the transformer ratio, such that any active circuit can then further be more easily realised, for example by means of the earlier described active switch capacitor filter techniques, or even using digital filter techniques or synthesising the active circuit with discrete elements.
Claim 12 states that the active circuit is coupled to at least one of the output terminals via a capacitor in between this output terminal and an input terminal of this active circuit. This configuration has the advantage that DC voltage or low frequency signals are filtered out by the capacitor before reaching the active circuit, and is thus particularly useful for realising an overall low-pass transfer function where the active circuit is thus to include a high pass filter.
It is to be noticed that the term ‘coupled’, used in the claims, should not be interpreted as being limitative to direct connections only. Thus, the scope of the expression ‘a device A coupled to a device B’ should not be limited to devices or systems wherein an output of device A is directly connected to an input of device B. It means that there exists a path between an output of A and an input of B which may be a path including other devices or means.
It is to be noticed that the term ‘comprising’, used in the claims, should not be interpreted as being limitative to the means listed thereafter. Thus, the scope of the expression ‘a device comprising means A and B’ should not be limited to devices consisting only of components A and B. It means that with respect to the present invention, the only relevant components of the device are A and B.
The above and other objects and features of the invention will become more apparent and the invention itself will be best understood by referring to the following description of an embodiment taken in conjunction with the accompanying drawings wherein
a-3c schematically depicts the method for realising particular embodiments of the electronic circuit architecture according to the invention
The present invention concerns a method for realising or designing an electronic circuit F having a particular desired filter characteristic F(s), as well as the electronic circuit architecture itself. The basic principle schematic is shown in
Vout1(s)−Vout2(s)=F(s) [Vin1(s)−Vin2(s)] (1)
With Vout1 being the voltage at output terminal OUT1
With Vout2 being the voltage at output terminal OUT2
With Vin1 being the voltage at output terminal IN1
With Vin2 being the voltage at output terminal IN2 and
With s being the Laplace variable
Various methods have been described in the past for realising such a filter, either based on lumped discrete elements, either based on active or even digital filters. The method of the present invention is based on the principle as depicted in
In
It can be mathematically shown, and examples will be given in a further paragraph, that, for a desired total low pass system filter characteristic F(s), the parallel active circuit G has to include a high pass filter. Similarly, in order to realise a desired bandpass filter function F(s), the parallel circuit has to include a band-reject filter function H(s); in order to realise a desired low pass filter function, the active parallel circuit has to include a high pass filter, and in order to realise a desired band-reject filter function, the active parallel circuit has to include a band pass filter function. In this respect it is further to be remarked that the filter characteristics H(s) of the filters H included in the active circuits G are not the pure inverse characteristic of the desired overall filter transfer function F(s), as will become clear from the examples in a further paragraph.
The passive series impedance elements Z1, Z10 can comprise one element such as a resistor, inductor or capacitor, or comprise a series connection of several components. The total series impedance of the series connection is represented by Z1, respectively Z10.
In a preferred embodiment the series impedance in the path between OUT1 and IN1 has to match the series impedance in the path between OUT2 and IN2. A balanced solution then requires Z1 to equal Z10. In case Z1 differs from Z10 the solution is unbalanced.
This method for designing a filter is particularly interesting for these applications wherein the filter has to withstand high currents and voltages. This is for instance the case in the low pass filter part of splitter circuits in telephone circuits. These circuits were extensively discussed in the already mentioned published PCT patent application WO 03/061121 and their function will therefore not be repeated in this document. A further requirement of these low-pass filters in splitters is that the low-pass filter characteristic has to be bi-directional, meaning that both in the direction from input to output, as well as in the direction from output to input, a filter characteristic is observed. Both observed filter characteristics may be identical, but this is not mandatory.
This bi-directionality is denoted on
Although the examples depicted in
For a splitter circuit, the series passive elements may comprise simple resistors, lumped inductors or capacitors a combination of all of them
Examples of embodiments for further realising a low-pass function into detail are shown in
The overall transfer function for F is denoted as
F(s)=Vout/Vin=1/(1+gm.Z.H(s)) (2)
For a desired F(s), such as a desired low-pass function, one can calculate the corresponding H(s):
H(s)=(1−F(s))/gm.Z.F(s) (3)
once H(s) is mathematically known, it has to be synthesised, using known techniques.
In general an iterative approach is used, by first selecting a particular and simply realisable H(s), taking into account the already mentioned principles, and then further calculating the resulting F(s):
This is shown in example 1 in
According to the definition of a low-pass filter function
H1(s) approximates1 for frequencies larger than the cut-off frequency ωo
For these high frequencies, equation (2) will result in
F1(s)=Vout/Vin=1/(1+gm.Z) (4)
thus for high frequencies larger than ωo, F1 acts as an impedance divider, with a high attenuation provided gm.Z is large.
According to the definition of a low-pass filter H1(s)=0 for frequencies lower than the cut-off frequency ωo
For these low frequencies eauation (2) results in
F1(s)=Vout/Vin=1 (5)
Indicative of the pass band of F1.
At intermediate frequencies: F1(s) will show a transistion region from pass band to stop band transition, with a slope related to the order of the high pass filter H1.
For the particular embodiment depicted in
F1(s)=Vout/Vin=(1+s/ω0)/(1+s(1+gm.R1)/ω0) (6)
For low frequencies (s=0): Vout/Vin=1, showing the pass band
For high frequencies (s=∞): Vout/Vin=1/1+gm.R1, showing the resistive voltage divider with 1/gm and R, thus corresponding to the stop band attenuation
F1(s) has thus a pole at ωp=ω0/(1+gm.R1) and a zero at ωz=ω0 and a transition region corresponding to a first order slope between ωp and ωz as can be observed from
In example 2, depicted in
H2(s)=s2/(a+b.s+s2) (7)
having the characteristics of a 2nd order high pass filter
The resulting F2(s) has then, according to equation (2) the following tranfer function:
F2(s)=Vout/Vin=(a+b.s+s2)/((a+b.s+(1+gm.R)s2) (8)
for low frequencies (s=0) F2(s)=Vout/Vin=1, showing the pass band of F2. for high frequencies (s=∞) F2(s)=: Vout/Vin=1/(1+gm.R) being indicative of a resistive voltage divider with 1/gm and R, thus corresponding to the stop band attenuation of F2.
Poles of F2(s) at located at a location with a factor (1+gm.R) lower than the zeros
Zeroes of F2(s) are located at the same location as the poles of the high pass filter H2(s).
A second order slope between poles and zeros is then present.
In
In other embodiments of the invention, such as circuit F20 shown in
In yet other embodiments, such as F30 shown in
Nothwithstanding the shown examples, the invention is not linked to any particular filter implementation, but is generally useably to any kind of filter. Thus any combination of low,high, bandpass filter is as well realisable using the the previously described principles.
While the principles of the invention have been described above in connection with specific apparatus, it is to be clearly understood that this description is made only by way of example and not as a limitation on the scope of the invention, as defined in the appended claims.
Number | Date | Country | Kind |
---|---|---|---|
04291764.1 | Jul 2004 | EP | regional |