Amplifiers – With semiconductor amplifying device – Including particular biasing arrangement
Reexamination Certificate
2002-06-25
2004-09-07
Nguyen, Patricia (Department: 2817)
Amplifiers
With semiconductor amplifying device
Including particular biasing arrangement
C330S054000
Reexamination Certificate
active
06788149
ABSTRACT:
CROSS-REFERENCE TO RELATED APPLICATIONS
This application is based on French Patent Application No. 01 08543 filed Jun. 28, 2001, the disclosure of which is hereby incorporated by reference thereto in its entirety, and the priority of which is hereby claimed under 35 U.S.C. §119.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a biasing system for biasing an electronic circuit. The invention relates more particularly to a broad-band distributed active load used as a biasing supply for biasing electronic circuits for microwave broad-band low-noise power applications, for example power amplifier electronic circuits.
2. Description of the Prior Art
Broad-band integrated power amplifiers operating at frequencies from DC to frequencies in excess of 100 GHz are very widely used in high bit rate (greater than 40 Gbit/s) optical fiber transmission systems.
For example, they are used to control electro-optical modulators, which are optical devices in which the propagation of light can be modified by applying a particular electrical field, typically with a peak-to-peak amplitude of 3 volts in the case of an electro-absorption modulator and more than 6 volts in the case of a lithium-niobate modulator.
High-voltage control circuits providing broad-band power amplification are required for controlling optical devices of this type.
These control circuits are generally based on distributed amplifier architectures.
A distributed power amplifier has an inherently high product of the gain and the bandwidth.
FIG. 1
shows the operating principle of this kind of amplifier. Amplifier stages T
1
to Tn are regularly distributed along an input transmission line GL and an output transmission line DL. Each transmission line comprises a respective conductive line GL and DL and components connected along it.
Each amplifier stage comprises a field-effect transistor, for example a metal semiconductor field-effect transistor (MESFET).
The two transmission lines GL and DL are therefore called the gate line and the drain line.
However, bipolar transistors can be substituted for the field-effect transistors in each amplifier stage T
1
to Tn.
In this case, the two transmission lines along which the amplifier stages are distributed are called the base line and the collector line.
The distributed amplifier operates as follows: the input signal to be amplified propagates along the gate line GL and is absorbed by the load Rg, whose impedance is substantially equal to the characteristic impedance of the transmission line. The characteristic impedance of the transmission line includes the impedance of the conductive line and the equivalent impedances of the components connected to it.
If the load on a transmission line has an impedance equal to the characteristic impedance of the line, the line typically behaves like a line of infinite length.
In other words, no energy can be reflected back to the input of the gate line GL.
A voltage wave propagating from left to right therefore passes the gate G of each transistor T
1
to Tn constituting the amplifier stages.
A current is then generated by each transistor and feeds the drain line DL. Half of this current propagates to the load Rd and the other half to the output. The half of the current from each transistor that propagates to the output is superposed with that from each of the other transistors with the same phase provided that the inter-cell propagation times on the drain line DL and the gate line GL are identical. The overall amplification factor is then the sum of the respective amplification factors of each transistor.
The signal propagated back toward the internal load Rd of the drain line DL is totally absorbed provided that the load Rd is tuned to (i.e. has an impedance equal to) the characteristic impedance of the transmission line DL.
Accordingly, in this distributed power amplifier output line, the portion of the injected current flowing to the internal load Rd is dissipated and lost and no energy is reflected toward the output load.
This condition is essential for preventing ripple on the frequency response and echo in the impulse response and also for preserving the performance of the distributed power amplifier, in particular its linearity. Non-linearity in an amplifier is a major cause of distortion.
Distributed amplifiers like those previously described require a high bias current, in excess of 100 mA, to deliver the required output voltage.
They must supply a high current and many transistors must therefore be used, as a result of which the bias currents are high.
In many applications, the problem then arises of biasing the transistors constituting the amplifier stages distributed along two transmission lines without degrading the performance of the distributed amplifier.
Until now, two prior art solutions have been used for biasing a distributed amplifier. They are described in “Fundamentals of distributed amplification”, by Thomas T. Y. Wong, Artech House, Boston & London, 1993.
However, in the context of the present invention, relating more particularly to broad-band power applications, none of the prior art solutions described in the above work is entirely satisfactory.
First of all, a first solution consists of biasing the drain of the transistors forming each amplifier stage of the distributed amplifier with the output of the drain line DL (see FIG.
2
).
This figure does not show the amplifier stages or the gate transmission line.
Accordingly, a bias voltage is applied via a standard “T” bias circuit
1
at the output of the distributed amplifier.
The bias circuit includes a decoupling capacitor
2
and a high-inductance choke coil
3
.
The function of the decoupling capacitor
2
is to isolate the output of the distributed amplifier from the direct current bias voltage.
At the output of the amplifier are various modules, such as electro-optical modulators, for example.
The function of the choke coil
3
is to block the high-frequency component of the current whilst allowing the low-frequency component to pass through it, so as not to interfere with the operation of the broad-band power amplifier at microwave frequencies.
However, although this conventional circuit for biasing a distributed amplifier supplies a satisfactory bias current to the transistors forming the amplifier stages along the transmission line, it has a number of drawbacks.
First of all, the circuit is not integrated. It is therefore external to the distributed amplifier integrated circuit. Apart from the space it requires, due essentially to the large size of the choke coil, this type of circuit necessitates the use of a separate module.
Moreover, in some applications, for example applications in which the distributed amplifier output controls an electro-optical modulator, a second “T” bias circuit is needed to bias the electro-optical modulator.
This solution is therefore nowhere near optimum in terms of integration capacity.
Another standard prior art solution consists of biasing the transistors directly via the drain resistance Rd of the drain line DL (see FIG.
3
).
As in
FIG. 2
, the amplifier stages of the distributed amplifier and the gate line are not shown.
This solution is fully integrated and avoids the need to use a “T” bias circuit at the output of the distributed amplifier.
However, as already explained, these distributed power amplifiers necessitate very high bias currents, in excess of 100 mA.
These very high bias currents necessarily lead to constraints regarding heating, making it necessary to use a very large drain resistance Rd, which has a very high stray capacitance. This high stray capacitance causes problems, especially at microwave frequencies, at which the resistance Rd no longer behaves as a pure resistance and the effect of the stray capacitance is no longer negligible.
Accordingly, over time, because of the heating of the resistance caused by the high bias current, the resistance no longer operates in its linear area. This behavior is also encountered in HF operation.
The performance of the resistance Rd
Lefevre Rene
Mouzannar Wissam
Avanex Corporation
Moser, Patterson & Sheridan L.L.P.
Nguyen Patricia
LandOfFree
Broad-band biasing system for biasing an electronic circuit... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Broad-band biasing system for biasing an electronic circuit..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Broad-band biasing system for biasing an electronic circuit... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3232216