Preface:
"From Transmitter to Antenna" started out as a set of documents intended to explain how the process of antenna impedance matching can be treated as a set of operations in the impedance plane or ' Z-plane'. Such a view leads to the idea that the best way to track and monitor the impedance matching process, particularly when the antenna is to be used over a wide range of frequencies (i.e., the 1.6 - 30MHz HF spectrum), is to use a set of bridges which can give null indications of the load impedance in relation to unity power-factor and target values of resistance and conductance. This three bridge R, G, φ system, first proposed by Underhill and Lewis in 1973 [1], has the virtue that it gives an unambiguous and generally optimal solution to the problem of how to adjust a given matching network, and provides directional information in relation to the adjustment of the network reactances. The system is therefore vastly superior to the conventional SWR monitoring approach used in manual impedance matching; and, being free from ambiguity, is the basis for the design of maximally dependable automatic matching systems.
An attempt to popularise the R, G, φ approach was made by Nigel Williams, G3GFC, in about 1980. Nigel designed and produced prototypes for a multi-function bridge which he called the "M50 Match Meter", but the associated commercial project never came to fruition. Many years after the event, he discussed his ideas with me (DWK), and subsequently passed over his notes and documents so that I could work them into a publishable form. The subsequent review and updating ot the material however proved not to be straightforward. The bridges were inaccurate due to lack of compensation for current-transformer non-idealities and needed to be improved; and it seemed that the audience for a set of articles about operations in the Z-plane would be limited to those with a good grasp of traditional circuit-analysis techniques. Hence I set out to do two things: The first, which turned out to be a significant undertaking in its own right, was to develop the background material needed for a good understanding of the subject. The second was to review the subject of bridge design, with a view to discovering best practice so that the Match Meter could be optimised.
To say that reviewing the subject of current-transformer bridge design proved not to be a simple matter would be something of an understatement. Here we enter a world fraught by lack of experimental data, naive theory, unchecked calculations and defective simulation models. It turned out that Nigel's bridges were fairly good after all, at least in an overall scheme where published (and some commercial) designs can turn out to have amplitude errors of 70% and phase errors of -720°. It is no wonder that people complain that they can't get their antennas to work over a wide range of frequencies. With monitoring equipment like that who could?
So I ended up doing a great deal of work on the theory and practice of bridge design. There is much which remains to be reported, but the gist is this: Bridges can be evaluated by choosing two independent (or nearly independent) adjustable parameters of the bridge itself and treating their deviations from the design values vs frequency as perturbations for the purpose of finding the phase and amplitude tracking performance.
The current transformer itself can be treated as a transmission-line device, and compensations for its imperfections deduced accordingly. When the problem is correctly understood, and the bridge is properly adjusted, it is easy to achieve amplitude tracking better than ±0.3% and phase accuracy better than ±0.2° over the 1.6 to 30MHz range (Appendix 6.4). Such accuracy is not needed for the purpose of setting-up radio stations, but since obtaining it requires no exotic test equipment or laborious procedure, there seems little point in not having it.
So here is my attempt at teaching AC theory, electrical materials science, inductance and capacitance; and at using those ideas to make radio stations work.
DWK 30th June 2007.
© D W Knight 2003 - 2010. © N Williams. 1983 and 2005.
David Knight asserts the right to be recognised as the principal author of this work and is the person to whom all enquiries should be directed.
Acknowledgements:
I would like to thank Andy Cowley M1EBV for reading and commenting on the developing articles and for considerable assistance in library work and the loan of test equipment and reference materials.
Refs:
[1] "Automatic Tuning of Antennae". M J Underhill [G3LHZ] and P A Lewis.
SERT Journal, Vol 8, Sept 1974, p183-184. Reprint of paper in Mullard Research Labs Annual Review, 1973.
"From Transmitter to Antenna" started out as a set of documents intended to explain how the process of antenna impedance matching can be treated as a set of operations in the impedance plane or ' Z-plane'. Such a view leads to the idea that the best way to track and monitor the impedance matching process, particularly when the antenna is to be used over a wide range of frequencies (i.e., the 1.6 - 30MHz HF spectrum), is to use a set of bridges which can give null indications of the load impedance in relation to unity power-factor and target values of resistance and conductance. This three bridge R, G, φ system, first proposed by Underhill and Lewis in 1973 [1], has the virtue that it gives an unambiguous and generally optimal solution to the problem of how to adjust a given matching network, and provides directional information in relation to the adjustment of the network reactances. The system is therefore vastly superior to the conventional SWR monitoring approach used in manual impedance matching; and, being free from ambiguity, is the basis for the design of maximally dependable automatic matching systems.
An attempt to popularise the R, G, φ approach was made by Nigel Williams, G3GFC, in about 1980. Nigel designed and produced prototypes for a multi-function bridge which he called the "M50 Match Meter", but the associated commercial project never came to fruition. Many years after the event, he discussed his ideas with me (DWK), and subsequently passed over his notes and documents so that I could work them into a publishable form. The subsequent review and updating ot the material however proved not to be straightforward. The bridges were inaccurate due to lack of compensation for current-transformer non-idealities and needed to be improved; and it seemed that the audience for a set of articles about operations in the Z-plane would be limited to those with a good grasp of traditional circuit-analysis techniques. Hence I set out to do two things: The first, which turned out to be a significant undertaking in its own right, was to develop the background material needed for a good understanding of the subject. The second was to review the subject of bridge design, with a view to discovering best practice so that the Match Meter could be optimised.
To say that reviewing the subject of current-transformer bridge design proved not to be a simple matter would be something of an understatement. Here we enter a world fraught by lack of experimental data, naive theory, unchecked calculations and defective simulation models. It turned out that Nigel's bridges were fairly good after all, at least in an overall scheme where published (and some commercial) designs can turn out to have amplitude errors of 70% and phase errors of -720°. It is no wonder that people complain that they can't get their antennas to work over a wide range of frequencies. With monitoring equipment like that who could?
So I ended up doing a great deal of work on the theory and practice of bridge design. There is much which remains to be reported, but the gist is this: Bridges can be evaluated by choosing two independent (or nearly independent) adjustable parameters of the bridge itself and treating their deviations from the design values vs frequency as perturbations for the purpose of finding the phase and amplitude tracking performance.
The current transformer itself can be treated as a transmission-line device, and compensations for its imperfections deduced accordingly. When the problem is correctly understood, and the bridge is properly adjusted, it is easy to achieve amplitude tracking better than ±0.3% and phase accuracy better than ±0.2° over the 1.6 to 30MHz range (Appendix 6.4). Such accuracy is not needed for the purpose of setting-up radio stations, but since obtaining it requires no exotic test equipment or laborious procedure, there seems little point in not having it.
So here is my attempt at teaching AC theory, electrical materials science, inductance and capacitance; and at using those ideas to make radio stations work.
DWK 30th June 2007.
© D W Knight 2003 - 2010. © N Williams. 1983 and 2005.
David Knight asserts the right to be recognised as the principal author of this work and is the person to whom all enquiries should be directed.
Acknowledgements:
I would like to thank Andy Cowley M1EBV for reading and commenting on the developing articles and for considerable assistance in library work and the loan of test equipment and reference materials.
Refs:
[1] "Automatic Tuning of Antennae". M J Underhill [G3LHZ] and P A Lewis.
SERT Journal, Vol 8, Sept 1974, p183-184. Reprint of paper in Mullard Research Labs Annual Review, 1973.
No comments:
Post a Comment