latest news

28.06.2011

Last weekly reports added

read more ...

30.04.2011

New weekly reports added

read more ...

20.03.2011

Weekly reports added

read more ...

20.03.2011

Website online.

introduction

Before the invention of the transistor in 1947, valves, also known as vacuum tubes, were used in almost every purpose that involved active electronics, being amplification one their most important applications. Although solid-state devices clearly brought many advantages, after more than 60 years from their genesis, there has been a growing interest in valve amplifiers at the high-end audio sector, all over the world. Many audiophiles claim that valve amplifiers have a different chime that is somehow better than the semiconductor counterparts. In the musical instrument amplifying sector, specially for the electric guitar, valve amplifiers were always (and still is) at the top of guitarist preferences, being a fact that three quarters of the actual production of valves are sold to this market. The reason for this preference is not however clear, being a subject that must be addressed by psychoacoustics since there is no evidence for such from the amplifier metrics.

Innovation in this area has been insignificant or almost null for at least the last three or four decades. Curiously, in last few years some new ideas appeared with a goal to improve amplifier characteristics by complementing this vintage technology with digital devices. The concept focus on the use of microprocessors to monitor and control the dynamic conditions of a valve amplifier. It tries to bring out the best of both worlds, without introducing, however, solid-state devices in the audio signal path. This type of control may be particularly advantageous because electron valves age faster than transistors. This phenomenon makes their characteristics to vary with the time of use. The digital machine can then compensate the changes through permanent monitoring and dynamic actuation over operating conditions.

This dissertation proposes the study of different known amplifier topologies. It spans from the preamplifier to the power amplifier stages and the design of a digitally controlled High-Fidelity valve amplifier. Existent CAD simulation models for valves will be analyzed and used during the process of design, avoiding long experimental cycles. These models will enable the use of electric simulator tools for valve circuit simulation, an important intermediate step for optimal design. A study on alternative dynamic control methodologies for valve amplifiers, like the introduction of digital distortion control, is also one of the proposals of this work.

goals

  • Study of the different electrical topologies for the pre and power amplifier sections of an audio frequency valve amplifier;
  • Study of the existent dynamic monitoring and control processes that involve electron valves;
  • Study of valve SPICE models;
  • Design, simulation and implementation of an integrated Hi-Fi valve amplifier (pre, power amplifier sections and power supply);
  • Development of a digital module for monitoring and control of the produced valve amplifier.

metodology

This work will involve the use of components like electron valves, transformers, a microcontroller and other off-the-shelf circuitry. At the begin, options will be made, trying to do tradeoffs between different components, in order to establish the best conditions for further work. This part of the work will be made with the aid of SPICE models and an electrical circuit simulator like PSpice.

After the amplifier design, printed circuit boards for the developed circuit will be developed. The digital module will be placed in breadboards.

Having assembled the physical circuits, the development of the digital part of the amplifier will take place. The tools involved in the work will depend on the available tool-chain for the chosen microcontroller.

Tests will be made during the algorithm development phase and also after it in order to guarantee all the specifications are met. Laboratory measuring devices will be necessary. Examples: oscilloscope, multimeter, network analyzer for audio frequencies.

This plan considers it is important to write the dissertation along with the development of work, starting it right after the begin of the algorithm phase. For writing, a typesetting system like LaTeX will be used.