Skip to main content
Top

2011 | Book

The Sound of Silence

Lowest-Noise RIAA Phono-Amps: Designer’s Guide

Author: Dipl.-Ing. Burkhard Vogel

Publisher: Springer Berlin Heidelberg

insite
SEARCH

About this book

There is a wide field of tasks left that can only be satisfyingly attacked with the help of old-fashioned analogue technology, and one of the most important are amplifiers for analogue signals. The strongly expanded content of the second edition of "the sound of silence" leads to affordable amplifier design approaches which will end up in lowest-noise solutions not far away from the edge of physical boundaries set by room temperature and given cartridges - thus, fully compatible with very expensive so called "high-end" or "state-of-the-art" offers on today markets - and, from a noise point of view in most cases outperforming them!

With easy to follow mathematical treatment it is demonstrated as well that theory is not far away from reality. Measured SNs will be found within 1dB off the calculated ones and deviations from the exact amplifier transfer won't cross the ± 0.1dB tolerance lines. Additionally, the book presents measurement set-ups and results. Consequently, comparisons with measurement results of test magazine will soon become easier to perform.

This new edition includes a new chapters about reference levels, Noise in Amp Input sections, Humming Problems, and much more.

Table of Contents

Frontmatter

Essentials

Frontmatter
Chapter 1. Amps, Pre-amps, Pre-pre-amps
Abstract
The purpose of this book is an economic one: to enable the reader to calculate certain noise related aspects of a phono-amp before building it up, thus, saving a lot of energy and time as well as avoiding needless expenses. That’s why I won’t debate whether the black box between the output of a turn-table and the input of an amplifier (amp) with the volume control is a pre-amp or a pre-pre-amp or what ever it will be. This book follows the purpose and that will be the only story to tell.
Burkhard Vogel
Chapter 2. RIAA Transfer/Anti-RIAA Transfer
Abstract
The signal on VRs is coded according to the rules set by the Record Industry Association of America (RIAA). This code is determined by three time constants: T1, T2 and T3. Cutting a VR means that the three time constants encode the signal in a specific way. The reason for this is to handle overloading and noise issues the optimal way.
Burkhard Vogel
Chapter 3. Vinyl Record Reference Levels
Abstract
Based on the fact that in many cases the manufacturer’s cartridge data specs do not provide satisfactory data the aim of this chapter is to make the handling and understanding of the many different vinyl record reference levels and modulation possibilities easier. In four sections and with the help of typical cartridge representatives I will go through this rather complex field.
Burkhard Vogel

Noise Basics

Frontmatter
Chapter 4. Noise in Components and Other General Noise Effects
Abstract
In fact, the reason why I wrote this book is noise, noise from electronic devices as well as noise as the signal-disturbing output of loudspeakers. The music I like should come out of the NOTHING. Clear, undistorted and close to the original.
Distortion matters of amplifiers can be wonderfully tackled and accompanying problems as well be wonderfully solved with the help of Douglas Self’s Audio book. Because of the high-gain and low-noise op-amps, on today’s markets we find that it is rather easy to design low-distortion amps of any kind. That is why this book does not treat this matter in depth.
Burkhard Vogel
Chapter 5. Noise in Bipolar Junction Transistors
Abstract
The circuit of Fig. 4.1 becomes a bit more complex when dealing with a real life amplifier and with a transistor as the first amplifying stage. In most cases, the noise production of FETs or valves becomes higher than that of Bipolar Junction Transistors (BJTs). Therefore, we have to dive rather deep into the noise making elements of BJTs. I do not discuss all the theories of modelling transistors. This can be done by studying some other publications After many trials, I found out that the adapted model for audio frequencies of Fig. 5.2 is very useful for practical noise calculations. For the audio band, it is derived from the so-called π-model that is given in Fig. 5.1.
Burkhard Vogel
Chapter 6. Noise in Field Effect Transistors
Abstract
When looking at Field Effect Transistors (FETs) I only mean JFETs (Junction Field Effect Transistors) because, when talking about low-noise, they are the only workable ones in the audio field. Whereas they are very good for MM cartridge purposes they are – in singleton configuration – a very bad choice for MC amplification. Like the BJTs, with their three leads to the outer world (Gate, Source, Drain), they can be placed in a circuitry in so-called common source CS, common drain CD and common gate CG configurations.
Burkhard Vogel
Chapter 7. Noise in Valves (US: Tubes)
Abstract
Concerning electronic noise, we cannot compare the world of valves with the one of the solid-state world. Whereas the latter offers a rather narrow bandwidth of noise voltage tolerance, respectively, valves come along with a very broad tolerance noise voltage bandwidth. Sometimes the noise voltage difference between a selected low-noise type and a non-selected, “normal” one can reach up to 20 dB in the audio band. That is why the calculation of the audio band noise of a valve gain stage requires exact operating data of the chosen valve.
Burkhard Vogel
Chapter 8. Noise in Operational Amplifiers
Abstract
From a noise point of view Operational Amplifiers (Op-Amps) are very much easier to handle than transistors or valves1,2,3,4. In this book, I only refer to low-noise voltage feedback op-amps. Usually, current feedback op-amps produce much more noise.
Burkhard Vogel
Chapter 9. Noise in Instrumentation Amps
Abstract
The difference between the op-amp and the instrumentation amps (in-amps) is the fact that, usually, the input of an op-amp is configured for single ended (non-symmetrical, un-balanced) purposes and the in-amp is configured as an amp with a true differential (symmetrical, balanced) input. However, both types of amps can be configured the opposite way – but coming along with certain disadvantages. In addition, in most IC cases, the in-amp’s gain can be set by only one resistor whereas the op-amp’s gain can be set by a range of different possibilities.
Burkhard Vogel
Chapter 10. Noise in Transformers
Abstract
Transformers (Trafos) are a very good substitute for MC cartridge pre-pre-amps (Fig. 1.1). Without big development efforts, they can be connected to a MM phono-amp’s input (Fig. 10.1). Unfortunately, and despite the many claims one can read from time to time in test magazines, transformers substantially add noise to the cartridge-transformer-phono-amp-chain.
Burkhard Vogel
Chapter 11. Noise of Vinyl Records: On How Much Phono-Amp SN is Needed?
Abstract
It is hard to get precise figures on the headline’s topic. Several sources1 claim that SNs range from −60 to −70 dB(A) for 33 1/3 LP records and −63 to −73 dB(A) for 45 Single or Maxi records, both with reference to a peak velocity of 8 cm/s/1 kHz. This is equal to an rms velocity of 5.66 cm/s/1 kHz. In addition, SNs of records heavily depend on the manufacturing process as well as on the vinyl material itself. Today, only three vinyl production companies exist worldwide in USA, in France and in Japan. The chemical formulae of the materials are the big secrets and they are the basic responsible factors for the noise of the records on the market - besides noisy phono-amps, of course.
Burkhard Vogel
Chapter 12. Noise Basics: Mathcad Worksheets
Burkhard Vogel

Best Practice

Frontmatter
Chapter 13. Noise of MM Cartridges
Abstract
The design of hum- and noiseless RIAA pre-amplifiers is really a kind of art. The existence of a great variety of circuit designs tells many stories about chases after high signal-to-noise ratios (SN), overload matters, and fights for precision of the RIAA transfer. Solutions proposed in the past to use – as substitute – resistors or inductors1 at the input of a RIAA amplifier for noise measurement purposes or to set on mathematical octave-band analysis2 to get SN results close to the measured ones (with MM cartridge as input load) do not satisfy. In addition, hum spoils many measurement attempts and the optimal loading capacitance of the MM cartridge is quite often not taken into account of calculations or measurements.
Burkhard Vogel
Chapter 14. Noise of MM Cartridges: Mathcad Worksheets
Burkhard Vogel
Chapter 15. Noise of Solid-State MC Phono-Amps
Abstract
Despite the fast growing digitisation of the electronic world the vinyl and analog aficionado is still alive. Moreover, it’s a world-wide growing market with increasing demand for excellent phono amplifiers for moving coil (MC) cartridges. Unfortunately, many so-called high-end (and high-price) products show lousy signal-to-noise (SN) figures, far away from what is theoretically achievable. Table 11.6 gives an overview on some of the good, the bad and best products.
Burkhard Vogel
Chapter 16. Noise of Solid-State MC Phono-Amps: Mathcad Worksheets
Burkhard Vogel
Chapter 17. Noise of Triode Driven Phono-Amps
Abstract
In the third Quarter of 2009 I had a long and intensive discussion with a Dutch friend on the noise production of his two-stage phono-amplifier. I didn’t believe in his output referred SN measurement results (FFT diagram shown in Fig. 17.1), because they always came up ~5–6 dB worse than the ones I have calculated. This fact became the starting point of a rather long lasting and extensive study on the noise of triodes and on how to handle them in a practical mathematical way. My goal: the difference between calculated and measured results should come-up within ±2 dB.
Burkhard Vogel
Chapter 18. Noise of Triode Driven Phono-Amps: Mathcad Worksheets
Burkhard Vogel
Chapter 19. RIAA Networks
Abstract
Masterpieces of phono-amps are the result of precision design calculation approaches followed by precision building-up of a prototype of the designed circuitry – before it goes to manufacturing. But this is not the end. It is also necessary to select the right sound-relevant passive components, like e.g. capacitors, resistors, and inductors. Although, this is a listening based subjective selection, it cannot be part of this book. Nevertheless, any RIAA network has a great influence on SNs of phono-amps. The relating mathematical proof is illustrated with (15.8)–(15.17).
Burkhard Vogel
Chapter 20. RIAA Networks: Mathcad Worksheets
Burkhard Vogel

Noise and Hum Measurement System

Frontmatter
Chapter 21. System Overview
Abstract
In Chap. 13 “Noise in MM cartridges” I’ve given a first insight into the measurement system (MS) to measure noise performances of MM cartridges. Figure 13.7 gives the details. Specifically, MM cartridge related parts of the shown measurement arrangement were covered by Chap. 13. The following additional topics will be subject of Chaps. 21–26.
Burkhard Vogel
Chapter 22. Measurement Amps, Noise Generators and FFT Noise Measurements
Abstract
In 1989 M. Wilfried Adam published a great article on the design of low-noise audio amps. The complete circuitry of a noise measurement system (MS) also became part of that article. M. Adam described all necessary measurement filters too. Instead of buying a very expensive all-in-one measurement instrument a few months later I decided to develop my own measurement equipment, mainly based on M Adam’s ideas, but with much less noisy op-amps and some improvements, of course.
Burkhard Vogel
Chapter 23. Measurement Filters and Networks
Abstract
Measurement filters and electronic transfer function reproducing networks play an essential role for any audio measurement set-up – especially when talking about noise measurements and RIAA phono-amps. In addition, the noise measurement of non-equalized amps, such as microphone pre-amps or any other linear amp, bears some challenge as well. In this chapter I’ll focus rather extensively on these issues. I’ll split it in two halves: the first one deals with the measurement approach already shown in the first edition of this book, the second half (Sect. 23.5ff) digs a bit deeper: it is the search for the ideal brick wall filter that allows the “wyciwym” approach in B20k: “what you have calculated is what you measure”. The maximum deviation between measurement and the calculated result should be found within ±0.25 dB, at least for solid-state based phono-amps.
Burkhard Vogel
Chapter 24. The Hum Figure of Linear Amps
Abstract
The frustration that immediately follows hum and noise (H&N) that may come out of the loudspeaker-amplifier chain in the gap between music titles begs the question of “how to measure or calculate the hum artefact content of the output noise of a power (or any other linear) amplifier with input shorted and output loaded with the rated load?”.
Gaps between CD music titles or from other low-noise sources should be hum- and noise-free. There could be one exception: vinyl records plus phono-amp. In most cases, their H&N level “marches” 1:1 through the whole amp chain, thus, producing the Signal-to-Noise ratio at the phono-amp’s output without major increase at the output of the power amplifier (PA) as well.
Burkhard Vogel
Chapter 25. The Hum Figure of Phono-Amps
Abstract
This chapter deals with the measurement of the Hum Figure of hum-infected phono-amplifiers. There are many reasons why it is worth insisting on the identification and elimination of hum in audio equipment. It can be demonstrated simply in Figs. 25.1 and 25.2.
Burkhard Vogel
Chapter 26. Mathcad Worksheets of Chaps. 22–25
Burkhard Vogel

The RIAA Phono-Amp Engine

Frontmatter
Chapter 27. Overview
Abstract
The purpose of the RIAA Phono-Amp Engine is to have every thing on one spot what is needed to demonstrate that measurement results (carried out in a home-office in a typical city housing area and not in a lab environment) are not far away, from what could be figured out by theory-based calculations. At the same time, it helps to check and test a great variety of connectors and connection possibilities. Additionally, also fulfilling the needs of people who want to listen to music with rather low-impedance MC cartridges I have included the design of a special trafo driven MC phono-amp.
Burkhard Vogel
Chapter 28. Modules 1–4
Abstract
The detailed circuit diagrams of the Module 1 phono-amp section are shown in Figs. 15.3a–b and the respective – in-case – power supply unit circuit diagram is shown in Fig. 15.3c. I guess these diagrams are self-explanatory. Figure 27.1 shows the wiring of the input connectors and Fig. 27.2 shows the output stage.
The calibration of the OP4 DC offset setting P3 needs some time because of the rather high time constant of R27, 28 and C25, 26. P6 sets the stage gain at +6dB – or any other gain to get an output voltage as indicated. It makes no sense to further increase the output voltage of Tr1 because of growing distortion. S4 allows a 180° phase change of the balanced output signal. S2 switches the two outputs on/off and R31 + C28 ensure a delayed on-switch of the two outputs when turning the module on. Rel 1–4 are TQ2-12V relays from Matsushita.
Burkhard Vogel
Chapter 29. Engine Performance
Abstract
To demonstrate the low-noise quality of the Module 1 and 2 MC phono-amps their FFT spectral noise voltage density graphs were already given in Figs. 15.8 and 15.9. The following graphs show the engine’s broadband quality with frequency (F) and phase (P) responses (F = red/top trace + P = blue/bottom trace) of several input–output combinations as well as some additional spectral noise voltage density charts for the whole amp chain: from input connector via phono-amp and output stage to source-selector balanced output with +6.02 dB gain.
Burkhard Vogel
Backmatter
Metadata
Title
The Sound of Silence
Author
Dipl.-Ing. Burkhard Vogel
Copyright Year
2011
Publisher
Springer Berlin Heidelberg
Electronic ISBN
978-3-642-19774-1
Print ISBN
978-3-642-19773-4
DOI
https://doi.org/10.1007/978-3-642-19774-1