Audio signal reproduction

By designing an adaptive harmonic spectrum amplifier and adjusting the harmonic distortion spectrum of the audio signal, the problem of unreliable THD specifications in the prior art is solved, and a more reliable indication of the perceived quality of the audio signal and a more balanced harmonic distortion spectrum are achieved.

CN119732077APending Publication Date: 2025-03-28TENO CO
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202280099307.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2022-07-04
Publication Date
2025-03-28

AI Technical Summary

Technical Problem

In the prior art, THD specifications are unreliable as a measure of subjective sound quality, and the structural relationship of the harmonic distortion spectrum is not fully considered, affecting the perceived quality of the audio signal.

Method used

By designing an adaptive harmonic spectrum amplifier called Harmonic Balanced Amplifier (HBA), the circuit can adjust the harmonic distortion spectrum of the basic order, and various harmonic distortion spectrums with a composite effect through the adjustment of the input impedance load.

Benefits of technology

A more meaningful and reliable perceived quality indication of the audio signal is achieved, which can make subjective changes according to the physiological functions of the target audience, and provide a more balanced harmonic distortion spectrum, thereby improving the perceived quality of the audio signal.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119732077A_ABST
    Figure CN119732077A_ABST
Patent Text Reader

Abstract

An amplifier stage uses a load transistor amplifier circuit, the load of which results in generation in the load transistor amplifier circuit of a second order harmonic distortion energy greater than a third order harmonic distortion energy for amplifying a source audio signal to produce an audio output signal. The fundamental order harmonic distortion spectrum is adjusted to improve perceived sound quality or listening enjoyment.
Need to check novelty before this filing date? Find Prior Art

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 217,846, filed on July 2, 2021, the contents of which are incorporated herein by reference. Technical Field

[0002] This patent application relates to the field of audio signal reproduction. Background Art

[0003] The Total Harmonic Distortion (THD) specification of an audio amplifier is often used as a definitive quality measure when comparing different amplifiers. THD defines the combination of all orders of harmonic distortion. When an audio amplifier circuit amplifies a single frequency tone (say at a frequency of 1kHz), the output signal will typically contain the amplified 1kHz signal as well as distortion energy of different orders, i.e., 2kHz, 3kHz, 4kHz, 5kHz, etc. While harmonic distortion energy typically decreases as the order increases, many amplifiers reduce THD by using amplifier circuit designs that reduce second-order distortion, which contains the largest amount of energy. These types of amplifiers have relatively large odd-order harmonic distortions compared to even-order ones.

[0004] Amplifiers with lower THD specifications are generally considered optimal. Various techniques are employed to further reduce the basic THD measurement of an amplifier, thereby providing an even lower THD specification for publication. Depending on the magnitude of the technique used to reduce the THD of an amplifier, it is questionable whether there is a corresponding improvement in subjective quality. In fact, many times subjective quality is actually reduced due to these techniques. Therefore, the THD specification is unreliable as a measure of subjective sound quality.

[0005] Vacuum tube amplifiers with a dominant even-order THD spectrum generally sound warm and smooth. In contrast, solid-state amplifiers with a dominant odd-order THD spectrum sound less warm and less smooth. Hybrid amplifiers consisting of both vacuum tube and solid-state technology have a mix of tube and solid-state orders of harmonic distortion, and thus a combined subjective sound quality of vacuum tube and solid-state. Whether vacuum tube, solid-state, or hybrid sounds best is highly subjective, but the harmonic distortion spectrum of each technology is very unique and identifiable.

[0006] The harmonic distortion spectra associated with each stage of production, post-production, and reproduction of an audio media file combine to form a complex final output harmonic distortion spectrum. The production of a master media file includes the harmonic distortion associated with the recording equipment. Post-production includes production harmonic distortion as well as harmonic distortion associated with the mixing and manufacturing of the final analog or digital media file. Reproduction of the media file includes all of the foregoing harmonic distortion, as well as harmonic distortion associated with the equipment used for playback, including the transducers used to create the final acoustic interface. Transmission technology, whether wired or wireless, at any stage also contributes unique harmonic distortion to the overall composite harmonic distortion spectrum.

[0007] The THD formula takes into account the magnitude of the overall composite harmonic distortion spectrum, but does not take into account the nuances of the structural relationship between one order of harmonic distortion and another. In order to properly understand the perceived quality of an audio signal, it is necessary to know the magnitude of each order of harmonic distortion in the harmonic distortion spectrum that makes up THD. Once this structural relationship is understood and presented alongside the THD measurement, a more meaningful and reliable indication of perceived quality can be made. Summary of the invention

[0008] The sound quality of a reproduced audio signal is a combined function of the harmonic distortion of the audio signal, the transmission technology employed, and the circuit topology of the amplifier, including the cumulative effect of all amplifiers in the reproduction chain. The distortion created and recorded in the original master file is combined with the harmonic distortion of the transmission technology and the amplification circuit to produce a composite harmonic distortion spectrum. These composite harmonic distortions contribute to the overall distortion of the system and determine the overall sound characteristics and the resulting perceived sound quality. These distortions can be measured as individual orders of harmonic distortion, which are multiples of the reproduced audio signal. Typically, the second, third, fourth, and fifth order harmonic distortion of the audio signal contribute most to the sound characteristics of the amplifier, but there are also cases where higher orders of distortion above the fifth order can contribute most to the sound characteristics of the amplifier.

[0009] Different technologies exhibit different spectra. For example, vacuum tube amplifiers dominate in even-order harmonics, but also taper off odd-order harmonics. Solid-state amplifiers dominate in odd-order harmonics, but also taper off even-order harmonics. Amplifiers with dominant even-order harmonics generally sound subjectively smoother, and therefore more natural. Solid-state amplifiers with dominant odd-order harmonics are less smooth and less natural. The composition of the harmonic distortion spectrum requires both even and odd harmonic orders to be present and appear in a structured order in order to achieve a sound that is considered balanced.

[0010] A balanced harmonic distortion spectrum follows certain principles, but is also somewhat subjective and personal, as it depends on the physiology of one's auditory system and normal age-related changes that occur over time. Having the ability to shape the harmonic distortion spectrum allows for subjective changes based on the physiology of the target audience, and also allows amplifier designs to have perceived sonic signatures that are not strictly limited by today's amplifier technology. For example, the distortion harmonics of a vacuum tube amplifier can be made to sound like a solid-state amplifier, and vice versa. The benefits of this are very compelling, as the reliability and energy efficiency of a solid-state amplifier can now perform like a vacuum tube amplifier.

[0011] An embodiment of the applicant, referred to as a Harmonic Balancing Amplifier (HBA), is an adaptive harmonic spectrum amplifier for adjusting the harmonic distortion spectrum of a fundamental order. An input impedance load within the circuit produces an output impedance load on the audio signal that provides various harmonic distortion spectra that have a compound effect on the overall spectrum. A single HBA circuit can produce an adjustable harmonic spectrum, and multiple HBA circuits can provide additional complex adjustability.

[0012] The HBA can be cascaded or paralleled using multiple circuits to further tailor the best fit harmonic distortion spectrum. The output gain of a single or multiple HBAs can be configured as an audio source, preamplifier or power amplifier, or in unity gain as part of an amplifier or standalone amplifier to provide pre- or post-processing for system and audio signals.

[0013] Further embodiments provide overall harmonic reduction of the audio signal, which reduces all orders of harmonic distortion to allow for greater flexibility in adjustability using single or multiple HBA circuits.

[0014] In some embodiments, an audio signal nonlinear power amplifier may include: a high impedance input; a low impedance output; an input stage connected to the input, the input stage including a transistor amplifier circuit, the transistor amplifier circuit including a load, which, in use, causes the input stage to exhibit a second order harmonic distortion energy greater than a third order harmonic distortion energy; and an output stage connected to the input stage and the low impedance output.

[0015] The load of the input stage may include a transistor string to reduce the magnitude of harmonic distortion by bending the transfer function of the transistor amplifier circuit. The transistor string may include a selector switch for selecting a number of transistors active in the transistor string. The load of the input stage may include a variable resistor capable of varying the load to vary the balance of second order harmonic distortion energy relative to third order harmonic distortion energy. The total harmonic distortion generated by the amplifier may be less than 0.5% of the signal amplitude at the low impedance output, or even less than 0.3%.

[0016] In some embodiments, an audio signal nonlinear amplifier may include: a high impedance input; a low impedance output; an input stage connected to the input, the input stage including a transistor amplifier circuit including an adjustable load, the adjustable load causing the input stage to exhibit different relative magnitudes of second order harmonic distortion energy and third order harmonic distortion energy according to the adjustable load in use; and an output stage connected to the input stage and the low impedance output. The second order harmonic distortion energy may be greater than the third order harmonic distortion energy. The adjustable load of the input stage may include a transistor string to reduce the magnitude of the harmonic distortion by bending the transfer function of the transistor amplifier circuit. The transistor string may include a selector switch for selecting a number of transistors activated in the transistor string. The adjustable load of the input stage may include a variable resistor capable of changing the load to change the balance of the second order harmonic distortion energy relative to the third order harmonic distortion energy. The total harmonic distortion generated by the amplifier may be less than 0.5% of the signal amplitude at the low impedance output, or even less than 0.3%.

[0017] In some embodiments, an audio signal power amplifier system may include: a nonlinear preamplifier having a gain close to unity or a low gain, the nonlinear preamplifier including: a high impedance input; a low impedance output; an input stage connected to the input, the input stage including a transistor amplifier circuit including an adjustable load, the adjustable load causing the input stage to exhibit different relative amounts of second order harmonic distortion energy and third order harmonic distortion energy according to the adjustable load in use; and an output stage connected to the input stage and the low impedance output. The amplifier system may also include a nonlinear power amplifier receiving the output stage, the nonlinear power amplifier including a transistor circuit that generates negligible second and fourth order harmonic distortion and non-negligible third and fifth order harmonic distortion. In this way, the system can provide a second order harmonic distortion energy higher than the third order harmonic distortion energy at the output of the nonlinear power amplifier in response to an input audio signal provided to the high impedance input, wherein the energy at progressively higher order harmonics is reduced.

[0018] The load of the input stage may include a transistor string to reduce the magnitude of harmonic distortion by bending the transfer function of the transistor amplifier circuit. The transistor string may include a selector switch for selecting a number of transistors to be activated in the transistor string. The load of the input stage may include a variable resistor that can change the load to change the balance of second order harmonic distortion energy relative to third order harmonic distortion energy. The total harmonic distortion generated by the amplifier may be less than 0.5% of the signal amplitude at the low impedance output, or even less than 0.3%.

[0019] In some embodiments, an audio signal playback device may include: a processor configured to receive an input audio track and generate a second-order harmonic copy thereof; a mixer configured to mix the input audio track with the second-order harmonic copy to generate a simulation of the second-order harmonic distortion energy within the input audio track in the mixed audio track; a nonlinear power amplifier receiving the mixed audio track, including a transistor circuit that generates negligible second and fourth-order harmonic distortion energy and non-negligible third and fifth-order harmonic distortion energy. In this way, the audio signal playback device provides a higher second-order harmonic distortion energy than the third-order harmonic distortion energy, wherein the energy at progressively higher order harmonics decreases. The total harmonic distortion generated by the device may be less than 0.5% of the signal amplitude at the low impedance output, or even less than 0.3%. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Certain embodiments may be better understood through the drawings and discussion.

[0021] Figure 1 A typical Fast Fourier Transform (FFT) of the output signal of a common Class AB solid-state audio signal amplifier using the Federal Trade Commission (FTC) 1kHz uniform test tone standard is shown. The even-order harmonic distortion is very low compared to the odd-order harmonic distortion. This spectrum is typical of solid-state amplifiers, which are often characterized as being more fatigue-prone and less clear than vacuum tube amplifiers.

[0022] Figure 2 A typical FFT of the output signal of a common class AB vacuum tube audio signal amplifier using the same FTC 1kHz unified test tone standard is shown. Even-order harmonic distortion is higher than odd-order harmonic distortion and is therefore considered to be dominant. The dominant even-order harmonic distortion (whose amplitude decreases gradually with increasing order) compared to odd-order provides a sound that is characterized as less fatiguing and clearer than solid-state technology.

[0023] Figure 3Variable input load VR1 / C2 impedance varies inversely with the output impedance load on transistors Q1 / Q2, with a gain of approximately 20 times or 26 dB, thereby changing the order of distortion of the audio signal output.

[0024] Figure 4A Based on the above, a single basic test frequency input signal f0 is used. Figure 3 Schematic diagram of distortion measurements of the circuit mentioned in , where the variable load VR1 / C2 impedance is set high, resulting in a very low output impedance across transistors Q1 / Q2. At the output, the composition of the harmonic distortion h2, h3 and h4 of each order shows a sequential decreasing trend of even and odd order harmonics as the order increases.

[0025] Figure 4B Based on the above, a single basic test frequency input signal f0 is used. Figure 3 Schematic diagram of distortion measurements of the circuit mentioned in , where the variable load VR1 / C2 impedance is set to medium (approximately equal to the source impedance of the amplifier), thereby producing a medium impedance on the transistors Q1 / Q2. At the output, the composition of the harmonic distortion orders h2, h3 and h4 shows the dominant odd third order harmonics and the reduced even order harmonics.

[0026] Figure 4C Based on the above, a single basic test frequency input signal f0 is used. Figure 3 Schematic diagram of distortion measurements of the circuit mentioned in , where the variable load VR1 / C2 impedance is set slightly higher than the source impedance, thereby producing a slightly lower impedance on the transistors Q1 / Q2. At the output, the composition of the harmonic distortion h2, h3 and h4 of each order shows the dominance of even-order harmonics and the reduction of odd-order harmonics, and their amplitudes decrease as the order of the harmonic distortion increases. Figure 4C The THD in Figure 4B THD in.

[0027] Figure 4D Based on the above, a single basic test frequency input signal f0 is used. Figure 3 Schematic diagram of distortion measurements for the circuit mentioned in , where VR1 / C2 impedance is set very low, creating a very high loading effect on transistors Q1 / Q2. At the output, the composition of the harmonic distortion h2, h3 and h4 of each order shows the dominance of odd-order harmonics, and the amplitudes of both odd and even numbers decrease with increasing order. Figure 4D The THD in Figure 4C THD in.

[0028] Figures 4E to 4Gare graphs of measured THD residuals of a power amplifier prototype, showing the minimum THD of the preamplifier, the optimized THD of the preamplifier, and the overall amplifier THD resulting from the minimum preamplifier THD, respectively.

[0029] Figure 5A is a circuit diagram showing one of the applicant's alternative embodiments, in which a plurality of Figure 3 Amplifier circuits of the type shown are cascaded and arranged and adapted to form a single composite audio amplifier circuit for creating more than Figure 4A , Figure 4B , Figure 4C and Figure 4D The variability of the harmonic spectrum is shown for customization purposes.

[0030] Figure 5B is a circuit diagram showing one of the applicant's alternative embodiments, in which a plurality of Figure 3 Amplifier circuits of the type shown are connected in parallel, arranged and adjusted to form a single composite audio amplifier circuit for creating the desired balance of the overall harmonic distortion spectrum beyond Figure 4A , Figure 4B , Figure 4C and Figure 4D as well as Figure 5A Options for customization purposes.

[0031] Fig. 6A The present invention is a schematic circuit diagram showing an alternative embodiment using the pre-compensation capability and post-compensation capability designed by the applicant. Using multiple audio amplifier devices, including a low-distortion audio source, a preamplifier, a power amplifier, and a transducer (such as a speaker), an alternative embodiment of a stand-alone device is designed, which is intended to supplement the overall distortion spectrum of the entire system so that the output at the power amplifier output has balanced distortion of various orders.

[0032] Figure 6B Figure 1 is a graphic of the basic distortion of a typical solid-state preamplifier reproducing a 1kHz sine wave input signal.

[0033] Figure 6C yes Figure 6B A graph depicting the basic distortion of the same aforementioned solid-state preamplifier.

[0034] Fig.6D This is a graph of the basic distortion of a typical 100 watt total power off-the-shelf solid-state power amplifier.

[0035] Fig. 6E is the composite effect of all harmonic distortion measured at the power amplifier output.

[0036] Figure 7is a schematic circuit diagram of a more complex circuit using an adjustable current source and a harmonic reduction circuit as another embodiment designed by the applicant.

[0037] Figure 8 is a circuit diagram of a digital audio playback device including an audio signal processor that generates harmonic distortion of a desired order and combines it into an audio track. In this simplified example, a conventional solid-state power amplifier (which has dominant odd-order harmonic distortion) can be balanced to have a harmonic distortion similar to Figure 2 The output of the amplifier is modified to have a dominant even-order harmonic spectrum in the amplifier to provide a sound characterized as less fatiguing and clearer. By this example, the solid-state amplifier now has the harmonic distortion spectrum of a vacuum tube amplifier. DETAILED DESCRIPTION

[0038] Figure 1 Harmonic distortion behavior in a conventional solid-state power audio amplifier using a single uniform 1kHz test tone is shown. Even-order harmonic distortion is produced at 2kHz, 4kHz, 6kHz, and 8kHz, while odd-order harmonic distortion is produced at 3kHz, 5kHz, 7kHz, and 9kHz. The odd-order harmonics are dominant and uniform compared to the lesser even-order harmonic distortion. At 35dB output, the odd-order harmonic distortion measurements are all within a range of approximately -17dB. The even-order harmonic distortion ranges from -55dB for the second order to -45dB for the eighth order, showing a gradual increase in energy with increasing order. This spectrum is typical of solid-state amplifiers, which are often characterized as being more fatiguing and less clear than vacuum tube amplifiers with a dominant even-order spectrum and both even and odd amplitudes gradually decreasing with increasing order (where each odd order is smaller than the previous even order).

[0039] Figure 2 A comparison of vacuum tube based audio power amplifiers using a single uniform 1kHz test tone is shown. Even order harmonic distortion is produced at 2kHz, 4kHz, 6kHz and 8kHz, while odd order harmonic distortion is produced at 3kHz, 5kHz, 7kHz and 9kHz. The even order harmonics are dominant, with each odd order being smaller than the previous even order. At 30dB output, the even orders gradually decrease with each successive even order, with measurements ranging from approximately -2dB for the second order to -30dB for the eighth order. Each successive odd order harmonic is smaller than the previous even order, with measurements ranging from approximately -16dB for the third order to -24dB for the ninth order. The spectrum of odd orders gradually decreases with each successive order from the third to the seventh order, with only a slight increase at the ninth order which is usually unnoticeable.

[0040] Figure 2 The overall harmonic distortion is higher than Figure 1 Overall harmonic distortion, but consistent with being considered a balanced spectrum, while Figure 1 So, although the overall harmonic distortion is higher, Figure 2 The perceived sound quality is better for less fatigue and greater clarity.

[0041] Figure 3 is a schematic circuit diagram of an audio amplifier circuit having a gain of approximately 20 times or 26 dB. The amplifier has a first stage comprising transistor Q1 powered by Q2. The amplifier has an output stage comprising transistor Q3. The gain of the circuit can be varied without affecting the harmonic distortion spectrum. The gain can also be close to unity gain when the function of the amplifier is simply to alter the harmonic distortion spectrum in the audio signal.

[0042] Figure 3 The audio amplifier circuit can use the VR1 / C2 combination to change the output impedance by adjusting the input impedance. This load can be fixed, or it can be adjustable, using a variable resistor at VR1 and / or a variable capacitor at the position of capacitor C2 to obtain different harmonic distortions required to balance the overall harmonic distortion spectrum of the amplifier.

[0043] By adjusting Figure 3 By changing the values ​​of VR1 / C2 in circuit 15 and using a frequency dependent load circuit, various frequency dependent distortion spectrum profiles can also be generated. For example, these configurations can be used to change the distortion characteristics of the intermediate frequency without changing the low frequency distortion and high frequency distortion.

[0044] When cascade Figure 3 When two or more circuits in circuit 15 (with or without gain) are combined, the distortion profile can be changed by adding or eliminating even-order or odd-order harmonic distortion.

[0045] exist Figure 4A In the middle, high input impedance VR1 / C2 provides a low output impedance across transistors Q1 / Q2 to provide a sequential decrease in even and odd harmonics as the order increases. This provides the harmonic content of the decreasing order to balance the overall harmonic distortion spectrum that requires even and odd orders.

[0046] exist Figure 4B In the example, the input impedance VR1 / C2 is approximately the same as the source impedance, resulting in a moderate output impedance across transistors Q1 / Q2 to provide dominant third order harmonic distortion and very low even order harmonic distortion. This has limited application except when the third order harmonic needs to be specifically added to balance the spectrum.

[0047] exist Figure 4C In the example, the input impedance VR1 / C2 is slightly higher than the source impedance, resulting in a less than medium output impedance on transistors Q1 / Q2 to provide dominant even-order harmonics and reduced odd-order harmonics, with amplitudes decreasing as the order of harmonic distortion increases. This provides additional content to an already balanced spectrum or where additional even-order harmonics are needed to balance the spectrum.

[0048] exist Figure 4D In the example, low input impedance VR1 / C2 provides a high output impedance across transistors Q1 / Q2 to provide dominant odd-order harmonics, with both odd and even amplitudes decreasing as the order increases. In extreme cases, this condition can overload the audio signal with excessive harmonic distortion, so applications are limited to adding increasing orders of even and odd harmonic distortion when it is deemed necessary to balance the harmonic distortion spectrum.

[0049] It should be noted that Figure 4A , Figure 4B , Figure 4C and Figure 4D are simplified examples and their harmonic behavior continues beyond h4. It is also important to note that Figure 4A , Figure 4B , Figure 4C and Figure 4D is an example of the variability of the HBA circuit, which is not strictly limited to these four load examples and can therefore have any load required to achieve the desired harmonic distortion spectrum.

[0050] Figure 4E , Figure 4F and Figure 4G The measured THD of a test power amplifier circuit having a preamplifier circuit and a 100 Watt power amplifier circuit is shown. The preamplifier has a sine wave generator with a THD of 0.00002% at 2Vrms (Volts Root Mean Square) under a 30kOhm load and an internal impedance of 20Ohm. The supply voltage of the preamplifier circuit is + / -20VDC. The output voltage is 7.5Vrms, the load is 15kOhm, the input impedance is 30kOhm, the gain is 20dB and the bandwidth is 300kHz. The power amplifier circuit is 100 Watts at 4Ohm for providing a power output of 25Wrms (WattsRoot Mean Square) under a 4Ohm load. As shown in FIG. Figure 4E As shown, the measured THD of the preamplifier is 0.00892%, and its distortion at 3kHz is greater than that at 2kHz. Figure 4F In the optimized preamplifier (such as Figure 3(middle) can provide a THD of 0.10874%, which is greater than the minimum THD of the preamplifier, however, its harmonic distortion has energy that decreases with increasing orders 2, 3, 4 and 5. Figure 4G The power amplifier output shown in maintains this harmonic distortion with decreasing energy as orders 2, 3, 4 and 5 increase.

[0051] As will be appreciated, having a structure similar to Figure 4B The output characteristics of the audio power amplifier circuit are similar to Figure 4C The preamplifier circuit combination will provide a combination effect similar to the 4A, and the combination is similar to Figure 4B Compared with power amplifiers and preamplifiers, the perceived quality will be less fatigue-prone and higher in clarity. Figure 5A and Figure 5B Shows Figure 3 Two different topologies are adapted to achieve this.

[0052] Figure 5A is a circuit diagram showing other embodiments in which amplifiers 15, 15' and 15" are cascaded and arranged using different gain configurations. By giving different harmonic distortion characteristics to each of the series-connected amplifiers 15, 15' and 15", more complex configurations can be used to balance the harmonic distortion spectrum. Figure 5A The circuit can be operated as a power amplifier or a preamplifier by changing the gain design of each stage.

[0053] Figure 5B Also shown is a circuit diagram of an alternative embodiment in which amplifiers 15" are arranged in parallel with adjustable gain levels and vector summing circuits to generate various distortion spectrum characteristics required to balance the overall harmonic distortion spectrum. Figure 5B The circuit can be operated as a power amplifier or a preamplifier by changing the gain design of each stage.

[0054] Fig. 6A The pre-compensation and post-compensation capabilities of the applicant's design are shown to be usable regardless of whether the source audio signal is transmitted via wired or wireless transmission. In addition to the receiver and amplifier at the receiving end, considering the harmonic distortion of each order of the transmission medium allows the transmission of a pre-compensated spectrum, thereby achieving balanced distortion of each order at the speaker output of the receiving amplifier. Post-compensation can alternatively be implemented to provide a similar balanced order harmonic spectrum at the speaker output.

[0055] Fig. 6APre-compensation of harmonic distortion of various orders at the transducer level using applicant's design is also shown. Considering the harmonic distortion of various orders of the transducer allows a pre-compensated spectrum to be transmitted to the transducer in addition to other amplifiers in the overall system, thereby producing an acoustic output that conforms to a balanced distortion spectrum.

[0056] You can use Figure 6B , Figure 6C , Fig.6D and Fig. 6E The actual test results measurements provided in the Fig. 6A .

[0057] Figure 6B The following is a graph of the basic distortion of a typical solid-state preamplifier reproducing a 1kHz sine wave input signal. Its THD is a negligible 0.00002% at 2 volts RMS into a 30kOhm load, and the preamplifier internal impedance is 20Ohms. The circuit power supply is + / -20 volts DC, and its output voltage is 7.5 volts RMS into a 15kOhm load. The gain of the circuit is 20dB, and the bandwidth of the preamplifier is 300kHz. As previously shown in Figure 1 Solid-state amplifier distortion graphs are similar to those depicted in Figure 6B Shown is the unbalanced distortion spectrum common to most solid-state preamplifiers.

[0058] Figure 6C yes Figure 6B The same previously described solid-state preamplifier basic distortion graph depicted in FIG, but applicant's embodiment is adjusted to balance the overall output of the system. Providing compensation at the preamplifier also provides pre-compensation for the power amplifier, thereby achieving a balanced output at the speaker terminals. Figure 4A The effect is similar to that of , where the distortion has shifted from a dominant odd-order spectrum to a spectrum with a sequential decreasing trend of even and odd-order harmonics as the order increases.

[0059] Fig.6D This is a graph of the basic distortion of a typical 100 Watt total power off-the-shelf solid-state power amplifier designed to produce a power output of 25 Watts RMS into a 4 Ohm load. The even order harmonic distortion is not dominant compared to the corresponding odd order harmonic distortion. Although typical for solid-state amplifiers, the harmonic distortion orders are considered to be unbalanced, as is the case with most solid-state amplifiers.

[0060] Fig. 6Eis the composite effect of all harmonic distortion measured at the power amplifier output, with applicant's implementation for balancing the entire composite system at the preamplifier. The combined orders of distortion of the system exhibit a more balanced overall spectrum measured at the power amplifier output. Even though all equipment is solid state, the composite spectrum of the various orders of distortion is closer to that measured for a vacuum tube system. The subjective performance of such a composite system will be better than without Figure 6C The harmonic adjustments provided in are less fatiguing and clearer.

[0061] Figure 7 An example of an implementation of the applicant's HBA circuit is shown, which provides a harmonic distortion characteristic adjustment that can reduce the overall harmonic content. The topology is a J-FET / bipolar cascade stage with Q9 and Q8, with bias provided by a J-FET Q10 operating as a constant current source and biased by R4 and R14. The Q8 base potential is generated by R3 and R5 and stabilized by C4. There is a bias return resistor R11 and a network consisting of R2, R12 and C3 at the gate circuit of Q9. This network is a local negative feedback around Q8 and Q9, and it provides gain reduction and stabilization to change the current / voltage (I / V) curve of the circuit. The change in the I / V curve forces the circuit to produce a dominant even harmonic distortion characteristic by bending its transfer function.

[0062] The collector load resistor R1 of Q8 and the bipolar transistors Q1, Q2, Q5, Q6 and Q7 form the output circuit of the amplifier. The transistor string in the output circuit is actually an active load. By using Figure 8 The "harmonic reduction" selector shown in can gradually reduce the amount of harmonic distortion by continuously bending the transfer function to adjust the amount of harmonic residuals produced by the circuit.

[0063] By selecting the amount of negative feedback around the Q8 and Q9 stages, the characteristics can be adjusted to be predominantly odd harmonics or predominantly even harmonics. Q8 and Q9 are selected for their nonlinearities in their I / V curves (determined with a curve tracer). Transistor Q9 can be replaced by a bipolar transistor to obtain different distortion characteristics. Control VR1 is used to gradually increase the amount of even harmonics by gradually loading the collector of Q8.

[0064] Q3 and Q11 form a linearized emitter follower whose purpose is to transform a high input impedance into a low output impedance. This linearization is provided by a constant current source Q11 and associated components.

[0065] The overall negative feedback at R10 and R13 is used to control the overall gain of the complete circuit and is coupled to the source of Q9 through emitter follower Q4. Emitter follower Q4 provides isolation between the input circuit and the output circuit.

[0066] Utilizing this nonlinear amplifier circuit, the overall amplitude of the distortion can be adjusted, and even and odd order harmonic distortion can be adjusted almost independently to achieve an overall balanced harmonic spectrum.

[0067] Figure 8 Also a circuit diagram of a digital audio playback device including an audio signal processor that generates and combines a desired order of harmonic distortion, in this case second order harmonic distortion, into an audio track. Such a device is not limited to second order harmonic distortion, but can be used for any order of harmonic distortion. This can be used to master original media files to pre-compensate for harmonic distortion deficiencies in typical playback devices, or to improve the actual harmonic distortion spectrum in a mastered media file.

[0068] Using Digital Signal Processor (DSP) Figure 3 The digital performance implementation of the analog circuit depicted in the Figure 4A , Figure 4B , Figure 4C and Figure 4D , Figure 5A and Figure 5B as well as Fig. 6A A distortion spectrum similar to that shown in to balance the overall spectrum. Using DSP, a second order harmonic signal can be generated and then added to the original trajectory. This can be achieved by processing the original trajectory to obtain an FFT data set, which can then be shifted to generate the second order harmonics. The amplitude of the data set can be attenuated to correspond to the desired second order harmonic requirement. In order to combine the original trajectory with the second order harmonic distortion trajectory, an inverse transform can be used to convert the FFT data set into a time trajectory and then mixed with the original trajectory. The synthesized trajectory can be stored in its modified digital form for subsequent playback, or played in real time without long-term storage. The synthesized modified trajectory now contains second order harmonic distortion. A method similar to Figure 4A , Figure 4B , Figure 4C and Figure 4D , Figure 5A and Figure 5B as well as Fig. 6A Any other significant orders of harmonic distortion as shown in the figure are removed to balance the overall harmonic spectrum.

[0069] Whether in the analog or digital domain, this technique can be used to correct for unbalanced orders of harmonic distortion associated with the various techniques involved in the original mastering. A remastered copy with balanced orders of harmonic distortion will sound subjectively less fatiguing and clearer.

[0070] In certain embodiments, the improved audio amplifier may be used in a hearing aid. A hearing aid microphone provides the audio signal source, and an amplified audio signal with a correctly balanced THD is provided to the output audio transducer of the hearing aid to provide less fatiguing and clearer hearing enhancement.

Claims

1. A method for reproducing an audio signal, comprising: providing a source audio signal from an audio signal source; providing at least one amplifier stage and amplifying the source audio signal to produce an audio output signal; as well as feeding said audio output signal to an audio transducer for listening purposes, The composition of total harmonic distortion (THD) in the audio output signal is changed to improve the listening quality by at least one of the following ways: processing the source audio signal using digital signal processing to introduce second order harmonic distortion energy; as well as The at least one amplifier stage uses a load transistor amplifier circuit including an output load that causes a second order harmonic distortion energy greater than a third order harmonic distortion energy to be generated in the load transistor amplifier circuit.

2. The method of claim 1 , wherein the providing at least one amplifier stage comprises providing the load transistor amplifier circuit including the output load that causes a second order harmonic distortion energy greater than a third order harmonic distortion energy to be generated in the load transistor amplifier circuit.

3. The method of claim 2, wherein the second order harmonic distortion energy produced by the load transistor amplifier circuit is greater than 5 dB greater than the third order harmonic distortion energy.

4. The method of claim 2, wherein the load is adjusted until the second-order harmonic distortion energy is greater than the third-order harmonic distortion energy in the output audio signal.

5. The method of claim 4, wherein the load of the input stage comprises a string of transistors to reduce the magnitude of harmonic distortion by bending the transfer function of the transistor amplifier circuit, and the load is adjusted by selecting a number of transistors in the string.

6. The method of any one of claims 2 to 5, wherein the amplifying further comprises amplifying the source audio signal using a solid-state amplifier circuit that produces negligible second and fourth order harmonic distortion and non-negligible third and fifth order harmonic distortion, wherein the composition of total harmonic distortion (THD) in the audio output signal has a balance of odd and even order harmonic distortion energies including second order harmonic distortion energy greater than third order harmonic distortion energy.

7. The method of claim 6, wherein the solid-state amplifier circuit provides power amplification and the load transistor amplifier circuit provides pre-amplification.

8. A method according to any one of claims 2 to 7, wherein the load transistor amplifier circuit is a negative feedback amplifier circuit, the magnitude of the negative feedback of which is selected to change the relative magnitude of second-order and third-order harmonic distortion energy in the audio output signal.

9. The method according to any one of claims 1 to 8, wherein the fourth order harmonic distortion energy of the output audio signal is less than the third order harmonic distortion energy.

10. The method of any one of claims 1 to 9, wherein fifth and higher order harmonic distortion energy of the output audio signal is imperceptible to normal adult human hearing.

11. The method of claim 1, 9 or 10, wherein the processing comprises: providing the source audio signal as an input audio track; Using Fourier transform to generate a second-order harmonic copy of the input audio track, and shifting the frequency of the input audio track to twice the original to simulate second-order harmonic distortion; mixing the input audio track with the second-order harmonic replica to produce in the mixed audio track a simulation of second-order harmonic distortion energy within the input audio track; The mixed audio track is amplified using a transistor circuit that generates negligible second and fourth order harmonic distortion energy and non-negligible third and fifth order harmonic distortion energy, wherein the composition of total harmonic distortion (THD) in the audio output signal is changed to improve listening quality.

12. The method according to any one of claims 1 to 10, wherein said providing the source audio signal comprises providing a hearing aid microphone, said audio transducer being an output audio transducer of the hearing aid.

13. The method of any one of claims 1 to 12, wherein the amplifier generates a total harmonic distortion of less than 0.3% of the signal amplitude at the audio output signal.

14. An audio signal power amplifier system, comprising: A nonlinear preamplifier having a gain close to unity or low gain, the nonlinear preamplifier comprising: High impedance input; Low impedance output; an input stage connected to the input, the input stage comprising a transistor amplifier circuit including an adjustable load which, in use, causes the input stage to exhibit different relative amounts of second order harmonic distortion energy and third order harmonic distortion energy depending on the adjustable load; an output stage connected to the input stage and to the low impedance output; as well as a non-linear power amplifier receiving said output stage, comprising a transistor circuit generating negligible second and fourth order harmonic distortion and non-negligible third and fifth order harmonic distortion, wherein the system provides higher second order harmonic distortion energy than third order harmonic distortion energy at the output of the nonlinear power amplifier in response to an input audio signal provided to the high impedance input, wherein the energy at progressively higher order harmonics decreases.

15. The amplifier system of claim 14, wherein the load of the input stage comprises a transistor string to reduce the magnitude of harmonic distortion by bending the transfer function of the transistor amplifier circuit.

16. The amplifier of claim 15, wherein the transistor string comprises a selector switch for selecting a number of transistors to activate in the transistor string.

17. An amplifier as claimed in claim 14, 15 or 16, wherein the load of the input stage comprises a variable resistor capable of varying the load to change the balance of the second order harmonic distortion energy relative to the third order harmonic distortion energy.

18. An amplifier as claimed in any one of claims 14 to 17, wherein the amplifier generates a total harmonic distortion of less than 0.5% of the amplitude of the signal at the low impedance output.

19. An amplifier according to any one of claims 14 to 17, wherein the amplifier generates a total harmonic distortion of less than 0.3% of the amplitude of the signal at the low impedance output.

20. An audio signal playback device, comprising: a processor configured to receive an input audio track and generate a second-order harmonic copy thereof; a mixer configured to mix the input audio track with the second order harmonic replica to produce in the mixed audio track a simulation of the second order harmonic distortion energy within the input audio track; a non-linear power amplifier receiving the mixed audio track, comprising a transistor circuit that generates negligible second and fourth order harmonic distortion energies and non-negligible third and fifth order harmonic distortion energies, The system provides higher second order harmonic distortion energy than third order harmonic distortion energy, wherein the energy at progressively higher order harmonics decreases.

21. The amplifier of claim 20, wherein the amplifier generates a total harmonic distortion of less than 0.5% of the amplitude of the signal at the output of the non-linear power amplifier.

22. The amplifier of claim 20, wherein the amplifier generates a total harmonic distortion of less than 0.3% of the amplitude of the signal at the output of the non-linear power amplifier.