Semiconductor emulation of tube amplifiers

What is claimed:

1. A semiconductor distortion system comprising: input means for receiving an input signal; a first distortion means for introducing distortion to produce a first distorted signal; a tone control means for altering the tonality of the first distorted signal and producing a tone control signal; a second distortion means for clipping the tone control signal at a level which is a function of a resulting clipped signal; and an output means.

2. The semiconductor distortion system of claim 1, wherein said input means and output means each include amplifiers.

3. The semiconductor distortion system of claim 1, wherein said input means includes a preamplifier and filtering means for filtering the input signal to produce a fundamental tonality of the amplifier.

4. The semiconductor distortion system of claim 1, wherein said first distortion means includes means for limiting said input signal and introducing even harmonics.

5. The semiconductor distortion system of claim 1, wherein said first distortion means includes an asymmetrical bias shifting means and a limiting means.

6. The semiconductor distortion system of claim 5, wherein said bias shifting means includes an asymmetrical impedance and a capacitor.

7. The semiconductor distortion system of claim 1, wherein said second distortion means includes an input terminal and an output terminal; amplifier means having a first input connected to said input terminal, a second input and an output; limiter means connected to the output of said amplifier means for limiting the output of said amplifier means; filter means connecting said limiter means to said output terminal; and feedback means connecting said output means terminal to said second input of said amplifier means.

8. The semiconductor distortion system of claim 7, wherein said first distortion means includes bias shifter means for asymmetrically transmitting input signals above a preselected value.

9. The semiconductor distortion system of claim 1, wherein said second distortion means includes clipping means for clipping the resultant signal as a function of the frequency content of said resultant signal.

10. The semiconductor distortion system of claim 1, wherein said second distortion means includes a transformerless filter means having a region of frequencies which are passed preferentially; and means for reducing said region as a function of increasing amplitude of an input to said semiconductor distortion system.

11. A semiconductor distortion system comprising:

input means including conversion means for converting an analog input signal to a digital input signal;

a digital processor having a first distortion program for introducing distortion to produce a first distorted signal by clipping, a tone control program for altering the tonality of the first distorted signal and producing a tone controlled signal, and a second distortion program for clipping the tone controlled signal:

said first distortion program including a filtering program to filter the input signal prior to being clipped by said first distortion program; and

an output means.

12. The semiconductor distortion system of claim 11, wherein said filtering includes resonances.

13. A semiconductor distortion system comprising:

input means including conversion means for converting an analog input signal to a digital input signal;

a digital processor having a first distortion program for introducing distortion to produce a first distorted signal, a tone control program for altering the tonality of the first distorted signal and producing a tone controlled signal, and a second distortion program for clipping the tone controlled signal at a level which is a function of a resultant clipped signal; and

an output means.

14. The semiconductor distortion system of claim 13, wherein said clipping is also a function of time.

15. The semiconductor distortion system of claim 13, wherein said clipping is also a function of the frequency content of the resultant signal.

16. A semiconductor distortion system comprising:

input means including conversion means for converting an analog input signal to a digital input signal;

a digital processor having a first distortion program for introducing distortion to produce a first distorted signal, a tone control program for altering the tonality of the first distorted signal and producing a tone controlled signal, and a second distortion program for clipping the tone controlled signal; and

an output means.

17. The semiconductor distortion system of claim 16, wherein said first distortion program compresses said input signal.

18. The semiconductor distortion system of claim 16, wherein said input means includes a preamplifier for receiving an analog input and producing an analog signal for said conversion means, said preamplifier includes means for increasing the input capacitance of the amplifier above the stray wiring capacitance for properly loading the input source.

19. The semiconductor distortion system of claim 16, wherein said second distortion program receives a first value, computes a second value as a function of a first value wherein the ratio of the second value to the first value increases for increasing magnitude of the first value.

20. The semiconductor distortion system of claim 16, wherein said input means includes a multiplexer means for selecting which of a multiplicity of input signals to be processed.

21. The semiconductor distortion system of claim 16, wherein said tone control program alters the tone parametrically.

22. The semiconductor distortion system of claim 16, wherein said digital processor further includes emulation program for emulating the signal transformation of a speaker audibly driving a microphone wherein said emulation program includes a filter program, a time delay program and mixing program.

23. The semiconductor distortion system of claim 22, wherein said filter program includes a resonance approximating the bass resonance of a speaker.

24. The semiconductor distortion system of claim 22, wherein said time delay program includes a time delay approximating the difference in the transit times of sounds from one side of a speaker to a microphone and from another side of said speaker to said microphone.

25. The semiconductor distortion system of claim 22, wherein said time delay program includes a time delay approximating the difference in the transit times of sounds from a first speaker to a microphone and a second speaker to said microphone.

26. A semiconductor distortion system comprising:

input means including conversion means for converting an analog input signal to a digital input signal;

a digital processor having a first distortion program for introducing even harmonic distortion to produce a first distorted signal, a tone control program for altering the tonality of the first distorted signal and producing a tone controlled signal, and a second distortion program for clipping the tone controlled signal; and

an output means.

27. A semiconductor distortion system comprising:

input means including conversion means for converting an analog input signal to a digital input signal;

a digital processor having a first distortion program for introducing distortion to produce a first distorted signal, a tone control program for altering the tonality of the first distorted signal and producing a tone controlled signal, and a second distortion program for clipping the tone controlled signal;

said second distortion program includes limiting, filtering and feedback program; and

an output means.

28. A semiconductor distortion system comprising:

input means for receiving an input signal;

a first distortion means for introducing distortion to produce a first distorted signal;

a tone control means for altering the tonality of the first distorted signal and producing a tone controlled signal;

a second distortion means including transfer means having an input and output for introducing harmonics into said tone controlled signal and increasing the ratio of the output signal to the input signal for an increasing input signal and means for clipping the resultant signal; and

an output means.

29. A semiconductor distortion system of claim 28, wherein said transfer means includes a plurality of switch means responsive to said input of said second distortion means for progressively adding parallel resistors to progressively increase the output current.

30. The semiconductor distortion system of claim 28, wherein said input means includes a preamplifier and filtering means for filtering said input signal to produce a fundamental tonality of the amplifier.

31. The semiconductor distortion system of claim 28, wherein said first distortion means includes means for limiting said input signal and introducing even harmonics.

32. The semiconductor distortion system of claim 28, wherein said first distortion means includes an asymmetrical bias shifting means and a limiting means.

33. The semiconductor distortion system of claim 32 wherein said bias shifting means includes an asymmetrical impedance and a capacitor.

34. A semiconductor distortion system comprising:

input means for receiving an input signal;

a first distortion means for introducing distortion to produce a first distorted signal;

a tone control means for altering the tonality of the first distorted signal and producing a tone controlled signal;

a second distortion means including an input and an output terminal, a clipping means, an amplifier means, a limiter means, a filter means and a feedback means interconnected between said input and output terminals; and

an output means.

35. The semiconductor distortion system of claim 34, wherein said input means includes a preamplifier and filtering means for filtering the input signal to produce a fundamental tonality of the amplifier.

36. The semiconductor distortion system of claim 34, wherein said first distortion means includes means for limiting said input signal and introducing even harmonics.

37. The semiconductor distortion system of claim 34, wherein said first distortion means includes an asymmetrical bias shifting means and a limiting means.

38. The semiconductor distortion system of claim 37, wherein said bias shifting means includes an asymmetrical impedance and a capacitor.

Description Submit all comments and votes

BACKGROUND AND SUMMARY OF THE INVENTION

The present invention relates generally to amplifiers and more specifically to all forms of audio amplifiers and guitar amplifiers. It further relates to a distortion synthesizer which replicates the sounds produced by overdriven vacuum tube amplifiers.

Since semiconductor devices have become viable components for amplifiers, there has been a debate upon the virtues of semiconductor or solid state amplifiers versus the vacuum tube amplifiers. Some believe that tube amplifiers work better because vacuum tubes are more natural amplifiers than semiconductor devices. Some think that semiconductor amplifiers produce a sound that has no warmth; they are too clear until the semiconductor devices saturate, then they are too noisy. Tube amplifiers seem not to give up when overdriven, they seem to try to reach the impossible.

The vacuum tube amplifiers, however, do have some limitations. For some people, the limitation is simply the warmup time and the fragility of vacuum tubes. For guitarists the problems are more serious. A powerful amplifier does not sound right when operating at low levels such as those needed to fill a small room.

The prior art is filled with va ious attempts to satisfy the guitarists need for the tube amplifier sound with the more reliable semiconductor devices or just to enhance the sound from vacuum tube amplifiers. Moog, in U.S. Pat. No. 4,180,707, simulates the overdriven amplifier with a compressor and a clipper that can produce even harmonics as well as odd harmonics to produce the guitar sound at preamplifier power levels. Claret, in U.S. Pat. No. 4,286,492, modifies the operating point of the amplifier to clip at lower Power. Woods, in U.S. Pat. No. 3,860,876, heavily modifies the frequency response of a distorted input. Smith, in U.S. Pat. No. 4,211,893, selectively adds gain in a preamplifier stage to get distortion. Sondermeyer, in U.S. Pat. No. 4,405,832, switchable forces odd harmonic distortion and, in U.S. Pat. No. 4,439,742, creates a soft crossover distortion, also an odd harmonic phenomenon. Scholz, in U.S. Pat. No. 4,143,245, creates distortion at any sound level by operating an amplifier at maximum levels with resistive loads and driving the speaker with only a portion of the amplifier output.

In another vein, Todokoro, in U.S. Pat. No. 4,000,474, simulates a triode tube amplifier with junction field effect transistors.

The prior art is also filled with many examples of design built by many manufacturers of guitar amplifiers. Technically, the guitar amplifier is a poor example of vacuum tube amplifier design. Certainly, the guitar amplifier of today does not have the frequency response nor the clarity of the high fidelity amplifier of yesterday. However, the reason is not technical but lies in the art. The sound of inexpensive, overdriven amplifiers has become part of the art.

Thus, to simulate the vacuum tube amplifier, one must fully appreciate the nature of the components of said amplifier. One of the key components of a vacuum tube amplifier is the output transformer. The transformer passes a relatively narrow band of frequencies in the middle of the audio spectrum. The feedback in the power stage of the amplifier broadens the transformer response so that the amplifier operates effectively over a wider range of frequencies. However, when the amplifier tubes are saturated, they cannot perform the feedback function and the response narrows to the transformer response.

Of course, another key element of the vacuum tube amplifier is the tube itself. The various stages of a vacuum tube amplifier are usually coupled with capacitors that carry the signal from the output of one stage to the input of the next while blocking the constant or DC voltage of the output from the input. When the input of a stage is driven so that the grid of the vacuum tube becomes more positive than its cathode, then significant currents will flow in the grid circuit. Some of the grid current charges the coupling capacitor and thereby alters the operating point of the vacuum to amplify more asymmetrically than it usually does. When this asymmetrical waveform is amplified and clipped symmetrically by a push-pull output stage, as usually found in powerful amplifiers, it produces even as well as odd distortion harmonics. It is the even harmonics that seem to be more musical than the harsher odd harmonics.

Vacuum tubes, as all devices, has an input capacitance. The significant component of this capacitance is from the grid to the plate. This is the Miller capacitance and is multiplied by the gain of the tube. Consequentially, this input capacitance operates with grid driving impedance to limit the frequency response and to thereby change the tone of the connected musical instrument as a function of the setting of the volume control.

The gain of tubes is not constant with respect to grid voltage. The gain generally increases for increasing grid voltages. This is important when analyzing class B or AB push-pull output stages. The gain change produces harmonics in the output.

The output stage cannot use significant feedback because of the phase shifting in the transformer and other circuit components. Consequently, the output impedance of the output stage is comparatively high. This high output impedance reacts with the speaker load differently than a low output impedance. The output impedance that is associated with the proper sound is approximately that of the speaker impedance.

The output tubes of a class AB or B amplifier draw larger currents from the power supply as the input signal increases. The power supply reacts to this by lowering its output voltage according to its output impedance. However, since this output impedance is generally capacitive, a suddenly appearing large input signal will be amplified at a high clipping level for a short time and then will progress to a lower clipping level. This effect is part of the punch of the amplifier and makes the amplifier sound less compressed.

There are digital devices currently available for producing audio effects such as chorus, flanging, reverberation, vibrato, sampling, pitch change, etc. The delay effects, such as flanging, reverberation and sampling, simply record the signal and play it back later. Pitch change records the signal and plays it back at a different rate. Harmonic analysis of these effects show that all extra frequencies that are generated are created by sampling. None of these effects intentionally introduce harmonics of the signal into the signal.

Another view of these effects is that their basic intent is to recombine a signal that has been delayed and may have been attenuated with itself. Thus, the only harmonics that can be generated are due to the sampling process.

Thus, the primary object of the invention is to provide a semiconductor amplifier which simulates the distortion of a vacuum tube amplifier.

Another object of the invention is to simulate the effects of grid current flowing. This produces even harmonic distortion which is a more pleasant and musical distortion than one made of solely odd harmonics. Further, the grid current effects in a capacitively coupled circuit produces the desirable attach on a note.

A further object of the invention is a guitar amplifier effects preamplifier which may be elegantly professional or may be a simple effects pedal. This is quite possible from the teachings herein because the tube simulation may be done either at low or at high power levels.

A still further objection of the invention is to achieve the general improvement of guitar amplifiers of all types to provide the high power distortion effect at all power levels.

An even further object of the invention is to provide a general use power amplifier which graciously handles excessive inputs.

Another object of the invention is to provide an amplifier with the correct input and output impedances to load the guitar or other source correctly and to drive the speaker correctly.

Another object of the invention is to provide an amplifier with a non-constant or variable gain stage to emulate the distortion of a Class B or AB output stage.

A still further object of the invention is to provide control of the output level of an amplifier as a function of the output level to emulate the effect of the power supply impedance.

An even further object of the invention is to provide an amplifier with a combination bias shifter and distortion stage that operates effectively at low input levels.

A still further object of the invention is to provide mathematical modeling of an amplifier so that the amplifier effect may be created via a digital signal processor.

An even further object of the invention is to provide analog and digital signal processors which intentionally introduces harmonics of the signal into the signal.

A still further object of the invention is the structure of the analog amplifier or the order of calculations of the digital signal processor.

These and other objects of the invention are attained by a distortion synthesizer, having a first distortion circuit, a tone control circuit for altering the tonality of the first distorted signal and a second distorted circuit for introducing harmonics into said tone controlled signal and for clipping the resultant signal as a function of said resultant signal. The system may be analog or a programmed digital processor. The first distortion circuit limits and introduces even harmonics by an asymmetrical bias shifting circuit. The first distortion circuit can also compress the signal. The second distortion circuit includes a variable gain stage which increases with increased input using a plurality of switches which progressively add parallel resistors to progressively increase output current. The clipping circuit in the second distortion circuit clips as a function of amplitude, time or frequency content of the input signal. A direct equalization circuit is provided which emulates a speaker audibly driving a microphone and includes a filter, delay and mixer.

The distortion system provides a distorted signal whose harmonic content increases with increasing input signal. The even harmonics content is initially increased with each cycle of an input signal. Also, the gain and total harmonic distortion increase for a first range of input signal amplitudes, and the gain decreases, and total harmonic distortion increases for a second range of input signal amplitude. In addition, the amplitude of the distorted signal initially decrease and subsequently increases the amplitude of the distorted signal.

Other objects, advantages and novel features of the present invention will become apparent from the following detailed description of the invention when considered in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of an audio amplifier incorporating the principles of the present invention.

FIG. 2 is a block diagram of a distortion synthesized according to the principles of the present invention.

FIG. 3 is a schematic of a distortion synthesized low power level operation according to the principles of the present invention.

FIG. 4 is a schematic of another distortion synthesized according to the principles of the present invention.

FIG. 5 is a schematic of another bias shifter according to the principles of the present invention.

FIG. 6 is a schematic of the input of a guitar amplifier according to the principles of the present invention.

FIG. 7 is a schematic of the variable gain stage according to the principles of the present invention.

FIG. 8 is a schematic of the variable attention stage according to the principles of the present invention.

FIG. 9 is a schematic of the power supply impedance effect emulator according to the principles of the present invention.

FIG. 10 is a schematic of an alternative to FIG. 9.

FIG. 11 is a schematic of the combination bias shifter and distortion stage according to the principles of the present invention.

FIG. 12 is a block diagram of the overall structure of the amplifier according to the principles of the present invention.

FIG. 13 is a block diagram of a speaker emulating circuit according to the principles of the present invention.

FIG. 14 is a block diagram of a computer system for emulating a guitar amplifier according to the principles of the present invention.

DETAILED DESCRIPTION OF THE DRAWINGS

A distortion synthesizer which replicates a vacuum tube amplifier having a transformer output would include asymmetrical attenuation of inputs signal an bias point shifting to produce even harmonics for large input signals, an asymmetrical, non-constant gain to produce both even and odd harmonics for small input signals, and a clipping circuit whose clipping level is a function of the output to decompress large input signals.

An amplifier incorporating the principles of the present invention is illustrated in FIG. 1 as including an input 15 connected to a preamplification or input stage 20. A plurality of adjustments 25 are provided on the preamplifier 20. These include frequency controls, for example, bass and treble, as well as gain control. The output 26 of the preamplifier 20 is connected to the input 32 of distortion synthesizer 30 which replicates the distortion of a vacuum tube amplifier. A variable control 33 is provided to select the distortion tone or the frequency of the distortion synthesizer's filter. The first output 36 of the distortion synthesizer 30 is connected to the input 42 of the power amplifier 40. A second output 35 of the distortion synthesizer 30 is available for connection to other types of effect devices available in the industry. Input 45 provides a second input to the power amplifier stage 40 from other sources other than the source 15 and more particularly the outputs of the effect devices driven by the signal from output 35. The output of the power amplifier output stage 40 is provided on terminal 50.

The block diagram of a preferred embodiment of the distortion synthesizer 30 is shown in FIG. 2. The input 32, after any required preamplification, drives the bias shifter 60 which drives the non-inverting input of amplifier 61. The output of amplifier 61 is limited by limiter 62. The limiter output is then filtered by filter 63 to produce the output 36 and feedback to the inverting input of amplifier 61. The physical realization of these functions is quite well defined by the electronics arts, for example, in a book by Tobey, Graeme, and Huelsman, "Operational Amplifiers, Design and Applications", McGraw-Hill.

The response of this embodiment when the signal is significantly limited, is that of the filter. The gain of the filter can be expressed in FORTRAN as

F=s*h*x/[ (s+w)*(s+x)]

where

s= the Laplace transfer operation

h= the maximum gain of the filter

w= the low frequency cutoff in radians/sec.

x= the high frequency cutoff in radians/sec.

Furthermore, let the output of the amplifier can also be expressed in FORTRAN be

A=Vp*b-Vm*a

where

Vp= the non-inverting input voltage

b= the non-inverting gain

Vm= the inverting input voltage

a= the inverting output gain

Then the response for the above system, when there is no limiting, is

R=s*b*h*x /((S+w)*(s+x)+s*a*h*x)

which has the following approximate characteristics:

center frequency gain=b*h/(1+a*h)

low frequency cutoff=w/(1+a*h)

high frequency cutoff=x*(1+a*h)

This amplifying system then produces a larger bandwidth with lower low frequency cutoff and higher high frequency cutoff than the filter 63 when a*h is larger than zero and preferably three or larger.

Notice that this block diagram consists of functions that are each well known in the electronics arts that may be implemented with variety of technologies, vacuum tubes, semiconductors, linear integrated circuits, and digital signal processors.

An embodiment for the block diagram of FIG. 2 is shown in FIG. 3. The bias shifter 70 is similar to bias shifter 60. The amplifier 71 is similar to amplifier 61 except that actual amplifiers exhibit limiting as their output voltages approach their power supply voltages The limiter 72 is similar to limiter 62 except that it depends upon the limiting nature of the amplifier 71 and that it provides a reasonably large output impedance for the filtering circuit 73. Again, the filter 73 is similar to filter 63.

The input signal on input 32 is from a preamplifier 80 which is configured as required. The bias shifter 70 is capacitively coupled by capacitor 81. Resistors 82, 84 and 85 form a standard inverting amplifier with operational amplifier 87 until the signal becomes greater than the conduction voltage of diode 86. Then the signal faces the same alterations as the signal through a tube that has grid current, namely, the signal is asymmetrically attenuated and the signal is offset by the charge on capacitor 81. Resistor 83 limits the charging rate of the capacitor 81 and thereby prolongs the bias shifting process gradually on a plurality of cycles. Limiting the charging rate is important because the attack of a note is then not offset while later portions of the note are offset. The limiter 72 can thus create waveforms with more power for the attack on the beginning than for the later portions of the note which is highly distorted. This additional power is needed to achieve the desired musical effects.

The amplifier 71 includes resistors 90 through 93 and an operational amplifier 94. Appropriate choices in resistor values can create the required gains a and b above. The operational amplifier 94 also helps the limiting function 62 by limiting its own output to be between the power supply voltages.

The function of limiter 62 is completed by limiter 72 which uses two resistors 100, 101 and two diodes 102, 103. This specific construction of resistors and diodes not only limits or clips the signal, but also produced harmonics with smaller amplitudes than clippers without resistor 101. The resistor 100 is similar to the unsaturated plate resistance of the output tube, while resistor 101 is similar to the saturated plate resistance. The diodes have a voltage drop similar to the maximum voltage excursion of the plate voltage. These diodes are preferably light emitting diodes chosen for their voltage versus current characteristics. Of course, these similarities are in a proportional sense because inexpensive operational amplifiers have smaller voltages and currents than amplifier output tubes.

The filter 73 is realized with high-pass components 110 and 111, and low-pass components 112 and 113 connected to operational amplifier 114. Capacitor 115 loads the limiter to produce another rolloff in the filter response. This additional reactive component and its consequential additional rolloff produces a twelve decibel per octave rolloff in the audio range for a sweeter distortion toneality.

This filter 73 is a generic representation of a generalized filter. In practice is desirable to make the frequency cutoffs variable. This