System for High Efficiency Vibratory Acoustic Stimulation

Abstract

A system and method of driving a floating mass transducer with an analog input signal uIN(t), uIN(t) being between ground and VCC, is provided. The method includes converting uIN(t) to a binary rectangular signal uR(t) with two levels VCC and GND. A switching network is driven with uR(t) so as to switch nodes N1 and N2 between VCC and ground. The floating mass transducer is coupled between nodes N1 and N2 to a capacitor C in parallel, and further to a coil L in series.

Description

TECHNICAL FIELD

The present invention relates to a system and method for high efficiency vibratory acoustic stimulation, and more particularly to a system and method for efficiently driving a floating mass transducer.

BACKGROUND ART

The standard treatment of hearing impaired persons is to use conventional hearing aids, which are essentially based on filtering and amplifying the acoustic signal. Another possibility is to employ so called “middle ear implants” which are based on vibratory systems. A vibratory system is an actuator driven by a signal derived from the acoustic signal and causes mechanical movements of structures in the middle ear or inner ear, which cause sound-like sensations. One example of such a vibratory system is the so called “Floating Mass Transducer (FMT)” described, for example, in U.S. Pat. No. 5,456,654 (Ball), which is hereby incorporated herein by reference in its entirety.

A FMT illustratively may include a magnet positioned inside a housing. The housing is proportioned to be disposed in the ear and in contact with middle ear or internal ear structures such as the ossicles, or the oval window. A coil is also disposed inside the housing. The coil and magnet are each connected to the housing, and the coil is typically more rigidly connected to the housing than the magnet. When alternating current is delivered to the coil, the magnetic field generated by the coil interacts with the magnetic field of the magnet causing both the magnet and the coil to vibrate. As the current alternates, the magnet, and the coil and housing alternately move towards and away from each other. The vibrations produce actual side-to-side displacement of the housing and thereby vibrate the structure in the ear to which the housing is connected.

The electrical equivalent circuit of an FMT as described above is approximated by an ohmic resistor of about R_L=50Ω. From the engineering point of view, R_L is a low impedance load, and one of the problems is to drive such a load at a high overall power efficiency.

One textbook approach of driving R_L is to use a push-pull emitter follower as shown in FIG. 1 (prior art). The system is supplied symmetrically with +VCC and −VCC, and input and output voltages u_IN(t) and u_R(t) are referred to ground potential GND. The circuit consists of npn-transistor T₁, pnp-transistor T₂, and R_L. T₁ conducts on positive swings of the input signal u_IN(t), T₂ on negative swings. Voltage u_L(t) and input voltage u_IN(t) are approximately related via

u _L(t)≈u_IN(t)+U_F for u_IN(t)<−U_F

u _L(t)≈0 for −U_F<u_IN(t)<U_F

u _L(t)≈u_IN(t)−U_F for u_IN(t)>U_F  (1)

where U_F denotes the base-emitter voltage of about U_F≈0.7 V.

For the estimation of the efficiency of such a push-pull amplifier, the base-emitter voltage is neglected. The output voltage then is equal to the input voltage, i.e., u_L(t)≅=u_IN(t).

For a sinusoidal input voltage

u _IN(t)=a₀ sin Ωt  (2)

with frequency

$\omega =\frac{2\pi }{T}\left(\mathrm{period}T\right),$

the mean power consumption P_L in R_L is given by

$\begin{array}{cc}{P}_L=\frac{1}{T}{\int }_T\frac{{u_{\mathrm{IN}}\left(t\right)}^2}{R_L}t=\frac{a_0^2}{2R_L}& \left(3\right)\end{array}$

The overall mean power P₀₀ used in R_L and the two transistors T₁ and T₂ is given by

$\begin{array}{cc}{P}_{00}=\frac{2}{T}{\int }_{T/2}\frac{V_{\mathrm{CC}}u_{\mathrm{IN}}\left(t\right)}{R_L}t=\frac{2}{\pi }\frac{V_{\mathrm{CC}}a_0}{R_L}& \left(4\right)\end{array}$

The overall efficiency η defined as the ratio of the power delivered to the load in the signal band and the overall power is obtained as

$\begin{array}{cc}\eta =\frac{\pi }{4}\frac{a_0}{V_{\mathrm{CC}}}& \left(5\right)\end{array}$

Clearly, the maximum efficiency of about η≈0.78 is reached for the maximum input voltage swing with amplitude a_o=V_CC. Note that for decreasing amplitudes a₀, the efficiency is decreasing linearly.

SUMMARY OF THE EMBODIMENTS

In accordance with an embodiment of the invention, a method of driving a floating mass transducer with an analog input signal u_IN(t) is provided. The method includes converting u_IN(t) to a binary rectangular signal u_R(t) with two levels V_CC and GND. A switching network is driven with u_R(t) so as to switch nodes N₁ and N₂ between V_CC and ground. The floating mass transducer is coupled between nodes N₁ and N₂ to a capacitor C in parallel, and further to a coil L in series.

In accordance with related embodiments of the invention, converting u_IN(t) may include ΔΣ-modulation or pulse width modulation (PWM). Driving the switching network with u_R(t) so as to switch nodes N₁ and N₂ between V_CC and ground may include connecting N₁ to ground when N₂ is connected to V_CC, and connecting N₁ to V_CC when N₂ is connected to ground. For example, N₁ may be coupled to V_CC via a PMOS-transistor T₁, N₁ may be coupled to ground via a NMOS-transistor T₂, N₂ may be coupled to Vcc via a PMOS-transistor T₃, N₂ may be coupled to ground via a NMOS-transistor T₄, and wherein u_R(t) drives T₁ and T₂, and u_R(t) drives T₃ and T₄. The power efficiency of driving the floating mass transducer may be independent of the amplitude of the analog input signal u_IN(t).

In accordance with another embodiment of the invention, a system for high efficiency vibratory acoustic stimulation is provided. The system includes a modulator having an input for receiving an analog signal u_IN(t), and providing at an output, as a function of u_IN(t), a binary rectangular signal output u_R(t) with two levels V_CC and GND. A switching network is coupled to u_R(t) so as to switch nodes N₁ and N₂ between V_CC and ground. A floating mass transducer is coupled between nodes N₁ and N₂ to a capacitor C in parallel, and further to a coil L in series.

In accordance with related embodiments of the invention, the modulator may be a ΔΣ-modulator or a pulse width modulator. The switching network may connect N₁ to ground when N₂ is connected to V_CC, and connect N₁ to V_CC when N₂ is connected to ground. For example, N₁ may be coupled to V_CC via a PMOS-transistor T₁, N₁ may be coupled to ground via a NMOS-transistor T₂, N₂ may be coupled to Vcc via a PMOS-transistor T₃, N₂ may be coupled to ground via a NMOS-transistor T₄, and wherein u_R(t) drives T₁ and T₂, and u_R(t) drives T₃ and T₄. The power efficiency of driving the floating mass transducer may be independent of the amplitude of the analog input signal u_IN(t).

BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing features of embodiments will be more readily understood by reference to the following detailed description, taken with reference to the accompanying drawings, in which:

FIG. 1 shows a system for driving an FMT that uses a push-pull emitter follower (prior art);

FIG. 2 shows a system for driving an FMT, in accordance with an embodiment of the invention; and

FIG. 3 shows a R_L, L, and C network between nodes N₁ and N₂ driven by an ideal voltage source u_E(t), in accordance with an embodiment of the invention.

DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS

A system and method for high efficiency vibratory acoustic stimulation is presented. The system, which illustratively may be used to drive a floating mass transducer, converts an analog input signal into a rectangular signal. The rectangular signal is used to drive a switching network that is further coupled to an RCL circuit including the floating mass transducer. The floating mass transducer may be employed, for example, in a middle ear implant. Details are described below.

FIG. 2 shows a class-D amplifier driving an FMT in an H-bridge configuration, in accordance with an embodiment of the invention. Class-D drivers in combination with H-bridges can be found in audio applications, e.g., Junle Pan, Libin Yao, Yong Lian, “A Sigma-Delta class-D audio power amplifier in 0.35 μm CMOS technology,” SoC Design Conference, 2008, ISOCC '08, Digital Object Identifier: 10.1109/SOCDC.2008.4815561, pp. I-5-I-8, 2008, which is hereby incorporated herein by reference in its entirety.

The system includes, without limitation, four transistors T₁, T₂, T₃, and T₄, which are operated as switches. Transistors T₁, T₂, T₃, and T₄, may be, for example, MOS transistors. Load resistor R_L representing the FMT is connected to a coil L and a capacitor C. The circuit is operated between supply voltage V_CC and ground potential GND.

The input u_IN(t) is converted to a rectangular signal u_R(t) with two levels +V_CC and GND. This may be achieved, for example, using a ΔΣ-modulator at a particular sampling rate f_s (see, for example, J. C. Candy and G. C. Temes, Oversampled Delta-Sigma Data Converters, Piscataway, N.J.: IEEE-press, 1992, which is hereby incorporated herein by reference in its entirety. The sampling rate typically is much higher than twice the bandwidth of u_IN(t). For example, if u_IN(t) is an audio signal with spectral components smaller than 20 kHz, the sampling rate typically could be f_s=1 MHz. Signal u_R(t) is a superposition of a dc-component V_CC/2, input signal u_IN(t), and a noise signal γ(t), i.e.,

u _R(t)=V_cc/2+u_IN(t)+γ(t)  (6)

Applying ΔΣ-modulation, the spectrum of γ(t) is noise shaped, i.e., the amount of noise in the signal band is very small. If a ΔΣ-modulator of 1^st order is used, the amplitudes of the noise spectrum is substantially zero at ω=0 (dc) and increasing with about +6 dB/octave within the signal band. A description of such noise spectra is given, for example, in C. M. Zierhofer, “Frequency modulation and first order delta sigma modulation: signal representation with unity weight Dirac impulses,” IEEE Sig. Proc. Lett., vol. 15, pp. 825-828, 2008, which is hereby incorporated herein by reference in its entirety.

Alternative binary representations of u_IN(t) may be, without limitation, based on Pulse Width Modulation (PWM). For PWM, u_IN(t) is represented by a train of pulses with constant amplitudes and constant rate, where the widths of the pulses are proportional to the instantaneous amplitude of u_IN(t).

The rectangular signals u_R(t) and it's inverse u_R(t) at the output the inverter are driving the switching transistors T₁, T₂, T₃, and T₄. The purpose of the transistors is to switch nodes N₁ and N₂ between the supply voltage rails. If N₁ is connected to V_CC (T₁ conductive), N₂ is connected to GND (T₄ conductive), and vice versa, if N₁ is connected to GND (T₂ conductive), N₂ is connected to V_CC (T₃ conductive). Of course, other switching networks, as known, in the art may be used to achieve this function. Assuming ideal switching performance it can be assumed that the network R_L, L, and C between nodes N₁ and N₂ is driven by an ideal voltage source u_E(t), as shown by FIG. 3, in accordance with an embodiment of the invention. Voltage u_E(t) is given by

u _E(t)=2u_IN(t)+2γ(t)  (7)

and is again rectangular with two voltage levels +V_CC and −V_CC. Because of the push-pull configuration, the dc-component of u_E(t) is zero.

For steady state sinusoidal analysis, voltages u_L(t) and u_E(t) can be represented by the complex pointers U_L(jω) and U_E(jω). A short calculation yields transfer function

$\begin{array}{cc}H\left(j\omega \right)=\frac{U_L\left(j\omega \right)}{U_E\left(j\omega \right)}=\frac{1}{1-{\omega }^2\mathrm{LC}+\mathrm{j\omega }\frac{L}{R}}\mathrm{and}\mathrm{its}\mathrm{magnitude}& \left(8\right)\\ H\left(\mathrm{j\omega }\right)=\frac{1}{\sqrt{1+{\omega }^2\left(\frac{L^2}{R^2}-2\mathrm{LC}\right)+{\omega }^4L^2C^2}}& \left(9\right)\end{array}$

H(jω) represents a low pass filter. For large frequencies, this expression is approximated by

$\begin{array}{cc}\underset{\omega \to \infty }{\lim }H\left(\mathrm{j\omega }\right)=\frac{1}{\sqrt{{\omega }^4L^2C^2}}=\frac{1}{{\omega }^2\mathrm{LC}}& \left(10\right)\end{array}$

that it is converging towards zero with −12 dB/octave. Since the noise spectrum of the input signal γ(t) is increasing with +6 dB in the signal band, the filtered noise spectrum at R_L, is decaying with −6 dB/octave.

The voltage across R_L is approximately twice the input signal without dc-component, if some residual noise is neglected, i.e.,

u _L(t)≅2u_IN(t)  (11)

Assuming ideal components L and C, the power efficiency of the circuit shown FIG. 3 theoretically is

η≅1  (12)

because R_I, is the only component that is able to absorb power. Because of (12), the power consumption almost entirely occurs within the signal band. One of the fundamental differences to the push-pull emitter follower FIG. 1 is that the efficiency is independent of the signal amplitude. It remains high even at very small amplitudes of the input which is not the case for the push-pull emitter follower.

The fact that efficiency η is almost independent from the input signal amplitude is mainly due to the passive network L and C around R_L. This can easily be understood considering the case that the network is missing, i.e., L=0 and C=0. Then u_L(t) is equal to the rectangular voltage u_E(t) as defined in (8), i.e.,

u _L(t)=u_E(t)  (13)

That is, a voltage twice the rectangular signal without dc-component applies at R_L. In this case, the overall power used in R_L is constant, and the fraction of power used within the signal band is proportional to the input signal amplitude. Thus the overall efficiency would be a function similar to (5).

The embodiments of the invention described above are intended to be merely exemplary; numerous variations and modifications will be apparent to those skilled in the art. All such variations and modifications are intended to be within the scope of the present invention.

Claims

1. A method of driving a floating mass transducer with an analog input signal uIN(t), uIN(t) being between ground and VCC, the method comprising: converting uIN(t) to a binary rectangular signal uR(t) with two levels VCC and GND;driving a switching network with uR(t) so as to switch nodes N1 and N2 between VCC and ground, wherein the floating mass transducer is coupled between nodes N1 and N2 to a capacitor C in parallel, and further to a coil L in series.

2. The method according to claim 1, wherein converting uIN(t) includes ΔΣ-modulation.

3. The method according to claim 1, wherein converting uIN(t) includes pulse width modulation.

4. The method according to claim 1, wherein driving a switching network with uR(t) so as to switch nodes N1 and N2 between VCC and ground includes connecting N1 to ground when N2 is connected to VCC, and connecting N1 to VCC when N2 is connected to ground.

5. The method according to claim 4, wherein N1 is coupled to VCC via a PMOS-transistor T1, wherein N1 is coupled to ground via a NMOS-transistor T2, wherein N2 is coupled to Vcc via a PMOS-transistor T3, wherein N2 is coupled to ground via a NMOS-transistor T4, and wherein uR(t) drives T1 and T2, and uR(t) drives T3 and T4.

6. The method according to claim 1, wherein power efficiency of driving the floating mass transducer is independent of analog input signal uIN(t).

7. A system for high efficiency vibratory acoustic stimulation, the system comprising: a modulator having an input for receiving an analog signal uIN(t), and providing at an output, as a function of uIN(t), a binary rectangular signal output uR(t) with two levels VCC and GND;a switching network coupled to uR(t) so as to switch nodes N1 and N2 between VCC and ground; anda floating mass transducer coupled between nodes N1 and N2 to a capacitor C in parallel, and further to a coil L in series.

8. The system according to claim 7, wherein the modulator is a ΔΣ-modulator.

9. The system according to claim 7, wherein the modulator is a pulse width modulator.

10. The system according to claim 7, wherein the switching network connects N1 to ground when N2 is connected to VCC, and connects N1 to VCC when N2 is connected to ground.

11. The system according to claim 10, wherein N1 is coupled to Vcc via a PMOS-transistor T1, wherein N1 is coupled to ground via a NMOS-transistor T2, wherein N2 is coupled to Vcc via a PMOS-transistor T3, wherein N2 is coupled to ground via a NMOS-transistor T4, and wherein uR(t) drives T1 and T2, and uR(t) drives T3 and T4.

12. The system according to claim 7, wherein power efficiency of driving the floating mass transducer is independent of analog input signal uIN(t).