Current driving



In all Hi-fi systems loudspeakers are voltage-driven: that is, an amplifier with low output impedance is used to keep the voltage applied to the load at a constant level. This approach has the following consequences:

As the load (the loudspeaker) is reactive and not linear, the current in the voice coil results distorted. Because the current is converted into acoustic pressure, the pressure (what we hear) is distorted as well.
There is a very strong feedback: in fact, according with Lenz law, during the operation of a loudspeaker at the ends of the voice coil there is a voltage that opposes to the cause that has generated it. Since the voice coil is connected through low resistance cables to an amplifier with low output impedance [Rout], we have a great inverse current [i'], that is a very strong feedback that brings the membrane to the zero position.
This force due to the feedback current is proportional to membrane velocity. Whereas the force of elastic spring due to suspensions is proportional to the shift from the rest point. The contemporary presence of two different forces gives non-linearity and therefore distortion.


All hi-fi systems have these problems, as all EMPLOY THIS SAME TECHNOLOGY.

To overcome these problems, we must use a different technology, that is the CURRENT DRIVING.

The current driving consists of using a current generator instead of a voltage generator. The difference respect to a voltage generator lies in the output impedance, that is ideally zero (in practice very low) in the voltage generator, while in the current generator the output impedance is ideally infinite (in practice very high).

The use of a suitable current generator has the following advantages:

The current in the voice coil of the loudspeaker can be controlled directly, avoiding the distortion due to voltage driving;
The inverse current due to feedback can be canceled, since the very high value of the output impedance inhibits inverse current circulation;
By blocking inverse current circulation, non-linearity due to the contemporary presence of different forces can be eliminated.

The best system to obtain loudspeaker current driving is to use a device that has as input a voltage and gives as output a current proportional to the input voltage: such an amplifier is called TRANSCONDUCTANCE AMPLIFIER.



The transconductance amplifier


In the transconductance amplifier the voltage is translated directly into current without unnecessary complications: a transconductance amplifier could be mono-stage and - to the limit – mono-component. Mono-component means that the signal crosses only a component, with immediate advantages over any other amplifier: in a conventional amplifier (that is in a voltage generator) the signal crosses at best 3 - 4 stages and a dozen of components. The difference with transitory signals (that is with the music) between crossing a component and crossing a dozen is enormous; the same applies to phase relationships.

The high output impedance virtually opens the driving circuit, preventing the circulation of the inverse current (due to the feedback) and drastically lowering the distortion. The feedback elimination has one more important advantage: it allows a correct management of the transitory signals (they are instead compressed and distorted by the conventional amplifiers).

Since the transconductance amplifier directly controls the current in the voice coil - current that increases with the shift of the membrane - a better dynamics is achieved respect to voltage amplifiers.

In the transconductance amplifier - thanks to the high output impedance, the current - not the voltage – is maintained constant. This involves that power increases with the rise of the load impedance and that the frequency response is therefore conditioned by impedance’s modulus. Generally the loudspeakers' impedance increases with the frequency because of the inductance of the voice coil; therefore a better power output can be achieved by increasing the frequency. On the other hand, as the loudspeakers' response is limited in frequency by inertia, current driving allows a kind of automatic compensation; in practice, an extension toward the high frequencies can be achieved.

The transconductance amplifier allows a better interfacing with the loudspeaker. With voltage driving the loudspeaker must have low impedance to obtain a good power transfer; therefore the voice coil must be wound with high section copper, with consequent elevated weight that affects the efficiency and the dynamic performances. Using current driving instead the power transfer is better with high impedance, so very light voice coils can be used, improving efficiency, dynamic behaviour and response extension toward the high frequencies: therefore a more elevated synergy between amplifier and load can be achieved.

Thanks to these characteristics, a transconductance amplifier can spur much better results than conventional amplifiers, independently from the cost: finally, it is a very more intelligent way to deal with the load, and this makes the all difference.

The current generator, although offering the best results with an optimized load, can replace the voltage generator in almost all the cases in which impedance modulus doesn't present great alterations in frequency; it should be noted that moving the loudspeakers in a listening room can result in frequency response variations much higher than those due to current driving.

The current generator, however, is not the universal amplifier; his resolution in the time dominion is much higher than conventional devices and the way it manages transitory signals is very different from conventional amplifiers; therefore it is not advisable to couple it with cheap products whose design completely lacks the time dominion: generally it is better to go for a compatibility proof.


We report some examples to underline the differences when a loudspeaker is current driven.


System impedance. At low frequencies there are 3 peaks; at higher frequencies the modulus increases due to voice coil inductance.


voltage driven


current driven


Frequency responses, at 1 meter with sinusoidal signal.

In theory, we will expect to find in the response of the system, current-driven, the same 3 peaks of the impedance at low frequency; the entity is instead very lower. The peaks appear also in the measure of the system when is voltage driven. Comparing the two frequency responses, one can note that the first and the third peak have substantially the same entity in the two cases, while the central one is around 4 dB more pronounced in the case of current driving. In substance, there is not great difference; if the measure instead than at 1 meter were taken at 3 meters (that is, in typical listening conditions), the two curves at low frequency would be substantially identical. Things are different at high frequency, where the difference is of 10 dB: whereas the response begins to lower because of the inertia, with the current driving we obtain a kind of automatic equalization, with consequent notable extension of the operating limits. Differently from what could be thought, therefore, in the examined case (dynamic woofer characterized from notable variations of impedance at low frequency) the alterations in frequency dominion interest more the middle-high range that the low one: generally, the alterations in low frequency are lower than those due to a different room positioning.

The time dominion behaviour is instead strongly different, as demonstrated by the following example (air operating woofer).


voltage driven current driven


The system, if current-driven, has a much better impulse response: impulse is narrow, is more linear and, above all, has a shorter decay. The woofer get faster (it seems like a tweeter); all this with damping factor zero.

The following example is related to a 8" wideband.

voltage driven


current driven


The loudpseaker has some irregularity in the 3 - 5 kHz region; the low extension is 50 Hz, whereas inertia not allows a complete response at higher frequencies (the speaker has a whizzer, but the moving mass is around 13 grams). When current driven, the speaker become a full range, with 20 kHz well aligned.


Conclusion 1: the current driving allows extraordinary musical performances due to his behaviour in time domain.

Conclusion 2: the current driving allows higher results respect to voltage driving, because it is possible to use also impedance variations to model total response; complexity increases, but performances too.


Vicolo T. Aspetti, 18
35100 Padova (ITALY)
Tel./Fax: +39 049 8644085

Via Bachelet, 20
Busa di Vigonza (PD)
Tel.: 049 8935095

Copyright - Ultrasound - All rights reserved - 1999-2000