A headphone Transconductance Amp for a change!

[Antoinel]

A Moving Coil Transconductance Headphone Driver

Why build another amplifier to drive your headphones? Instead, use your existing amp. But; your existing amp is preferably a Transconductance Power Amp instead of a classical Voltage Source Amp [VSA]. Please visit thread # 35 of the topic in Headphone Systems entitled "Does anyone make their own headphones?" This thread has details to bring new life and flavor to your music. Please find attached the same schematic as in thread # 35.

 

[Antoinel]

Please allow me to mull [the bolded] further and answer you at a latter time with a schematic showing voltage and current values on/at the primary winding. I recall having it to do with creating a summing junction [at inverting OA] so as to satisfy zero volts [ac and dc] at the center tap of the primary as it must be for symmetry. Also I did not want to stress the output stages of both OAs by having a too low output impedance load for them. The output load for each was 500 Ohm [transformer]// 560 Ohms;~250 Ohms. Noting also my concern that the input signal to the non-inverting OA [from CD player] may be as high as 8 Vp-p!.

Hello nereis. The choice of a 560 Ohm per your question above was to supply as much current to the CT of the output power transformer without stressing the output stages of the op amps. A compromise of sorts. The new attachement TransAmp1, is the same as shown earlier; except I added the measured voltage value [1.3 Vp-p] at the output of OA1. It is also the same value at the output of OA2; but 180 degrees out of phase. The output voltage at the secondary was 0.26 Vp-p. Thus, the step down ratio is clearly 5:1. Why these specific voltage values? They reflected the maximum value at 20 Hz before I saw a spurious oscillation riding on top of the 20 Hz signal. The output signal at 20 Hz was visually distorted [scope]. The visual distortion disappeared upon lowering the output signal to ~0.12-0.13 Vp-p. This circuit is an attenuator and not a voltage gain block. I am not sure what loop feedback [to the summing junction of OA2] means. The circuit does not have excess open loop gain to sacrifice towards this end.

The second attached schematic [TransAmp2] shows another option to use a transformer-coupled headphone amp. Straightforward, painless, and works beautifully. An automotive bridging power amp [PYRAMID PB-440X] was used to drive a 25 VA toroid power transformer [TRIAD VPT24-1040] in a step down mode [~10:1]. The undistorted signal at the secondary was 2 Vp-p in the range of 20 Hz to 20 KHz. This level is way high [overkill] for my hearing using the GRADO SR80i headphones. Frequency response was also flat. Clearly, this circuit does not have the ills of that of TransfAmp1.

 

[Antoinel]

This thread wraps up the experiments begun above using a bridge power amplifier [PYRAMID PB440X] driving a step-down toroid power transformer to create a great sounding output to GRADO SR80i headphones. The attached schematic has two Figures. Figure 1 is the same overall schematic I used above [TransAmp2], but with the following two modifications.

• The primary windings of the transformer were connected in series so as to create a center tap. Ditto for the secondary windings, and thus generated 2 a differential output signals. The shaded triangles signify the absolute phase of the input and output signals.
• The centar taps of the primary and secondary windings were connected with a 1,000 uF capacitor [note its polarity] and then referenced to ground via a 25 Ohm resistor. A kind of feedback [thus voiding the isolation between in and out] whereby the feedback loops of the power amplifiers may ensure that the differential signals at the primary and secondary windings are symmetrical. Alternatively the center tap of the primary can be grounded with the capacitor in place and with shown polarity. Ditto for the secondary center tap; but without a capacitor as there is no DC voltage on it.

Figure 2 shows the connection of the headphones to the secondary windings of two transformers [in stereo]. It works great. But notice that the common lead of the headphones [green] shorts out 1/2 the differential outputs of the secondaries. It also generates an imbalance [signal] at the center taps of the secondary where there should none. This signal is further communicated to the primary and the power amplifiers. Clearly, the common lead [greens] of the headphones must be electrically separated into one for the R phone and another for the L Phone to fully benefit from the differential outputs at the secondaries.

The GRADO headphones sounded great. The amp delivered ample bass energy, level, and frequency response.

The Helmet Headphones [many threads ago, 8 ohms per side] were plugged in instead of GRADO. The 8 Ohm non-inductive resistor{s} across the secondary [Fig. 1] were removed. The return ground leads of the helmet headphones [green] are already separated. It sounded different and acceptable. The power output at the secondary was plenty to drive these speakers to loud levels.

This headphone amp is powerful and detailed. It is easy to assemble, experiment with, and use to reproduce music.

I will continue to nurture this topic of "A headphone Transconductance amp for a change !".