diyAudio (
-   Headphone Systems (
-   -   A headphone Transconductance Amp for a change! (

Antoinel 5th February 2012 12:08 AM

A headphone Transconductance Amp for a change!
2 Attachment(s)
Attachment 264309

Attachment 264310

A headphone Transconductance Amp for a change!
The attached schematic shows one channel of an OP Amp driving a buffer pair of complementary output transistors (TIP 120, and TIP 125 Darlingtons). This output stage is operating in the common emitter rather than the customary common collector configuration. Thus, it is an inverting amplifier with voltage gain if desired. This transconductance or current source amp requires 2 separate power supplies for proper operation. The schematic shows one +/- 6V supply to the OP Amp and another higher voltage +/-15 V for the output stage. The output of this amplifier is the common port or center point of the +/- 15V supply. Thus the headphones are connected between this port and ground. By ground I mean the center point of the +/-6V supply. The prototype was assembled on an experimenter's proto board. The attached images are not pretty; but they confirmed its feasibility. One image is a top view and the second is taken from the side of the volume control. The components are discernible. The above power supplies I used happen to be (ginormous) surplus from spent computers.

This amp is capable of a high outupt current into headphones. Note the resistor of 220 K which attenuates the input signal so as protect one's hearing. The quiescent conditions of the amp are:
  • 10 mV dc offset at output port
  • 5 mVpp noise/hum at output port
  • Output transistors idle at ~23 mA each
  • Op Amp is RC 4560 which sounded slightly better than OPA2134
  • Must have output Zobel, a 10 pF ceramic cap (between inverting input and output of Op Amp) and the 100 K feedback resistors to prevent oscillation
  • Headphone was Grado SR 80 connected in parallel (mono). It must be connected before turn on so as to complete the circuit of the output stage.
Two independent power supplies of +/-15 V and one common lower voltage +/-6-9 V for the OP Amps are needed to assemble a headphone stereo transconductance amp. Clearly, these power supplies can be compact in size.

I believe DIYers (and I) will run with this "square one" amplifier and tweak it so as assemble in the longer range the best there will be.

kevinkr 5th February 2012 10:29 PM

:cop: Post moved to new thread, please do not post unrelated projects to the O2 thread.

Note that the link to the file which is presumably a schematic is broken.

Antoinel 5th February 2012 11:20 PM

Please help. I am not able to upload a pdf of an important schematic which goes with images. Thank you

Antoinel 5th February 2012 11:24 PM

A headphone Transconductance amp for a change!
1 Attachment(s)
Here is the schematic for the headphone transconductance amp which goes with the images shown 2 threads earlier.

kevinkr 5th February 2012 11:24 PM


Originally Posted by Antoinel (
Please help. I am not able to upload a pdf of an important schematic which goes with images. Thank you

Have you checked to see whether the pdf is too large? Break it into a number of smaller pdfs - if converting from jpg reduce the color depth or even make it black and white, increase the compression levels before converting to pdf.

Edit: I see you succeeded. :D

Antoinel 5th February 2012 11:27 PM

kevinkr. Thanks for the info. I was able to upload it in a separate thread. Best regards.

Antoinel 9th February 2012 01:45 AM

It is a fact of life. Some of us DIYers are averse to using IC Op Amps in the path of a music signal. Our fellow DYIers have a highly critical and discriminating sense of hearing. One fall back to avoiding such signal IC Op Amps is to use Op Amps which are assembled from discrete components. A power amp is an example of a discrete albeit powerful Op Amp which has the capability to deliver both a high voltage and a high current into a loudspeaker (power). Music signals pass right through and out of it! A second fall back is to use IC Op Amps in a role which "supports" an amplifier which is neither an IC nor a discrete Op Amp. An example is a servo DC amp which is used around the signal amplifier so as to primarily remove a DC offset from its output port; for example as it drives headphones. This supportive IC Op Amp may also need to be a high performer e.g. OPA2134 or RC4560. The signal amplifier needs the highest quality support it can get. The last fall back is not to use either an IC or a discrete Op Amp at all. This approach presents its own design challenges and trade offs.

In the preceeding thread(s) , I have shown the schematic of a working Headphone TransConductance Amplifier (HTCA) which used a high quality IC Op Amp in the path of the musical path. Some fellow DYIers may have tuned out in a spontaneous gesture of disapproval! In an upcoming thread, I will show two schematics of a Class A (HTCA) which is based on a discrete Diamond Buffer for processing music signal and also embodies:
  • A high quality IC Op Amp as a DC servo Amp to null DC offset voltage at its output
  • No IC or discrete Op Amp in any role therein
Thereafter I'll compare the attritubes and ills (perceived or not) of the 3 approaches which were used.

Antoinel 9th February 2012 07:40 PM

A headphone Transconductance amp based on a Diamond Buffer
1 Attachment(s)
The attached pdf: Headphone 003 is a schematic of a refined Headphone Transconductance amp. It does not utilize IC OP Amps in the path of the music signal and around it. It is based on the Diamond Buffer; popularized in its hay day by National Semiconductor. The input signal Vi is presented to the bases of complemetary symmetry transistors 2N3440 (NPN) and 2N5415 (PNP). These transistors are working in the common collector configuation; meaning the buffered signal is extracted from their emitter ports. Lets focus on the positive half of the circuit. This begins with 2N5415 (PNP) and is traced all the way up to the +15 V supply rail. The discussion applies fully to the bottom half which begins with 2N3440 (NPN) and is traced to the -15 V supply rail due to complementary symmetry. Notice that I labelled only the values of the circuit elements for the upper half. And I labelled the equivalent circuit elements of the bottom half with the quiescent voltage drops and currents passing through them. These quiescent conditions for the upper half are identical to the lower half due to complementary symmetry. This gives the broadest picture of the schematic

In the ideal world of semiconductors, the front end transistors are highly matched. Thus, the base current entering the NPN transistor is equal in value to that leaving the base of the PNP complement. They cancel to a net zero current passing through the input resistor (47K) from the ground node. This resistor can be as high as 1Meg, and thus there is no input offset voltage. In the real world; the base currents of the input transistors are mismatched, and did cause a non-zero input offset voltage in the actual circuit (+0.712 V). The circuit shown in the bottom left of the schematic is used to null as best as possible this input offset, and it simulataneously nulled the voltage offset at the output node of the amplifier. Piece of cake to implement! Still I used a DC coupling capacitor in the input circuit to block the residual offset voltage (- 26 mV) so as to protect the output source of the input signal.

The upper N-channel JFET (2N3819) sources or (sinks) a constant current (~3.1 mA). This current flows through the red LED to cause a voltage drop of (1.875V), and then through the Si diode to cause an additional voltage drop of (0.68 V), and then through the body of 2N5415 to cause an additional voltage drop Vbe of (0.68 V). The constant current finally exits the 2N5415 transistor out of the collector (mostly) and base (trivial) ports. The sum of the above diode voltage drops is equal to (3.24 V). This is the bias voltage for the NPN Darlington Transistor TIP 120 and its emitter resistor (30 Ohm). The calculated emitter current is equal to 65 mA. This transistor is working in Class A and in the common emitter configuration. Because a lot of current is flowing through it and its emitter port goes to ground via the reistor (30 Ohm). This NPN TIP 120 is powered by a separate +6 V supply; modelled as a battery. In its operation, its emitter current Ie is equal to its collector current Ic; by neglecting the trivial current entering the base port and adding to Ie. So, current Ic begins to flow from the positive terminal of the upper 6 V battery, enters the collector port, passes through the physical body of the transistor, exits the emitter port, passes through the resistor (30 Ohm), does not stop at ground, but continues its journey through the load resistor (15 Ohm = headphones) and finally reaches the negative terminal of the upper 6 V battery where it belongs (clearly exhausted after all this work).

The impedance at the output port [Vo] is high because of the opposed collector connection of the TIP transistors. The batteries are dead shorts for a music signal. This output impedance can be several hundred Ohms or more. The "product" of this output stage is current and not voltage. Thus it is called a current source amp or a transconductance amp.

I chose a +/-6 V power supply for the output stage because it is fully adequate. The oscilloscope trace registered at best 200 mV peak to peak music signals across the 15 Ohm headphones I used. Protect your ears.

Antoinel 11th February 2012 06:19 PM

A headphone Transconductance Amp based on a Diamond Buffer
2 Attachment(s)
It is important to determine the DC voltage offsets which appear at both the input [Vi] and output [Vo] ports of the amplifier. These voltage offsets are a direct consequence of mismatches in the characteristics of the complementary devices and differences in the voltage drops across diodes and resistors used therein the circuit. An excessive level of DC voltage at the output relative to ground and across the headphone may damage it and/or move its diaphragm off its center or midpoint. This limits the excursion of the diaphragm one way or the other so as to possibly cause distortion. At the input port, one needs to know the polarity of the voltage relative to ground so as to properly connect a DC blocking electrolytic capacitor (if needed) between the input port and the music signal source.

The attached schematic PDF Headphone004 shows the circuit which was used to determine the DC voltage offsets at [Vi] and [Vo]. Note the following:
  • The triangle symbolizes the amplifier circuit so as to simplify the schematic and thus make it clear (less congested). This triangle implicitely includes all of the circuit elements which were used in PDF Headphone003 (previous thread) like the transistors, resistors, diodes and power supplies.
  • Use a ZOBEL at [Vo] to minimize ultrasonic broadband noise.
  • The switch at the input port [Vi] presents two options. The OFF position connects the variable input resistor [R] to ground. The value of this input resistor is the Input Impedance which the amplifier presents to the music signal source. The ON position of the switch connects the input resistor [R] to the outupt node [Vo]. The Input Impedance of the amplifier is still equal to the numerical value of [R].
The attached Word document Headphone004A shows the influence of the value of the input resisistor [R] on the the DC voltage offsets at [Vi] and [Vo]. Table 1 shows the results with the input switch in the OFF position. My results may differ from those a DIYer will get due to differences in the specific mismatches of our actual circuits. BUT; the trend of [Vo] as a function of [Vi] or its reverse must be the same. As [Vi] becomes more positive in value relative to ground, [Vo] becomes more negative in value relative to ground. If I chose a tolerable DC voltage offset at [Vo = 50 mV] to protect the headphones, then the Input Impedance of the amplifier is [R=6.8 K Ohm]. It is agreeable to sources of music signals which are characterized as Low Output Impedance.

Table 2 shows the DC voltage offset results with the input switch in the ON position. The trend of [Vo] as a function of [Vi] is exactly like shown in Table 1. The Input Impedance of the amplifier is now [R = 15 K Ohm] which gives a tolerable [Vo ~ 50 mV] offset. The connection of [R] between the input and output ports gives its circuit a weak DC Servo reaction; because the amplifier is inherently phase inverting. Resistor [R] appears as a feedback resistor; clearly from [Vo] back to [Vi]. However, the DC servo reaction and this global loop feedback are trivial compared with those enabled by an Operational Amplifier for comparison. The reason is simple, the indicated transconductance amp does not have an open loop (R = infinite, input switch is off] or a closed loop [input switch is on] voltage gain to speak of; like the many hundreds of an Op Amp.

I prefer the circuit with the input switch ON and thus R = 15 K Ohm [Table 2]. It has the mild protection of the DC servo seesaw effect.

Listening to the amplifier. I have a SONY CDP-C515 compact disc player. It has a remote volume control for its output RCA port and simultaneously for its internal headphone output. The output RCA ports need to look into an input source impedance which is greater than 10 K Ohm; which is the case here. I used either the Right Or the Left RCA outputs. Merging the 2 output RCA ports with a Y connector to make a mono signal is doable; but it did not sound well. The headphones were Grado SR-80 connected in mono (R = 16 Ohm). They sounded detailed, and tonally balanced.

The headphone output on the SONY is suspected to be driven by an IC Op Amp working as buffer of a low output impedance. Ironically, I used it as the model for the low impedance source of the music signal to drive the headphone transconductance amplifier. I used a conversion adapter from a stereo 1/4 inch plug to a stereo female RCA. Like above, I used either R or L ports, and did not merge the ports via a Y connector to make a mono signal. Again, the headphones sounded detailed and great. Maybe slightly different in tonal balance relative to the above case.

Antoinel 16th February 2012 08:17 PM

Simplified the headphone Transconductance Amp.
1 Attachment(s)
There is French proverb which translates to: "simplicity makes beauty". The simplified transconductance amp is shown in PDF Headphone005. Three objectives were sought and met. First, the input complementary symmetry transistors 2N3440 (NPN) and 2N5415 were removed. They were replaced by the 2 diodes labelled 1N914* so as to establish the original bias of the output transistors. Second, the input resistor was raised to 47K; why dump valuable input signal to ground by using a smaller one like 15 K?. This resistor can be either grounded or connected to the output port Vo. This latter connection is a mild DC servo and is preferred. It maintains the DC offset at the output port ~ zero volts. Thirdly, the resultant DC offset at Vo slowly cycled over several minutes (due to servo action) between + 15 mV and -15 mV. Note the resistor (267 K) which is connected between minus 15 V and the input port Vi. Without it, the DC offset at Vo before servo action was about minus 1.3 VDC; this is unacceptibly high. With it and without DC servo (i.e. 47 K grounded) the offset at Vo came down to ~50 mV. The value of this resistor was arrived at by trial and error. Start with 1 Meg Ohm and lower its value by sequentially using 680K, 470K, 330K etc until a target 50 mVDC is obtained at Vo. I show a connection to minus 15 VDC. In the hands of another DIYer the connection may need to be to plus 15 VDC instead; because of different element mismatches in our respective prototypes. This modified amp sounded more relaxed than earlier versions.

All times are GMT. The time now is 10:05 PM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2