The Psychoacoustic Bass Enhancer.

by Jan Meier

meier5_0.gif

[Editor: The author has applied for a patent in Germany for the invention described in this article. The German Patent Application number is 10119094.8. Individuals are authorized to make this device for their own personal use, but must obtain permission from Jan Meier for commercial applications.]

Our senses, especially our eyes and ears, are remarkably precise instruments. We can distinguish the slightest gradations in intensity, color, frequency, etc. Nonetheless, eyes and ears are also easily fooled. For instance, in well known optical experiments straight lines may look bent or equally long lines seem to have different lengths These experiments tell us, that the brain processes involved in perception also play an important role in the way we see and hear our world.

The study of the physiological and mental processes of hearing is called psychoacoustics. Principles of psychoacoustics are widely used in audio technology. An example of psychoacoustic processing is data compression with MPEG-3, which removes information from the signal without (or almost without) affecting sound quality. Another example is the loudness button on many amplifiers that compensates for the reduced sensitivity of the ear for the highest and lowest frequencies at low sound pressures.

In this article a device is presented that makes signals below 60 Hz audible in loudspeakers and headphones that normally, by their mere physical construction, are not able to reproduce these frequencies. The device combines two psychoacoustic phenomena.

BACKGROUND

The principle of the missing fundamental:

The sound of a single note of a music instrument is the summation of its fundamental tone (say 200 Hz) and a number of harmonics (400, 600, 800, 1000, …. Hz). If we electronically remove the 200 Hz fundamental tone our ear only hears the harmonic frequencies at 400 Hz and higher. Nonetheless our brain tells us that the pitch of the note is 200 Hz. Since the 600 and 1000 Hz frequency components are no fundamentals of the lowest (400 Hz) frequency component present, our brain knows that something is missing and “adds” an imaginary fundamental tone of 200 Hz. However, the “color” of the note is lighter when the fundamental tone is missing.

Mechanical harmonic distortion in the inner ear:

A pure 200 Hz sine wave not only makes the basillary membrane inside the ear vibrate at 200 Hz, but also at 400, 600, 800, 1000, 1200…. Hz. For sine waves between 200 and 3000 Hz these overtones have amplitudes of 33%, 13%, 6%, 4%, 2%, …. of the amplitude of the fundamental. Harmonic distortion in the inner-ear thus sums up to approximately 60%!

Nonetheless, we only hear a pure sine-wave at the fundamental frequency since our brain has learned that this specific frequency spectrum belongs to a pure tone.

To my knowledge, nobody has ever investigated the frequency spectrum of the basillary membrane at very low frequencies. However, the decreasing sensitivity of the ear indicates that the membrane is relatively “stiff” for this frequency range. Therefore it can be expected that the relative amount of overtones at the lowest frequencies strongly increases. This phenomenon is seen with many music instruments. A “cheap” piano does not really produce a 27.5 Hz tone when one strikes the lowest key, because the resonance board is simply too small to swing at 27.5 Hz, but it produces a tone that appears to be 27.5Hz due to the principle of the missing fundamental.

At their lowest frequencies corpses are more ready to vibrate at the overtones than at the fundamental. My guess was that in the inner ear the information on the lowest tones thus will be merely transmitted by the overtones produced.

Most loudspeakers and headphones are not able to make the air move at frequencies between 20 and 50 Hz and therefore these frequencies will not be heard. However, if we electronically create harmonics of these lowest tones and add these signals to the original audio-signal, we suddenly will hear the low fundamentals, due to the principal of the missing fundamental. Moreover, my speculation was, that if the spectrum of these overtones was chosen so as to create an energy spectrum on the basillary membrane that, except for the fundamental tone, resembles that of a pure sine wave, then we will hear something that is very close to this sine tone. This idea is illustrated in figure 1.

meier5_1.gif
Figure 1

The most accurate way to generate a spectrum of overtones is to have the audio signal analyzed and frequency components added by using the technique of Fourier analysis. However, this requires quite a lot of computational power and off-line evaluation before the music can be played. For a more practical use, I wanted an analogue, real-time solution.

One analogue solution to create overtones already exists. The Philips company uses it for portable equipment (boomboxes and the like) and calls it Ultrabass. Basically this solution looks at the low frequency content of the signal and restarts a continuously increasing triangular signal at each zero-crossing of the signal.

meier5_2.gif
Figure 2

The resulting triangular waveform has a fundamental frequency twice that of the original waveform (see figure 2) and is added to the original signal. Ultrabass has a number of disadvantages that makes it hard to call it HiFi:

  • With a sinusoidal input signal, only 2nd, 4th, 6th, …. (even order) harmonics are created, because UltraBass generates even and odd  harmonics from a sawtooth waveform, whose basic frequency is twice the frequency of the original input signal.So f(1) = 2 x f(0)The harmonics are even harmonics of the original input signal

    f(2) = 2 x f(1) = 4 x f(0)

    f(3) = 3 x f(1) = 6 x f(0)

    f(4) = 4 x f(1) = 8 x f(0)

    f(5) = 5 x f(1) = 10 x f(0)
    …..

    However, to “hear” the missing fundamental we also need the 3rd, 5th, …. (odd order) harmonics.

  • Short term variations (cycle-to-cycle) in signal amplitude are not reproduced in the Ultrabass signal, because the amplitude of the Ultrabass signal is set by its envelope and only is allowed to vary slowly. These variations are very important to our ears for recognizing signals as being from a non-artificial origin. (Figure 2a)
  • Short term variations in cycle length are reflected in the Ultrabass signal but the longer cycles also have larger amplitudes. (Figure 2b) This is unnatural. Short term variations are very important to recognize an instrument as not being artificial. However, adding short term variations that are not there in reality will change our perception of the instrument.
  • With a mixture of two signals of different frequencies the Ultrabass signal merely represents the strongest signal. It is the strongest signal that sets the number of zero crossings in the Ultrabass harmonics generator. However, the presence of the second signal results in a number of aberrations. One is shown in figure 2c.

meier5_3.gif
Figure 3

Those interested in more details on Ultrabass can take a look at the US patent no. 6111960 (“Circuit, Audio System and Method for Processing Signals, and a Harmonics Generator” by Aarts et al.).

I decided to design my own solution that had to fulfill the following requirements:

  • all lower harmonics (2nd, 3rd, 4th, 5th, 6th, 7th, …..) should be calculated.
  • The amplitude of the harmonics should decrease with increasing order. The rate of decrease should be adjustable (I did not have any indication yet, what would be the most suitable ratios of the various amplitudes so making it adjustable allowed me to experiment).
  • Cycle-to-cycle variations in signal amplitude and cycle length should be properly reproduced in the calculated harmonics.
  • With a mix of two signals, harmonics of both fundamentals should be calculated.

Such a design is not very straightforward. I have tried several solutions – none being ideal – but the one presented in this article is close to optimal and still relatively easy to build. The basic schematics are shown in Fig 4.

THE CIRCUIT

meier5_4.gif
Figure 4

The input signal, Vin(t), is buffered by two parallel buffer stages. Input buffer 1 leaves the input signal unaltered. Input buffer 2 filters the signal and removes the high frequency (and the very low frequency) components.

A harmonics generator takes the filtered buffer signal, V’in(t), and uses it as input for a high order mathematical function.

Vgen (t) = a2.Vin2(t) + a3.Vin3(t) + a4.Vin4(t) + ……….

It is a multiplication series of the form: f(out) = a.f(in) + b.f(in).f(in) + c.f(in).f(in).f(in) + ….

If the filtered input signal is a pure sine wave, this procedure generates all the major harmonics. For example the square of a sinusoidal signal generates a signal with the double frequency (plus an offset):

sin(2p.f.t) = 0.5 (1 – cos(4p.f.t))

If the amplitude of a signal decreases by a factor two, then the squared signal would decrease by a factor four! Thus, weak signals that pass through the generator would not be heard and strong signals would be “amplified” too strongly. To correct for this phenomenon a separate circuit continuously monitors the signal envelope and adds this information to the harmonics generator. The generator corrects the amplitude of its output signal accordingly.

The last stage of the device is an adder to sum the original input signal, the generator signal, and (for offset correction reasons) the envelope signal.

meier5_5a.gif
meier5_5b.gif
Figure 5

Figure 5 shows the schematics in detail.

Input buffer 1:

This input buffer is inverting. Since the adder is also inverting, the total signal path is non-inverting

Input buffer 2:

meier5_6.gif
Figure 6

This input stage combines a 1st order high-pass filter with a second order low-pass filter. The filter characteristics can be changed by adjusting the resistance and the capacitance inside the feedback loop. Figure 6 shows the filter characteristics for the various settings.

The envelope calculator:

The envelope signal, A(t), should increase instantaneously at a sudden increase of the amplitude of the input signal and slowly decrease after a single wave. This is achieved by having two opamps charging a 47uF capacitor as soon as the amplitude of the input signal is higher than the voltage across this capacitor. One opamps charges at the large positive signal amplitudes, the other inverts the signal and charges at the large negative signal amplitudes. Two 22 Ohm resistors limit the envelope signal A(t) at the capacitor to frequencies up to 150 Hz. The capacitor slowly discharges via three 12 kOhm resistors that connect the envelope calculator and the harmonics generator.

The harmonics generator:

Central to the harmonics generator is an analogue current multiplier/divider RC4200. This IC has three inputs I1, I2, I4 that are actively kept at ground potential and one output I3. The input and output currents are allowed positive values only and are related according to:

I3 = (I1 . I2) / I4

Basically, with the harmonics generator the input signals are given by

I1 = A(t)/12000 +V’in(t)/2200

I2 = A(t)/12000 – V’in(t)/2200  (the filtered buffer signal has passed a converter)

I4 = A(t)/12000

For a sinusoidal input signal the envelope signal is approximately 5.7 times (amplification factor of the envelope converter) the sine amplitude.

V’in (t) = K.sin(2p.f.t)

A(t) ~ 5.7 K

I3 ~ (K/2100 + K.sin(2p.f.t)/2200)(K/2100 – K.sin(2p.f.t)/2200)/(K/2100) = K/2100 (1 – 0.91.sin2(2p.f.t))

The output signal of the RC4200 thus is a cosine signal with twice the frequency of the input sine wave plus an offset. The amplitude of the cosine and the offset are proportional to the amplitude of the input signal.

The ratio of the amplitudes of the envelope signal and the input signals has been taken with care to create an offset in the output signal that is slightly higher than the time varying signal part. Thus it is made sure that all currents are positive.

The envelope generator is connected via a 1M ohm resistor to the 12 Volts power supply. This guarantees that I1, I2 and I4 have small positive values, even when the input signal is zero. If the input signals are zero the RC4200 becomes instable.

The output signal of the RC4200 could well be used as input signal for another RC4200 to multiply it with the original sinus signal. Thus signal components with three times the frequency of the input signal can be calculated. However, to continue this way to construct the higher order harmonics is quite cumbersome. Moreover, the RC4200 is an expensive part, so using a single RC4200 for each higher order harmonics would become quite expensive. I therefore used a trick and connected the output signal of the RC4200 (after buffering with a opamp) to its own two inputs by means of variable resistors Rf. Thus a feedback is created that generates the higher harmonics. The feedback resistors Rf can be varied to create different harmonic spectra (a higher Rf produces more harmonics).

meier5_7a.gif
Figure 7a

Figure 7a shows the output signals and their amplitude spectra for a sinusoidal input signal with the three resistor value settings as shown in the schematics. They are very similar (figure 7b shows the three outputs from 7a close to each other for comparison). The differences are more easy to discern sonically and in the frequency spectra.

meier5_7b.gif
Figure 7b

The higher harmonics created are not expected to contribute strongly to the impression of deep bass tones. At the contrary, having many overtones could make the sound rather “sharp”. Therefore the buffer at the output of the RC4200 constitutes a low-pass filter with an upper frequency of 340 Hz.

With no signal at the input the output signal of the RC4200 is very low. The sudden onset of an input signal instantaneously increases the output signal to a specific mean value. This signal jump is corrected by subtracting (half of) the envelope signal from the output of the RC4200. The subtractor for this purpose also constitutes a low-pass filter with an upper frequency of 340 Hz.

The output of the harmonics generator is added to the original input signal via a (linear) potentiometer that allows a continuous control of its amplitude.

CONSTRUCTION

meier5_9.gif
Figure 8

People that have read my article on the Analoguer will note a similarity. The Analoguer too has an input buffer at the front and an adder at its end. This allows for an easy integration of the two systems into one device (figure 8). Somebody who already has build the Analoguer only has to add the second buffer 2, the envelope calculator, the harmonics generator and the potentiometer.

For the direct signal buffer and the adder high quality opamps are recommended. I use LM6171 forced into class-A. For the other blocks sound quality is of less importance, since our signal is artificial anyway. I used relatively cheap dual N5534 opamps, not driven into class-A.

The RC4200 is a multiplier made by Fairchild and by JRC and costs about $7 US. There are more analogue multipliers on the market but these are generally much more expensive. There are no pin-to-pin compatible substitutes, however. I used an IC socket for the RC4200, but this is simply because it was a first version that was expected to need quite some “debugging”. It allowed me to remove the chips to prevent frying them when soldering other components.

The rotary switches are made by Lorlin. I used metal-film 0.2 W no-name versions that I have on stock in abundance and some decent polyester capacitors. The potentiometer is a no-name type and should be linear taper.

meier5_11.gif

In the picture, I have two enclosures that were put on top on each other. The Analoguer circuitry and the bass-enhancer circuitry were connected by cables. The bass-enhancer was built on experimental circuit board only and looks rather messy. The lower unit is a normal Analoguer. The upper unit is the bass-enhancer. The big dial on the right sets the volume of the bass signal added (potentiometer). The left small dial sets the filter frequency of the input buffer 2. The dial at the back of the unit sets the feedback resistance in the loop of the multiplier that sets the shape of the spectrum.

Due to the larger number of ICs the current demand is rather high. The power supply should be ±15 Volts, 200 mA minimum. I used two Analoguer power supplies (one power supply for the Analoguer and one for the bass enhancer branch), but people also simply could take a larger transformer. One Analoguer power supply will power both the Analoguer and the bass-enhancer if the transformer power handling is increased to 15VA.

meier5_10.gif
Figure 9

To check the proper function of the envelope calculator and the harmonics generator not only requires the use of at least an oscilloscope and a frequency generator (or CD-player with test-tones), but also a thorough understanding of the function of the various components. I do not have any specific test procedure. Figure 9 shows an example of the output signal for various sittings of the harmonic composition switch. The output of the harmonics generator can be measured at the 4.7K ohm potentiometer.

meier5_8.gif
Figure 10

Figure 10 shows an example of the output of the envelope signal. The envelope signal can be measured at the 47uF capacitor inside the envelope calculator circuit. I simply took a variable signal at the input and looked at both the input signal and the output signal of the envelope calculator at a 2-channel oscilloscope. It immediately worked as I hoped for, and as it didn’t need any debugging I never developed any test procedure.

THE RESULTS

I recommend setting the psychoacoustic bass enhancer just by ear and good taste. No specific instructions. Figures 5 and 8 show the effects of the Filter Threshold and Harmonic Spectrum controls respectively. People should experiment with the bass enhancer. If the Analoguer and the psychoacoustic bass enhancer are built together, remember that they are independent of each other. One works at the low frequency domain, the other at the high frequency domain. The Analoguer does have its own bass enhancement section, and it can be used with the psychoacoustic bass enhancer for best effect.

The following WAV sound clip demonstrates the effect of the psychoacoustic bass enhancer. The file is mono, 8-bit resolution to keep its size limited. There is some high-frequency noise which comes from my laptop PC. I did clear up some of the noise, but wasn’t able to remove it completely. The file contains six test tones:

5 seconds 100 Hz sine wave
5 seconds 50 Hz sine wave«
5 seconds 25 Hz sine wave
5 seconds 100 Hz sine wave + bass enhancer
5 seconds 50 Hz sine wave + bass enhancer
5 seconds 25 Hz sine wave + bass enhancer

Download bass enhancer demo WAV file (73Kbytes)

The first three sections will show how the 100 Hz can be easily reproduced by most loudspeakers/headphones whereas the 50 Hz already is less strong. The 25 Hz normally is inaudible. After passing the tones through the bass enhancer with the Filter Threshold set at position 1, the 100 Hz signal has hardly changed, the 50 Hz signal has a slightly changed timbre, and the 25 Hz becomes audible!

Note 1: I used the rather cheap microphone input of my laptop to create the files. Despite filtering some “noise” is still present.

Note 2: To decrease file size the wav-file has a sampling rate of 22 kHz and a resolution of 8-bit.

My Sennheiser HD600 headphones are said to have a deep and tight bass. However, testing with a frequency generator revealed that hardly anything happens below 30~40 Hz. These phones are simply not able to reproduce these low frequencies, even with the volume cranked up. However, when I turned up the potentiometer of the bass-enhancer things changed dramatically. I even could feel and “hear” tones down to 20 Hz wobbling my eardrums! Not surprisingly these tones tend to have a relatively “light” color, but they were definitely recognized as 20 Herz tones and not 40 or 60 Hz. Frequencies between 30 and 50 Hz were reproduced remarkably well, sounding round and “weighty”.

The amount of feedback in the harmonics generator did have a notable influence on the sound. Personally I preferred the 22k ohm settings. If you do not like to add a switch with various feedback values I suggest using this value

Testing with real music first revealed that only little music has substantial frequency components below 50 Hz. Sure, certain organs go down to 16 Hz. and a piano grand goes down to 27.5 Hz but these tones are very rarely used. Actually, I’m a piano-player myself and I only know of one piece that uses the lowest octave; a piece by Bartok called “With Drums and Pipes” (fantastic music by the way, from the piano suite “Out of Doors”).

So, with most music, my impressions were not staggering at first hearing. There was only a rather subtle effect, if any at all. However, after selecting pieces that really go deep, the effect was found to be most satisfying. Suddenly I became aware of an acoustic environment in which the deeper tones developed, a texture from which the lowest frequencies evolved in a most natural way. As a result one simply gets drawn more into the musical scenery.

Of course I also tested with loudspeakers. A friend of mine has Quad ESL63 loudspeakers. Very nice indeed, but due to their working principle, little sound pressure is present below 60 Hz. Imagine how he looked when I suddenly made a 25 Hz note audible! Again tones below 60 Hz sounded very substantial, albeit a little bit light in color. Testing with music in speakers confirmed my experiences with my headphones. The effect normally is very small but if it’s there, it really can be very involving.

People might argue that the bass tones produced are artificial and never can represent real uncolored bass tones as produced by big and mighty loudspeakers. However, there are many arguments for the psychoacoustic bass enhancer. First of all, big loudspeakers do not produce a real uncolored bass. A 30 Hz tone has a fundamental wavelength of 11 meters. No such frequencies can be properly reproduced in a normal-sized living room due to cancellation effects. Reflections to walls, floor and ceiling will cancel a large part of the original sound waves. We do feel the deep bass tones since everything starts to shake, but hearing is a different thing.

Moreover, room resonances result in a very uneven frequency response and the decay time of the acoustic energy is very long. The major problems with room acoustics generally are found in the lowest bass region, not in the treble (unless you listen to music in your bathroom). The psychoacoustic bass enhancer shifts the acoustic energy into a frequency region where room acoustics are much less problematic. The music is heard and not merely felt.

In headphones, the distance between ear and driver is much smaller than the distance between ears and walls and therefore the reflected waves are much smaller. Of course there are reflections at the skin, etc., but due to the long wavelength, reflected waves and direct wave are in phase. Only at very high frequencies does this cavity influence sound characteristics by interferences.

I’m aware that my description of the circuitry is rather short and that at a first glance you will come up with many questions. That’s on purpose smile.gif . If you’re not able to answer most of these questions by yourself you will not be able to build this device.

Have fun,

c. 2002 Jan Meier.
The author’s website: Meier Audio.

The Analoguer – A Remedy for Digitalitis.

by Jan Meier

meier4_0.gif

I remember well the time when the first CD-players came on the market with specifications no other system could match. Over the years, however, both the quality-standard of my equipment and my demands on sound reproduction grew and I got more and more irritated. Although the CD outperforms the record (IMHO) by miles, I often noticed an edginess and harshness that was unnatural to my ears. It appears that I had been infected by digitalitis.

Recently I professionally got involved in the various aspects of digital recording of biomedical signals, and I started to understand that the CD is not as perfect a medium as manufacturers always claimed. Digital recording introduces a number of anomalies. Digitalitis is caused by the relatively low sampling rate of CDs, such that high frequency signals are recorded with poor resolution. The Analoguer circuit described in this article corrects this harshness of CD sound by attenuating frequencies above 15kHz with a signal processing technique called windowing. It can also be used to “tame” edgy analogue recordings.

BACKGROUND

meier4_1.gif
Figure 1

The major anomaly of the CD is introduced by the relatively low sampling frequency of 44 kHz. Although this sampling frequency allows us to record signals up to 22 kHz the upper frequencies are not very well presented. Figure 1 compares the sampling of a 2.5 kHz and of a 21 kHz waveform. After sampling of the original signals I connected the consecutive measurement-points by straight lines. The image-signal of the 1 kHz signals thus constructed equals the original signal pretty well.

The image of the 21 kHz signal, however, continuously varies in amplitude; it “wobbles”. This phenomenon can be seen with all the higher frequencies above 15 kHz and is an inherent property of the system. Although such high frequencies hardly can be heard by most people (myself, I don’t hear anything above 15.5 kHz) the continuously varying amplitudes in combination with non-linearities of our reproduction system and our ears introduce lower frequency by-products that might well be discernible.

meier4_2.gif
Figure 2

In a recording session, before a digital image of an analog audio-signal (microphone-signal, analog master-tape, …) is made, all the frequency components above 22 kHz are electronically removed with a sharp low-pass filter – a so-called brick-wall filter. Without this filtering higher frequency signals would be represented as low frequency signals and “distort” the sound. Figure 2 shows the sampling of a 41.5 kHz waveform and how its resulting image represents a 2.5 kHz signal. This effect is called aliasing.

meier4_3.gif
Figure 3

The removal of these high frequencies results in another anomaly of the CD. Figure 3a demonstrates how a square wave signal of 2 kHz is composed of an infinite series of sinus-waves with frequencies of 2, 6, 10, 14, 18, 22, 26 …. kHz.. However, if all the components above 22 kHz are removed, the resulting waveform is no longer square, but shows a high frequency ringing. In signal-analysis, it is called the Gibbs phenomenon.

meier4_4.gif
Figure 4

This phenomenon might not be a major problem as long as a continuous signal is used, but the ringing also shows with short pulses, as demonstrated in figure 4. What you can see from the picture is, that in the image of the pulse there is already a signal present before the pulse actually starts and that there is also signal left after the end of the pulse. The so-called pulse response has deteriorated considerably.

To minimize wobbling and to improve the pulse response in many (biomedical as well as other) applications not a brick-wall filter (as with CD) is used but a filter that gradually attenuates the amplitude of the higher frequencies. The second part of figure 3 shows the image of the square wave after such filtering. Ringing in the image has almost disappeared. However, the price that is paid is a less sharp transition between the two discrete signal levels.

The technique of such frequency dependent attenuation of the input signals is called windowing and, with proper filter settings, is able to eliminate ringing completely. Unfortunately, with audio-signals sampled at 44 kHz (as with the CD) these windowing settings would require strong attenuation of the signals between 5 kHz and 15 kHz and thus would affect the treble most negatively.

However, by only reducing the frequencies above 15 kHz, wobbling and ringing can already strongly be reduced without effecting the signals inside the audio-band. This is what the digital filters of the CD-players by WADIA (and some of the systems of T+A and Sony) do in the digital domain and the Digital Antidote by Taddeo in the analogue domain. Both reduce the signals in the upper frequency-band in order to improve pulse-response.

Unfortunately, I don’t have the money to buy myself a WADIA (or better said, I’m not willing to spend so much on a CD-player) and at the time when I decided to attack digitalitis I didn’t know about the Digital Antidote yet. So I had to make my own solution.

THE DESIGN

In order to achieve this goal, I needed a relatively steep low-pass filter. Standard solutions can be found in every text-book on electronics but for audio-applications standard solutions have one major drawback; they change the phase at the higher frequencies and our ear is very sensitive to phase-changes (so I’m told). I therefore decided to build a filter that does not introduce any phase changes over the audio-band. Not surprisingly (as I found out later) the solution I came up with resembles that of the Digital Antidote (well, the wheel was also invented more than once). Only the technical implementation differs.

meier4_5.gif
Figure 5

The higher frequencies are removed by adding a small frequency independent time-delay, Tdelay, to the audio signal and adding this delayed signal to the original input signal. After division by a factor of 2, we have the average of the input and the delayed signals. For low-frequency signals this small delay has little effect and the average (output) signal nearly equals the original input signal. For higher frequencies, however, the delayed and the original signals are no longer in phase and the amplitude of the average signal is much lower than that of the original input. This is demonstrated in figure 5.

It can be easily shown that the ratio of the amplitudes of the average (output) and of the input signal is given by:

        • A(f) = cos ( pi*f*T

      delay

         )

The amplification factor stays close to 1 over a long frequency range and next drops to zero relatively fast.

The total delay of the output signal (as referenced to the input signal) is Tdelay/2 and is frequency-independent. Thus no phase-shifts are present between the various frequency components in the output signal.

meier4_6.gif
Figure 6

A block-diagram of the required circuit is shown in figure 6. First the input signal is buffered. Next the signal is time-delayed and in a third step both input and delayed signals are added (and buffered) by an adder.

meier4_7.gif
Figure 7

The buffer and the adder are relatively easy to build using opamps. Standard solutions can be found in any textbook. However, since inverting adders are intrinsically more stable then non-inverting adders (one of the inputs of the opamp is directly connected to ground which minimizes problems with DC-drift) I decided to use both an inverting buffer and an inverting buffer. The output signal thus is non-inverted (figure 7).

Calculations showed that the delay-circuitry (called an “all-pass filter”) should be able to delay the input signal up to (approximately) 18 microseconds over the complete frequency band from 0 to 22 kHz. There exist solutions that achieve such a delay using a single opamp. The Digital Antidote (US Patent No. 5436882) for example uses such an approach.

These systems, so-called second-order, all-pass filters, use the intrinsic system-resonances at higher frequencies to broaden their bandwidth. To use resonances, however, in my understanding also implies an impairment of the pulse-response. I never did, however, a thorough analysis and testing of these circuits for my application.

I decided to use first order all-pass filters instead. These filters are only able to delay the signal up to 10 microseconds (over the band-width required) and therefore two of these circuitries have to be placed in series to achieve the required overall delay of 18 microseconds. The price that we have to pay is that two opamps are to be used instead of one.

meier4_8.gif
Figure 8

A basic first order all-pass filter is shown in figure 8. Simple mathematics show that this filter has a gain factor:

    A(f) = ( R – 1/(2*pi*f*C ) / ( R + 1/(2*pi*f*C )

The absolute gain is 1 for all frequencies but there is a frequency-dependent phase shift that results in a constant time-delay at the lower frequencies of 2RC. Placing two equal filters in series the time shift Tdelay becomes 4RC.

By interchanging the resistor and the capacitor, each all-pass filter not only delays the signal but also inverts it. A signal inversion is unwanted, but this is automatically corrected using two filters in series. The advantage of interchanging these components is, that the non-inverting input of each opamp is DC-coupled to ground which (again) increases stability and minimizes problems with DC-shifts.

At frequencies far beyond the audio range, the phase shift of the delayed signal is maximally 360 degrees, which means that the input and the delayed signals are in-phase again. This implies that these very high frequencies are not attenuated by the system. However, although there are no audio signals in this range, most CD-players produce various high-frequency “noise”-components (quantisation noise, RF-interferences from the microprocessors in the system, etc.) that are not heard but still make life hard for our amplifiers. To eliminate these components I placed capacitors in parallel to the feedback resistors of both the buffer and the adder. This results in an additional second order low-pass filtering with a filter-frequency of 72 kHz, which is beyond the audio range, and therefore has no effect on the sound.

As for the exact delay time we have to compromise. A larger delay not only results in a stronger reduction of the anomalies but also in a stronger reduction of the bandwidth at the higher frequencies. A trade-off has to be made and the optimal value might well depend on your ears, your equipment, and your taste. I therefore decided to make the filter settings variable. The time delay can be changed with a switch that changes the resistances in the delay line. Five different settings are provided.

CONSTRUCTION

meier4_9a.gif
Figure 9a

meier4_9b.gif
Figure 9b

The complete schematics of the filter are shown in figures 9a and 9b. The opamps used are the LM6171 from National Semiconductor. By additionally connecting the output of each opamp via a resistance to one of the power-rails the output stage of each amp is forced into class A functionality. The power supply has a ground loop breaker, so the audio inputs and outputs MUST have floating grounds – their grounds cannot be directly connected to the enclosure. (See A Precision Preamplifier-Power Amplifier System with Natural Crossfeed Processing for more discussion about biasing opamps to function in class A and ground-breakers in power supplies).

meier4_10.gif
Figure 10

The frequency characteristics of the various filter settings are shown in figure 10. Setting 1 only attenuates the very high frequencies and can be used to reduce the anomalies of systems with a high sampling rate (DVD-Audio or SACD). The other settings are meant for CD, DAT, minidisc, DCC, etc.

meier4_11.gif
Figure 11

With little effort I was also able to incorporate a bass-enhancement similar to the one implemented in the headphone amplifier. It electronically increases the lower frequencies to compensate for the natural decrease of bass-response that every acoustic driver has. The frequency characteristics of the bass-enhancement are shown in figure 11.

meier4_12.gif

The Analoguer is placed inside a sturdy aluminum case and only parts of premium quality are used (a PC board with a 70 micrometer copper-layer, 1% metal-film resistors, polystyrol and polycarbonate film capacitors, LM6171 opamps, torroidal transformer, heavy-duty silver-plated switches, gold-plated jackets, etc.).

meier4_13.gif

The filter is pretty straighforward to make but note that quick and dirty solutions using cheap parts and a non-optimal mechanical construction (bread-board, no shielding) tend to degrade sound quality. The gains might well be counterbalanced by the losses.

Setting the Analoguer

The optimal settings for removal of the higher frequencies depends on various factors such as the listening environment, the characteristics of your loudspeakers and, above all, your personal upper limit of hearing (which may vary from 14 to 20 kHz). Therefore the filter function of the Analoguer can be adjusted by the filter switch.

The filtering effect at the first position is very weak and is specially intended for SACD and DVD-A. Due to a higher sampling frequency, the anomalies in these systems are less pronounced and require less reduction of the higher frequencies.

It is recommended that you start listening with the right switch in its middle position and leave it there for several hours of listening. The effect of the Analoguer is subtle, and few people will notice an immediate effect. However, after a few hours you will notice that listening has become more relaxed and less strenuous. If after a while you find some of the upper frequencies missing, reduce the filter action by one step and again spend several hours of listening before you make any further adjustments. If, on the other hand, you don’t notice the absence of upper frequencies, increase the filter action by one step. Again, please wait before making further adjustments. The effect of the Analoguer is most readily heard and felt with music that has a large share of overtones. The female voice, as well as violins, oboes, etc. are well suited for finding your personal optimal settings.

Bass-enhancement

Some headphones and smaller loudspeakers have a restricted bass response. To compensate for this natural loss of the lower frequencies, the Analoguer can electronically enhance the bas signals. The bass-enhancement should be used with care. The lowest frequencies are amplified by 10 dB and the power that your amplifier must deliver increases accordingly. If your music starts to sound harsh, please switch off the bass-enhancement or reduce the sound level immediately. The bass-enhancement switch has five positions. In the first position the enhancement is switched off.

For loudspeakers, The benefits of bass-enhancement strongly depends on the type of loudspeaker. Some speakers, especially bass-reflex systems, have a fast roll-off at lower frequencies that can not be properly compensated for by the Analoguer. Loudspeakers with a closed cabinet tend to roll-off more gently and generally will benefit more from the bass-enhancement. Enhancement of the lower frequencies may also increase colourations and make the sound muddy. Always be aware that sonically less can be more. With bass-heavy music at high sound levels distortion can be produced that will damage your loudspeakers.

The Results

In a direct A-B comparison, using medium filter settings and bass-heavy music no evident differences can be discerned. However, using music with a substantial share of upper frequencies (soprano, hobo, upper strings) one notices that the sounds gets less brittle and that the harshness at the treble has gone. The sound simply becomes more relaxed. I’m now able to listen to music at a much louder sound-level than I did before without getting annoyed. This, for me, is the best proof that I’m cured from digitalitis and that is just what the filter is supposed to do. Sound has a more analogue quality (in the best sense of the word) and that’s why I called this device “Analoguer”.

I want to emphasize that the effect of the filter is very subtle. Most people probably will hear the effect more noticeably with their headphones since these are normally more revealing. It will not make a Wadia out of your Samsung CD-player or make your $ 200,- HiFi sound like something state-of-the-art. If you use cheap equipment I strongly advice you to spent your money elsewhere before you build this filter. If, however, you’re an audiophile and if you have high-resolution equipment, then you might well consider building this filter.

For people that are interested to build their own filter I’m offering a DIY-kit for approximately $260 US. I know that it is cheaper to buy all the electronic parts by yourself, but please note that for the money you have a professional PC board added (with soldering mask, tinned soldering eyes, and all the holes drilled) as well as the aluminum case with a 4 mm (!) front- and a 2 mm back-plate with all the holes milled. It also should be noted though that, in order to bring this project to a good end, you need at least to have some basic soldering experience and as well as proper tools (soldering iron with a 0.4 ~ 0.5 mm pencil tip). Although construction is rather straight-forward (a component plan is added), this project is not intended for real novices. If you doubt your own skills please contact the author for the possibilities to obtain a finished device.

As always, have fun.

c. 2002 Jan Meier.

A Precision Preamplifier-Power Amplifier System with Natural Crossfeed Processing.

by Jan Meier

meier3_0

“The better is the enemy of the good.”

The headphone amplifier with the natural crossfeed filter published on HeadWize in fact was my first DIY-project. I built this device because I was unsatisfied with the sound reproduction via the headphone-socket of my CD-player. But, as things go, I started to like constructing and decided to design and build some power amplifiers also. Having finished these, they sounded so good that I also made a new matching preamplifier with integrated headphone-amp.

Although the circuit of this new preamp basically is the same as that of the original headphone-amp, some modifications made it possible to increase the sound-quality substantially. The new preamp also has some more options as far as inputs and outputs are concerned. In this article I’ll briefly discuss each modification and leave it to the reader which modifications/options he wants to realize. For the basic preamp/headphone amplifier circuit, the reader is referred to the original article.

The matching 35W stereo power amplifier has 44 output stage opamps per channel and is not intended for a DIY novice. [Editor: the author also includes instructions for building a less ambitious 10W stereo amplifier.] In my opinion it really requires quite a lot of experience to build this amp properly. I had to drill/solder over 2000 holes/connections per amplifier. I made three of them, two for myself for biamping purposes and one for a friend. However, I have found the sound quality of the amplifier to be very rewarding. I was able to compare it with some very decent commercial amplifiers (DENON, LINN, NAIM), but these were completely outclassed by the new preamp-poweramp combo (an opinion shared by others).

THE PREAMPLIFIER-HEADPHONE AMPLIFIER

meier3_1

meier3_2.gif

Modifications to the original headphone amp circuit:

1. Breaking the ground loop
The preamp incorporates a ground-loop breaker. The ground-connection of the mains-socket is directly connected to the case of the preamp. It is connected to the ground-plane of the audio-circuit via a 4.7 ohm resistor in parallel with a 100 nF capacitor. This resistor prevents 50/60 Hz currents from flowing freely along the ground connections between the various audio components in a system, and thus eliminates the 50/60 Hz hum. Even if one component does not have a loop breaker, but all the other ones have, then there are also no ground loops and there is no problem. The 100 nF provides adequate RF-shielding. To prevent high voltages on the interconnect cables in case of a defective transformer, both inputs of the transformer are secured by a fuse (you never know which input is connected to neutral and which is connected to the alternating high potential).

The metal housing of the mains filter and the enclosure are both directly connected and are grounded to the mains. Normally the “signal ground” is also directly connected to this ground; however, in such situations a ground loop will occur if other equipment is connected to the preamp. By connecting the signal ground through a 4.7 ohm resistor, loop currents (and thereby hum) are greatly reduced. This implies that the preamp audio inputs and outputs MUST have floating grounds – their grounds cannot be directly connected to the enclosure.

2. Driving the opamps into class A operation
The output of each opamp is connected via a 1.5K ohm, 0.6 Watt resistor to one of the voltage-rails to drive the opamps into class A operation. At zero voltage output each output-stage now has to drive a 10 mA current and effectively works in class A. Only driving a low-impedance headphone at high volumes will result in the output stages leaving the class-A range.

By comparison, the output stage of a class B amplifier has two transistors that act like switches. One is opened to deliver the positive output currents, the other is opened to deliver the negative output currents. The switching behaviour going from positive to negative output currents (and vice versa) introduces distortion (for a very short moment the opamp is not able to “control” the signal) in the output that is readily heard (TIM-distortion).

With the output of the opamp connected via a resistor to one of the voltage rails, the DC output voltage will not change but one of the two output transistors will be opened to “dissipate” the current that flows through the resistor. As long as this current is higher then the current demand to drive the load, this output transistor will stay opened (and the other one will stay closed). There is no switching and therefore no distortion added.

This technique in principle does not limit voltage-swing, but it does limit the current swing. However, this should be no problem with my design. I enforce a DC output current of 10 mA. If higher currents are demanded by the circuitry (headphone) driven, the opamp will turn to class AB-operation. It is rather unlikely though that the preamp will need to output currents in excess of 10 mA, and if it does, sound levels will be so high that the distortion will not be heard. This modification resulted in a substantially improvement of sound quality, and can be easily added to the original design. Strongly recommended.

3. RF-shielding and prevention of oscillation
The + input of the first-stage opamps are connected to the potentiometer via two 1.5K ohm resistors. In the middle these two resistors are connected to ground by a 47 pF capacitor. Also 10 pF capacitors are added between the outputs and the inverting inputs of each opamp. These measures prevent high-frequency signals from entering the circuit and thereby increase stability and prevent high-frequency oscillations. I used polystryrol capacitors, but any other film-capacitors will also do.

4. Bass-enhancement circuit

meier3_3.gif

I slightly modified the bass-enhancement circuit. The functionality has not changed, but now the feedback resistors are 10K ohms, and the outputs of the opamps are always connected by a 150 nF capacitor. This does not improve sound quality, but it does prevent annoying clicks when changing the settings of the bass-enhancement.

5. Decreased impedance of the potentiometer
Originally, a 50K ohm potentiometer was used. I found a lower impedance to sound marginally better – but only marginally. It is not worthwhile replacing a 50K ohms pot, if you already built the original circuit.

meier3_10.gif

meier3_11.gif

Additions to the original headphone amp circuit:

1. Inputs
There are now five inputs connected via a switch to the potentiometer. At the input jacks, the signal-pathways are connected to ground by 47K ohm resistors. This decreased the capacitive and inductive crosstalk between the various channels/inputs both audibly and measurably. Actually, I was rather surprised how much these resistors added to the sound quality.

2. Line out
For recording purposes, a non-volume-controlled output was added. The audio source can be chosen independently from the source being listened to. Note that there is no signal buffer and that it might be advantageous not to have these switches set to the same position, if a recording device is connected. Otherwise, the same source will be loaded by both the preamp and the recording device and cables.

3. Preamp out
A volume controlled output to drive a power amplifier. This output signal is not processed by the natural crossfeed filter.

4. Processor out
A volume controlled output to drive an amplifier (e.g., an electrostatic headphone amplifier). This audio-signal is processed by the natural crossfeed filter.

5. Headphone out
For connecting a dynamic headphone. The headphone jack I used has a built-in switch that disconnects the processor outputs, if a headphone is connected. It is made by Lumberg (part-number is KLBRSS 3 L) and can be ordered at Farnell in Germany (ordering number 838 550). The jack is directly mounted to the board.

6. Increased headphone output impedance
The headphone output impedance is normally near zero ohms. Optionally, the output impedance can be increased to 120 ohm by adding a resistor. Many headphones are designed to be connected to a source with a 120 ohm output impedance. Personally, I did not add these resistors to my preamp, but built a plug to connect preamp and headphone that has these resistors incorporated. My Sony headphones reacted very favorably to this increased impedance, whereas my Sennheiser HD600 became rather muddy. Simply try which suits your headphones/taste best. Since most dynamic headphones have a higher impedance at lower frequencies the increased output impedance results in an increased bass (with my Sony + 3dB!, Sennheiser + 1.5 dB).

THE POWER AMPLIFIER

meier3_4.gif

meier3_5.gif

The design

This all-opamp power amplifier has 44 output opamps per channel! Why 44 output opamps? My design goals were:

  • The output stage should be very fast.
  • The output stage should be linear, so the “control-opamps” would have an easy job.
  • The amplifier should be driven by a regulated power supply (unregulated supplies, as used in conventional amplifiers are IMHO a major source for a decrease in sound quality, because there is no infinite PSRR).

I wanted a completely regulated power supply for the output stage for currents up to 4 Amps. This implied using 4 pairs of LM317/LM337, since one voltage regulator only handles 1 Amp. I, therefore, would also need at least 4 pairs of output transistors per channel, since you can’t put voltage regulators in parallel to supply the same component with current. (There are voltage regulators that handle more than 1 Amp, but these are very expensive and require lots of heatsinking). So the choice was between [8 transistors + 4 opamps + heatsinks] or 44 opamps.

I also wanted to have a linear output stage with no distortion, which implied local feedback using one opamp per transistor-pair. For semi-class-A operation, 4 pairs of transistors allow for 4 different “switching” points (or more precisely, the output voltage where the opamp switches between the two transistors in the output stage). 44 Opamps allow for 44 different “switching points” (actually I only use 24 different points but this still is far better than 4). Each transistor-pair that is to be driven in class-AB dissipates heat and requires a heat sink. Opamps like the LM6171 don’t need a heatsink (unless you use the dual version at high currents).

Opamps are an ideal solution, but their current capabilities are too limited. I, therefore, placed 44 of them in parallel. To drive them in pure class A would demand a high DC-current (per opamp) and increase power dissipation. I decided to inject only a relatively small DC-current, so each opamp works in class AB.

By using different current values for the various opamps, each opamp will switch at a different overall current demand. At any time of operation, the major part of the output opamps will be in a non-switching state, and the “control-opamp” (which is working in class-A) is able to control the output signal continously. TIM-distortion is eliminated, although class-AB functionality is used. In contrast to a conventional class AB device, where there is no control during switching, there always will be “control” using many parallel output stages, each switching at different points. That’s why I called it semi-class A.

I wanted biasing currents for the output stage opamps between approximately 1.5 and 5 mA – not too high to cause excessive current drain and power dissipation, and not too low to start switching at very low sound levels. So RP resistors should be approximately within the range 3K ohms to 12K ohms. Then I simply selected values that were available in the catalogue. No sophisticated calculations.

When you look at the costs, I don’t think that the transistor solution is much cheaper than the all-opamp solution. High quality, high speed transistors as well as decent heatsinks do cost. Of course the second solution is more elaborate, but the way is our goal, so time is for free.

Construction

meier3_6.gif

There are no special new construction techniques. This is not a project for novices. The enclosure shown is a real nice part but unfortunately also very expensive ($150). Like the preamp, the audio input and output jacks MUST have floating grounds, because of the loop-breaker circuit in the power supply.

The opamps that have the highest power dissipation in “standby” are the input opamps. They have to dissipate a current of 18V/1.5K ohms = 12 mA. Power dissipation is 216 mW. This is fully acceptable. Power dissipation of the output stage opamps is much lower, except when driving large signals into low loads. However, such power demands are transient. Power handling is one of the reasons I decided to use the LM6171 instead of its dual version LM6172. Note that 2 x 44 amps have quite a large total body surface so any heat is easily transferred to the air.

Originally, part of the amp got quite hot, not because of the heat dissipated in the opamps, but because of the heat dissipated in the voltage regulators. The total standby current for both channels is approximately 0.34 A. The power transformers are 2 x 18 VAC (50 Watts). With a voltage drop across the positive voltage regulators of 25V – 18V = 7V, power dissipation becomes 2.4 Watts. The negative regulators have to dissipate an equal amount of heat, so total power dissipation in the voltage regulators comes up to approximately 5 Watts. This is easily handled by the heatsink I made out of a aluminum sheeting. Even at high sound levels, the output voltage of the regulators does not drop below 23V, which means that the regulators can still do their jobs most adequately.

The amplifier uses 2 power transformers, not to have a completely independent supply for both channels (because they aren’t), but because I use a very slim enclosure and one big transformer would not fit. Each transformer “drives” 4 positive and 4 negative voltage regulators (4 pairs). Each voltage regulator pair, consisting of a positive and a negative voltage regulator, “drives” a group of 11 output stage opamps (44 opamps total). One additional pair of positive and negative voltage regulators (after a thorough LC-filtering) powers the the four input stage opamps of both channels. There are separate fuses for both transformers. The values of the fuses shown in the schematic are for 230VAC mains. For 110VAC, a value of 800 mA would be more appropriate.

A simpler 10W amplifier

The maximum output voltage of the amplifier is approximately 16V and the maximum current is about 6.6 Amps. To build a “smaller” power amplifier, reduce the number of output stage opamps to limit the current capability of the output stage. There will be a point where the amplifier will not be able to deliver the 16V. It all depends on the impedance of the speaker. For example, using 20 opamps will limit the maximum output current to approximately 20 x 0.15 = 3 Amps. With a 4 ohm loudspeaker, the maximum voltage is 12 V and maximum music power is 0.5 * I * I * R = 18 Watts per channel.

Which output stage opamps should be removed to reduce the output power? I would take the ones with the higher impedances to the power rails first, since this would drive the output stage for a longer period of time in pure class A. However, I think the sonic difference will be small, if some of the other opamps were removed.

Operation

The maximum power of the amplifier is primarily set by the supply voltage (18V). The maximum output voltage is approximately 16V. With 8 ohm speakers, the power rating is 16 Watts per channel. With 4 ohm speakers, the power rating doubles (32W per channel). Continuous power output equals the peak power output since, except for the supply voltage, power supply is “over-dimensioned”. Each LM6171 opamp is able to deliver up to 150 mA of current, we have 6.6 Amps per channel. With 16V output, the amp should drive loudspeakers down to 2.5 ohm. In this case, the output impedance effectively seen by each separate opamp is 44 x 2.5 + 10 = 120 ohms and does not represent a major problem (LM6171 is specified for impedances down to 50 ohms).

The noise of preamp and power amp measured at the loudspeaker connections with the volume at maximum was heardly noticable (no hum due to the regulated power supply) and unmeasurable for my multimeter (less then 0.1 mV!). SNR thus by far exceeds the specifications of the CD and is estimated to be better than 120 dB.

The “Phase” switch can be used to configure the both channels of the amplifier for biamping or can convert the amplifier into a monoblock with double the output power. It has three positions:

  • Position 1: each channel is driven by its own input buffer. This is the normal stereo mode.
  • Position 2: each channel is driven by one and the same input buffer. This can be used for biamping when both channels drive different units of the same loudspeaker. (Alternatively, one can connect the same output of the preamp to both inputs, but this solution saves cable and was given for free by the phase-switch.
  • Position 3: Each channel is driven by the same input buffer but the phase of one of the output channels is reversed. Connecting one single loudspeaker to the positive terminals of both channels (instead to one positive and one negative (ground) terminal) doubles the signal amplitude. This option is specially designed for high impedance, low efficiency loudspeakers. Advantage: Maximum output power per loudspeaker has increased by a factor 4 (approximately 64 Watts at 8 ohms instead of 16 Watts). Disadvantage: The stereo amp is converted to a mono amp (double costs).

Last week I listened to some real loud music. Due to the Analoguer filter (described in another article on HeadWize), I am simply able to sustain much louder sound levels now. I have used the amps with various loudspeakers (CHORD, KEF, QUAD, etc.) and although some of these speakers are quite hard to drive, the amp did not seem to have any problems with them. Especially the excellent bass-control was one of the first characteristics noticed by most listeners.

meier3_7.gif
meier3_8.gif
meier3_9.gif

I am offering a DIY-kit for the headphone amplifier (NOT the preamplifier) with the updates discussed in this article. The completed headphone amp is shown above. If you are interested in the kit, please e-mail me.

I strongly recommend experimenting with these designs. As always, have fun!

Addendum

10/11/2000: Corrected ground-loop breaker section in power supply schematics for preamp and power amp. For the ground-loop breaker to work properly, the circuit ground must be isolated from the metal enclosure, which is connected to the mains ground.

11/6/2000: Repositioned 10pF feedback capacitor around IC2 in preamp for greater stability. Also added pictures of headphone amplifier kit and updated picture of preamp-power amp combo at beginning of article.

c. 2002 Jan Meier.

A Compact 50W Integrated Amplifier with Meier Headphone Section.

by Tim Harrison

harris2_0.jpg

My reason for constructing this project was to develop a design for a compact integrated stereo amplifier suitable for use by a poor (but sound quality conscious!) student living in a university or college dorm. The amplifier drives a pair of loudspeakers using two LM3876 integrated power amp ICs (50 watts per channel), or a pair of headphones via a Meier crossfeed filter and an OPA2134 dual opamp. It provides four switchable line level inputs, and an unbuffered line level output for recording purposes. The design uses readily available good quality components, and is based around four separate PCBs; one for each power amp channel, one for the power supply board, and one for the preamp/headphone driver.

THE CIRCUIT

harris2_10.gif
Figure 1

The block schematic for one channel of the design is shown above (figure 1). The preamp and the first stage of the headphone amp are separate in this application, ‘straddling’ the gain across the volume control. There is an initial gain of 2.5 before the control, followed by a further gain stage of x3 after it. This arrangement allows the power amp to be driven directly from the output of the volume control without further gain, and makes for lower noise operation of the headphones. The input selector switch is a 4-way, 3-gang type, so one gang isvused for each channel, and one gang is used to switch the input indicator LEDs.

harris2_1.gif
Figure 2

Above is the schematic for part of the preamp board (figure 2). The output of the selector switch is sent to pins J1 and J3. Looking at the left channel, C1 and R2 form a low pass filter with a -3dB point of 40kHz, which rejects any RF interference picked up on the interconnects. R2 also sets the input impedance of the unit, in this case 47k ohms. R1 ensures the opamp U1 is presented with an equal impedance at both its inputs, helping improve its distortion performance as outlined on the OPA2134 datasheet. The value of R1 (9k1) is the nearest commonly available value to the parallel combination of R3 and R4 (22k and 15k respectively). R3 and R4 set the gain of this stage, just under 2.5 in this case. This value allows ample headroom for a wide range of source signals, which could be as much as 3VRMS. In this case, the peak output voltage of 10.6V would be fine with the suggested ±15V power supply.

This initial gain brings the signal up to a level whereby the output from the volume control can drive the power amp circuits directly, with no further gain, and allows the headphone driver circuit to operate with a lower gain, giving lower noise performance. C7 forms a 100kHz low pass filter with R3, rolling off the gain to unity at very high frequencies, and helping promote stability of the opamp. It is not strictly necessary with the suggested OPA2134 device, but allows the drop-in substitution of a cheaper but more oscillation prone device, such as the NE5532, if budgets are tight. C19 AC couples the output from this stage to the volume control, and with a 50k potentiometer, sets the -3dB point of the headphone amp’s response at 1.4Hz (the power amp has further high pass filtering). This capacitor is very important, as all the other stages are DC coupled, and C19 prevents any DC offsets from source components being amplified and presented to the headphones or speakers.

The resistor R9 links the output of the input selector to a recording device, such as a tape deck or minidisc recorder. It helps prevent the source becoming too loaded down feeding both the input gain stage and the recording device, and protects the source should the output become shorted to ground for any reason. The outputs from J5 and J6 are fed into the volume control pot, which should be a good quality type. Finally, C3 to C6 provide local decoupling of both the power supply rails, C5 and C6 decoupling the high frequencies, with C3 and C4 decoupling the lower ones.

harris2_2.gif
Figure 3

The output of the pot feeds the power amp and the headphone driver, which is also mounted on the preamp board. Looking at the above schematic for the headphone driver (figure 3), we can see that the opamp U2 is used in a similar configuration to the input amp U1. In this case, R24 matches as closely as possible the parallel combination of R11 and R12, helping reduce distortion as before. Again, C21 allows compatibility with cheaper opamps. R11 and R12 set the gain of the stage at just over 3, bringing the signal up to a level sufficient to drive a pair of headphones. This stage also acts as a buffer, isolating the Meier crossfeed filter from the varying output impedance of the volume control. C8, R14, (with C10, R21, and R15) form a crossfeed filter, which in this case is permanently wired in circuit. A detailed description of the operation of this circuit can be found in Jan Meier’s article A DIY Headphone Amplifier with Natural Crossfeed.

Basically, the circuit performs a frequency selective mix of the two channels into each other, allowing recordings meant for speaker listening to sound natural on headphones. I had built projects with the filter made switchable in the past, but I never turned it off, so the switch was omitted here. Finally, the opamp U3 forms a simple noninverting buffer to drive the headphones. R17 forms a minimum load when the phones are disconnected, and helps prevent pops and clicks when they are connected with the unit powered up. While it is possible to substitute cheaper opamps in other parts of the circuit, the device used here needs to have a high output current capacity, and must remain stable when driving difficult loads. J10 and J12 are the output to the headphone socket, which should have its ground isolated from the chassis so as not to defeat the ground loop breaker circuit. Again, C11 to C18 provide local supply decoupling for the opamps.

You can find more information on the detailed operation of opamp based circuits, such as the preamp and headphone amp circuits presented here, in Chu Moy’s article Designing an Opamp Headphone Amplifier. Figure 4 is the power amp schematic for one channel (both channels are identical – and use one power amp board each).

harris2_3.gif
Figure 4

The circuit given here is similar to the one presented in the article, Single Chip 50 Watt / 8 Ohm Power Amplifier, on Rod Elliott’s site, ESP. The LM3876 is a good quality component capable of delivering 56W continuously into an 8 ohm load and 100W peak – enough for any dorm! It has a quoted distortion figure of 0.06% at 40W output, and offers good sound quality in a simple design. It has comprehensive output protection circuitry, preventing not only thermal runaway, but protecting the device from short circuits on the output, and voltage spikes from inductive loads.

Looking at the circuit, R3 and R1 set the gain of the power opamp at 23, and C1 limits the DC gain to unity. It also forms a low pass filter with a -3dB point of 7.2Hz. R2 draws roughly 1.5mA from pin 8, disabling the internal muting function of the LM3876, and C2 provides a large time constant for the action of the muting circuit. R4 should be a 1W resistor, and has 10 turns of 0.4mm enamelled wire wound round it, with its ends soldered to the resistor leads, giving a roughly 0.7uH inductor in series with the 10 ohm resistance. The inductor acts to promote stability of the power opamp, by ensuring a minimum 10 ohm load at higher frequencies. Likewise, the low pass zobel network formed by C7 and R5 (which should also be a 1W type), helps prevent oscillation should any RF appear on the output. C3 to C6 provide local supply decoupling for the power amp IC.

To enable the power amplifier to deliver its full rated power (56W/ch) continuously, and to cater for the potential 100W peaks, I decided to build a good quality power supply for the project, capable of supplying 200W. The main power supply for the speaker amps was built directly into the chassis, and is a fairly standard design. It supplies ±35V, and is capable of just over 3A continuous per rail for both the power amps. A ±15V supply for the preamp and headphone driver is provided from the main supply by the PSU board. Firstly, I will describe the main power supply, whose schematic is shown below (figure 5):

harris2_4.gif
Figure 5

The mains enters the chassis via a filtered IEC inlet, and the live line is fed through a 1A antisurge type fuse mounted in an insulated chassis fuse holder, before both the live and neutral lines are fed to a DPST rocker switch mounted on the front panel. The mains feed from the switch is connected to the primary of the power transformer, and a pair of transient suppressors are wired in parallel with it (only one is shown in the diagram). They should be rated for the mains voltage where you are, and should be mounted securely on the base of the chassis, I used two sections from an insulated terminal block.

The secondaries of the transformer are wired in series, and the wires from the toroidal types can be connected directly to a heavy duty chassis mounted bridge rectifier. The output of the bridge rectifier is sent to a pair of reservoir capacitors, C2 and C3, connected in parallel with C4 and C5, which provide high frequency decoupling. The only other point about the power supply that needs explaining is the ground loop breaker circuit. The 0V rail is connected to chassis ground and mains earth via R1, a 10 ohm wire wound resistor, in parallel with C1, a mains rated 100nF capacitor. The resistor prevents any currents flowing round the loop created by the mains earth and the ground in unbalanced phono interconnects. The 100nF capacitor shorts the resistor at high frequencies, allowing any RF to flow to ground in the normal way. I placed C1 and R1 on the underside of the stripboard I used to mount the reservoir and decoupling capacitors.

The output from the main PSU is fed to the power amp boards via a front panel DPST switch, allowing the speaker amps to be switched off for headphone only listening, and also (unswitched) to the preamp PSU board. Below is the schematic for the preamp PSU board (figure 6):

harris2_5.gif
Figure 6

The ±35V rails must be reduced to around ±20-22V before they can be fed to standard three pin regulators. I simply used a potential divider comprising R3 and R6 for the positive rail, and R4 and R7 for the negative rail. Simply placing a reverse biased 12-15V zener diode in series with the supply, i.e. in place of R3 and R4 (and omitting R6 and R7), would be an alternative option, and probably simpler – this option didn’t occur to me until after I built the prototype! C1 to C4 decouple the output of the regulator, and R1, R2, and R5 set the current flowing though the LED indicators, around 15mA in this case. The stabilised ±15V supply is presented on pins J1-J3, and the remainder of the pins provide supplies for a pair of power on indicator LEDs (mounted next to the mains rocker switch), and the input selector LEDs. These are mounted above the input selector switch, and light to show which input has been selected. They are controlled by the remaining gang of the three gang rotary switch.

CONSTRUCTION

I have provided my PCB artwork for you to use to make your own PCBs if you are interested in building all or just part of this project. Below are links to the artwork files, and the relevant placement guides (all GIF format):

harris2_9.gif

Download PC Board Artwork and Placement Guides 1
Download PC Board Artwork and Placement Guides 2
Download PC Board Artwork and Placement Guides 3
Download PC Board Artwork and Placement Guides 4
Download PC Board Artwork and Placement Guides 5
Download PC Board Artwork and Placement Guides 6

The artwork will print the correct size if you set your graphics software to output 600dpi to your printer. The placement guides should be printed at 300dpi. If you have trouble getting them the right size, the power amp boards should be 41mm wide, the PSU board should be 113mm wide, and the preamp board 132mm. I made the artwork for the lead pitch and size of the components I could source, so I suggest you print out the placement guides real-size (300dpi), and compare the sizes and lead pitches of the components you can source, selecting the ones that best fit the board.

As a guide to component selection, I used 0.6W metal film resistors throughout, except for R3, R4, R6, and R7 in the PSU, which should be 5W wire wound radial lead types, and R4 and R5 in the power amp, which should be 1W wire wound types. For the decoupling capacitors use ceramic disc types, and for capacitors in the signal path (C19, C20, C8 and C10 in the preamp), I used the Wima MKS4 250V series, although any metal film type will do (but may not fit).

harris2_6.jpg

The line level signal from the sources is received via an array of gold plated phono plugs on the rear of the unit. The plugs I used had red and black identification bands on them to indicate which channel should be connected to them. This was important, as I was not planning to print any lettering onto the case, so the connections and controls had to be fairly self-explanatory. The phono plugs should be mounted using an insulating bush, as the design uses a ground loop breaker circuit, and the signal and earth (chassis) grounds are separated.

The source signals are routed via screened cable to a rotary selector switch mounted on the front panel which is used to select the source to be listened to (and recorded from). The switch should be a good quality part, as a positive tactile response from it enhances the feel of the finished project. The part I used was a 3-gang 4-way type, allowing 4 stereo inputs to be accommodated, leaving one gang free to switch the source indicator LEDs mounted above the control. I used a cheap part by a company called Alpha, their SR2611 series. This switch works fine and only cost a few pounds (roughly $5 US).

For a volume control, I used a 50k ALPS pot, but a cheaper type of any value between 10-100k could be used. A conductive plastic track type is preferable to a carbon track, and should be logarithmic law (also called audio taper). The ALPS RK27 series pots (the blue ones), while pricey, come highly recommended, as they have a very nice tactile feel to them, and exhibit good tracking between the gangs.

harris2_11.jpg

For the preamp and headphone section opamps, I recommend the OPA2134 by Burr Brown, and DIL sockets are a good idea to help prevent heat/static damage during soldering. Note, the LM3876T power opamp in figure 4 must be used with my PCBs, the T suffix denotes the package type. The power opamps share a large 2 degrees C per watt heatsink mounted on the rear panel in the prototype and are mounted using greaseless silicone insulators and insulating bushes. Make sure the metal tab of both the power amp ICs is isolated from the chassis – this is very important.

Power supply

The value of the mains fuse in figure 4 varies depending on what type of transformer you use, and the supply voltage in your country. Since I live in the UK where the mains supply is 230V, and I am using a 225VA rated toroidal transformer, a 1A antisurge fuse was used. Take care to get this value right, as if it is too low, you will suffer nuisance blowing, and if it is too high, you will not get proper protection in the event of a fault. The fuse rating can be calculated in the normal way using I = P / V. A double pole type switch is preferable to a single pole type, as it allows the unit to be completely isolated from the mains when it is switched off. The mains rocker switch used should be rated to handle the in-rush current of the transformer, anything over 4-5A should be fine in this case.

harris2_8.jpg

I don’t normally include transient suppressors in the power supplies of audio projects I build, as I run the mains supply to my home audio system through a filter which includes them. However, as this integrated amp was designed to be used away from home, they are included here. You can use a pair together in parallel as suggested to increase their dissipation capacity.

The toroidal power transformer was made by Nuvotem and sourced from RS components in the UK, at a cost of about 23UKP (or about $35 US). In general, the power transformer itself should be a good quality type, and I recommend a 225VA toroidal part for this project. If the budget is tight, a lower value toroidal (say, 160VA) could be used, or even a conventional EI laminate type. Although the standard EI transformers are cheaper than toroidal types, and exhibit a lower inrush current, they are less than ideal for a compact unit. They tend to be quite bulky, and emit strong electromagnetic fields, leading to hum pickup in adjacent circuitry. A toroidal transformer is both compact, and emits a far less strong field.

Although the rated current of the power supply is only 3A, the charging current of the reservoir capacitors will be much higher than this at times. I recommend using a 35A type bridge rectifier, such as the KBPC3506. A pair of heavy duty insulated terminal blocks should be mounted nearby, and the centre tap of the secondaries connected to this. The terminal block will now form a star grounding point, and should be the place all the 0V rails in the unit are connected together. This method of grounding ensures hum free operation. Save yourself a lot of grief and use this method the first time – hum free results are almost guaranteed.

If the budget is tight, 4,700uF capacitors can be substituted for the 10,000uF ones specified in figure 4, especially if a 160VA transformer is used. I had trouble fitting 10,000uF caps into the chassis, so I used two 4,700uF caps in parallel per rail. I couldn’t get any capacitor mounting brackets, so I simply soldered C2 – C5 onto a small piece of stripboard. You could use either method, but be sure to take your DC output from the capacitors and not the rectifier.

For the 5W resistors in figure 5, I used vertically mounting ceramic, wire wound resistors, but you could use standard axial types, with one leg bent down the side, if you find the radial types hard to get. C5 to C8 decouple and stabilise the output of the potential divider (or zener diode), before it is fed into a pair of standard voltage regulators. These should be mounted with a pair of small flag type (clip-on) heat sinks with a thermal resistance of around 20-25 deg. C per watt. I used Redpoint Theramalloy PF752.

Chassis

I mounted the project in a compact instrument case (300mm W x 150mm D x 100mm H), which has a removable internal chassis. The case was supplied painted grey, but once I had drilled it, I decided to repaint the chassis blue to make the project look more individual. I prepared the chassis by sanding it down thoroughly, making sure that all surfaces would provide a good key the paint could adhere to. I then cleaned all the surfaces with white spirit, and applied three thin coats of standard car spray paint. I used diffused blue 3mm LEDs, black rocker switches with blue markings, and black aluminium knobs to complete the effect.

You can see the layout I used pretty clearly from the pics inside the unit, all the boards were mounted on the base of the chassis, except for the two power amp boards which were mounted on the rear panel. The bridge rectifier is bolted to the bottom of the metal chassis. I used 4mm binding posts for the speaker terminals, two black ones for the ground connection, a green one for the left channel, and red for the right. All the signal wiring should be done using shielded cable, with the screen grounded at one end only. Ribbon cable can be used for LED wiring, 32/0.2mm hookup wire should be used for power amp supply and speaker connections, and 7/0.2mm hookup wire can be used for other low power connections.

If there is hum on the output of the completed project, the problem is almost certainly to do with the ground scheme used. Make sure that there is a 10 ohm resistance between the chassis and signal ground (i.e. that you have not defeated the ground loop breaker), and make sure you have not accidentally grounded a point by two paths simultaneously. The star grounding scheme as outlined earlier is highly recommended. The path to ground on the volume control pot is particularly critical, in the prototype the unit refused to stop humming until the far end of the wiper had its own separate connection to the star ground point. It should be possible to set the volume control to zero and, with the unit on, put your ear to the speaker and hear nothing but a faint hiss.

RESULTS

My impression of the project overall is very good, it sounds good, and is very compact. The performance from the IC power opamp is impressive, and I think my prototype looks nice, too! Listen to your favourite cans through it late into the night, or let it provide some serious slam through speakers for a small room or dorm.

c. 2002 Tim Harrison.

An Enhanced-Bass Natural Crossfeed Filter.

by Jan Meier

meier2_0

[Editor: This article is a follow-up to A DIY Headphone Amplifier with Natural Crossfeed by Jan Meier and resulted from an e-mail discussion with Chu Moy – the author of An Acoustic Simulator for Headphone Amplifiers.]

If you have an amplifier with a mono-switch, then here is a little experiment: listen to a stereo recording (by headphone) in stereo mode, and then press the mono-button and watch the bass. If you hear the same way I do, then you will notice that the bass suddenly seems to have weakened – it has become less pronounced. The effect is similar to that what is heard with a crossfeed filter, only much stronger. Listening in mono does introduce cancellation of low frequencies, but there is also cancellation higher frequencies (which is generally is even stronger since, with normal stereo recordings, low frequencies are more in phase). With the crossfeed activated, a weak cancellation will only be present at low frequencies, but at all frequencies, the sum of the sound pressures at both eardrums always equals the sum of the pressures in stereo mode!

At first I also wondered about the apparent loss of bass, but actually, it is this unnatural, larger then life-size, uni-directional bass, that counts for most of the annoying effects of headphone listening. I know, the crossfeed sound is nothing for a bass-freak. One should not expect a punchy bass, only a relaxation of the sound. It’s like listening to loudspeakers – a balanced speaker does not jump at you at first hearing but is rather colourless/neutral/unobtrusive. The rewards come while listening for longer periods of time. A good speaker does not fatigue, and this exactly is the strength of the natural crossfeed filter.

To add bass or not to add bass….that is the question. I believe that most bass-losses are due to psychoacoustic effects, but after thinking it over more carefully, enhancing the bass response of the natural crossfeed filter could be legitimate, because headphone sound is optimized without using crossfeed. If there really is a psychoacoustic effect (a uni-directional bass is unnatural and I believe that, with headphones, this emphasizes its existence), then the effect has been (unconsciously) corrected for in the sonic design of the transducers.

meier2_1.gif
Figure 1

The crossfeed design by Siegfried Linkwitz (see An Acoustic Simulator for Headphone Amplifiers by Chu Moy) includes a bass boost to compensate for low frequency cancellation. Figure 1 is a graph of the frequency response of both the direct-signal and of a mono-signal that is given on both signals simultaneously. Responses were calculated for a 60 Ohm load (such as headphones) and for a very high output load (e.g., a headphone amplifier).

meier2_2.gif
Figure 2

As with the natural crossfeed filter, the direct signal with the Linkwitz filter shows a signal loss at lower frequencies, (-1.0 dB at 60 Ohms, -0.35 dB at 50k Ohms). However, more important is that a mono-signal at frequencies below 700 Hz is increased by up to 1.3 dB at a 60-Ohm load and up to 1.9 dB (!) at 50k Ohms. The delay times for the Linkwitz design (figure 2) are fairly natural, as the crossfeed signal has similar filter frequencies and thereby should have similar delay times as the natural crossover filter.

meier2_3.gif
Figure 3

I designed a modified version of the natural crossfeed filter that has a frequency response very similar to the Linkwitz filter. It can be found in figure 3. The crossfeed level is medium. It easily can be tested between a CD-player and headphone amplifier (there is no insertion loss). I guess it sounds very similar to the Linkwitz filter, but is a little bit easier to realize. I have never implemented this Linkwitz equivalent, being fully satisfied with the original natural crossfeed filter.

The Linkwitz equivalent can be substituted for the filter in my headphone amplifier design. The bass EQ switch (S1) was not intended to compensate for any apparent loss of bass due to the crossfeed. It simply should compensate for the natural roll-off of the transducers. Such filtering is nothing new. With my headphones, I use position “3.” In this position, only the very lowest frequencies are amplified. Even with the Linkwitz equivalent, it could be very nice to keep the bass extension switch. Since it can be switched off, it will not hurt and the extra work/costs are little. It is, however, a matter of personal taste. Hi-fi purists do not like unnecessary equalization and might want to leave it out.

meier2_4
Figure 4

With the Linkwitz equivalent filter, it is not possible to set the crossfeed level with a 2 x 6 switch. Not only do the resistors and capacitors in the two outer networks (Z1) have to be switched, but also the central capacitor to ground. A possible option is to use a 3 x 4 switch (see figure 4) and leave the first two crossfeed positions out (they have a very weak effect, and I never use them).

meier2_5.gif
Figure 5

To customize the time delay and amplitude profiles of the Linkwitz equivalent, download the circuit simulation spreadsheets below (in Quattro Pro 1 and 3 formats). Change the values B4..C9 at the first page according to figure 5 and the corresponding lines in the pictures will change. Similarly you can change the values in D4..E9 for a direct comparison of the different options.

Download circuit simulation spreadsheet (in Quattro Pro 1 and 3 formats)
Download circuit simulation spreadsheet (in Quattro Pro 1 and 3 formats)

(Please remove the .xls extension on the files above.)

Yesterday, I did some listening with popular music with heavy bass using the original natural crossfeed filter. The bass in these recordings was more “centered”. As expected, I could not/hardly notice any specific loss of bass. I present the Linkwitz equivalent design as an example of how the various filters can be tested to personal taste, by putting them between a CD player and a power/headphone amplifier, before eventually building the headphone amp project.

Addendum

6/22/99: Added figure 2.

7/26/99: Added figure 5 and instructions for using spreadsheet circuit simulator.

5/4/00: Jasmin Levallois built this version of the Pocket Headphone Amplifier (see article by Chu Moy). It features an input gain stage, the Meier enhanced-bass natural crossfeed filter and an output buffer. He writes:

Finally I got some free time to complete my project…. I got a lot of work to do for school during the last few weeks and I didn’t have time to work on my amp. This weekend I decided to take one day to transfer the amp from the breadboard to the pc board. I used about the same circuit as Jeff Medin. The input stage has a gain of 10, the output stage is a voltage follower, and in the middle I put the Meier bass-enhanced crossfeed circuit.

I used 2 OPA2132 opamps, but if I had to do it again I would use 2 OPA2134. An OPA2132 costs $6.99 while an OPA2134 costs $2.67. Since there is almost no audible difference between both opamps, I would go with the OPA2134 to save money. Since the second stage has no voltage gain, I decided to omit the capacitor in front of the output stage. I also removed the resistor in front of the output stage, and I don’t hear any noise from the output stage. The only noise I can hear, sometimes, is coming from my CD player.

As you’ll see on the photos, the inside of my amp is very messy, but, hey, its my first electronic project. Fortunately, even if it’s messy inside, the outside looks pretty good. I really like this Serpac Enclosure (Digikey part no. SRH65-9VB-ND); it looks ways better than the PacTec case.

The photo of the battery compartment is to show that the Serpac enclosure has a 9v Battery compartment with battery contacts. It’s easier to remove the battery with that kind of battery compartment than the PacTec Enclosure. Also the Serpac enclosure is just about the same size as the Pactec enclosure except that it’s a bit longer, and the height is a little bit less. This might be a problem for the electrolytic capacitors. I would recommend the Philips ones with this enclosure rather than the Panasonic Z series because the Philips electrolytic caps are much smaller.


leval1.jpg
leval2.jpg
leval3.jpg

Download parts list for Levallois Amplifier (MS Excel format)

5/6/00: Gus Wanner developed a low impedance version of the enhanced-bass filter to drive his Sennheiser HD600 headphones directly from a power amplifier, and has prepared a MS-Excel application to model the enhanced-bass filter (both low- and high-impedance versions). DIYers can change component values and instantly see the effects of their changes on the filter’s frequency response, time delay profiles, etc. Wanner writes:

Compared to my Sennheiser HD25s, the HD600s have a lower sensitivity (the HD25s produce about 105 db SPL at 1mW into 70 ohms, while the HD600s produce about 97 db SPL with 1mW into 300 ohms), a higher impedance, and a maximum input level of 200mW. With the HD600s, 200mW requires about 7.75 volts across each phone. I measured the impedance magnitude versus frequency for both the HD25 and HD600 headphones. The measurements were made using an audio oscillator in series with a 300 ohm 1% metal film resistor. By measuring the voltage across each headphone and the voltage across the 300 ohm resistor, it is possible to compute the impedance.

I have attached a spreadsheet which provides my data and the impedance plots. It is interesting to note that both headphones have an impedance peak around 100 Hertz; in the case of the HD600s this peak has a magnitude of almost 600 ohms! I also looked at both the current and voltage waveforms on a scope; for both phones the waveforms are in phase for the entire audio range. Slight phase shift can be seen at 20 Hz and at 20,000 Hz, but I would say that the impedance of both of these phones is predominantly resistive.

To handle the HD600 inefficiency, I decided to look at Jan Meier’s crossfeed networks. Jan’s networks have the advantage of low insertion loss, but tend to be sensitive to both source and load impedances. I created a new spreadsheet with an analysis of Jan’s enhanced bass crossfeed filter, using his original component values (designed for use within his headphone amplifier) and a low impedance version designed for use with a nominal 300 ohm load impedance.

This spreadsheet is attached hereto as well; you can see a performance comparison between my final low impedance compenent values and Jan’s original. Because of the large impedance variation of the HD600s at low frequencies, I incorporated their impedance versus frequency into the design as wel. I did not assume a constant headphone load impedance in my analysis of the network, but actually put in my measured data of impedance versus frequency to see how the response of the network would be affected by the Senn HD600 peak around 80 Hertz. For other headphones, you can substitute the impedance magnitude versus frequency (if you have it) into the appropriate column on the spreadsheet. Alternatively, you could substitute the manufacturer’s nominal impedance value (e.g., 120 ohms or whatever) at each of the test frequencies.

Components for my crossfeed networks were purchased from Digikey – the capacitors are Panasonic ECQ-E(F) series metalized polyester film types which are available in values up to 10 microfarads at 100 WVDC at reasonable prices. Non-polarized electrolytics are NOT recommended for this application. The resistors are Ohmite TA Series “Power Chip” ™ thick film on an alumina substrate 5 Watt rating. These resistors have basically no inductance (50 nanohenries at 1 MHz!). Non-inductive wirewound resistors could also be used, however standard wirewound resistors are not recommended due to their inductance.

With this network, the HD600s have plenty of volume with my 20 watt/channel monitor amp and sound wonderful – the closest thing to electrostatic speaker sound I have heard! Again, note that the low impedance version of Jan Meier’s network is more sensitive to load impedance than the low impedance version of the modified Linkwitz network. With the HD600s, to get 200mW requires about 24.5 volts into each channel of the modified Linkwitz crossfeed network, which corresponds to an amplifier output of about 75 watts into 8 ohms! I did, in fact, hook up the modified Linkwitz crossfeed network to my 150 watt/channel power amplifier, and the Hd600s sounded great. But I wanted to use the HD600s with my McIntosh C40, so I had to look for another solution.

wanner_meier.gif

Download Gus Wanner’s MS Excel modeler for the Meier enhanced-bass filter 1
Download Gus Wanner’s MS Excel modeler for the Meier enhanced-bass filter 2

c. 1999, 2000 Jan Meier.
The author’s website: Meier Audio.

A DIY Headphone Amplifier With Natural Crossfeed.

by Jan Meier

meier3_7.gif

For me, listening to music is a very private affair. Throughout my life, my musical tastes always have been somewhat different from that of the other people I lived with. Consequently, I learned to appreciate good quality “cans.” This appreciation has increased since, as a teenager and as a student, I never had the financial means to buy loudspeakers that could stand up to the sound quality offered by my Sennheiser and Beyerdynamic headphones.

Nonetheless, listening was not all heaven on earth. The in-head localization phenomenon did not please me. With recordings presenting a wide soundstage, some instruments are heard in one of the two audio channels only. This is most annoying, like a bee buzzing in one’s ear.

Some 15 years ago I experimented with electronic crossfeed by bridging the left and right outputs of the headphone channel of my amplifier with resistors. Although the crossfeed thus produced cured the “buzzing bees,” the sound became extremely dry and I dropped the idea.

In the last few years, a number of systems have appeared on the hi-fi market that also produce crossfeed, but in a much more refined way. The analogue headphone amplifiers of HeadRoom and the digital systems of Sony (VIP 1000), Sennheiser (Lucas), and AKG (Hearo 777) are well-known examples. These systems all more-or-less cure the problem of the in-head localization experienced with headphones.

These systems work by simulating the mechanisms that a person uses to locate and externalize sources of sound:

First, the sound of a source to the right side of the listener (e.g., the right loudspeaker in a stereo setup) not only reaches the right ear, but attenuated and delayed, is also heard by the left ear. The level of attenuation and the delay time of this crossfeed signal provide important directional information.

Second, the soundwaves are partly absorbed and partly reflected by the listener’s head. Especially, the reflections at the ear pinnae interfere with the soundwaves that directly enter the ear canal and amplify or attenuate specific frequency components. As these reflections depend on the direction of the soundwave, the “color” of the sound changes with the direction of the source.

Third, reflections of soundwaves on the walls of the listening room produce reverberation that conveys an extra feeling of space.

The information obtained by these mechanisms is further refined by movements of the head. Changes in sound levels, delay times and sound color refine our sense of direction. For a demonstration blindfold a friend and ask him to locate a ticking clock that you have hidden in the room. He will start turning his head, although he can’t see it. With his head in a fixed position, he will find an exact localization much more difficult.

All these mechanisms are missing when we hear music using headphones. The transducers are directly coupled to our ears. The sound of the right (left) transducer will not reach the left (right) ear and the reflections on the oracles have changed and hardly interfere with the original soundwave. Moreover, the sound-sources are attached to our head, so head movements no longer add information. Reverberation is not present.

In principle, digital sound processors can simulate the mechanisms described, but the results are, thus far, not fully satisfactory, because the reflections on the pinnae are very complex and listener-specific.

Fortunately, the mean directional information is provided by the differences between what we hear by our two ears. A “natural” crossfeed from the right (left) audio signal to the left (right) transducer, with an appropriate attenuation and delay, will reduce most of the adverse symptoms of headphone listening considerably.

A straightforward approach to mimic crossfeed is to take the original stereo signal, attenuate its amplitude and have it delayed. Then cross the two channels and add the processed signals to the original stereo signal. In a mathematical formula:

Vleft,out(t) = Vleft,in(t) + .Vright,in(t-t0a<1

Vright,out(t) = Vright,in(t) + .Vleft,in(t-t0)

The HeadRoom systems work like that (with a being a frequency dependent parameter). For a more detailed information just take a look at their most enjoyable homepage (http://www.headphone.com). The people at HeadRoom are very fine engineers, but have not revealed the schematics of their circuitry. So when I decided to build my own headphone-amplifier, I had to design an appropriate crossfeed-filter myself.

The crucial part in the crossfeed-filter is the realization of the required time delay of approximately 300 ms. Although standard solutions for signal delay can be found in many text books on electronics, a fixed frequency-independent delay with headphones has one major drawback: the so-called Comb-effect.

A conventional crossfeed filter, such as that realized by HeadRoom, mimics the sound of a left or a right sound source most adequately, but the frequency-spectrum of a source in front of the listener is unnecessarily disturbed. For this in-front source, the left and right audio signals are equal: a mono signal. In principle, these signals need no crossfeed. However, with conventional solutions, there still is, and the audio signals at the headphone-transducers become:

Vleft,out(t) = Vright,out(t) = Vleft,in(t) + a .Vleft,in(t-t0a <1

Especially in the high frequency range, the delayed crossfeed signal interferes with the original input and attenuates specific frequencies. The frequency-curve is no longer flat but shows a larger number of dips (the Comb-filter effect). HeadRoom compensates for the overall attenuation with a filter that gently lifts the higher frequencies, but the dips in the frequency-curve do not disappear.

With a fixed delay, the Comb effect can not be eliminated, so I decided to make the delay of my crossfeed filter frequency dependent. For localization of a sound source, the delays of the frequencies below 2 kHz are the most important, and therefore should have a natural 300 ms delay. For higher frequencies the delay is reduced. Moreover, by also slightly shifting the original audio signals in the other direction and by giving them a small, frequency dependent attenuation before the crossover signals are added, the mono signals will be left undisturbed! A music signal that simultaneously is found on both channels is left unchanged, and a signal on only one input channel is partly transferred to the other channel, with an appropriate time delay at low frequencies.

meier1

Figure 1

Take a look at the basic crossfeed circuit shown in figure 1. The crossfeed is performed by only three resistors and two capacitors! It’s hard to believe, but the circuit really does the job!

meier2.gif

Figure 2

meier3.gif

Figure 3

For those interested in the technical details, figures 2 and 3 show the amplitudes and the time delays of both the crossfeed-signal as well as the (direct) audio signal. The amplitude of the crossfeed-signal decreases with frequency, and thus mimics the shadowing effect of the head at higher frequencies. In the lower frequency-range, the time delay between the crossfeed and the direct signals is 320 ms, and thus mimics the natural delay for a loudspeaker seen at an angle of approximately 30 degree by the listener. By choosing different parameter values for resistances and capacities the crossover signal can be “tuned”, but with the values shown it works well for my ears with 95% of all my recordings.

SOME MATHEMATICAL STUFF

With a conventional crossfeed filter, the direct signal equals the original input signal, and the crossfeed signal is realized by attenuation and delay:

Vinput(t) = cos(2pft)

Vdirect = Vinput(t) = cos(2pft)

Vcrossover = a(f).cos(2pf(t-tdelay))

a(f) << 1

With a mono signal, both direct signals are equal and both crossfeed signals are equal. The output signals become:

Vout(t) = Vdirect + Vcrossover = Vinput(t) + a(f).cos(2pf(t-tdelay))

The second term interferes with the input/direct signal and results in a number of dips in the frequency spectrum at those frequencies where: 2pftdelay = (2n+1)p. This is the so-called Comb-effect.

meier4.gif

Figure 4

With the “natural crossfeed” filter, the crossfeed signal is also realized by attenuation and delay, but now the direct signal is also (slightly) attenuated and slightly time-shifted in the other direction (figure 4):

Vinput(t) = cos(2pft)

Vdirect = A(f).cos(2pf(t+tdirect))

Vcrossover = a(f).cos(2pf(t-tdelay))

A(f) = sin(2pftdelay)/(sin(2pf(tdirect+tdelay)) » < 1

a(f) = sin(2pftdirect)/(sin(2pf(tdirect+tdelay)) << 1

tdirect << tdelay

2pf(tdirect+tdelay) < p/2

(This last condition guarantees that a(f) and A(f) will change monotonically with the frequency.)

The result is that, with a mono signal, the sum of the direct and the crossfeed signals equals the original input signal and there is no Comb-effect:

Vout(t) = A(f) cos(2pf(t+tdirect)) + a(f).cos(2pf(t-tdelay)) = cos(2pft) = Vin(t)

The condition 2pf(tdirect+tdelay) < p/2 requires that the effective delay (tdirect+tdelay) be shortened for higher frequencies. Natural delay times of 300 ms only can be realized for lower frequencies.

THE NATURAL CROSSFEED NETWORK

meier5

Figure 5

The basic network is shown in the figure 5. It contains a chain of three passive networks, of which the outer two have the same impedance value. It easily can be shown that:

Vleft,out(t) = (Z1 + Z2)/(2*Z1 + Z2) * Vleft,in(t) + Z1/(2*Z1 + Z2) * Vright,in(t)

Vright,out(t) = (Z1 + Z2)/(2*Z1 + Z2) * Vright,in(t) + Z1/(2*Z1 + Z2) * Vleft,in(t)

The crossfeed signals are given by the last terms of these two equations.

Using

Z1 = R1 // C1 = R1 /(1+iwR1C1)

Z2 = R2

the transfer function of the crossover becomes:

Z1/(2*Z1 + Z2) = (R1/(2*R1 + R2))/(1+iwC1R1R2/(2R1 + R2))

This is the transfer function of a first order low-pass Bessel filter with a filter frequency of:

f = (2R1 + R2)/(2pC1R1R2)

Using:

R1 = 1000 Ohm
C1 = 470 nF
R2 = 2200 Ohm

the filter frequency becomes f = 650 Hz.

To derive the time delay of the crossover signal:

f (phase shift of the transformation) = arctan(C1R1R2/(2R1 + R2)) and
t (time shift) = /(2pf) = f/w

For low frequencies:

f = wC1R1R2/(2R1 + R2)
t(w~0) = C1R1R2/(2R1 + R2)

For high frequencies:

f = p/2
t = 1/4f

To derive the time shift of the direct signal:

Amplitude A(f) = 1 – Z1/(2Z1+Z2) = (R1+R2+iwC1R1R2)/((1+iwC1R1R2/(2R1+R2))

f = arctan (wC2R1R2/(R1+R2)) – arctan (wC2R1R2/(2R1+R2))

t = f/w

t(w~0) = C1 R1 R2 / ( R1 + R2 ) – C1 R1 R2 / ( 2 R1 + R2 )

The two filter frequencies are:

f1 = 1/((2R1+R2)(2pC1R1R2) = f

f2 = 1/((R1+R2)(2pC1R1R2)

Using the crossover frequency instead, the time delay at low frequencies is 1/(2pf) = 250 ms. Together with a time shift of 70 ms of the direct signal, the effective time delay is 320 ms. MS

meier_excel.gif

Excel circuit simulator.

Download circuit simulation spreadsheet (in MS Excel format)

The MS-Excel circuit simulator (above) lets you experiment with the filter’s component values for the standard crossfeed filter (see my article here for a simulator for my enhanced-bass filter). Component values can be changed in the upper left corner of the table. The pictures (the frequency response and time delay characteristics of the crossfeed filter) automatically adapt. The central branch in the spreadsheet consists of a resistor and a capacitor in series. For the standard version of the crossfeed filter, the capacitance has to be given a very high value so as to act as a short-circuit. Each side-branch has a parallel pair of a resistor and a capacitor in series. Each side branch thus consists of two resistors and two capacitors.

meier_excel_cir.gif

Circuit model for spreadsheet simulator

For the standard version of the crossfeed, one resistance has to be set to zero and its corresponding capacitance to the regular value (440nF / 320nF / 220nF). The other resistance has the regular crossfeed value (1.1k / 1.5k / 2.2 kOhm) and its corresponding capacitance is set to a very high value. The values in the spreadsheet are for a relatively low crossfeed level.

CONSTRUCTION DETAILS

Building a headphone amplifier is like building a power amp – only the current demand is just a little bit lower (about a factor 100). Various designs can be found in the internet, and it is relatively easy to integrate the proposed crossfeed filter. I designed my own (see the picture) which is op-amp based. I know, many hi-fi enthusiasts say “yuck” to opamps, but note that even in many so-called “High-End” CD-players, opamps are found in the signal path for amplification and filtering.

The op-amps should be chosen with care. They have to be able to deliver relatively high current values and to drive low impedance loads. I decided to use the National Semiconductor LM6171. This is a wide band (100 MHz) voltage-feedback opamp that is able to deliver 10V into a 100 ohm load. To prevent difficulties when driving low-impedance headphones (32 ohms or less), I placed a 47 ohm resistor at the output of each channel. With my Beyerdynamics (DT990/DT931, 600/250 ohm) and my Sennheiser (HD600, 300 ohm) headphones, the opamps perform most adequately. Other alternatives are the LM6172/6181/6182, the OPA604/627 by Burr-Brown (used in the HeadRoom systems) or the LTC1206/1207 by Linear Technology (able to drive 30 ohm loads!).

Schematic of headphone amp with natural crossfeed network.
Figure 6a

Schematic of headphone amp with enhanced-bass natural crossfeed network.
Figure 6b

The schematics shown in figures 6a/6b/6c represent the third generation of my original design. There are two versions of the headphone amplifier: one with the standard crossfeed and one with the enhanced bass crossfeed (see An Enhanced-Bass Natural Crossfeed Filter for more information). The standard crossfeed sound is nothing for a bass-freak. One should not expect a punchy bass, only a relaxation of the sound.

The two crossfeed settings of the original bass-enhanced filter are comparable to the low and the high crossfeed levels of the standard filter in this article. The enhanced-bass filter has a frequency response very similar to the modified Linkwitz filter to compensate for any apparent loss of bass due to the crossfeed. While I hardly notice any specific loss of bass with the standard filter, I present the enhanced-bass design to give DIYers an option of which filter to build. I have added a medium crossfeed level so that both the standard and bass-enhanced filters now have 4 settings. The 4.4K resistor bridges the switch at all settings to reduce any “blops” during switching.

The output stage of each opamp is connected to one of the voltage rails by a 1.5 kOhm resistor. This forces the output stage into class-A functionality and increases soundquality considerably. Also 10pF capacitors are added to the feedback loop to increase stability at high frequencies. Careful matching of all resistors prevents offset voltages and the need of coupling capacitors and the amplifier now is DC-coupled. The power supply has a ground loop breaker, so the audio inputs and outputs MUST have floating grounds – their grounds cannot be directly connected to the enclosure. (See A Precision Preamplifier-Power Amplifier System with Natural Crossfeed Processing for more discussion about biasing opamps to function in class A and ground-breakers in power supplies).

meier6c.gif
Figure 6c

A stable power supply to the opamps is crucial for optimal performance (figure 6c). I used a toroid transformer. Charge-reservoirs of 1000uF and two voltage regulators (L7815/7915) provide constant voltages of ±15VDC. Moreover, high frequency noise is extensively filtered by 100mH inductors. 2200uF capacitors and 100nF film capacitors further reduce any ripple and noise after the voltage regulators. I admit that the large capacitors are a little bit overdone. When I switch the device off it will still work for about 10 seconds. However, capacitors are relatively cheap, so why not.

The amplification and power supply circuitry is rather straightforward. I have just a few notes:

Both the standard and enhanced-bass filters have four crossfeed level settings: none (normal stereo mode), low, medium and high. I personally prefer the low and the medium crossfeed levels for most applications.

No opamp is perfect. To ensure a zero offset voltage at the output the impedance values in the circuitry have been carefully balanced. Do not use other resistor values than as indicated in the schematics, as this might lead to damage of your headphones.

The headamp has two sockets for connection to headphones. Both sockets will provide different sound characteristics. One socket has a very low output impedance and gives the amp tight control over the headphone action. However, many headphones have been sonically optimized to be driven by an output impedance of 120 Ohms and may sound better when connected to the other socket. Generally, the low impedance socket provides a clean sound whereas the high impedance socket yields a warmer sound. Use the one you like most. There is no risk of damage to your headphone by connecting it to either socket. You can also use the sockets to connect two headphones simultaneously. However, the volume produced by the high impedance socket will be slightly lower than that of the other socket.

meier3_8.gif
meier3_9.gif

As far as the signal paths are concerned, I used WIMA MKS-film capacitors (the red ones), electrolytic capacitors of ELNA, and ¼ Watt metal-film resistors.The 7 Watt toroidal power transformer is by Nuvotem-Talema, and the 10k logarithmic potentiometer by Alps (the blue one).

meier3_10.gif
Figure 8

meier3_11.gif
Figure 9

The frequency response and the time-delay of the standard crossfeed network are presented in the figures (figures 8 and 9). Turning the switch from position 1 to position 4 activates and increases the crossfeed. I personally prefer positions 2 and 3 for most recordings. To see the response graphs of the enhanced-bass filter, see my article here.

Last warning: do yourself a favor and don’t use cheap parts. Have a decent (logarithmic) potentiometer (Alps or equal), film capacitors, gold plated sockets and a proper metal housing. Even if you don’t like and use the crossfeed-filter (which I doubt), your headphones are most likely to sound much better than they ever did using the normal headphone socket of your CD-player or amplifier. Manufacturers generally do not care about the sound quality of headphone sockets and build them cheaply. They sound accordingly.

Now, before you start building, you surely want to know how the crossfeed network performs. I have made 3 WAV-recordings to demonstrate the effect of the crossfeed filter. The first recording is made without filtering. The second and third files are made with the filter switch in positions 2 and 4 respectively. The recordings contain a simple balance test. The effect of the filter on normal music is more subtle and only can be truly appreciated (in my humble opinion) while listening for longer periods than is possible with these files.

Balance test w/o filter (1.4Mb – wav)
Balance test w/filter in position 5 (1.4Mb – wav)
Balance test w/filter in position 6 (1.4Mb – wav)

The standard crossfeed amplifier is now commercially available as the Corda HA-1 from Meier Audio. For people that are interested to build their own headphone amplifier, I’m offering a DIY-kit of the standard crossfeed amplifier for approximately DM 450 ($203 US). I know that it is cheaper to buy all the electronic parts by yourself, but please note that for the money you have a professional PC board added (with soldering mask, tinned soldering eyes, and all the holes drilled) as well as the aluminum case with a 4 mm (!) front- and a 2 mm back-plate with all the holes milled. Although construction is rather straightforward (a component plan is added), this project is not intended for real novices. If you doubt your own skills please contact the author for the possibilities to obtain a finished device.

Well, I never heard any of the HeadRoom systems, but the reviews of these systems very well describe the impressions that I have with my own amplifier. The sound is not really externalized out of the head, because pinnae-reflections and reverberation are missing. However, the soundstage in our head becomes more continuous and has more inner-logic. With some recordings the effect is hardly heard; with other recordings, the effect is rather extreme. All recordings can be listened to for much longer periods of time, without your head screaming “Take those headphones off!”

Have fun!

c. 1999, 2000 Jan Meier.
The author’s website: Meier Audio.

Addendum

3/19/99: Capacitor values in figure 1 corrected from uF to nF.

3/20/99: S2 in figure 6a now has a bypass position.

5/11/99: Updated amplifier and power supply schematics to simplify design, avoid excessive DC offsets at the amplifier outputs and provide for a more flexible bass enhancement control. Figures 6 – 9 modified.

5/27/99: Updated PC board and component layouts with 47 ohm resistor at preamp outputs (after the coupling capacitors) to prevent oscillation at RF frequencies.

9/7/99: Tomohiko Ishigami made this Pocket Headphone Amplifier (see the article by Chu Moy) which uses the acoustic simulator circuit by Jan Meier (see A DIY Headphone Amplifier With Natural Crossfeed). He reduced the gain of the amp to unity to minimize problems with noise, which he later traced to the CD player itself.
The larger case is from Radio Shack (RS 270-213).

I feel it is very good idea to use modular approach. I used separate board for crossover and the buffer itself. This way, I did not have to go crazy load all the parts on one board which will result in a hay wire. Also, this approach is useful when I was trying to achieve smaller size.

I was able to use 1uF polymer capacitor for input…. These are so tiny. It is made by Phillips and you should be able to find it in Digikey [Digikey part nos. shown below]. I used this same type for my crossover circuit allowing me to conserve a lot of space:

3019PH-HD 1uF Metal Film Box ( 10mm (H) by 7mm (W) by 6mm (L) )
3015PH-ND .22uF Metal Film Box
3011PH-ND .047uF Metal Film Box

 

tomo4.gif

11/23/99: Added calculations for time delay of crossfeed signal.

1/7/00: Several DIYers have installed Jan Meier’s natural crossfeed filter as a front-end to the pocket amplifier by Chu Moy. Jan offers these tips re: selection and placement of a volume control for this combination:

It all depends on the specific circuitry. Generally it might be better to place the pot after the filter instead in front of it. The influence of impedance changes might be less pronounced. A 10 kOhm pot will certainly be too small. 50 kOhm will be a kind of minimum I think. However, note that with certain opamps this will result in changing offset voltages, since the DC impedance changes with volume.

2/8/00: Added calculations for time shift of direct signal.

2/16/00: Tomohiko Ishigami wrote:

Here is a picture of Mr. Meier’s stand alone acoustic simulator unit which I use with my Melos headphone amplifier. It sounds very transparent having almost nothing in the signal path. It is so small you can hide it anywhere. I just wanted to show you since it looks so cute!

tomo7.gif

5/1/00: Jeff Medin‘s version of the Pocket Headphone Amplifier (see the article by Chu Moy) has 3 sections: a gain stage, the crossfeed filter by Jan Meier and an output buffer stage. The power supply creates a virtual ground with a Texas Instruments TLE2426 voltage reference instead of a resistor divider network. The 1uF (or less) capacitors are Philips box-type metal film; capacitors larger than 1uF are Panasonic FC/Z series. All resistors are 1/4W Yaego metal film. Medin writes:

This is the FIRST amp I built after discovering HeadWize. It is a “basic” pocket amp with the natural crossfeed circuit by Jan Meier. ALL parts are from Digikey. It has very good decoupling with 3 capacitors per opamp and 3.9uH chokes (the 4 green things that look like resistors – they are connected in series with each V+ and V- lead). The first stage (on the left side of the first picture) is an OPA2132 with a gain of 10.

This then feeds a Meier crossfeed circuit (4 caps in a row) and you can see the crossfeed resistor on TOP of the board (2.2k) with long leads. The output from the filter feeds a voltage follower (OPA2132) stage. The switches are for low and high crossfeed, power, and bypass for binaural recordings. I used Philips Box style metal poly caps. The two large caps on top & bottom of board are 1uF input caps. The output is taken from the OPA2132… with a 100 ohm resistor… which is included in the feedback loop so it will drive very low z phones and to prevent oscillation due to capacitance from long cables. I used 100 ohm resistors in BOTH stages. If the resistor is OUTSIDE the loop, the impedance WILL have an effect on the sound of the phones, sometimes more bass, sometimes MUCH less signal based on the efficiency of the phones, etc. etc. Some phones as you know are spec’d to be run from an impedance of 100-150 ohms or so. I have a 15 year old APT/HOLMAN preamp (designed by same guy that invented THX-Tom Holman) and it’s Headphone Jack is driven by a 5532 with a 120 ohm resistor OUTSIDE loop right to the jack. I would suggest people can try both (like Jan did) and see what sounds better to them. I would DEFINITELY recommend that you include this resistor in at least the last stage.

Note that I did not have any problems, I always “over-build” opamp circuits so I don’t have to worry about problems later on. It’s just habit.

medin5.gif
medin1.gif
medin2
medin3.gif
medin4.gif

5/4/00: Added MS Excel circuit simulation application link and image.

1/16/00: Jan Meier has started a new company called Corda, which is offering a DIY headphone amp kit for approximately $250 US. This headphone amp has the latest circuit revisions as described in his article A Precision Preamplifier-Power Amplifier System with Natural Crossfeed Processing, such as opamps biased for extended operation in class A, ground-loop breaker topology, a crossfeed filter with 4 settings, and two headphone jacks with different output impedances. He writes:

The circuitry in the DIY kit represents the third generation of my original design. The output stage of each opamp is connected to one of the voltage rails by a 1.5 kOhm resistor. This forces the output stage into class-A functionality and increases sound quality considerably. Also 10pF capacitors are added to the feedback loop to increase stability at high frequencies. Careful matching of all resistors prevents offset voltages and the need of coupling capacitors. And the amplifier now is DC-coupled.

The power supply has a ground loop breaker, so the audio inputs and outputs MUST have floating grounds – their grounds cannot be directly connected to the enclosure. (See A Precision Preamplifier-Power Amplifier System with Natural Crossfeed Processing for more discussion about biasing opamps to function in class A and ground-breakers in power supplies).

The headamp has two sockets for connection of a headphone. Both sockets will provide different sound characteristics. One socket has a very low output impedance and gives the amp tight control over the headphone action. However, many headphones have been sonically optimized to be driven by an output impedance of 120 Ohms and may sound better when connected to the other socket. Generally, the low impedance socket provides a clean sound whereas the high impedance socket yields a warmer sound. Use the one you like most. There is no risk of damage to your headphone by connecting it to either socket. You can also use the sockets to connect two headphones simultaneously. However, the volume produced by the high impedance socket will be slightly lower than that of the right socket.

I know that it is cheaper to buy all the electronic parts by yourself, but please note that for the money you have a professional PC board added (with soldering mask, tinned soldering eyes, and all the holes drilled) as well as the aluminum case with a 4 mm (!) front- and a 2 mm back-plate with all the holes milled. Although construction is rather straight-forward (a component plan is added), this project is not intended for real novices. If you doubt your own skills, please contact me for the possibility of obtaining a finished device.

5/27/01: Headphone amp and power supply schematics updated to Corda HA-1 design. Added schematic for enhanced-bass crossfeed filter amplifier. Also added revisions in text discussing latest updates.