by Jan Meier
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) + a .Vright,in(t-t0) a<1
Vright,out(t) = Vright,in(t) + a .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-t0) a <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.
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!
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.
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
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.
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)
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(w C1R1R2/(2R1 + R2)) and
t (time shift) = f /(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
Excel circuit simulator.
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.
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.
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!).
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).
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.
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).
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.
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!”
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 by 6mm (L) )
3015PH-ND .22uF Metal Film Box
3011PH-ND .047uF Metal Film Box
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!
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.
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.