[SP Wild]

THE RESONANT FREQUENCY OF FOUR HEADPHONES

SOURCE:      Bryston BDA-1 DAC

AMPLIFICATION:     Lehmann Black Cube Linear

HEADPHONES:    HD595  HD650  K701  D7000

EXTRAS:        DC to 240v regenerator,  Furutech power lines,  AQ Cheetah IC

SONG USED:    My! My! Time Flies (And Winter Came, Enya)



What is the resonant frequency?
 
I believe that as with all objects, a headphone will contain a resonant frequency
which when impacted by sound waves will resonate back certain frequencies at
greater intensity than the rest of the spectrum.  This frequency thus affects, to a
significant degree, the tonal balance or "character" of a particular headphone.  
 
Due to this resonance - phasing annomalies will occur at the resonant frequency.  
It is this phasing annomaly, a form of distortion, that individuals will
subjectively find pleasant or unpleasant.  This phasing annomaly often enhance the
properties of "imaging" and "soundstaging" at the resonant frequency if it is
inherent in the design of a particular headphone.  To illustrate this hypothesis, I
have compiled my subjective experiences of four popular audiophile grade
headphones.
 
 
HD595
 
This headphone appears to resonate at "trombone" or "trumpet" frequencies (I
estimate these frequencies are surrounding the 1 khz mark).  It is no coincidence
that I find these instruments tend to stand out from the rest of the sound field.

Overall impression of these phones are good, not excellent as compared to other
phones in this comparison.  They can sound slightly congested and not quite as
transparent as the competition.  Soundstage is smallest of the bunch and imaging
(instrument location) is also more diffuse. Bass is adequate though not completely
filled out, due to the open design.  Treble is comparatively rolled of when
compared to the adjacent resonant frequency being very dominant, though the treble
is sufficient.  All these negatives are in comparison to pricier rivals - a good
headphone nevertheless.  As of now this headphone is eliminated by the competition.


HD650

This headphone is perhaps my personal favourite.  Without a doubt, this factor
may bias my opinions.  The RF (resonant frequency) is audible at the lower
midrange, approx 300hz - 600hz.  This RF sounds like the singer is a little too
close to the microphone if you do not like this "muddy" presentation.  For those
that do like this presentation it sounds like the singer is gently singing a
lullaby to the ears at an intimate distance.

These are excellent headphones.  Imaging is the best of the lot, soundstage is
second largest to the K701.  Treble is very good, at the same intensity as the
HD595 but subjectively more exquisite and extended, largely due to the lack of
resonance in the adjacent frequencies.  A full bodied and very transparent midrange
coupled to very articulate and full bodied midbass that is taut and of better impact
and presence than the HD595's.  Transients are slightly softened, helped by a
slightly less present upper midrange and treble as compared to the K701 - all
leading to a non fatigueing experience.


K701

This is a microscope of a headphone, often a favorite of mine for listening to
numerous recordings.  The RF seems to exist at the upper midrange 5khz and up all
the way to the lower treble.  This is the headphone that has the largest soundstage
of the lot and the most adept at recreating the large enclosed venue that a lot of
performances utilise when playing live.

The K701 involves the accurate rendering of musical details as amplified when
refracting off multiple solid surfaces at greater distances.  The imaging is 2nd
only to the HD650 - acceptable when considering imaging is also a little more
diffuse at larger enclosed venues.  Transient response is the best in this lineup,
aiding detail retrieval, also aided by its RF.  Often I view the treble response as
the best of the lot, more extended than the HD650 - However in this particular
setup and with this particular track - the lower treble along with the uppermid
range emphasis is obvious and detrimental to this piece and the rest of the album
also - on other recordings this is a none issue.  Bass is extended very low with
the least impact and presence - in line with a large venue performance.  Bass is
subjectively the most taut, possibly due to flaws that are hidden by the lower
volume.  Overall an airy open and spacious presentaion.


D7000

The most underrated headphone in the audio community.  I cannot isolate the RF of
this headphone after numerous attempts.  Being the only closed headphone I am
believing that all the frequencies are resonating within the earcups.
 
As part of the final top trio of headphones this headphone is the least
transparent.  I find the treble response mostly annoying due to excessive sibilance
in numerous recordings, so I was very surprised when it turned out this phone had
the nicest treble rendition in this track.  Exquisitely detailed and extended with
no trace of sibilance leaves the competition behind.  Imaging ranks third, possibly
due to the closed nature hence reverberance in every frequency clouding instrument
separation and positioning.  Often suffering from a slightly recessed midrange - in
this track it is not detrimental or even of significance - possibly because this
track does not convey a boomy bass on this phone.  The bass in many tracks can be
overpowering hence recessing the mids, but on this particular recording the bass is
bettered by none.  Tighter than the HD650 with a lower extension down to 20 hz with
no problems and at reference levels.  The fullest bodied bass with the most
realism.  Though less transparent, details are all present at all frequecies with
no trace of exageration with a soundstage only slightly smaller than the HD650,
though larger than the open HD595 which has dropped out of the comparison.  A
lively, punchy and full sized presentation.


Conclusion

This is not review hence no winners can be declared outright, though the HD595 did
lose.  This serves as an expression of opinion and as a means to encourage other
listeners of multiple phones to listen for these "resonant frequencies" and "phase
annomalies".  Each of the top three phones will excell and dominate other phones
depending on the source material and possibly mood.  I have enjoyed listening to
these phones and sharing my experiences at Head-Fi and I hope to get a camera one
day and post my setup along with a review of the BDA-1 DAC VS DACMAGIC in the near
future.  Cheers!

 

[Vergex2]

Interesting comparisons..

Though how can you say that the D7000s are the most underrated HPs on Head Fi (or the audio community in general)? They're mentioned with some reverence with much frequency.

 

[dBs]

I'm curious how you tested and the numerical results received.

 

[SP Wild]

 
Quote:

Interesting comparisons..

Though how can you say that the D7000s are the most underrated HPs on Head Fi (or the audio community in general)? They're mentioned with some reverence with much frequency.

They are generally accepted as "fun" cans, a definition that implies they are inaccurate in a pleasant way.  Far from it, these cans on this recording rendered the most accurate ruler flat frequency response than any other phone, resulting in arguably the most pleasant rendition of this particular track.  However I failed to mention that the D7000 were very diffuse at the extreme lower mid where the sound borders the upper bass this is obvious when different notes of different instruments get busy at that frequency - these phones then struggle - the other two open phones have no such dilemmas.  Nevertheless, far from being fun, I find these phones as a serious competitor with the most ruler flat frequency response by a large margin.
 
Quote:

I'm curious how you tested and the numerical results received.

The frequency resonances are but an approximation upon numerous listening with these phones for which I wholly own and have spent a lot of time comparing.  No equipment was used and level matching was approximated as I didn't feel minor volume variances affected the perception of the overall tone to the headphone.  My professional music experience is minimal with the only achievement of being a drummer in my high school band many years ago.  I reference most of my audio subjectivity with the audio memory of the repetitive rehearsals in a very intimate environment that did not necessitate high volumes in the percussion, electronic or acoustic instruments (The HD650 is my favorite for this reason).  I played live and on-stage only rarely and cannot rely on my memory for those audio comparisons, I can only estimate the sound of a concert-hall from an audiences perspective as I was never seated with the audience.

 

[Head Injury]

Quote:

The frequency resonances are but an approximation upon numerous listening with these phones for which I wholly own and have spent a lot of time comparing.  No equipment was used and level matching was approximated as I didn't feel minor volume variances affected the perception of the overall tone to the headphone.

1. How did you know that something was a "resonance" rather than just a spike or dip in the frequency response? In fact, looking at the Headroom graph, the K701 has a sudden recession right at the 5kHz mark, where you say it resonates.
 
2. Higher volume increases the perception of low bass and upper treble, so it could very much affect the overall tone of a headphone. Hence why certain headphones are considered good "low-level" headphones and others are recommended for higher volumes. Low-level headphones often have boosted treble and bass so their tone sounds more neutral at low volume.

 

[SP Wild]

 
Quote:

1. How did you know that something was a "resonance" rather than just a spike or dip in the frequency response? In fact, looking at the Headroom graph, the K701 has a sudden recession right at the 5kHz mark, where you say it resonates.
 
2. Higher volume increases the perception of low bass and upper treble, so it could very much affect the overall tone of a headphone. Hence why certain headphones are considered good "low-level" headphones and others are recommended for higher volumes. Low-level headphones often have boosted treble and bass so their tone sounds more neutral at low volume.

I don't know for sure if its a resonance or not - I may well just have an over creative mind.  But imagine we were to hang the naked drivers to our ears, I am sure the sound will very different to the actual phone.  So this thread may well be stating the bloody obvious when I think about it as the headphone construction is what controls driver resonance. 
 
With the HD595, I think its biggest weakness is the inability of the frame and cups ("chassis torsional rigidity" in car speak) to control vibration which is really obvious when compared to the higher end headphone.  As with the K701 I may be wrong with the 5khz estimation as I do not have measuring equipment, suffice to say the resonant peaks are at the upper midrange all the way up to the lower treble.
 
I guess I have been caught out with the volume issue.  When I said that I approximated volume matching, which I did at some point, my notes were based on the volume that I was comfortable with on each headphone.  The D7000s were played at the loudest, the HD650 less and the K701 lesser still.

 

[SP Wild]

Get this guys, I have some interesting findings that I found out right this minute. 
 
I was just curious about the damping issue of the HD595 and how it vibrates, I was gently tapping at the frame for looseness when I noticed that the thud that sounded from the tapping on the fleshy part of my finger had a tone.
 
Without music I can hear the tone of the chassis by taping on it gently, all over the earcup, all over the frame.  These are my results:
 
K701:      Higher tone
HD650:   Lower tone
HD595:   In between the other two
 
D7000:    This one was very hard to understand, tapping the phone it resonated very little, the chassis was very stiff and vibration control at a level far better than any other, I couldn't quite understand which tone stood out.  "What frequency is here that is not in the other phones"  Then I got-it, it resonated at very low 20-30hz - very faint, but It is there,  this sensation did not occur in the other phones.
 
This test adds some credibility to the resonance hypothesis.  The tones being generated by the chassis of the phone itself - corresponds to its sound signature. 

 

[Roger Strummer]

Now I see what you are trying to see, it was so not clear from your first post, I almost dismissed it thinking you were just talking about the frequency response.
 
It is a very interesting point, and your hypothesis seems very reasonable. To give it a more scientific approach I wonder how good one or more accelerometers  would be, it would accurately tell us what that frequency of maximum vibrations of the frame, and then we could see how correlated that is to peaks we see in the frequency response. Yet another fun project... I think I'll do that sometime next week.

 

[GaryJW]

Quote:

Now I see what you are trying to see, it was so not clear from your first post, I almost dismissed it thinking you were just talking about the frequency response.
 
It is a very interesting point, and your hypothesis seems very reasonable. To give it a more scientific approach I wonder how good one or more accelerometers  would be, it would accurately tell us what that frequency of maximum vibrations of the frame, and then we could see how correlated that is to peaks we see in the frequency response. Yet another fun project... I think I'll do that sometime next week.

I've seen this done on satellites (a little larger than headphones I agree).  Take a satellite and place it on the world's largest subwoofer and do a frequency sweep to check that the vibration modes are where they are expected.  (I say subwoofer deliberately, when I asked, I was told that the vibration table (I guess 4m x 4m in size (from memory)) was actually driven by a large set of coils. I am not 100% sure how the response was measured but may have been as simple as a microphone (since one could hear the resonances) or maybe through reactance measurements through the coils themselves.
 
An equivalent approach with headphones would be to use an external speaker to provide the sweep tone and a (small) microphone clamped to the headphone shell.  Calibration would take some doing....  (I would think use a sheet of metal of a simple shape for which the modes are easily calculable/predictable and use that as a reference).

 

[iponderous]

With respect SP Wild, I think that you are tilting at windmills here. The sound being produced by the headphone driver would overwhelm the negligible sonic influence of the headphone's "resonant frequency" as you describe it.
 
Closed headphone shells have dampening material, earpads, bits and pieces screwed and glued on to them. All of these additions would dampen any innate resonance that the material being used to form the shell might have. Your theory looks even shakier when applied to an open headphone.  
 
Tapping the outside of a headphone and hearing a tone is not analogous to tapping a head on a drum and listening for a note. Drums are empty vessels full of air. Tapping on the batter head moves the air between it and the resonant head, which causes the drum shell to vibrate or resonate in order to produce sound. It is free of internal dampening material to maximise resonance. Drummers usually find themselves adding dampening material to reduce the drum's resonance, and the unwanted overtones that it produces.
 
Whereas headphones are designed to minimise the unwanted resonances that could colour the sound between it leaving the driver and reaching the listener's ears, which is a very short journey indeed. I suspect that the wood used in closed headphones such as the D7000 and the W5000 is there to serve a cosmetic purpose rather than an acoustic one, regardless of what the companies marketing would have us believe. Those are the flagship headphones for Denon and Audio Technica so they should at least look the part.
 
A brick has a tone when tapped as does a piece of iron. So what? Wrap some cloth around them and their unique "resonant frequencies" have been silenced. I'm afraid that it's back to the drawing board for you.
  
 
​

 

[SP Wild]

 
Quote:

With respect SP Wild, I think that you are tilting at windmills here. The sound being produced by the headphone driver would overwhelm the negligible sonic influence of the headphone's "resonant frequency" as you describe it.
 
I call it how I hear it.  I hear resonance in the K701, HD595 and HD650 when playing music and it seems to correspond to the resonance when tapping the phone.
 
Closed headphone shells have dampening material, earpads, bits and pieces screwed and glued on to them. All of these additions would dampen any innate resonance that the material being used to form the shell might have. Your theory looks even shakier when applied to an open headphone.  
 
A large part of the LA5000/LA2000/LA7000 mods revolve around dampening of resonances.
 
Tapping the outside of a headphone and hearing a tone is not analogous to tapping a head on a drum and listening for a note. Drums are empty vessels full of air. Tapping on the batter head moves the air between it and the resonant head, which causes the drum shell to vibrate or resonate in order to produce sound. It is free of internal dampening material to maximise resonance. Drummers usually find themselves adding dampening material to reduce the drum's resonance, and the unwanted overtones that it produces.
 
Yes, dampening is a process of controlling resonance.
 
Whereas headphones are designed to minimise the unwanted resonances that could colour the sound between it leaving the driver and reaching the listener's ears, which is a very short journey indeed. I suspect that the wood used in closed headphones such as the D7000 and the W5000 is there to serve a cosmetic purpose rather than an acoustic one, regardless of what the companies marketing would have us believe. Those are the flagship headphones for Denon and Audio Technica so they should at least look the part.
 
I argue that phones are engineered to resonate at certain frequencies in order to enhance reproduction.
 
A brick has a tone when tapped as does a piece of iron. So what? Wrap some cloth around them and their unique "resonant frequencies" have been silenced. I'm afraid that it's back to the drawing board for you.
  
You have dampened its resonance and created another.  You are quick to put me down.  Just in case you believe I am a Denon shill - for the record I am a HD650 person.
​

 

 

[iponderous]

^ I'm not putting you down SP Wild. I'm just challenging your hypothesis, which strikes me as implausible. 

 

[Slaughter]

 

[tigermilk]

 
Quote:

I call it how I hear it.  I hear resonance in the K701, HD595 and HD650 when playing music and it seems to correspond to the resonance when tapping the phone.

If "tapping" on the phone like an impulse, you are exciting all frequencies at once.  Unless you have "magic ears" and a mind that can pick out the frequencies, this is meaningless.  Measurements (e.g., microphone for acoustic resonances or accelerometers for structural responses) are required to come to a hypothesis.  Otherwise you are just hearing things...

 

[Roger Strummer]

 
Quote:

I've seen this done on satellites (a little larger than headphones I agree).  Take a satellite and place it on the world's largest subwoofer and do a frequency sweep to check that the vibration modes are where they are expected.  (I say subwoofer deliberately, when I asked, I was told that the vibration table (I guess 4m x 4m in size (from memory)) was actually driven by a large set of coils. I am not 100% sure how the response was measured but may have been as simple as a microphone (since one could hear the resonances) or maybe through reactance measurements through the coils themselves.
 
An equivalent approach with headphones would be to use an external speaker to provide the sweep tone and a (small) microphone clamped to the headphone shell.  Calibration would take some doing....  (I would think use a sheet of metal of a simple shape for which the modes are easily calculable/predictable and use that as a reference).

VERY interesting stuff, thanks, I'm seeing what I can find about previous measurements of vibration modes of speakers. I'm doing the approach you suggested, just instead of using a mic I'm using accelerometers right now since it is what I have immediately available, but shortly thereafter I'll get mics, and maybe I can do something for measurements of the coils (its sweet having friends with good equipment or access to it).
 
I think I'll make a thread in the sound science forum to post the (rough and preliminary) first results I get and expose my thoughts of whatever can be read of the results, and also see what Tyll  might say about all this (although I bet he's very busy right now with the measurements at CJ).