back to Superlux
back to measurements

post separation

NO SMOOTHING is applied to the shown plots. Most measurement sites have some smoothing applied which ‘irons flat’ sharp peaks and ‘wiggles’. I do not use smoothing because some info about sound quality is lost when plots are smoothed.

Aside from a small correction of the microphone itself also some correction in the lowest frequencies is applied to the plots to compensate for the perceived loss of bass when using headphones. This is described HERE in more detail.
A ‘horizontal‘ frequency response curve on the shown frequency response plots on this website thus indicates a perceived ‘flat’ tonal signature.

ALL measurements are made with a good SEAL on a flatbed measurement rig.
shape of your head, bone structure, pad size, pad ‘softness,  (compliance), hair or no hair and or wearing glasses may (drastically) change the frequency response of some headphones, so… your personal experience may differ substantially from these plots.

Frequency response (tonal balance) is the most sound-determining aspect of headphones. A horizontal line shows audible neutral response in the plots on this website. Deviations in different severities at different frequency bands have an effect on the sound character.
The bigger the deviation the stronger the effect.

Below an aid to help determining the sound character of headphones with relation to the frequency response.

sound descriptions mine

post separation
Superlux HD687


The Superlux HD687 has a similar build as the HD681-EVO, but that’s where similarities end.

The cable on the HD687 is detachable. It comes with a 3 meter and a 1 meter cable which can also be combined to create an even longer cable. Also a clip is included so you can fix the headphone plug to the cord and can’t pull it out any more. Personally a find the removable cable just fine as it is. It doesn’t come off easily and when the wire gets caught it unplugs itself.

Alas, the cable is microphonic. This means that mechanical contact sounds, when the cable rubs against clothes for instance, are transferred to the cups and can be heard during quiet passages and silence.
This headphone also comes with gold-plated 3.5 to 6.3mm TRS jack connector. A soft carrying pouch is also included.

The headband has a thin layer of rubber on the underside with ‘air filled cushions’.
The shiny surface of the headphone parts are
not fingerprint magnets but scratches will be quite visible. Looks-wise it seems to be targeted at younger people or the classier looking Hi-Fi market though.
People with large heads may not be able to correctly place this headphone properly on the head. Just like the HD681-EVO, HD-662-EVO and HD688 the height adjustment range is too limited.
The height can only be varied 15mm. Most other headphones can be adjusted between 25mm and 60mm. The most common adjustment range is around 30 – 40mm.

There are small L/R markings and Braille markings are present on the outside.
The impedance is around 40Ω which is excellent for connecting it to portable equipment.
The efficiency is high enough so it plays decently loud from most portable devices such as mobile phones and small players. It has large 50mm drivers which differ substantially from the drivers found in the HD681-EVO. The dome and voice-coil are a lot bigger (25mm ø vs 18mm ø).

The attenuation from outside noises is poor due to the semi-open design. The difference between semi-open and fully open is that in a fully open system you can clearly see the drivers from the outside and the rear sound is directly sent to the outside where in semi-open the rear sound is coming out though ‘ports’. Outside noises can come in just slightly attenuated. People around you can hear what you are listening to. Just slightly more attenuated compared to a fully open headphone.

The clamping force of this headphone was perfect and did not have to bend the rods as in the prototype. The clamping force can easily be adjusted by locally bending the rods. A video showing how to do this is found in this instruction video

They are a bit cheap plastic looking/feeling. The steel rods connecting the ear-cups are thinner and can be bent and twisted without things breaking off or able to bend back. cheapish look about it you can bend the headphone in all kinds of positions and it always comes back in the original position (within limits of course).
The headphone seems sturdier than it appears to be based on its looks. I have had reports of headbands breaking off from HD681-EVO’s though which share the same headband and connection to it.

In the end it is always about the comfort and sound. The comfort part is excellent ! NO complaints there from me, but my head is average sized.


Type: Over ear, semi-open
Usage: Home, studio, transportable.
Driver type: dynamic
Pads: replaceable, pleather with buit-in dustfilter
Collapsible: No
Headphone cup connector: 3.5mm TRS ‘dongle’
Cable entry: single sided (Left)
Cable: 1,8m terminated in 3.5mm TRS jack.
Driver size: 51mm
Inner Pad dimensions: depth = 20mm, diameter= 60mm
Max. power rating: 0.2W
Max. voltage: 2.8 Vrms (8Vpp)
Max. current: 70 mA
Max. S.P.L.: 115 dB
Impedance: 40 Ω
Efficiency: 92 dB/1mW (106 dB/1V)
Weight: 250 g. (excl. cable)
Clamping force: medium
Accessories: 6.3mm adapter, 1mm cable, 3m cable, connector clip, pouch.

Subjective sound description:

The HD687 has a full bodied and warm signature with a good quality and quantity bass.
Mids are warm and a bit ‘laid back’. These headphones are missing the clarity/presence the prototype, I reviewed earlier, had and even though they lack clarity it isn’t as severe that it sounds ‘scooped’.
The elevated treble gives the headphone a sharp and sibilant sound while at the same time is lacking clarity/presence.
For longer listening sessions this will result in listening fatigue.
When ‘busy’ music is played with lots of instruments the sound is muddled and finer nuances are completely gone and are replaced with harshness a bit.
This headphones plays a lot nicer at lower listening levels. At higher levels the quality becomes quite poor.

Below the frequency response of the HD687 with pleather pads. (Left and Right)

FR HD687 stock

Channel matching is excellent. Only some differences around 4kHz. Bass starts to roll-off from 100Hz and is -7dB around 30Hz. Mids start to roll-off from 600Hz and drop to -10dB around 4kHz. The upper mids thus are about half as loud as the mids. Because the drivers are not angled the dip will remain audible. The result is a very laid back sound which won’t become shrill not even when recordings are crappy.
From 6kHz to 10kHz there is a substantial peak (+10dB) opposite the lower mids. From there the level drops down to neutral levels again till about 16kHz. Above 16kHz the level drops off yet again.

Below the differences between the (stock) prototype and the (stock) production version.

HD687 Proto vs production

The prototype had much better bass extension and clarity. The prototype also had a ‘laid back’ character but not as severe. The peak around 16kHz resulted in a bit more ‘air’ in the prototype.

Fitting HD681-EVO velours (Superlux calls it Velvet) pads, as these are similar in size, will give better results. That would, however, mean that the drivers would be much more ‘exposed’ to dust and hairs as these velours pads do not have a dust filter integrated which the pleather pads do.
Below the response with velours pads and stock pleather pads.

pleather vs velours

A nice improvement in comfort and sound can be had with the velours pads. The 3kHz dip is relatively smaller as the bass and mids levels drop a few dB. The treble peak around 7kHz is reduced well over 5dB and even the 9kHz peak is lower.

output resistance / damping-factor

As this is a dynamic headphone the frequency response can be amplifier output resistance dependent when certain higher output resistance amplifiers are used.
Instead of showing impedance plots, which are hard to ‘read’ when it comes to assessing the tonal balance change in the real world, the HD687 is measured via a low resistance output (
0.2Ω) and via a high resistance output (120Ω). On a higher output resistance amplifier the output level will be lower of course due to voltage division. To compensate for this the amplifier is cranked up to the same level (12dB at 1kHz in this case). This way the plots are overlaid and it is easier to see how the tonal balance changes.

R 120

Driving it from a much higher output resistance amplifier will give it slightly more (1dB) bass.

Seal can be an issue with closed-back headphones but is usually less of a problem for open headphones. Breaking the seal (improper fit on the head, glasses, hairs) usually means a loss of (sub)bass. The HD687 is not different than most other open headphones in this aspect. Below the effect of a seal breach is shown.
Perfect seal, Seal broken with thick arm glasses.

seal loss

Below the distortion plot of the HD687

Dist HD687 R

Compare this with the distortion plot of the prototype HD687 below and it is easy to see that not only the frequency response ‘suffered’ from the redesigned (cheaper ?) driver but also distortion levels. Especially the 3rd harmonics (indicating compression/clipping alike non linearities) are considerably (15dB) higher. but also the 2nd harmonic distortion is about 10dB higher in the production model.


Below the HD687 distortion plot but shown in a percentage scale.Dist HD687 R percent

10% distortion in the bass area is audible. Bass harmonics (which are present in a real bass as well) are reproduced a bit louder.
One would expect the bass to sound distorted and poor in quality but in reality sounds ‘good’ at (to me) normal listening levels.
As high 3rd and 5th order harmonics are an indication of ‘clipping/compression’ by flattening of the sinewave it was interesting to measure at lower levels as well.
The reason FR/distortion is measured at a higher level is to create enough signal to noise ratio with the room around you and the mic/pre-amp/ADC.
Measuring at 90dB SPL gave: 8% (3rd harm) and 1.8% (2nd harm) at 40Hz
Measuring at 80dB SPL gave: 6% and 0.8% at 40Hz
Measuring at 70dB SPL gave: 1.1% and 0.4% at 40Hz
So the distortion in the bass rises dramatically above 70dB SPL (which is a decent loud level already) and reaches audible levels.  The HD687 becomes coarser and more gritty/harsher sounding at higher levels but sounds fine at softer to normal listening levels.

The actual 2nd harmonic distortion above 100Hz may well be lower than 0.2% and above 2kHz even be lower than 0.1%. A shortcoming of my measurement rig.

Below the CSD (Waterfall plot) of the HD687 with stock pleather pads. (left and right are superimposed)


It is obvious that there are quite a few resonances here. At 4kHz (the dip in the frequency response) it becomes evident that the dip is caused by a resonance.
Usually, short resonances in that part of the frequency range are not that audible.
The elephant in the plot is the peak at 7kHz that is definitely audible and can give a sharp ‘edge’ to some recordings.

Below the Spectrum plot which is more or less a waterfall plot but looked at from the top where the level is colour coded with the time progressing to the top of the plot. In the CSD the time-scale is 5ms, in the plot below the time scale is 50ms.

Spectr HD687 LThere is some low level ringing around 700Hz. The 4kHz resonance as well as those around 7kHz are clearly visible.

Step response is another way of judging impulse behaviour.
The plot below shows a step response of the HD687 (Left channel).

Step L HD687

This is a pretty poor response. While the initial rise is just -4dB, which isn’t bad, there is quite some resonances visible (7kHz) for quite a while. The drop-off from the trace is an indication the subbass extension isn’t that great.

With some very simple modifications (no need to open the  headphone) the sound quality (and measurements) can be improved.
All that is needed is 2 foam discs (AKG Foam Net Piece K-270/ K-271) + (AKG K-240/270/271 Velour Pads)
or (Superlux HD-681 Evo Ear Pads Velour which comes with 2 almost similar foam discs). And 1 sheet of 2ply or 3 ply toilet paper.
The toilet paper damps the treble peak a little and the foam + toiletpaper cover the ‘anti-bass port.
When this port is fully open the frequencies below 100Hz are reduced. When this port is fully blocked there is way too much and muddy bass.
By blocking it slightly a nice balance between bass level and roll-off can be created.
The foam disc+ toilet paper do this quite well.

Below the frequency response of the modified HD687 versus the stock HD687.

stock vs vel+TP+foam

Considerable more bass extension and improved treble and a more natural ‘warmth’.

below a similar set of measurements except with Superlux HD681 Velours pads + supplied foam discs and 1 ply of soft 2-ply toilet paper.
The soft type toilet paper sheet is cut to the same size as the foam discs. Pull the individual plies it is made off must be pulled apart. Not always easy to do.
Pull off the original pads.
Lay the round ply of toilet paper onto the driver. Put the foam disc on top (foam towards the driver) and fit the velourspads. Tuck the plastic retaining ‘flap’ into the small space between baffle and cup.

Below the frequency response of the modified HD687. (Left and Right)

FR modified

A little bit more channel imbalance but extended bass response. A slow roll-off below 40Hz. The ‘clarity/presence’ dip is somewhat less deep giving a bit more but below ‘neutral’ clarity. This does work well with lesser quality recordings.
The treble response is a bit smoother but slightly evelated around 9kHz which is audible as a slight ‘sibilance’ and treble tonality being a bit ‘off’.
The overall treble level is good though. Slightly ‘soft’ and pleasant in general.

Seal is a slight issue only. Breaking the seal just slightly (thin armed reading glasses) results in a barely noticeable drop in bass levels. Even when a seal is broken by thick armed glasses the loss of subbass is not bad at all compared to perfect seal.

seal breach velours

Also driving the modified HD687 from a high output resistance source (120 Ω) is not a very big problem compared to a low (0.2 Ω) output resistance.

120 Ohm Left

Distortion levels are not affected that much but are somewhat lower. At normal listening levels the sound does not sound distorted at all. These levels are high because the driver has a high sensitivity and are driven a lot louder in this test than they will be at normal listening levels.

Dist L modif L percent

The CSD below also is looking a lot better. (Left and Right superimposed)

CSD modif velours

The 4kHz resonance did not change but the ones around 7kHz have been reduced considerably.

The step response below shows ringing is reduced. Clarity and ‘attack’ is also improved and bass extension is much better.

step modif L


In stock form the HD687 is listenable at low levels albeit being a bit sharp in the treble.
Clamping force and comfort is decent. Adjustability for different head sizes is pretty poor. This headphone won’t fit properly or comfortable on all head sizes for sure.
Isolation is poor as it is a semi-open design. Those around you can hear what you are playing and ambient sounds are attenuated just slightly.
This makes it not very suitable for portable usage.
The distortion creeping in at more than decent levels makes this headphone not really recommendable for studio usage.
The sound is full and a bit bassy and ‘laid back’ in the upper mids. It is lacking clarity and presence but not in a muffled way. Treble is (too) accentuated and is sibilant. At higher listenening levels the sound becomes ‘gritty / coarse’.

Modifications as described above change the sound (and comfort also a bit) for the better. This is an extra investment of around € 10.- when ordering these from Thomann along with HD687 itself. Quite worth the extra investment.
The sound becomes more balanced. Is still bassy/full/warm sounding. The modified version is still lacking some clarity but not as much.
Treble quality is improved a lot. When this headphone is not played loudly the sound is quite pleasant and sounds quite good. Good dynamics and decent ‘sound stage’.

The HD687 is an improvement over the HD681-EVO to me and a different flavour than the HD662-EVO where the HD687 is a bit more ‘relaxed’ sounding.
Only suited for low to medium sound levels but works well in that range.


  • Sound quality when modified at low to decent volume levels
  • good sensitivity, can be driven directly from portable equipment.
  • 2 cable lengths supplied (no microphone nor remote) 3.5mm + 6.3 mm TRS jack.


Less positive:

  • The inside pad diameter is on the small side for some people.
  • Spacing between the driver and ear is too small for some people.
  • Height adjustment is way too little (just 15 mm), people with larger than average head won’t get a good fit.
  • Cable is microphonic (touching the cable is quite audible)
    Looks and build quality are screaming ‘cheap’.
  • can’t play loud so not suited for studio usage.
  • The champaign pink accents may not appeal to the majority of users.


post separation

back to Superlux
back to measurements