D.I.Y. headphones

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.


post separation
On this page measurements of:

‘Stratocaster’ headphones
Fostex T50RP with HM5 pads
Fostex T50RP with Alpha pads
Fostex T20RP with Alpha pads
Beyerdynamic DT250-250 with original pads
NAD RP18-Kapton with HM5 pads
Sennheiser HD560-Ovation with custom pads
It should be noted that since these were measured ‘Stratocaster’ has made his own measurement rig and results have improved by leaps (no plots of newer versions)


 NO SMOOTHING nor HRTF ‘corrections‘ are applied to the shown plots.
Most other sites have between 1/3 to 1/6 octave smoothing applied which ‘irons flat’ sharp peaks and ‘wiggles’
Sharp peaks and wiggles MAY come from the measurement rig and/or from the headphones themselves.
I choose to show plots without smoothing and has to be taken into account.
A horizontal line in the plots shows equal soundpressure with a margin of a few dB.

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

post separation

T50RP in Karelian Birch wood with HE300 headband and HM5 pads.

T50 HM5-s

Pads are soft and comfortable. Clamping force and weight is high. The pads compressed too much and my Pinnae were touching the double layer of foam tucked inside to reduce the treble. Without these foam discs that problem was solved but the treble was too much.

With the foam inside the treble isn’t spiked any more so this foam works well. It sounds well expanded to both sides of the spectrum but bass reproduction is slightly off.

Best way to describe it is when a bass ‘walks’ over its tone ladder some notes are emphasized more than others.

It also sounds a bit ‘smothered’ at higher SPL as if the driver is ‘pinched off’ too much and can’t easily make the needed excursions. In technical terms it has higher distortion in the bass (didn’t measure it).

Surprisingly good extension and ‘presence’ compared to stock T50RP’s. It lacks some clarity and air compared to my reference. This version is definitely much better than a stock T50RP. Very smooth with no piercing treble.

Only gripes are bass that is slightly ‘off’ and weight and clamping force. The ‘missing’ top-end isn’t obvious, it is probably masked by the 10kHz peak. The small lack of clarity isn’t obvious nor deal-breaking when listened to stand alone.

Frequency response shows a slight bathtub curve, somewhat elevated bass and treble

T50RP-HM5 freq

CSD shows clean treble  but damping below 2kHz leaves something that may be improved upon.csd r t50rp-hm5

post separation

T50RP in black and white ebony wood in unknown headband with Alpha pads.

T50 Alpha-s

The Alpha pads are very comfortable. The clamping force is somewhat lower than the HE300 headband.
The weight is about the same as that of the HM5 padded one but for some reason (maybe the larger pad surface and or headband construction ?) not feeling as heavy.

This is closer to my ‘reference’ in sound signature and has better clarity than the HM5 pad version.
This is the best ‘all acoustical/mechanical mods’ T50RP I have heard till now and comes very close to my filter corrected T50RP. Good extension to both ends of the spectrum and no obvious lack of ‘air’ and ‘presence which is what characterises most stock T50RP’s.
Also the bass is better than the HM5 padded one but also shows the ‘pinched-off’ bass character (higher distortion), in a lesser degree though. Very smooth natural sound.
Alas the balance is slightly off to the left if that were cured I could be perfectly happy with it.

T50RP Alpha

The balance is off by about 4dB on average, drawing the stereo image to the left. Tonally pretty good and once adjusted for balance is quite enjoyable. VERY flat for a mechanically/acoustically tuned T50RP. The sharp dip at 7kHz doesn’t seem audible and may be caused by the angled pads.

csd t50rp-alpha r

Nice and clean CSD, the mids are also much more controlled in this version.

post separation

T20RP in black painted ebony wood in HS50 headband with Alpha pads.


The clamping force is slightly above what I find pleasant. The weight is also on the higher side of comfortable but figure I could listen to it for a longer period of time.I refer the yellow wood colour over all black myself.

I would say very close to the T50RP with Alpha pads but with a slight ‘edge’ and slightly less smooth sound but tonally balanced. Bass suffers from the same ‘distortion’ but in a somewhat lesser degree and imbalance. Good extension to both ends of the spectrum and am missing ‘nothing’. The treble is somewhat peaky though and fatiguing after some time.I would rate it just below the Alpha padded T50RP because of the small ‘edge’ (most obvious in strikes of drums and cymbals). Slightly ‘rougher’ sounding.

There is some imbalance in the lows only, this is audible by the bass being shifted to the right somewhat while the rest of the stereo image is normal. The 5kHz peak is probably giving this headphone its ‘edgy’ treble with some recordings.T20RP Alpha

The CSD is quite clean except for some very short lived ringing. I think the peak in the frequency range is more detrimental to the sound than that very short lived resonance.

csd t20rp-alpha r

post separation

Beyerdynamic DT250-250 in Karelian Birch wood and original headbands.


Lighter in weight than the other woodies and good clamping force. The velours pads of the HD650 are slightly more pleasant to the touch than the Beyer ones.

Tonally closest to my own reference. Very vivid and dynamic sounding.
Perfectly balanced sound. Bass is on the lean side and you hear it ‘drawing’ to the left.
Not as refined sounding as the orthos and my own reference but somewhat more ‘forward’ with some slight graininess to it.
Tight bass, toms and ‘clatter’ are very present which makes it open and vivid sounding I guess.
Extends to both ends of the spectrum. This helps in making a headphone sound realistic, and that’s what it does.
Not suited for bass heads.

The right channel is having slightly too little lows, the left channel somewhat too much. This may be due to too much damping on the right side or could be a driver issue. The frequency range itself is quite impressive and extends to both sides of the spectrum.DT250

The CSD is very clean with the exception of a ridge at 13kHz. Perhaps the ‘graininess’ originates there.

csd dt250 r

post separation

NAD RP18-Kapton in black and white Ebony, fitted with HM5 pads in original headband.

NAD RP18-s

Clamping force is leaning towards firm and a bit heavy in weight.

Tonally balanced, the midrange seems slightly ‘better’ than T50RP-Alpha, a bit more clarity.
Bass on this one is also somewhat ‘clippish’ but seems slightly better than that of T50RP-Alpha.
The cymbals are over-accentuated a bit, making it sound highly detailed.
No sibilance or harsh sound with the better recordings. A bit in-between the lusher T50RP-Alpha and the more direct DT250 treble.
Not as smooth/clean as T50RP though but certainly not rough sounding.
Mids are its strongest points. A very clean and pleasant sound leaning toward ‘cuppy/hollow’ sound in some recordings on direct comparison with a reference headphone.
Very open and not overly dynamic nor compressed sounding, just ‘right’ in the mids tonally albeit leaning towards some edginess with some recordings.
The Alpha padded T50RP is more bassy (warmer) and slightly smoother up top but the NAD sounds a bit more ‘detailed’ and ‘fresher’ be it somewhat ‘rougher’.

Frequency range right up to 4kHz is quite good. The dip at 5kHz and treble peak is less desirable.NAD RP18

Around 500Hz the NAD behaves better than the T50RP drivers BUT from 3kHz up there is obvious ringing. This may translate in feisty treble with bright recordings.

csd r

post separation

HD560 Ovation


Comfort wasn’t as high as HD650 but is very light clamping force, very light weight. The in appearance quite similar HD250 is the better sounding one.

I listened to it with the home made pads (leather on top of velour pads) which increased the upper mids.
It is a polite but not special sounding headphone that doesn’t appear to lack bass but obviously lacks bass extension.
No piercing highs and totally NOT having a vivid (dynamic) sound.

HD560 ovation

CSD is quite clean.

csd hd560 brown r

post separation
SFI-973F (home made ortho)

SFI-973FSFI driversSFI driver

I baptised this headphone SFI-973F as it consists of SFI drivers, DT990 housing, DT770 pads and DT330 baffles and it is Filtered.

below green trace:  SFI driver mounted on DT330 baffle in a DT990 housing using DT770 pads.
purple trace:    same but filtered with quite a complicated filter.
sfi dt770 stock vs filtered
Below the result of the filtered SFI driver (orange = R, purple = L), 1/3 octave smoothed.

sfi dt770 filtered smoothed 1-3

30Hz to 18kHz isn’t that bad for what WAS sold as a tweeter.
I don’t think they are on the market any more so more of a curiosity than a serious project.
The T50RP-mkII is better suited as parts are widely available for such a build.
Seeing as these drivers were sold in 32Ω and 120Ω versions with holes on both sides and also in 8Ω but with holes on one side only I suspect these drivers (32Ω and 120Ω) were not intended as tweeters at all but as headphone drivers.Stand alone I tried it with various pads and configurations but won’t measure nor sound well at all.I have to say with the filter (3 notch sections, had to add one to the filter PCB) it sounds excellent and with a slightly raised bass and treble it sounds excellent and I would even go as far as by saying it rivals the filtered T50’s !Have been enjoying it over the last few days.
It needs quite a bit of filtering to get it right but those ‘few €’ drivers sound like a > 1k€ system and is very comfortable.
Modified CHAmp (so it has sh!tloads of power but no configurable output resistance any more) + filter PCB + add-on
Those that may be interested in copying it can contact me for details (via contact form, reply or forum).
post separation


Yes… strange name and looks like an old Beyerdynamic DT990. Actually the drivers aren’t stock and come from a very cheap headphone of the brand ‘Skytronics‘ and is probably some cheap OEM driver. The headphone itself was pretty crappy and an old driver-less DT990 served as a ‘Fosterphone‘. Some older and thinner DT990 pads are used as they sound better than the newer pads. No… this headphone can not be bought anywhere and the only reason for posting this is because I can… Below the Frequency Response of this abomination.

skyerdynamic old dt990 pads

It sounds the way it measures (not surprisingly) bass-shy (thin) and a bit over-accentuated because of the rise around 2.5kHz. Not annoying though just a bit ‘fake’ clarity. It does sound airy and detailed but at the same time a bit boring by the lack of ‘body’ (lows). Jazz and instrumental actually doesn’t sound that bad but can have an unpleasant ‘edge’ sometimes. The reason for that is shown in the CSD below. Massive ringing from 2kHz to 4kHz. There are worse areas in the frequency range to have ringing though so isn’t as bad sounding as it looks.

l wf skyerdynamic old pads

Not amongst the greater headphones and a bit ‘boring’ and brightish/over detailed. Let’s just say the Fosterphone is awaiting some better drivers. Until then the Skytronics drivers have found asylum there.

post separation

back to measurements

  1. Raz says:

    I was wondering what is your test setup exactly? What software do you use? What is the signal path? Do you use a Foam Head for the headphones? If so, what microphones did you use in it?
    Thank you and keep up the great work! 🙂

    • Solderdude says:

      Some of the requested information is found here:

      pdf containing pictures and some tests:

      Click to access test_rig_tests.pdf

      pdf containing filtered amplifier schematics:

      Click to access WM61_mic_amp_B-PCB.pdf

      I use a home made rig, a picture of it can be found in a link on the page mentioned.
      It uses WM61A microphone capsules which are connected to a pre-amp with integrated correction needed for that microphone.
      The correction consists of a notch filter for high frequencies and a high pass filter.
      This way I get a line level input signal that is corrected to have a flat frequency response above 300Hz and an attenuated level for the bass.
      This means that if my plot shows a flat horizontal line it is perceived as ‘flat’.
      So these plots differ from those of Innerfidelity and Headroom as well as those from SBAF.
      My results are closer to those from Golden Ears.

      The rig is wood covered with dense rubber foam with mics flush mounted.
      WM61A -> pre-amp (filtered) -> EMU0204 (line level) -> PC with XP -> REW.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.