The Historic Beyerdynamic DT 48 E (25 ohm)

Yes, historic. The Beyerdynamic DT 48 model has been in continuous production in a variety of forms since 1937. That's pretty historic ... I can't even remotely think of another headphone like that.

Let's take a look ...

Intended Application
The Beyerdynamic DT 48 E ($529 MSRP) is a full-size, sealed headphone intended for use as a monitor headphone in pro-audio studio and location applications. These cans are designed to take the rigors of the road, and they certainly will. There are two impedance variants of this model (25 ohm and 200 ohm); and there is an audiometry version, the DT 48 A.00.

Styling and Build Quality
These headphones have all the styling of a precision stainless steel, cap-head, shoulder bolt, which is to say, I love the way they look. It's not often that the discussion of build quality and styling are identical, but such is the case with the DT 48 E. Extremely sturdy nickel-plated spring steel bands for the headband, extension arms, and earpiece gimbals are firmly held together and pre-tensioned for use by a tidy assortment of bolts, screws, spacers, washers, and nylon-insert locknuts. This headphone is meant to be able to take a beating, and be repairable when over beaten. To see it is to understand it; its beauty is in its utility. If you have the fashion sense of a ham radio operator like I do, you'll love the look of these cans.

Ergonomics and Comfort
You WILL listen to these headphones. From the moment they apply their death grip on your head, and the suction cup-like seal around your ears is made, you know your dealing with a headphone that intends to be heard, and really doesn't care how you feel about it. I'm not saying these headphones were uncomfortable, in fact, I was surprised how comfortable they were considering how much clamping pressure they had. There have been warnings from owners to regularly check to drain sweat. I'd say they were very effective at sealing up your ear-holes with little collateral discomfort, but it did feel a bit claustrophobic to me.

Isolation
I wasn't a bit surprised to hear for myself and see in the measurements that these headphones had great isolation. Among the best I've seen a sealed, full-sized headphone --- that's a good thing for a monitor headphone. Coupled with the rugged build quality, their ability to provide the listener very good isolation makes these headphones very good for portable professional recording applications where intelligibility rather than fidelity is required.

Fig 1 show the isolation from outside noise. This is a very good result; possibly best in class.

COMPANY INFO
beyerdynamic Inc. USA
56 Central Ave.
Farmingdale, NY 11735
(631) 293-3200
ARTICLE CONTENTS

COMMENTS
dalethorn's picture

I found myself amused at several points in the text. I recall entering a discussion on Head-Fi for the first time, and stating that the DT-48 models circa 1974 sounded much like a string-and-can setup - more refined of course, but basically that. Which did not go over well. And I was astonished to find that they still make this headset, and that they updated the sound to near-hifi standards. Pretty amazing. I can scarcely wait to read the comments on the DT-48 forums. Did I say forums? As in plural? The mind boggles...

dalethorn's picture

Now that the engrams are coalescing, two things come to mind. One is the explanation for the sibilants - that's bothered me since day one and I couldn't understand why. Having used equalizers many years ago, where sometimes an adjustment had to be the reverse of what my instinct said, the explanation makes sense.

Two, in reading the "fully sealed" response graph, I see the main part of the midrange at -1 to -5 db approximately, then a gradual rise to approximately +8 db at 80 hz. I'm thinking that would make the headphones sound bassy - not deep-bassy with impact, but bassy nonetheless. Yet what I hear with the best seal I could get sounds like the opposite of that - more like a gradual dip of 10 db rather than a rise.

Edit: Would this mean that I can't get a seal on my head like the one that produced this response?

mward's picture

Very interesting history lesson. Thanks Tyll. Also enjoyed your brief comment about the current high-end kings from Sennheiser, Audeze, and HiFiMan. Hopefully you'll have more on the subject in the future?

svyr's picture

Thanks Tyll!

>Two, in reading the "fully sealed" response graph, I see the main part of the midrange at -1 to -5 db approximately, then a gradual rise to approximately +8 db at 80 hz. I'm thinking that would make the headphones sound bassy - not deep-bassy with impact, but bassy nonetheless. Yet what I hear with the best seal I could get sounds like the opposite of that - more like a gradual dip of 10 db rather than a rise.

+1. I always have to EQ up for 100hz down. (about 4db at 80hz, 6 for 60, 10 for 30)

>Would this mean that I can't get a seal on my head like the one that produced this response?

quite possibly. These are very sensitive on how you position them around your ears. If have to poke around at them for about 30 seconds making sure there's no channel imbalance because of how I put them on ...

>This headphone is meant to be able to take a beating, and be repairable when over beaten. To see it is to understand it; its beauty is in its utility.

Or you can beat someone up in public transport if they give you a hard time about how you look wearing a pair :D

>Among the best I've seen a sealed, full-sized headphone --- that's a good thing for a monitor headphone.

I think the only other HP I've seen with more Isolation is Pioneer Monitor 10. These had larger cups and about as much pressure on the sides of the head, so they may have provided a better seal. Plus the sound was a lot more humane and suitable for music...

lol... Denon D2000 for monitoring. That's.... And calling them sealed :D ...(bass heavy and next to no isolation)

other than that, very curious to see the 3-7k dip. and the 10k peak. I suppose that makes them bright and recessed in highs at once :D

Kool Bubba Ice's picture

Thank you for the review & measurements. I agree with a lot of what you said, but I obviously disagree with your comments on sound. I am a bit bewildered that you suggest the D2000 over the DT48 for professional monitoring. How long did you listen to these headphones? I posted your review on my DT48 Facebook page. I ask all those who choose to comment on my page, to do so in a respectful manner. I do know the DT48 is the industry standard for field recordings, as NAGRA fully endorses them. They are un conventional & do a lot 'wrong' in the minds of many 'audiophiles.' IMO, they are truer then life..

dalethorn's picture

At $379, about one fourth the price of the Sennheiser HD-800, I felt the DT-48 was a pretty good deal. I gave away an HD-600 and HD-650 recently, and both may have measured better than the DT-48, but neither one could touch the DT-48 on musicality and clarity from the lower midrange through the highs, given good quality recordings. It wasn't particularly subtle to me either - quite dramatic in fact. That is what I understand to be the consensus of many users of the DT-48 models.

Kool Bubba Ice's picture

Dale, IMO, the DT48 is on par with the 'ultra' high end headphones like the K1000 & HE6.. My 50's NAGRA was not embarrassed in anyway. Albeit, the other 2 are more audiophile friendly & is more conventional sounding. & the K1000 does better the DT48 in many areas.

iainmaitland's picture

Hello, I'm selling a brand new pair for $320!

200ohm version.

http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=190528392939&ssPage...

Thanks!

Leaffan's picture

KBI: Please keep your DT48 prosthelytizing off this site too....they might have sounded good back in 1937...but for crying out loud...it is 2011! The Ford Model T was a great car back in the day...but now...well, not so much.

The measurements are OBJECTIVE evidence of pretty poor sounding headphones...period. And my ears confirm this too.

Tyll is the REASON for our hobby and has heard pretty much everything out there....his comments on the poor sound quality of the DT48s mean a whole lot.

dalethorn's picture

Apologies if this seems unnecessary, but as a long-time audiophile and expert headphone reviewer, the DT-48E does not compare to a model T Ford - on the contrary, and despite weakness in the bass (which measures as strength ironically) the general impression I have of the midrange is that it's far better quality than anything else that sells at B&H etc. in or below its price range. Asking someone to back off on excessive posting may be a reasonable thing when arguments become repetitive, but suggesting that KBI and others are listening to 1937 tech is clearly wrong, as the DT-48 has been updated sonically on a continuing basis to this very day. I have no doubt having owned a few DT-48's that Leaffan and others *might* find the DT-48 sound to be weird or unnatural, having been conditioned to the more usual audiophile colorations in other headphones, and to that I say no problemo. But I do have a Sennheiser 800, I love the sound, I feel it's very close to a good and neutral signature, and I also believe that the DT-48 equals it with recordings that don't have strong sibilants or are dependant on a lot of bass impact. The specific comments that the article made about the sound (the sibilants and the light bass) were exactly what I and others have described previously, so we are hearing the same thing! But the measurements show a 10 db rise in the lower midrange to bass that neither Tyll nor I heard, since we did not hear increased bass, but rather decreased bass. When you read beyond the specifics to comments that "I didn't like the sound", you shouldn't give that sort of comment the same weight as the specific sound attributes like "bass light" and "exaggerated sibilance".

shamu144's picture

Thank you for those measurements Tyll !

Funny to see such an erratic behavior of the frequency response. I mean, going from -10dB at 60Hz to +10dB at 80Hz is pretty insane and should result very obvious to confirm in subjective listening test, still I can't recall anyone raising similar issues so far. The DT48 are rather heavy, could it be possible that the weight and clamping force combined together did excite some low frequency resonances of the head system ?

As reference, Beyer do facilitate a frequency response chart with each DT48A, and this 80Hz bump is absolutely non existant. However, the rest of the frequency response do indeed show a somewhat similar pattern (dip between 3khz and 7khz, then this peak at 10khz and later a new deeper drop at 15khz), but to a lesser magnitude degree.

If you have a few minutes, it would be very nice to read some comments regarding the other datas gathered and measured, for example the square wave reponse at 300Hz and 30Hz for the 120 ohms impedance output... What do they tell you?

Finally, I agree that the 120ohms output impedance won't make a big difference, but if you look carefully, bass response below 80Hz seem to increase by 2 or 3 dB... That is certainly enough to be noticed I would say. The square wave responses do also look much cleaner with the 120 ohms output impedance, or am I fooling myself here ?

Cheers.

dalethorn's picture

There was a guy in New York named Harry who used to walk around after Sep. 11 '01 with some handouts saing "History - don't let it be mystery." I like the slogan, which is pertinent here. For example, I hear 15 and 16 khz tones clearly and unambiguously on the DT-48E, but not so clearly on the Sennheiser HD-800. I'm not sure how that equates with the measurements, but I assumed it had as much to do with a possible irritation factor in sibilants as the "missing" lower sibilant info. Opinion?

dalethorn's picture

B&H, not known as a discounter, sells the 25 ohm DT-48E regularly for $379, which is a lot less than the $529 shown in the article.

Kool Bubba Ice's picture

I hope you don't find my response as defensive in any way.

I posted this on my page. I meant every word of it.

"Tyll is a true pioneer in terms of measurements, & is willing to measure any headphone. A true class act."

Kool Bubba Ice's picture

Hi dale.. The MSRP seems to vary.. I usually see them sell for 449.00. Beyers MSRP. A couple sites do sell them for a little over 500.00.. I never seen the DT48 sell for the price Tyll put in his review. B&H used to sell them for 313.00.. I hope I don't sound too narcissistic, but sometimes I think I'm the reason for the DT48's price increase. I do know many people bought the DT48 based on my recommendations & blogs. Around 25 or so since 08..

Frank I's picture

Your not serious are you. They way you used to post I really do not think anyone paid any attention to you.

dalethorn's picture

I hope someone else with a new DT-48E can confirm, using test tones such as the famous Mosquito tones, the response of the DT-48 from 15 khz on up. Given that the chart here shows anything above 11.5 khz as nonexistent, I'd welcome an explanation. I can see the tone files I have playing on Foobar, Winamp and others in their respective positions and amplitudes, so I'm confident of the frequencies I'm hearing. I don't have any particular concern for the DT-48 as an actual device or product, but given how much coloration and bad sound is being sold for big $$$ these days, I do have a concern about not being able to get decent sound for a reasonable price in the future. Right now I have a Senn HD-800, but when I'm old and destitute and living under the bridge, I may have to settle for a lower-budget headset.

svyr's picture

well noticed re: > 11.5k

dalethorn's picture

Scratch that goof - this chart apparently goes to 100 khz, which is unusual for a headphone. So apparently the DT-48 response dips and then goes back up around 15 to 16 khz, so that looks accurate. Now then, the bass. Unless I'm also reading this wrong, it looks like the response is down less than 5 db at 10 to 40 hz. That's interesting. By and large, the curve says "bassy" to me, with a maximum deviation of -11 to -12 db at 60 hz.

svyr's picture

mmm, it'd be good if learned to read the axis labels first :D ...
Oh well, yes, it looks like the second peak is around 15-7k...

maverickronin's picture

The link to the full pdf is broken right now and I'm sort of interested in the square waves.

svyr's picture

Tyll just copy pasted it in wrong ;)
http://innerfidelity.com/images/DT48E.pdf

maverickronin's picture

I missed that somehow. Its sure interesting though. The 30hz waves are about the funniest I've ever seen and from the impulse response the driver looks like it rings like a bell.

shamu144's picture

Note that the square wave response looks cleaner to me with a 120 ohms impedance output.

I am also quite convinced the dip and bump we see in the 60/80 Hz is a measurment artifact. It makes no sense and noone has ever reported such a behavior. Additionnally, Beyer provides a frequency response of the DT48A and the response is ruler flat from 10Hz to 1000Hz.

Finally, I believe Tyll you should use the Free Field HRTF correction provided by your equipment, and not the more modern one used. I can link you an AES paper where you will see the Free Field objective design goal for the frequency response of headphone and it looks damn identical to your raw measurments of the DT48E... I don't know how to insert images but if someone explains me, I will share it here.

Cheers.

svyr's picture

upload to head-fi and cross post using a std html image tag...

<img src="link to image"/>

shamu144's picture

There you go, thanks Svyr...

As you can see, absolutely no dip/bump in the 60/80Hz region... I really have no clue at to what induced this odd measurment in the lower frequencies, but it is certainly not the headphones.

Tyll says: "Here you can see that there is a return of bass energy when you do get a seal as noticeable in the curves below 60Hz --- better seal gave higher bass energy. But you can also see a resonant peak of some kind that moves between 80Hz (with a good seal) and 300Hz with a poor seal. Acoustic half-wavelengths at this frequency are roughly between 18" and five feet, so the cause of this feature is not likely acoustic resonances in the ear-cup. More likely is a mechanical resonance in somewhere in the physical headphone itself. "

Well, maybe just another clue that something mechanical indeed is not behaving properly, but that is not coming from the DT48. Just more food for thoughts.

shamu144's picture

Now, here is the raw frequency response of the DT48E measured by Tyll (grey lines):

And here is the design goal for a free field calibrated headphone, as shown in the AES paper and including an average of 40 subjects. Look at the right/bottom chart for open ear canal:

I see many striking similarities in both pattern above 1khz, with a 10dB bump in the 1khz to 5khz range, then a sudden big dip reaching -10dB followed by a sharp peak again around 10khz and so on...

For me, the DT48 appear as a free field calibrated headphone and measurment adjusted for the HRTF should apply proper correction to get meaningfull results.

Cheers.

inarc's picture

shamu144,

if you look closely, you will see noticeably differences between the AES paper's and Tyll's raw graph. AES' begins to rise at 1 kHz, Tyll's at 600 Hz; AES' peaks at less than +10 db, Tyll's at about +15 db; AES' begins to fall at 6 kHz, Tyll's at 3 kHz; AES' reaches the bottom (-10 db) at 10 kHz, Tyll's (-15 db) at 6 kHz; etc.. (All values approximate.)

Tyll Hertsens's picture
... in particular:

"I am also quite convinced the dip and bump we see in the 60/80 Hz is a measurment artifact. It makes no sense and noone has ever reported such a behavior. Additionnally, Beyer provides a frequency response of the DT48A and the response is ruler flat from 10Hz to 1000Hz."

It's likely they were using a standard coupler where you simply press an earpiece against it. In that case, the normal mechanisms of the headphones clamping pressure, seal, and positioning on the head do not come into play. I think the big artifact in the bass is a resonance of the whole headphones themselves vibrating, or just each earpiece as it sits on the ear, but in either case the headphones must be placed on a head as they normally would be then left alone. So, Beyer's curve being without the artifact doesn't surprise me at all because those devices have a little arm that clamps the earpiece to the tester.

This is exactly why I use a head --- so that the testing is done in a way that as closely replicates the actual use of the cans, and so that things like the actual seal against an actual ear become part of the measured performance.

"Well, maybe just another clue that something mechanical indeed is not behaving properly, but that is not coming from the DT48. Just more food for thoughts."

Yes, it absolutely is coming from the DT 48. It was quite repeatable. It doesn't exist on the other headphones data I took that day. The food for thought here is that you have to deal with the measurements ... they're real. What the artifacts mean is another story; that they're there is now established fact.

"Finally, I believe Tyll you should use the Free Field HRTF correction provided by your equipment, and not the more modern one used. I can link you an AES paper where you will see the Free Field objective design goal for the frequency response of headphone and it looks damn identical to your raw measurments of the DT48E..."

Free-Field HRTF models sound as it approaches your head from directly in front and in an anechoic environment. Music is designed to be played on speakers that are 30 degrees to either side, where the HRTF is substantially different from the Free-Field HRTF. Headphones are closely coupled to the ear and do not act at all like sound arriving from directly in front of you. Why would you want to use the FF HRTF? Those papers you quote are quite old, and there have been many competing ideas over the years. Why do you pick that one?

shamu144's picture

Thank you Tyll for weighting in.

Regarding those odd results measured for the low frequency, I hope you did not misunderstand me : I do not question your set of measurments and I am indeed convinced they are accurate and repeatable. I don't think repeating the measurments with another set of DT48E would yield different results. However, I do question how those low frequencies measurments "in the context of this dummy head and DT48" relate to real life experience with human heads, because even with a poor seal, I can't hear this "return of bass energy" mentionned, and I believe almost every DT48 owner will agree with me here. Are you indeed saying that you can hear a corelation between your measurments and subjective listening tests ?

The DT48A Beyer measurement reflect for me better what I am hearing in the 50Hz to 1000Hz range, even though I have to admit that their measurment below 50Hz do not seem to translate well to the subjective listening experience. Your hypothesis of standard couplers used by Beyer and non consideration of the seal effect might explain this.

Tyll says: "Free-Field HRTF models sound as it approaches your head from directly in front and in an anechoic environment. Music is designed to be played on speakers that are 30 degrees to either side, where the HRTF is substantially different from the Free-Field HRTF"

It was my understanding that the HRTF with such an angle of 30º would not be so different from the free field HRTF. Here follows the design goal for a free field response with an alternative angle of 22,5º (bottom/right), again, according to this AES document:

Basically, there seem to be some slight variations in the high frequency perception when comparing with the pure free field response.

If you look at the design goal for a diffuse field (bottom/right), you realize the theorical frequency response is then completely different as shown here:

We must consider at this point that the DT48 was engineered and designed more than 60 years ago... I bet you those new competing ideas on HRTF had not even been invented. Free Field and Diffuse Field was probably all there was to choose from. Why would you use a modern HRTF model for such an old headphone?

In fact, I feel this is a trully crucial aspect to understand the raw frequency response of any headphones: you must first understand what was their original design goal, be it free field, free field 30º, diffuse field, etc.... The raw frequency response measured will heavily depend on it.

I would go as far as to say that adjusting every headphones raw measurments with the same HRTF is leading to serious confusion, as people expect the "magical" HRTF adjusted frequency response to translate to their listening experience. Well, I understand this is simply not true. If a headphone was designed to recreate a diffuse field but that you apply a free field HRTF on raw measurments, you end up with a useless frequency response chart.

When I first looked at the DT48 frequency response, I was completely shoked by its aspect as the listening experience suggested a rather extremely flat frequency response... Now I really feel the free field frequency response comes very close to explain it.

Cheers.

Pages