2-Way Headphones with Passive Crossover?

[mike7877]

So I've been thinking about the biggest shortcoming of headphones with dynamic drivers: HF response, usually above 3kHz or so.

For example, consider the design of the Sennheiser HD 650s - they're 42mm.

The voice coil is the red ring. This driver is something of a combination ring radiator/standard driver. I assume they did this to help with HF response and, also, physically flatten the thing a bit. With this driver, at least half of the highs will be coming from the ~21mm dome in the center which is somewhat centered on the ear canal - a bit of a compromise (but isn't that all speaker design is?...)

Back before I knew much about audio, I came across these Panasonic Shockwave RP-HT65 headphones:

You might recognize them c. 1992-2010 in 3 forms (foldable, foldable with inline pot for volume, not foldable...)

They sounded extremely good compared to every other headphone I'd heard under $30 (and many that cost more, though they might not have reached their SP levels).

I was listening to them recently, and I found that in the 3-12kHz range(especially 3-8), compared to my HD 650s, they are superior. Not all aspects, but frequency response flatness of the Panasonic brings its average performance in the 3-8kHz range higher than the HD 650s.

I have a pretty strong feeling that most of this benefit is just down to the smaller size of the Panasonic driver -it's 28mm.

I imagine that if the design for whatever's in the RP-HT65 was shrunk to 10 or 12mm (it might already exist in the earbud versions of the Panasonic Shockwave series lol), high frequency performance would be exceptional all the way up to 18kHz.


Sennheiser's 800 S driver looks like this:

And it is a 56mm driver...

It looks like there's a perfect spot right in the middle for a 10-12mm driver...


--------tl;dr

A 56mm driver with with a hole in it (see above) with a 12mm driver in its center: crossover frequency in the 250-300Hz range.
Why isn't it done???

The headphones could have an external box for the crossover - you plug the box into your headphone amp, and the headphones have a proprietary cable that goes from the cans to the box!

Has anyone thought of this before?
We do it with speakers... Imagine we were all still using 8 inch midbass drivers with whizzer cones on the 1.5" voice coils...

 

[DVDdoug]

I don't know anything about headphone design but a small driver can cover the whole frequency range if you don't need to fill a room with bass.

There are headphones that cover the whole audio range with good Harman Compliance (and more headphones with "bad" response/compliance).

And it doesn't seem that any one design approach is better than another. There are good and bad dynamic headphones, planer headphones and electrostatic headphones, etc. If you sort the reviews here by "recommended" that becomes clear. I also becomes clear that there is very little correlation between price and sound quality.

But the high frequencies get "screwy" because of acoustic resonances between the headphone and ear.

Understanding Headphone Measurements (video)

 

[mike7877]

I don't know anything about headphone design but a small driver can cover the whole frequency range if you don't need to fill a room with bass.

There are headphones that cover the whole audio range with good Harman Compliance (and more headphones with "bad" response/compliance).

And it doesn't seem that any one design approach is better than another. There are good and bad dynamic headphones, planer headphones and electrostatic headphones, etc. If you sort the reviews here by "recommended" that becomes clear. I also becomes clear that there is very little correlation between price and sound quality.

But the high frequencies get "screwy" because of acoustic resonances between the headphone and ear.

Understanding Headphone Measurements (video)

Yes, that too - it seems to be a particular problem for 7-11kHz. Increasing distance does seem like it would fix maybe 3/4 of the problem (simply put, from personal preliminary findings)

I think the problem with an increasing distance solution vs. what I suggested above is, ultimately, driver position relative to the ear becomes more and more important for imaging. Also the things get bigger and heavier... If you could position the drivers within the can portion x-y by about 15mm each way, that'd probably cover positioning for treble/imaging. Upper mids might suffer from that movement though. Also the driver might need to be even bigger to handle +100% power (I'd imagine the design with the cans being +60-80% diameter and depth)


I wonder what a crossover would look like for a 300 ohm woofer and 150 ohm tweeter crossed at 400Hz... Would they be small values? If the box can be small that's one thing, but if it'll need to be 10 pounds and 2 liters.... Lol
Then active would be the only option

 

[DVDdoug]

I wonder what a crossover would look like for a 300 ohm woofer and 150 ohm tweeter crossed at 400Hz.

Actually... You do need a larger inductor at higher impedance, and at lower fequencies. (But you can use smaller capacitors).

Inductive Reactance
Capacitive Reactance

 

[Curvature]

So I've been thinking about the biggest shortcoming of headphones with dynamic drivers: HF response, usually above 3kHz or so.

For example, consider the design of the Sennheiser HD 650s - they're 42mm.

View attachment 487589

The voice coil is the red ring. This driver is something of a combination ring radiator/standard driver. I assume they did this to help with HF response and, also, physically flatten the thing a bit. With this driver, at least half of the highs will be coming from the ~21mm dome in the center which is somewhat centered on the ear canal - a bit of a compromise (but isn't that all speaker design is?...)

Back before I knew much about audio, I came across these Panasonic Shockwave RP-HT65 headphones:
View attachment 487591

You might recognize them c. 1992-2010 in 3 forms (foldable, foldable with inline pot for volume, not foldable...)

They sounded extremely good compared to every other headphone I'd heard under $30 (and many that cost more, though they might not have reached their SP levels).

I was listening to them recently, and I found that in the 3-12kHz range(especially 3-8), compared to my HD 650s, they are superior. Not all aspects, but frequency response flatness of the Panasonic brings its average performance in the 3-8kHz range higher than the HD 650s.

I have a pretty strong feeling that most of this benefit is just down to the smaller size of the Panasonic driver -it's 28mm.

I imagine that if the design for whatever's in the RP-HT65 was shrunk to 10 or 12mm (it might already exist in the earbud versions of the Panasonic Shockwave series lol), high frequency performance would be exceptional all the way up to 18kHz.


Sennheiser's 800 S driver looks like this:

View attachment 487596

And it is a 56mm driver...

It looks like there's a perfect spot right in the middle for a 10-12mm driver...


--------tl;dr

A 56mm driver with with a hole in it (see above) with a 12mm driver in its center: crossover frequency in the 250-300Hz range.
Why isn't it done???

The headphones could have an external box for the crossover - you plug the box into your headphone amp, and the headphones have a proprietary cable that goes from the cans to the box!

Has anyone thought of this before?
We do it with speakers... Imagine we were all still using 8 inch midbass drivers with whizzer cones on the 1.5" voice coils...

It's been done by Sennheiser's former main designer, Axel Grell. There's no acoustical need to do so as far as I know. Issues in the high frequencies are not caused by breakup, but ear interactions.

HEAVYS - Headphones, Earbuds, Gaming, Audio, for Rock & Heavy Music

Forged for Superfans, Headphones with 2 frequency ranges and 8 speakers. listen to Heavy Metal the way it’s meant to be heard. Personalized sound field & design for optimized superior experience. Trusted by over 150,000 music fans. Engineered by Axel Grell. Headphones, tws Earbuds, Gaming...

www.heavys.com

These headphones are different because the HF drivers are meant to cause externalization and frontal localization. Whether or not they do so, I don't know.