Real Headphone Bass at Home
Mar 13, 2002 at 1:43 AM Post #16 of 44
haha!
biggrin.gif
That's pretty funny Chalky, er...I mean Dusty...
rolleyes.gif


Anyway, I think I understand where you're coming from now. Thanks for the further explanation.

Why are crossovers bad? Why do we not like them?
No, please, let me answer...
smily_headphones1.gif


The typical cheesebag bunko crossover of yesteryear messed up the sound through time smear by hosing up phase allignment of the audio.

Typical 2 or 3 way speakers had/have a woofer which is fed by a low pass filter with typically a one, two, or three, or four pole (6db, 12db, 18db or 24db per octave slope) cutoff. The midrange driver was/is usually fed through a pass band filter with cutoffs above and below the range of the driver's capacity, and then the tweeter is similarly fed through a highpass filter which cuts off on the bottom end below the tweeter's response capability.

The typical designer looks at the response curve and listens to each driver in a system, often using pink noise as a source to determine where to set the bounds of a driver's frequency response, avoiding resonance nodes and other response anomalies by keeping the "problems" outside of the frequencies you send the given driver.

If you design a great crossover, but push your drivers beyond where they perform well, or if you use great drivers, but skimp on the crossover design, either way, you end up with less than great sound.

Using a sub with cans...

Human hearing is very sensitive to phase allignment in the midrange and lower HF range. Even minor shifts of even a few degrees can be detected by trained ears. Minor phase shifts at low (bass) frequencies are much harder to detect. Most folks don't notice it until it begins to introduce comb filtering or other sonic cancellation effects.

Wavelengths of low frequencies are measured in FEET, not inches so for way more than adequate phase allignment between sub and can, things are just fine to have the sub located in the same room, but it is not necessarry to be sitting on it!
wink.gif
Since the K1000's are running full range (they naturally run out of gas below 45hz so that's how I chose that xover frequency for the sub) there is no alteration to the full range response characteristics of the cans.

BTW, the sub level is very low in amplitude to match well with the cans, so low in fact that even though it's noticable without wearing the headphones, it wouldn't bother anybody. When wearing the cans, you cannot tell that the sound is coming from an external source. It integrates very nicely. I'm using a "cheap" modified Sony SA-WM40. Can't go wrong for the bux.
see: http://www.epinions.com/Sony_SA_WM40_626383
http://www.audioreview.com/PRD_124407_2741crx.aspx

For the record, the K1000's drop off very rapidly below 40Hz due as much as anything else to self cancellation since they're bipolar radiators. (Like Maggies for your ears) The xover on my sub is 24db/octave. I've tried other subs including a very expensive REL with my K1000's and since the needs are so modest, the sub I have sounds absoultely as good and maybe even better than the REL for this application.

I have a couple pair of HD580's, a pair of HD600's and a pair of K-501's, and use the Max and the SAC to drive them besides the K1000's and the K1000 with the sub is absolutely more satisfying than any of the other options.

Well, that's my $0.02!
Cheers!
 
Mar 13, 2002 at 2:31 AM Post #17 of 44
Quote:

Originally posted by kwkarth
I'm using a "cheap" modified Sony SA-WM40. Can't go wrong for the bux.
see: http://www.epinions.com/Sony_SA_WM40_626383
http://www.audioreview.com/PRD_124407_2741crx.aspx

For the record, the K1000's drop off very rapidly below 40Hz due as much as anything else to self cancellation since they're bipolar radiators. (Like Maggies for your ears) The xover on my sub is 24db/octave. I've tried other subs including a very expensive REL with my K1000's and since the needs are so modest, the sub I have sounds absoultely as good and maybe even better than the REL for this application.


For those interested in following kwkarth's lead, the SA WM40 is on sale here for $129.

http://www.inetshopping.com
 
Mar 13, 2002 at 2:36 AM Post #18 of 44
Wow! Good price!
 
Mar 13, 2002 at 4:36 AM Post #20 of 44
Quote:

Originally posted by irlsanders
I may someday talk myself into a pair of Lowthers.
Note to Dusty, check these out. Also the nOrhs.


Thanks for that, am already aware of them (both), but I'm talking full-range. I really need something that goes down to 20Hz-ish. Quote:

That said, I don't really buy the folks who say they find nothing lacking in the bass from their cans.


Well, that would be me. I really don't (find anything lacking). Quote:

No set of cans with less than an 8" driver can produce realistic bass.


Sorry, that's just wrong. It has nothing to do with the size, it only has to do with how slowly they can move without too much DC-like current burning them out. Things that make sense in speakers don't have the same validity in headphones.

kwkarth -- thanks for all that, will have to digest. I would like the opportunity to actually hear such a thing, but alas, I do not foresee that in my future. I suspect my problem lies with that comb-filtering effect you are talking about. Every time I have ever accidentally left my speakers on while listening to headphones, it's been an unpleasant experience. Maybe I am incorrectly extrapolating this, and combining it with what you said about old crossover design (I thought it was in the theory, not in the implementation -- can you really design a crossover without phase problems? It was my understanding that you could not.), and leading me to this bias, but there I stand.
 
Mar 13, 2002 at 2:33 PM Post #21 of 44
Hi Dusty,
I wonder if what you were hearing when you left your speakers on was more of the unpleasant interaction between your headphones and the mids and hf spectrum? If you were to be able to filter out everything above 45Hz from your speakers, there shouldn't have been anything to conflict.

RE crossover design... Certainly in the digital domain, one can implement a crossover without deleterious effect. (of course we still have the brick wall in the D2A to contend with...

Modern day analog crossover design has indeed come a long way and from what I read, does a very credible job without too much effect on phase allignment.

There was a good article I just read last night in this month's issue of audioeXpress entitled: "e-Bass: An application of Phase Linear Sub-Bass Equalization" which discusses this in part. The article was a reprint from Electronics World Feb. 2001. According to the author, Graham Maynard, phase alignment even in low bass frequencies is important for ultimate realism! So I guess I have to start strapping my sub woofer to my head now...
tongue.gif


Cheers!
 
Mar 13, 2002 at 4:45 PM Post #23 of 44
Quote:

Originally posted by kwkarth

RE crossover design... Certainly in the digital domain, one can implement a crossover without deleterious effect. (of course we still have the brick wall in the D2A to contend with...



I use 'infinite impulse response' based digital equalizers when mastering, and the difference is incredible. Plain old EQ now sounds like a phasey swarm of bees to me. There is a new generation of digital speakers coming online now that will benefit greatly from advances in digital crossover design. Also with 96 to 192KHz sampling rates begining to become a reality, only bats will be concerned with where the brick wall is now.

Hey Dusty - I don't grok your reference to the drivers "moving slowly enough" to reproduce bass. If there is a burn out problem, how come we don't see ferreofluid cooled drivers, or heat sink fins on headphones? I supose that a slow moving driver is needed to produce long wavelengths, but bass has transients too. That adds up to the need for light, stiff, long excursion drivers in speakers. Also, I can hear (and feel) a quiet 40Hz coming from an 8" driver across the room, but NOT from a headphone driver. There is a quantitative difference in what's going on down there.
 
Mar 13, 2002 at 5:37 PM Post #24 of 44
Quote:

KW, the K1000s have what looks to be an LC network in each speaker panel. Any guess as to what it might be? I suspect either a high pass to limit driver excursions, or a notch filter to tame the driver resonance.


Yup it's there and if I remember right I read somewhere that it's used to voice the drivers. I've joked about moving the network outboard and using oil filled capacitors for better sound quality.

I have noticed at higher volumes the K1000's tend to get a little shouty sounding and have attributed this to non linearity in either the drivers themselves or the netorks. My bet is on the networks.
 
Mar 14, 2002 at 4:26 AM Post #25 of 44
Quote:

Originally posted by kwkarth
I wonder if what you were hearing when you left your speakers on was more of the unpleasant interaction between your headphones and the mids and hf spectrum? If you were to be able to filter out everything above 45Hz from your speakers, there shouldn't have been anything to conflict.


Heh-heh, that would be silence (I have Spendor S3/5's)...no, seriously, I match the Spendor's with a sub (Sunfire True Sub Jr.), and I don't know what I don't like about it, I just don't...will investigate it at some time in the future. Quote:

RE crossover design... Certainly in the digital domain, one can implement a crossover without deleterious effect. (of course we still have the brick wall in the D2A to contend with...


Meh, not to be a luddite or anything, but you have way too much faith in the digital domain. I'm a synthesist, and there's a huge difference between filters in physical modeling digital synthesizers and real analog synthesizers. Just for example. Quote:

There was a good article I just read last night in this month's issue of audioeXpress entitled: "e-Bass: An application of Phase Linear Sub-Bass Equalization" which discusses this in part.


Haven't finished reading that article yet. Hey, are you more organized than I am (I moved recently, and all my magazines are in storage)? There is another article in audioXpress, some time within the last year (earlier than November 2001, definitely), I think, where the author tries to compensate for crossovers by re-recording his CD's (the trick was, apparently to run them through the filters "backwards", I.E., in time, hence the inability to do this in real-time). I think he got it wrong, he tried to run one signal through both crossovers -- I personally think he needed to run them separately on a bi-amped speaker. Wouldn't mind knowing which issue this was so I know what to look for.

So anyway, why would the article need to do such a thing, if crossovers still have no negative effect? Quote:

Originally posted by irlsanders
Hey Dusty - I don't grok your reference to the drivers "moving slowly enough" to reproduce bass.


I'm saying that's all it takes. For them to move slowly enough. That's how bass frequencies are reproduced, by moving slowly in and out ("slowly" -- WRT treble frequencies, I don't mean in and out, like you can watch them move, I mean like on the order of 20 - 40 times per second, maybe even less). Quote:

If there is a burn out problem, how come we don't see ferreofluid cooled drivers, or heat sink fins on headphones?


Not sure about this. I'm not sure we don't see ferrofluid cooled drivers. Many contemporary drivers are ferrofluid cooled. As for the heat sink fins -- well, we don't see those on speakers either, now do we? Probably because they are so fragile (I'm not talking power amp heat fins, I'm talking enough to prevent the heat generated by running DC through the driver from building up and ruining the coil). Quote:

I supose that a slow moving driver is needed to produce long wavelengths, but bass has transients too.


No, it doesn't. Those would be higher frequencies. If you believe in Fourier's Theorem. Quote:

That adds up to the need for light, stiff, long excursion drivers in speakers.


Not sure about the long excursion part, but the rest of that statement is true. Quote:

Also, I can hear (and feel) a quiet 40Hz coming from an 8" driver across the room, but NOT from a headphone driver.


That's called the inverse square law. The amount of energy dissipates with proportion to the inverse of the square of the distance as you move away from the speaker. Why? Think about it, when you're right up against a driver, there's basically only one place you can have your ear. When you're 10 times further away, there's about a hundred places you can have your ear. The rules are entirely different when your ear is right up against a driver. I agree that they are quantitatively different, but a lot of the mathematical limits that are taken for granted in speakers are completely inappropriate in headphones -- I.E., you can't ignore (as much) the point- or line-source problem of multiple drivers.
 
Mar 14, 2002 at 7:04 AM Post #26 of 44
Yikes! Where's a good telephone when you need one! There are so many things I want to comment upon but am too lazy to type about them all.... Where to begin...

Sunfire 'eh? I never laid ear to one, but I've hear that they were good! And Spendors...Well it goes without saying... The Spendors are probably so phase coherent that when you try to mix the relatively slow Sunfire, you hear problems. Much the same with Maggies. They're so fast that it is VERY difficult to find a sub that properly mates with them.
Quote:

Meh, not to be a luddite or anything, but you have way too much faith in the digital domain. I'm a synthesist, and there's a huge difference between filters in physical modeling digital synthesizers and real analog synthesizers. Just for example.


Naw, I'm just an optimist at heart. I agree with you but some day we'll get it right! Need more bits, need more samples. Sound is more discrete than folks want to give it credit for!
wink.gif
We are talking WAAY more than just 20-20KHz to capture all the subtleties.
Quote:

Haven't finished reading that article yet. Hey, are you more organized than I am (I moved recently, and all my magazines are in storage)? There is another article in audioXpress, some time within the last year (earlier than November 2001, definitely), I think, where the author tries to compensate for crossovers by re-recording his CD's (the trick was, apparently to run them through the filters "backwards", I.E., in time, hence the inability to do this in real-time). I think he got it wrong, he tried to run one signal through both crossovers -- I personally think he needed to run them separately on a bi-amped speaker. Wouldn't mind knowing which issue this was so I know what to look for.


No, certainly not more organized, I just buy them at the news stand and then stack them when I'm done! X-over pre-emphasis sounds loony to me!
wink.gif


Quote:

I'm saying that's all it takes. For them to move slowly enough. That's how bass frequencies are reproduced, by moving slowly in and out ("slowly" -- WRT treble frequencies, I don't mean in and out, like you can watch them move, I mean like on the order of 20 - 40 times per second, maybe even less).


Well, that's not all there is to it. You must be a software or digital kinda guy!
wink.gif
There is the physics of moving al that air effectively. Compression and rarification. Ety's have a lot less air to move in your ear canal than any conventional can does and they in turn much less air to move than any conventional speaker.
It all comes down to how effectively the drivers can cause your eardrum to wiggle.
Quote:

No, it doesn't. Those would be higher frequencies. If you believe in Fourier's Theorem.


Allright, one of you guys is talking time domain and the other is talking frequency domain. We have waveforms and we have discrete frequencies. Can't we all just get along?
wink.gif


Bass reproduction had less to do with the inverse square law than the simple ability to move quanities of air. At very low frequency, the Q of any driver is very low, i.e. omnispheric radiation unless you horn load with a BIG horn!

Cheers!
 
Mar 14, 2002 at 5:09 PM Post #27 of 44
Quote:

Well, that's not all there is to it. You must be a software or digital kinda guy!
wink.gif
There is the physics of moving all that air effectively. Compression and rarification. ....
It all comes down to how effectively the drivers can cause your eardrum to wiggle.
Allright, one of you guys is talking time domain and the other is talking frequency domain. We have waveforms and we have discrete frequencies. Can't we all just get along?
wink.gif


Bass reproduction had less to do with the inverse square law than the simple ability to move quanities of air. At very low frequency, the Q of any driver is very low, i.e. omnispheric radiation unless you horn load with a BIG horn!

Cheers! [/B]


What I'm trying to get across is that bass is a physical presence in the room. A significant pressure wave is actually felt by your whole body, not just the fragile little cillia waving away in your eardrums. Etys or any can will fry your cillia to cinders by the time you've cranked them up far enough to fell the bass in your skull. Not to mention the fact that you now have little fireballs in your ears as the DC heat causes them to ignite!
biggrin.gif


The wind farms up in NoCal used to have an 'eggbeater' windmill design that put out a 3-4 Hz tone that was inaudible, but had both the amplitude and wavelength to compress peoples internal organs and make them sick. Now that's the kind of bass I'm talking about! As a tuba player, I may be over fond of the low range, but from experience I know a tuba doesn't sound very good in a small room, as its long wavelengths don't have space to form and propagate before running into boundarys that start messing it up. I suspect the same problem is in effect in the VERY small room between a headphone driver and your eardrum - even more so with a canal type set.

I recall a gimmicky old product called the BonePhone, or somesuch, that you wore around your neck. I was supposed to allow you to get tactile bass sensations, but really just felt more like a vibrator, which is fun in it's own right, but perhaps not approriate for this forum!

Anyway, this brings me back to my contention that headphone listening, despite all its many enjoyable strengths, will always have the weakness of non-immersive bass, which I counter through the addition of a sub when feasible.
 
Mar 15, 2002 at 12:23 PM Post #28 of 44
irlsanders,

if the headphones you have listed in your profile were the only options, I'd be just as critical of the bass performance of headphones as you are. As a matter of fact, I feel that 90 percent of all headphones leave a lot to be desired in that area.

But there are other possibilities. The Beyerdynamic DT 770 Pro has become a kind of reference to me when it comes to bass performance. It's a closed design that has more bass extension than most speakers and that has enough slam and conveys enough raw energy to rattle your skull. I am certain it doesn't reach as low as 4 Hz, but it definitely has ample response at 20 Hz. Incidentially, Dusty Chalk uses 770 Pros as well and is very pleased with them. I strongly recommend you give them a try.

Nowadays I feel that the Audio Technica ATH-W100 has even more realistic bass response, but since it is distributed exclusively in Asia, you couldn't try it before you buy it.
 
Mar 15, 2002 at 2:23 PM Post #29 of 44
Quote:

Originally posted by Dusty Chalk
No, it doesn't. Those would be higher frequencies. If you believe in Fourier's Theorem.


OK, I can't see how to quote a quoted section, but
the above refers to bass having transients.

Let me try to restate islanders point.

Bass notes have transients.

After Fourier decomposition these transients might
not show up as low frequencies, but they were made
by an instrument designed to create low frequencies (tuba, string bass, big drum etc). These transients
are crucial in my opinion. The impact of many
instruments comes from the transients. I'm not sure
that these transients are *always* higher frequency
than the fundimental. Idealy the transients should be reproduced on the same driver that makes the fundimental and harmonics of the note. Speakers and headphones are
inherently limited in repoducing real instruments.
Real intruments make all their own frequencies.
headphones and speakers try to make all the instruments frequncies at the same time. This leads to one instrument doppler shifting the sound of another. Most headphones, and some speakers, at least have the advantage of each intrument's sound coming out the same driver.

On headphones and bass: Headphones can never reproduce bass properly. Like islander and others
have said, bass is not just an ear thing, it is a body thing as well. Also it is difficult to
properly reproduce a bass note in a typical living room since the reflection hits you too soon. You'd need low frequency absorbing walls to get close and this isn't simple (cheap) to do.
Sitting on a subwoofer is no answer, it vibrates
the wrong body parts.

Life is a comprimise, so is sound reproduction.
There is no right answer, just the one that works
best for us.
 
Mar 15, 2002 at 4:39 PM Post #30 of 44
Quote:

Originally posted by dknightd

Life is a comprimise, so is sound reproduction.
There is no right answer, just the one that works
best for us.


Someone ought to breakout a hammer and chisel for this. You should consider using it for a sig, dknightd.
smily_headphones1.gif
 

Users who are viewing this thread

Back
Top