Head-Fi.org › Forums › Equipment Forums › Sound Science › Could a pair of headphones be deisgned like this to completely eliminate harmonic response distortion?
New Posts  All Forums:Forum Nav:

Could a pair of headphones be deisgned like this to completely eliminate harmonic response...

post #1 of 29
5/1/11 at 9:32am
Thread Starter 

If harmonic response distortion is the driver producing frequencies in multiples of the desired tone, couldnt a pair of headphones, or a speaker setup be designed to eliminate this by making it physically impossible for the driver to produce anything above a specified tone, and then incorporate multiple drivers? It would make more sense if i list driver numbers and frequency range.

 

Driver 1 - 10 Hz - 19 Hz

Driver 2 - 20 Hz - 39 Hz

Driver 3 - 40 Hz - 79 Hz

Driver 4 - 80 Hz - 179 Hz

Driver 5 - 160 Hz - 319 Hz

Driver 6 - 320 Hz - 639 Hz

Driver 7 - 640 Hz - 1279 Hz

Driver 8 - 1280 Hz - 2559 Hz

Driver 9 - 2560 Hz - 5119 Hz

Driver 10 - 5120 Hz - 10239 Hz

Driver 11 - 10240 Hz - 20479 Hz

 

Due to the massive quantity of drivers, this would best be reserved for speakers, or a custom IEM at the rate that they add drivers. But would a system like this theoretically make it impossible to experience HRD? None of the drivers support frequencies that are a multiple of other frequencies, so it looks like it would would to me, especially with an open pair of headphones. 

 

And an unrelated question to the rest of this post, why do headphone designers put a grill or a mesh on the back of open headphones? Also, why does nobody design a closed pair of headphones with the sound baffles so that no sound waves are reflected off the inside of the casing?

Reply
post #2 of 29
5/1/11 at 1:56pm

Interesting idea, i think it would work best for speakers. Do you know if someone has already tried this?

Reply
post #3 of 29
5/1/11 at 2:25pm
Thread Starter 

Well there are supposed to be a fair number of four way speaker designs, which, might ideally be set up like this to minimize THD (not HRD, my bad) -

driver 1 - 20 Hz - 2,499 Hz

driver 2 - 2,500 Hz - 4,999 Hz

driver 3 - 5,000 Hz - 9,999 Hz

driver 4 - 10,000 Hz - 19,999 Hz

 

Which theoretically eliminates harmonic distortion from 2,500 Hz and up, but anything below that is almost as prone to harmonic distortion as usual. You could probably adjust these numbers so that any given driver has a maximum possible frequency that is one less than 3 times the lowest frequency, which means that the lowest tone will produce 2 "spikes" on a THD graph, and everything else will only show one. However, these spikes would be the closest to the original tone, and as such, would be the most noticeable ones. Perhaps a design where You take a given driver, measure its THD, and then find the "distance" where it becomes nearly negligible, then make the driver service two frequency bands?

Reply
post #4 of 29
5/1/11 at 4:15pm

I thought distortion is caused because amplifer is not a perfectly linear device.  I've experimented in the lab, when we raise the gain up, distortions become apparent.  I don't know how this will effect amplifier's distortion.

Reply
post #5 of 29
5/1/11 at 4:26pm
Thread Starter 

Harmonic distortion is caused by the drivers when you put a 500 Hz tone through and the drivers in a pair of headphones or speakers produces tones other than 500 Hz, typically in multiples of the original tone. Go to headroom, click on a pair of headphones, click the graph, go to the bottom, select "harmonic distortion" and thats what im talking about.

Reply
post #6 of 29
5/1/11 at 4:30pm

Just out of curiosity, do you know how they measure the THD of the headphone drivers?  

Reply
post #7 of 29
5/1/11 at 8:01pm
Thread Starter 

Using a mic in a dummy head that may or may not amplify the effects? Is this a rhetorical question? The point isnt that you cant *hear* a clear 1000 hz tone if the headphones have bad THD, but it affects the way everything else sounds, theoretically making it less clear and / or coloring the sound. 

Reply
post #8 of 29
5/1/11 at 9:08pm

Most headphones of above average quality (and even some of average) already have below 1% THD.

 

This would best be used in the case of speakers as mentioned, but I imagine it would take a lot of work getting an active crossover and various amplifiers to do it.  Otherwise you risk the passive crossover adding distortion with that many xover points.

Reply
post #9 of 29
5/1/11 at 10:03pm
Quote:
Originally Posted by High_Q View Post

I thought distortion is caused because amplifer is not a perfectly linear device.  I've experimented in the lab, when we raise the gain up, distortions become apparent.  I don't know how this will effect amplifier's distortion.


The harmonic distortions produced by drivers are usually an order of magnitude superior to that of an amplifier.

 



Quote:
Originally Posted by Shike View Post

Most headphones of above average quality (and even some of average) already have below 1% THD.

 

This would best be used in the case of speakers as mentioned, but I imagine it would take a lot of work getting an active crossover and various amplifiers to do it.  Otherwise you risk the passive crossover adding distortion with that many xover points.


I don't even want to imagine the efficiency of such a speaker, ten way passive? say hello to your 70 dB/W speaker.

 

 

Reply
post #10 of 29
5/2/11 at 5:54am
Thread Starter 

Well the premise here is how to get that number to virtually 0%, not even as a viable idea, but as a hypothetical "ideal configuration".

Now how is it that crossovers work?

Reply
post #11 of 29
5/2/11 at 6:04am

The real trick would be creating said drivers.  I imagine it would be an amazing feat of engineering to fabricate a vibrating body that did not produce multiples

Reply
post #12 of 29
5/2/11 at 6:18am
Thread Starter 

So the vibrating body should typically create multiples anyway, independent of its "possible" frequency range?

Reply
post #13 of 29
5/2/11 at 6:43am

Not even the best digital active crossover slopes aren't steep enough. If we want, for example, that a driver would output a fairly even amplitude from 40-79 Hz, it will unavoidably also output almost the same amplitude at 39 or 80 Hz. It also seems logical to me, that the distortions would just add up between the drivers no matter how steep is the crossover slope and THD would be almost the same. Not to mention the utter impracticality and other inherent problems (time-alignment, phase etc).

Reply
post #14 of 29
5/2/11 at 8:32am
Thread Starter 

Well couldnt an active crossover altogether stop the current if it exceeds a certain frequency?

Reply
post #15 of 29
5/2/11 at 8:41am

Honestly, I'm not even sure why you'd need to do this.  Most high-end headphones seem to have a THD of >.1%.  I honestly doubt that human hearing can even detect that.

Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Could a pair of headphones be deisgned like this to completely eliminate harmonic response distortion?