Well, a measurement of the headphone should SHOW THE RESONANCE. Once I had a pair of Beyer DT-48S phones, that I flew to LA from SF just to buy in the 60's. They were 'clean' but bright in the midrange and I could hear HISS with these phones better than any other way. We measured these phones in 1974 in our lab in Switzerland with a B&K artificial ear that we had in the lab. I found this amazing midrange resonance that was both a notch and then a peak. It certainly set the midrange for these phones.
As with Demian, a colleague, and even more in-tune with modern headphone design, I hardly know what you folks have been arguing about for the last day or so.
Yesterday I heard a pair of Sennheiser headphones (very expensive) and found them free of any lack of smoothness, at least on a short term listening.
As with Demian, a colleague, and even more in-tune with modern headphone design, I hardly know what you folks have been arguing about for the last day or so.
Yesterday I heard a pair of Sennheiser headphones (very expensive) and found them free of any lack of smoothness, at least on a short term listening.
it does....
no idea what Frank has been arguing about, hes not heard them before...
not looking for smoothness, just lack of painfulness
Last edited:
It seems our ability to tolerate large accidents in response curves of headphones is amazingly large. May-be our brain, able to discriminate rooms resonances, is doing some auto correction because no room ?
More than that, if i can correlate the response curve of a loudspeaker with listening impression, it is not so true with headphones. I make big mistakes when i try to equalize some instruments with some headphones.
On the contrary, i'm unable to suffer any earphone and cannot afford closed headphone for a long time.
I wonder why it seems so difficult to build good headphones, while it seems easier than speakers (little membranes, low power).
The resonance is most likely a dimensional issue and may not be fixable with damping. Its so high Q I doubt you can actually hear it with music, maybe with an oscillator. The sound issues are more likely the back cavity material or the diaphragm. The big problem with diaphragms for these is that they are very thin to get the sensitivity and extended highs and they breakup at pretty low drive levels. I have not found a way to measure this since it changes a lot between on ear and free space. Cut and try seems to be the way to get a good sounding headphone. You won't get there with instruments alone.
If you look at the different plots even the very big change (1 CM pad) makes a very small change in the response and magnitude of the peak. The changes I saw could be from removing and replacing the headphone on the fixture with no changes to it. Usually you need to make several plots to be sure of your measurement.
The different pads will make a difference, more from absorbing energy in the cavity than anything else. There are some exotic damping materials that might work better. If you have the headphone apart the "tap test" may be the best. Tap and listen to the parts. Metal parts tend to be high Q and seem to be pretty audible. But plastic looks cheap. Fortunately we are getting plastics that really look like metal today.
Speakers as well. Welcome to www.SoundStageAV.com
Look through the curves and you will see little correlation between response and perceived quality. There seems a higher correlation between distortion and perceived quality even when the response is really weird: SoundStage! Measurements - Zu Cable Druid Loudspeakers (9/2002)
I suspect response issues are the same as color balance, The colors change when the light source is incandescent or open sky but you can still tell red from green from blue. Your hearing can pull the pattern of a voice and allow you to recognize who is calling regardless of the response variations and horrible distortions of a cell phone. Perhaps the best gauge of an audio system is the reaction in the first 100 mS before your brain has processed and removed the overall signature of the sound system. Listen to a speaker for an hour and it will sound pretty correct and whatever is next will be wrong. This may be why a-b tests are do hard to do.
I dont think its the resonance/peak thats actually heard directly, more a sort of anomaly; a diffraction/interference/cancellation caused by that resonance, sort of comb filtering and its the peaks and dips within all that complexity bouncing around and canceling out what should be there that is 'heard' sort of an absense of the bass or mids that should be there, rather than a peak in the treble, but translates as a peak/resonance in the treble.
I dont, nor would I ever own them, but i've spent a fair bit of time with different peoples (including several in my own home on my gear) and I have the same or similar issue with all of them. they are mostly a very rigid hitech plastic, except for the diaphragm and the piece that seems to cause the problem, which they have made an attempt to damp, maybe by ear so it worked for those in the test grpoup, or the designers themselves, but shows up randomly among users depending on the shape of the ear and its localized response.
dont know if that makes any sense, its odd, even without it I wouldnt buy them, but its my main critique of them.
I dont, nor would I ever own them, but i've spent a fair bit of time with different peoples (including several in my own home on my gear) and I have the same or similar issue with all of them. they are mostly a very rigid hitech plastic, except for the diaphragm and the piece that seems to cause the problem, which they have made an attempt to damp, maybe by ear so it worked for those in the test grpoup, or the designers themselves, but shows up randomly among users depending on the shape of the ear and its localized response.
dont know if that makes any sense, its odd, even without it I wouldnt buy them, but its my main critique of them.
Last edited:
http://audiosystemsgroup.com/Ferrites-Ham.pdf
Ferrites are often used between power interstages on digital boards, I have also seen them used on every power pin on some designs. Quite often used to filter the power for an oscillator, forming a pi filter with a couple of low value caps. Thats the quick answer, but they are used extensively and are almost a universal panacea for high frequency noise.
Some interesting views on power line communications, its adding EMC!
http://www.compliance-club.com/default.aspx?id=17
Another good article on ferrites, goto log in though:
http://www.bethesignal.net/bogatin/bts305-battling-ferrite-superstition-p-667.html
http://www.ipblox.com/papers.html about 5th down
Ferrites are often used between power interstages on digital boards, I have also seen them used on every power pin on some designs. Quite often used to filter the power for an oscillator, forming a pi filter with a couple of low value caps. Thats the quick answer, but they are used extensively and are almost a universal panacea for high frequency noise.
Some interesting views on power line communications, its adding EMC!
http://www.compliance-club.com/default.aspx?id=17
Another good article on ferrites, goto log in though:
http://www.bethesignal.net/bogatin/bts305-battling-ferrite-superstition-p-667.html
http://www.ipblox.com/papers.html about 5th down
Last edited:
The problem you are referring to is the "2" i listed.
Modulation of the signal, due to mechanical vibrations.
The causes can be different.
Bad contacts can produce R witch will vary with vibrations: Level modulation.
Capacitances (even parasitic) can change their value: Level plus/or Phase modulation.
Inductances (even parasitic) can change their values: Level plus/or Phase modulation.
All those distortions are due to a defect. THEY ARE NOT supposed to exist and have to be cured at the source.
Adding some mechanical isolation to an amplifier to cure any of those vibration problems is just like working on the suspension of your car when your reels are not well balanced: you address the problem the wrong way.
If vibration is a percieved problem use SMD...
What worries me more having seen vibration and ballistics tests on equipement (not valve obviously) is what levels are people listening to create measurable vibration distortion in their gear! and how near is the gear to the speaker. Myself I keep my valve amp well away from the speakers, and use longer speaker cable, best quality orange mains cable at the moment, recovered from the hedge trimmer that went up in smoke
Springs are not a good choice for vibration reduction, no damping, us AV mounts if you have to.
What worries me more having seen vibration and ballistics tests on equipement (not valve obviously) is what levels are people listening to create measurable vibration distortion in their gear! and how near is the gear to the speaker. Myself I keep my valve amp well away from the speakers, and use longer speaker cable, best quality orange mains cable at the moment, recovered from the hedge trimmer that went up in smoke
Springs are not a good choice for vibration reduction, no damping, us AV mounts if you have to.
Motional Feedback...
Amplifiers can have internal mechanical behaviors as has been discussed (heatsink ringing and component mechanical ringing) but higher up the list would be 'common' CD players that typically exhibit audible modulation due to vibration.
Stone kitchen top sink cutouts are a ready and zero cost high mass platform that works especially well when spring mounted (car valve springs).
Adding further mass (sand bags or stone shelf) on top of the component in question can totally ameliorate acoustic vibrational feedback.
Dan.
Hello Christophe, surely we all live in the real world, with real world components and real world construction techniques, and so we should expect some vibration dependencies in real world mass produced equipment.
Amplifiers can have internal mechanical behaviors as has been discussed (heatsink ringing and component mechanical ringing) but higher up the list would be 'common' CD players that typically exhibit audible modulation due to vibration.
Stone kitchen top sink cutouts are a ready and zero cost high mass platform that works especially well when spring mounted (car valve springs).
Adding further mass (sand bags or stone shelf) on top of the component in question can totally ameliorate acoustic vibrational feedback.
Dan.
Tubes are microphones because their parts (Plate, Cathode, grid) can change their distances when they vibrate. And, indeed, with them, you can change the color of your amp with different supports and dumping. I think this is the reason why people still believe in those magic cones
Modern electronic components are carefully encapsulated under high pressure. They use supports like ceramic, not subject to size changes. They resist very well to huge vibrations.
The use of printed boards, where the connections are not subject to changes their sizes or position is a guarantee too. We use them in rockets, imagine the "G"s !
If the wiring between boards is far away from steel covers, cables tight enough, there is no reason to get a change of their inductance/ parasitic capacitances neither.
The effects of those variations are proportional to the impedance: Modern amps use low impedance everywhere.
The parasitic inductances/capacitances of a modern good (fast) amplifier varies at very high frequencies (MHz) so, even little changes of their values with vibrations will not affect any signal in the audio bandwidth. Neither their level, neither their phase.
If we notice a change of the sound when we modify the feet of our amplifier/preamplifier, their is only two explanations: or we fool ourselves, or there is a defect somewhere inside (or in plugs), and this part need repair.
[edit]"heatsink ringing"
This can change the acoustic of your room, and you can listen to this, but, if power transistors are tightly mounted, there is no reason they suffer from this: they are NOT at all microphonic.
May-be Scott will tell-us more about the way active devices, like ICs or transistors behave with vibrations ?
Modern electronic components are carefully encapsulated under high pressure. They use supports like ceramic, not subject to size changes. They resist very well to huge vibrations.
The use of printed boards, where the connections are not subject to changes their sizes or position is a guarantee too. We use them in rockets, imagine the "G"s !
If the wiring between boards is far away from steel covers, cables tight enough, there is no reason to get a change of their inductance/ parasitic capacitances neither.
The effects of those variations are proportional to the impedance: Modern amps use low impedance everywhere.
The parasitic inductances/capacitances of a modern good (fast) amplifier varies at very high frequencies (MHz) so, even little changes of their values with vibrations will not affect any signal in the audio bandwidth. Neither their level, neither their phase.
If we notice a change of the sound when we modify the feet of our amplifier/preamplifier, their is only two explanations: or we fool ourselves, or there is a defect somewhere inside (or in plugs), and this part need repair.
[edit]"heatsink ringing"
This can change the acoustic of your room, and you can listen to this, but, if power transistors are tightly mounted, there is no reason they suffer from this: they are NOT at all microphonic.
May-be Scott will tell-us more about the way active devices, like ICs or transistors behave with vibrations ?
Last edited:
Free Tweak....
Dan.
Hi Marce, read my above post.....Its all to to do with mass and spring rate...in this case there is no unsprung mass, so the only ringing is in the spring itself !. I'll bet you have not yet tried valve springs (just ask an automotive head shop and they will throw them at you).
Dan.
Not so much, no. You can tap on nearly any engineered (as opposed to "designed") preamp or amp or DAC and have zero output from it. And that tap is a much higher amplitude shock than simple vibration though the floor or air.
I dont use anything at home, but have used proper AV (anti vibration) mounts on the equipement I help design (think high vibration, ambient 110-115dB background noise, metal walls). But mass on springs would help, but springs on there own have a resonance, hit that and the fun starts
Esperado, Had to use embedded passives for some PCBs, due to vibration and acceleration. SMD and PTH electroylitics (especially SMD) are worse components for vibration (large body skinny legs), they fall off, hence the popularity of tants and high value MLCCs these days, for standard stuff though no real problem.
Esperado, Had to use embedded passives for some PCBs, due to vibration and acceleration. SMD and PTH electroylitics (especially SMD) are worse components for vibration (large body skinny legs), they fall off, hence the popularity of tants and high value MLCCs these days, for standard stuff though no real problem.
- Status
- Not open for further replies.