I don't use tweeters that low in frequency. In my opinion using a tweeter too low in frequency brings its own set of problems
.
Vau, that's quite a long time to focus on basilar membrane only
You may have found out that it serves as a prefilter, and the most interesting stuff on sound perception happends in the modulation domain in time-frequency-amplitude space at the deeper parts of auditory chain.
If you find interest, check my 'basilar membrane' wavelets aka Bark wavelets on my home page.
- Elias
Hi Elias,
I am fascinated how the basilar membrane processes the sounds we hear. I don't think that manufacturers have given it as much respect as they should have done when trying to designing speakers.
I will take a look at your website.
Many thanks,
Ian
You've an extra http in the link Ian. Works fine once it's removed though.
I confess I haven't had chance to read through it all, and while I appreciate that it's (presumably) been pitched toward beginners, FWIW purely as a bit of feedback, I think the breakup definitions on p.3 & 4 could stand a little clarification as it's a bit sweeping at present. You state this occurs in all bass / misbass units, but there's breakup & then there's breakup. Presumably you're referring to uncontrolled resonance, but many midbass units employ controlled resonance of the powertrain to extend the upper BW limit beyond that of the mechanical LF oscillation (piston) region, and it is far from invariable that they have a huge peak at the top end. Look at the classic Vifa P13 for e.g. Same goes for wideband drivers. There's a related point probably most commonly seen mentioned in DIY circles with the Elsinore speaker project that peaking at the top end of a given drive unit's response is not always a matter of mechanical breakup but can be an acoustical effect from a buildup of energy caused by cone / dustcap / frame etc. profiles. To be sure though, I think everybody agrees with the basic point that XOs do not want to be located within a region where there are major response aberrations, whatever the cause.
I'm also a little mystified by the comment on baffle-step. You say this is 'relative to driver diameter' and that it 'causes frequencies to rise.' Unless you're using the term in some different way to how it is usually employed (?) that is not the case. Step-loss, to the best of my knowledge, is merely the apparent reduction of LF SPLs at the listening position caused by the transition of a monopole radiator from 2pi to 4pi radiation space when wavelengths exceed the dimensions of the baffle. F3 can be roughly calculated by 4560/baffle width in inches. Am I missing something?
I confess I haven't had chance to read through it all, and while I appreciate that it's (presumably) been pitched toward beginners, FWIW purely as a bit of feedback, I think the breakup definitions on p.3 & 4 could stand a little clarification as it's a bit sweeping at present. You state this occurs in all bass / misbass units, but there's breakup & then there's breakup. Presumably you're referring to uncontrolled resonance, but many midbass units employ controlled resonance of the powertrain to extend the upper BW limit beyond that of the mechanical LF oscillation (piston) region, and it is far from invariable that they have a huge peak at the top end. Look at the classic Vifa P13 for e.g. Same goes for wideband drivers. There's a related point probably most commonly seen mentioned in DIY circles with the Elsinore speaker project that peaking at the top end of a given drive unit's response is not always a matter of mechanical breakup but can be an acoustical effect from a buildup of energy caused by cone / dustcap / frame etc. profiles. To be sure though, I think everybody agrees with the basic point that XOs do not want to be located within a region where there are major response aberrations, whatever the cause.
I'm also a little mystified by the comment on baffle-step. You say this is 'relative to driver diameter' and that it 'causes frequencies to rise.' Unless you're using the term in some different way to how it is usually employed (?) that is not the case. Step-loss, to the best of my knowledge, is merely the apparent reduction of LF SPLs at the listening position caused by the transition of a monopole radiator from 2pi to 4pi radiation space when wavelengths exceed the dimensions of the baffle. F3 can be roughly calculated by 4560/baffle width in inches. Am I missing something?
Last edited:
I don't use tweeters that low in frequency. In my opinion using a tweeter too low in frequency brings its own set of problems
well.. I wasn't thinking THAT low to be honest, definitely can't see any working down to 100-150hz. But a small midrange can! Will read your paper.
Basilar membrane has nothing to do with creating the variations in sound pressure over time occurring in the ear canal.
All discussion here is about how well speaker system is at performing as single source or as multiple source across desired bandwidth. Effective driver diameter, driver spacing and crossover choices are controlling variables. When effective acoustic centers of drivers remains below 1/4 wavelength of radiating frequency the drivers resolve as single source. When speaker system behaves as multiple source, lobing happens, and reflections become multi-path, breaking up timbrel and harmonic coherence, leading brain to work on weather reflection is part of existing source information stream, or start of a new information stream.
Above a few kHz, lobing may be very intense, but inconsequential do to head shadow dominating directional determination.
Hierarchy of directional cues is HF processed first. As lower frequency information decodes from basilar membrane, brain looks for association to higher harmonic content.
This follows physics; the closer two frequencies are, the longer the beat they form, and the longer a Fourier analysis it takes to resolve them.
Causal nature of sound production guarantees harmonic content. So localization cues from large bass source start with harmonics.
When a low frequency decode occurs that doesn't associate with existing HF decodes, source is interpreted as more distant/lower priority. Likewise HF decode with no LF follow up is interpreted as small source, typically less threatening.
Most sounds are identifiable with <3ms of source (opening transient), including direction and size. Lower brain is responding to direction and size long before frontal cortex comes up with "flute", "bee", or "lion".
Continuation of a source beyond 3ms up to about 50ms contributes primarily to interpretation of loudness. Much beyond this time frame effects are interpreted as reverberant or echo.
Use of crossover slopes below 24dB/oct is a waste of driver performance. Doing it passively using modern drivers produces speakers sounding no better than speakers from the 50's, 60's and 70's using drivers of similar diameter and moving mass.
All discussion here is about how well speaker system is at performing as single source or as multiple source across desired bandwidth. Effective driver diameter, driver spacing and crossover choices are controlling variables. When effective acoustic centers of drivers remains below 1/4 wavelength of radiating frequency the drivers resolve as single source. When speaker system behaves as multiple source, lobing happens, and reflections become multi-path, breaking up timbrel and harmonic coherence, leading brain to work on weather reflection is part of existing source information stream, or start of a new information stream.
Above a few kHz, lobing may be very intense, but inconsequential do to head shadow dominating directional determination.
Hierarchy of directional cues is HF processed first. As lower frequency information decodes from basilar membrane, brain looks for association to higher harmonic content.
This follows physics; the closer two frequencies are, the longer the beat they form, and the longer a Fourier analysis it takes to resolve them.
Causal nature of sound production guarantees harmonic content. So localization cues from large bass source start with harmonics.
When a low frequency decode occurs that doesn't associate with existing HF decodes, source is interpreted as more distant/lower priority. Likewise HF decode with no LF follow up is interpreted as small source, typically less threatening.
Most sounds are identifiable with <3ms of source (opening transient), including direction and size. Lower brain is responding to direction and size long before frontal cortex comes up with "flute", "bee", or "lion".
Continuation of a source beyond 3ms up to about 50ms contributes primarily to interpretation of loudness. Much beyond this time frame effects are interpreted as reverberant or echo.
Use of crossover slopes below 24dB/oct is a waste of driver performance. Doing it passively using modern drivers produces speakers sounding no better than speakers from the 50's, 60's and 70's using drivers of similar diameter and moving mass.
Hi Scottmoose,
Thanks for your comments.
Yes what you have said is correct.
I've pitched the articles in at beginners. I find it difficult sometimes not to give sweeping statements, I am really trying to avoid that!.
Regarding the point you raised about the breakup.....yes I am referring to the uncontrolled resonance area of the driver which is usually quite predictable.
Despite all the developments in driver manufacturing I don't try and push mid/bass drivers high in frequency ( I mean anything above 1.5kHz ) before rolling them off. I personally don't like the energy storage problems that can occur within the system ( by system I mean speakers and amplifier) .
The breakup article is one of a few articles that have been designed as introduction to the way I design my speakers.
There is a full blown ebook in the pipeline but that is something that I am working on in as well.
I suppose with the baffle step I what I was trying to say in a nutshell that it was relative to the speaker diameter if used on a front baffle of similar size to the driver.
Again this is where I just wanted to generalize for the purpose of the article. I might change that!
I appreciate your input.
Thanks Ian
Did you get the notification email ok?.....is it in your spam box?I did register but the download does not go through.
Thanks for your views.
Despite the fact I did already know how lobing works but the refresser course was welcome.
Use of crossover slopes below 24dB/oct is a waste of driver performance. Doing it passively using modern drivers produces speakers sounding no better than speakers from the 50's, 60's and 70's using drivers of similar diameter and moving mass.
I don't agree with the above but each to his or her own. I am not a fan of those high order crossovers personally, but I wouldn't ever that any other method was a waste of driver performance. Sometimes it can be horses for courses.
Cheers Ian
Ian,
All I got was a PDF labeled
"Where to start
Understanding
Phase Beating
Loudspeakers"
Which contained 14 pages explaining equal loudness contours, and nothing regarding "Phase Beating" loudspeakers, whatever they may be.
Do they sound better when wearing a wife beater
?Art
The page comes through when the link is corrected. No registration is required.
I notice that no mention was made of the correlation between a piston and a point source as it applies to a woofer.
As far as I can see based on this thread (I haven't had chance to look elsewhere), it seems to be an advocating taking a 1st order HP / LP & pulling them apart by x amount to introduce a large dip in the midband FR / power response. Shades of the so-called Solen-split variation on 1st order, where they've put the XO at the -6dB rather than the -3dB frequency, except there the object isn't really to cause a dip & what there is is relatively small, all other things being equal ~1.5dB.
There is a lot of wonderment inside a piano
May I ask about the strings that are made with nested coil-overs, whether their mechanical complexity is responsible for producing the non-harmonically related overtones?
My understanding is that the inharmonic property comes from the stiffness of the string. The ideal string would have distributed mass and will keep tension but not have stiffness. The stiffnes makes the overtones slightly sharp.
Overwraped bass strings are used to achieve much higher mass per length than the steel strings have. I think they have more flex than a fat steel string would have, perhaps for the same reason.
David S
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.