Magnesium can be very good, but WRT aluminium cones, I suspect you've either been using poor drivers or possibly using them incorrectly (I can't comment since I don't know what they were). An aluminium cone designed for piston operation over a given BW, used over that BW, with proper attention to attenuating stopband modes, shouldn't have any sound of its own, although the driver will have other colourations from the suspension & motor design of course. You can usually spot resonances as fluctuations in the impedance curve, and of course on HD plots.
Last edited:
I wouldn't think twice about an aluminum subwoofer. But usually when the response starts to narrow I can hear some kind of mild breakup. It tends to come out in the electric guitar. I probably shouldn't prejudge but with so many choices why bother. Ribbons are good choice if one wants pistonic top end.
In almost all cases, resonance of rigid/undamped diaphragms is accompanied by a peak in response of varying width. It also shows up on the waterfall. The sharpness of the peak depends on the material and geometry of the diaphragm.
Hificompass has a lot of great, easy-to-review measurements of various drivers.
Here’s a super obvious case. Should be easy to tell when the diaphragm is starting to break up: Dayton Audio RS52AN-8 | HiFiCompass
Last edited:
You can usually see the response of a driver go from smooth to rough as directivity narrows. I suspect that our ears are very sensitive to this and interpret the rough response as being a certain type of sound, maybe something we like or don't like (not trying to troll anyones preferences). Tweeters do for the most part have rough responses.
Well, leaving any argument over the audibility of that particular matter to one side for a moment (certainly, in multiways it has been demonstrated that an abrupt shift in polar response can affect perceived tonality), it has little to do with the cone / dome material as such, and for a given size is primarily a question of its geometry and that of the surround, basket & as relevant any central cap or polepiece extension.
Exactly. It's just a coincidence that directivity can start narrowing where the diaphragm starts to break up. Both of these phenomena happen on the same spectrum, and both are more likely to occur as frequency increases.
Let's look at a few examples for 33Polkhigh (cherry-picked to emphasis that they don't have to happen at the same point):
Satori TW29TXN - starts beaming (non-omni radiation) around 2.0-2.5 kHz, doesn't start breaking up until ~ 19 kHz. Remarkably, the breakup seems to be very well controlled.
Satori TW29TXN-4 | HiFiCompass
Scan-Speak D3004/664000 - starts beaming around 2.0 kHz, doesn't start breaking up until ~25 kHz
ScanSpeak D3004/664000 | HiFiCompass
Zaph Audio ZA14W08 - starts beaming around ~1.4 kHz, doesn't start breaking up until (arguably) ~6 kHz
Zaph Audio ZA14W08 | HiFiCompass
SB Acoustics SB15NBAC30 - starts beaming around ~1.5 kHz, doesn't start breaking up until ~5 kHz
SB Acoustics SB15NBAC30-4 | HiFiCompass
Accuton C173-6-096 - starts beaming around ~1.4 kHz, doesn't start breaking up until 3-3.5 kHz (and it is a *bad* peak)
Accuton C173-6-096 | HiFiCompass
Scan-Speak 22W/4851T00 - starts beaming around 1.0-1.2 kHz, doesn't start breaking up until 2.0-2.5 kHz. Because it's paper, there's a bit more self-damping compared to most metals and the peak isn't very pronounced. There are some paper-cone drivers that have worse peaks than this one, actually
https://hificompass.com/en/speakers/measurements/scan-speak/scanspeak-22w/4851t00
I'm running them in a pair of Audio Concepts Sapphire IIti clones fresh-built with Vintage Focal 7K011dbl woofers. They replace the aperiodic Focal T120 titanium tweeters that are unavailable. To tamp the resonance peak I zobeled them, and ran them without any padding resistors as in the originals. They are still a db or so shy, more so 30 degrees off axis. The net effect is not as bright as the originals, which are a bit bright anyway. I cannot hear the 27k peak or any artifacts, especially with the grill cloth in place, which is a must for the WAF. I would play a 27k tone to see what happens, but my generator died!
What was the room/setup you were using this in? The DA25Tx is a very wide dispersion driver, and I can imagine that in a smaller room it might be tough to strike an appropriate balance between on-axis and power response.
First let me clarify something. What is or isn't "break up" is in some ways semantics. But most will agree that its not a binary event. Some cones do it gradually over many octaves.
As far as breakup and beaming being coincidental, I would have to say no, theres no coincidence. At least based on what I've read from others at this site and my own experience.
Cone break up is related to how the cone responds to stress. The stress comes from playing wavelengths that are acoustically significant to the size of the cone (around 1/4 wavelength and bigger). DBmandrake who posts here knows quite a bit about it (I'm not saying he will agree with me, just that he knows about break up.)
Now look at the aluminum sb you posted, it loses its smoothness well before 1000hz. Then look at the waterfall of the accuton, its resonating early on. Also look at the satori tweeter waterfall. Its showing signs of resonance very early as well. THere is a lot of variation because there are variations in design and materials and how the cone is damped. But for a cone or dome to not have some break up as it gets higher in response it would have to be very unusual to say the least, and there is a lot of marketing and hype.
So bottom line, pick what you like. I'd say things like classical music tend to benefit from harder domes like ceramics but thats an opinion as well.
Last edited:
Since when?
Last I checked, the vast majority of people use it as a synonym for resonance in the powertrain (moving components), specifically the cone / subcones & surround. Er, yes. If they are designed to. In the case of metal cone drive units, that is limited to wideband drivers e.g. Jordan, Markaudio, EAD, Tang Band, Bandor et al. Midbass units often employ low Q resonance in softer cone drivers to extend the upper limit of their operating BW, in the same way wideband drivers using similar materials do, just to a lesser extent. No mysteries here, just well known operating characteristics & design principles.
You appear to be referring to concentric flexure; there are also radial modes to consider.
It does? I'm looking at the HFC measurements right now, and I'm seeing a rather linear midbass. No fluctuations in the impedance curve in this band at all that I can see, which is the
I was wondering if this would crop up. I'm afraid this is just a classic example of the limited value of CSD / waterfall displays -limited in that while they have some uses, unfortunately not everybody seems to quite understand how to read them properly or what they are showing about practical operating conditions. Rather than spending an age reinventing the wheel, I advise you to follow this link: Zaph|Audio and scroll down to the sub-section 'Usefulness of the CSD'. This is how it actually works.
See above. The same answer also applies to the query made a couple of posts back about the ETC. There are two other aspects WRT to the TW29TXN. Firstly, Satori dome tweeters use wide roll roll surrounds as active and significant parts of the radiating surface, especially low down -a trend largely established for better or worse by Vifa / Scanspeak about 20 years or so ago. And secondly, SB are using Textreme here as a hybrid between soft & hard domes (or an alternative to an aluminium alloy dome that doesn't need a Helmholtz / phase shield).
All cones or domes eventually reach a resonant mode. This is dictated by every law of relevant physics and mechanical engineering. Where did you see somebody claim otherwise? Inspecting the thread, I can't see any such claim or even implication made.
It is indeed -why you have singled out classical music though I'm at a loss to understand. There are excellent examples of all types of drive unit, which I don't seem to see anybody disputing either. We have been discussing the operating characteristics here, not subjective impressions.
Last edited:
In a manner of speaking.
No. You should understand it a bit further before fixing your opinion so firmly. What you are referring to is the relationship between diaphragm size and beaming, but clearly it's only a superficial comprehension on your part. As the wavelengths get shorter, the sound radiating from different areas of the diaphragm will combine more and more out of phase at the mic/listening position, so you end up with destructive interference off-axis.
I have not much more motivation to continue this conversation, so hopefully this will get you on your merry way:
https://2ccpsa3l1qjy41i2m31gqo0g-wp...-Stiffness-of-Loudspeaker-Cones-ALMA-2003.pdf
beaming
Last edited:
Since when?
My point was that there is no set threshold for how big a peak or dip must be to constitute break up. If we use your definition then cone resonance is break up. Maybe you want to be more specific.
Again, what I said is that break up, being cone resonance as you just described it, is not binary. It can be a little or a lot. Some cones break up slowly without obvious peaks. Is this not the case?
Feel free to explain, and why resonance from this particular kind of stress should be distinguished from "break up".
Look at the 500-1000 hz region. THere is a noticeable change. But lets look at some very high quality graphs like the seas excel.
Here is the 8" mg, Notice the trend up and a peak around 800 hz and a rippling of the response.
SEAS Excel W22EX-001 (E0022) 8" magnesium Cone Woofer
Now here is the smaller cone version, same thing but it happens higher, with the peak around 1200 hz. SEAS Excel W12CY-001 (E0021) 4.5" magnesium Cone Woofer
Now if what I am describing here isn't break up then that's okay. It might technically be a bell mode. Its certainly cone resonance that's related to frequency.
What's next? The revelation that its the same info as frequency response.
Obviously I was referring to it happening in the audible range. Which goes back to what I said before, cones sound like what they're made of, at least if they're played high enough. I don't see why this is controversial. You seem to be suggesting that this happens much higher and that the earlier resonances don't constitute break up.
Lets say earlier cone resonance from stresses shouldn't be referred to as break up. They are still audible resonances.
Last edited:
Oh dear.
I don't need to be 'more specific'. In terms of physics and engineering, cone resonance of any kind is 'break up'. It really is as simple as that.
I believe I've said that twice now.
Pardon? Who said it 'should be distinguished from "break up"'? What I said was that you had only bothered to mention one type of resonance. See above. Cone resonance of any kind is your precious 'break up'. End of story.
I'm looking at a drive unit with +/- 1dB variation in response between 500Hz - 1KHz based on the measurements in question, with an impedance graph devoid of any evidence of resonance in that band, and no sign in the HD plots either. Perhaps you could explain more clearly what 'noticable change' you are referring to, beyond, obviously, the merging of data plots at / just below the bottom end of that scale.
Sigh. With respect, you really need to learn something about what you're speaking of. First: repeat after me: any kind of cone resonance is 'break up'. Why you keep talking as though people have said otherwise is a trifle mysterious.
Next: you are clearly unaware of how Seas take driver measurements. Unlike most manufacturers, it is not on a large test baffle, but in small sealed test boxes. Information at the following link: DATASHEETS These boxes to be specific: http://seas.no/images/stories/about_seas/TEST_BOXES.pdf The rise you are seeing in the lower midrange, or more accurately, the drop in output in the lower registers is (drum roll please) ...nothing more than baffle step loss of the resulting sealed box system. Seas think this is a more representative means of showing the driver performance, but they provide (dotted line) the mathematical IB curve derived from T/S parameters assuming an infinitely large baffle in addition to the measured.
I'm getting the strong impression (never!) that you are deliberately chosing not to learn. No, I think the majority of people are aware that CSD plots are a different way of presenting the same data as a frequency / amplitude response graph. What you could have learned from it however, (which has particular reference to your complaint about the Accuton in this regard) is that a significant resonant mode in a driver's raw unfiltered response results in apparent ringing in the CSD plot lower down in frequency. And that as soon as this is addressed, said excess vanishes.
I don't need to be 'more specific'. In terms of physics and engineering, cone resonance of any kind is 'break up'. It really is as simple as that.
I believe I've said that twice now.
Pardon? Who said it 'should be distinguished from "break up"'? What I said was that you had only bothered to mention one type of resonance. See above. Cone resonance of any kind is your precious 'break up'. End of story.
I'm looking at a drive unit with +/- 1dB variation in response between 500Hz - 1KHz based on the measurements in question, with an impedance graph devoid of any evidence of resonance in that band, and no sign in the HD plots either. Perhaps you could explain more clearly what 'noticable change' you are referring to, beyond, obviously, the merging of data plots at / just below the bottom end of that scale.
Sigh. With respect, you really need to learn something about what you're speaking of. First: repeat after me: any kind of cone resonance is 'break up'. Why you keep talking as though people have said otherwise is a trifle mysterious.
Next: you are clearly unaware of how Seas take driver measurements. Unlike most manufacturers, it is not on a large test baffle, but in small sealed test boxes. Information at the following link: DATASHEETS These boxes to be specific: http://seas.no/images/stories/about_seas/TEST_BOXES.pdf The rise you are seeing in the lower midrange, or more accurately, the drop in output in the lower registers is (drum roll please) ...nothing more than baffle step loss of the resulting sealed box system. Seas think this is a more representative means of showing the driver performance, but they provide (dotted line) the mathematical IB curve derived from T/S parameters assuming an infinitely large baffle in addition to the measured.
I'm getting the strong impression (never!) that you are deliberately chosing not to learn. No, I think the majority of people are aware that CSD plots are a different way of presenting the same data as a frequency / amplitude response graph. What you could have learned from it however, (which has particular reference to your complaint about the Accuton in this regard) is that a significant resonant mode in a driver's raw unfiltered response results in apparent ringing in the CSD plot lower down in frequency. And that as soon as this is addressed, said excess vanishes.
Last edited:
Go to the off axis FR measurement, where you can see the FR become less smooth around 500hz. Just, like this hivi m5a which is a similar size metal cone that starts resonating at around same point.
https://www.parts-express.com/pedocs/specs/297-437-hi-vi-m5a-specifications-45110.pdf
This is what you've been saying:
If you want to stand by this statement then fine. However afaik normal domes leave the pistonic range and have resonances long before serious break up.
Honestly I post here with good intentions but you're completely trying to misconstrue this. I was referring to the shift in FR from being smooth to choppy. And yes it coincides with the driver response narrowing (no longer omnidirectional). And yes there is still a rise and slight peak even without the effect from baffle step.
There are tons of examples of driver measurements that have choppier responses around the point where response starts to narrow. You seem to be saying its all motor and measurement artifacts, and not related to stresses in the cone itself. If that's right then fine. I honestly don't care that much about it. I don't have your knowledge or experience in designing speakers, but I would appreciate if you didn't get such a bug up your a** because I disagreed with you.
Last edited:
Pistonic motion: YouTube
Breakup: YouTube
Anything that does not look like what is shown in the "pistonic motion" clip is breakup. End of story.
Pistonic motion speaker design. – Code Acoustics
Cone displacement - breakup on left, no breakup on right
Acoustic Horn Design – The Easy Way (Ath4)
Beaming is not caused by breakup. Breakup is not caused by beaming. Beaming is affected by breakup, but not vice versa.
Breakup: YouTube
Anything that does not look like what is shown in the "pistonic motion" clip is breakup. End of story.
Pistonic motion speaker design. – Code Acoustics
Cone displacement - breakup on left, no breakup on right
Acoustic Horn Design – The Easy Way (Ath4)
Beaming is not caused by breakup. Breakup is not caused by beaming. Beaming is affected by breakup, but not vice versa.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.