Absolutely for tweeters -potentially less so for other units, but depends on design priorities
Agreed -ideally. Although with the proliferation of wide-surround designs, < 2.5KHz or so we're losing out on the piston behaviour.
One reason I'm philosophically not convinced about ring-radiators etc., although admittedly the Satori variations are still very good, so I'm not a hard-liner in that sense. Given the option, I'd always prefer a hard dome model with a narrow surround though. 😉
I would be looking at a CD and waveguide for that price. Much better off axis.
Rob 🙂
That SS 2904 is basically the same as the venerable VIFA XT25 jobs, albeit with a refined motor assembly.
I've read that article before, and while I agree with a lot of it, there are also simple phenomena that can be easily observed that seem to go against some of the interpretations. For instance, by holding a neodymium magnet next to a vibrating copper tube (home-made xylophone), it can be observed that the vibrations decay at a faster rate as the magnet is brought closer.
The vibrations in this case were in the kHz range. Translating that to a speaker, the eddy current is guided along a long thin wire and the amplifier closes the loop, and I see no reason why the additional mechanics of bass tuning should somehow change the operating principle.
I think the tricky point with break-up modes is that the cone resonance is perfectly aligned in phase with what the voice coil is already doing, so any 'braking' is at a minimum. That would be similar to the bass resonance, which I see got Esa Merilainen so bothered because of people constantly talking about damping, even though the system is plainly 'freewheeling' with minimum damping.
What I'm not so sure about is the accuracy of the impedance plots, which most manufacturers seem to begrudgingly show for us engineering 'sticklers'. Ever seen a super-accurate one with delta impedance on a log-log plot? On a hunch, I'd be looking for <0.1 ohm ripples from 10-20 kHz, so the usual coarse 10-20 ohm rising impedance would have to be filtered out to be able to see what's going on. Alternatively, a much higher frequency resolution to see if there are any narrow spikes, which might be missed if the plotted points are too sparse. For there to be a bell mode, the impedance has to have some kind of peak. I see no way around it.
Without speculating further, if there's a known break-up mode, I'd try that fix with a parallel notch placed in series, measure the result and go from there.
So, it would seem that at the lower frequencies of a tweeter, electrical damping is very useful,
and at the higher frequencies of a tweeter, mechanical damping is very useful.
and at the higher frequencies of a tweeter, mechanical damping is very useful.
Hi,
jeah Esa's motivation and attention seems to be on the current drive vs world and its tough read, has angry tone to it. Put pink glasses on and ignore the style of his and concentrate on the science part.
You can put your imagination on it, think what impedance (reactance) literally means Z = U/I? Why speaker driver has impedance peak on the low frequency even though inductance of voice coil is very low? Its all in the article, and how the impedance differs at resonance and above and how it affects what the backEMF/eddy current relates.
I'm not professional on this, but could imagine the resonating xylophone tube is system at resonance just like loudspeaker driver at its main resonance; mass is cancelled by some spring, and inductance is low, so eddy current can work as a brake as per the article. Difference is that a dome / cone breakup / mode is independent of the voice coil, connected but happens regardless. You could probably dampen a tweeter bell mode with the magnet if the dome was made out of copper 🙂 In addition dome bell mode, or cone breakup of a woofer, is at frequency where the system has high inductance. Inductance of copper tube is quite low as its single turn with air core, also resistance, so impedance at kHz is low compared to loudspeaker voice coil.
Well, I really don't know how it plays out with the xylophone tube, but Esa's dissection off speaker driver is quite logical to me at least with no question marks and plays nicely with everything else I've poked, like purifi papers, or what happens to response if you play with circuit impedance in simulator, like xo parts, amplifier output impedance etc.
Well, as long as anyone is able to utilize the knowledge for their benefit its good, its not necessary to know everything what happens under he hood. We can and could and should EQ a dome / cone breakup peak by reducing current through the voice coil by increasing series impedance at this frequency, which equalizes frequency response nice and also reduces any other current at that frequency. By other current I mean for example harmonic distortion component that motor non-linearity at lower frequency has induced in the circuit. End result is that the distortion in electrical domain is not emitted fully into acoustic domain, let alone amplified by the breakup.
Last edited:
@tmuikku
The reason why is very similar to a standard RLC circuit (which is the electric analog equivalent)
https://en.wikipedia.org/wiki/RLC_circuit
Any resonances or oscillation follows this exact same principle.
So there needs to be some kind or Mass component and a compliance to make something resonating.
R is just nothing more than friction.
The BL is the factor to the electrical domain.
I never thought about your explanation about how to dampen it.
Like I said before, if I remember correctly it didn't seem to effect much with woofers?
But again, that has been a while.
So I am very curious to see some more measurements of this.
I think current feedback would work well here as well.
Especially when used for (low) mid-range and up.
It doesn't make any sense to apply if for the low frequency (sub) range at all.
It's only pretty hard to implement in modern Class-D (IC) amplifiers (that are also practical and affordable).
The reason why is very similar to a standard RLC circuit (which is the electric analog equivalent)
https://en.wikipedia.org/wiki/RLC_circuit
Any resonances or oscillation follows this exact same principle.
So there needs to be some kind or Mass component and a compliance to make something resonating.
R is just nothing more than friction.
The BL is the factor to the electrical domain.
I never thought about your explanation about how to dampen it.
Like I said before, if I remember correctly it didn't seem to effect much with woofers?
But again, that has been a while.
So I am very curious to see some more measurements of this.
I think current feedback would work well here as well.
Especially when used for (low) mid-range and up.
It doesn't make any sense to apply if for the low frequency (sub) range at all.
It's only pretty hard to implement in modern Class-D (IC) amplifiers (that are also practical and affordable).
As was noted above, it's basically the high end version of the R2604/8320 (Vifa XT25) with a different, lower-mass coating, a different central plug, and the Revelator motor. It has better higher-order HD performance on the low end thanks to that motor, although that's also partly skewed by it having a better-damped LF. You're paying a heck of a premium for the build quality & that better HD, though it is a better tweeter. Still has that elevated HD2 of all wide-surround (or 'only surround' in the case of ring radiators) tweeters though.
it's expensive for what it is, but you maybe can expect some consistancy on the building side and datasheet result
You could buy 10 XT25s for one pair; get perfect matching from 2 of them, and sell the other 8 🤣
You can. But you also get that from many much cheaper tweeters too -incuding the aforementioned 8320.
for sure, I can't affoard this iluminator SS. One that had my eyes is the 83000 at 45 euros (single magnet one) which seems to have a cleaner low end than the doublle magnet 833000. Seen also waveguided in the Heissmann wave guide-tweeters test ! The SB26CDC should be less hard sounding though I surmise.
Revelator. 😉
How do you classify 'cleaner'?
I don't know what that means, so no idea, although with the demise of the Peerless DA25TX it's technically speaking about the best affordable 1in hard-dome tweeter widely available -with the Seas 27TBC/G and 27TAC/GB lurking in the offing.
You do know you are giving advice to someone who believes cement resistors prevent tweeters from reproducing details. 🙂
cleaner waterfall, look at Dibirama the waterfall of the 83000, its low end above 2k hz is cleaner (faster and better damping) than the 833000
Not a believing, nore a biass. But if you cannot hear a difference with a tweeter between a basic cement vs MOX vs Ayrton Perry Mundorf Supreme resistor by yourself, there may be several cause : not so resolving cable (between source/amp,pre or loudspeakers as well), and or source resolution, and or amp resolution & drive, and or tweeter itself or the quality of choice of passive other parts -coils and caps-. With my current aluminum tweeter and mid, it is clearly different ! Yes, more veillled sound with basic cement in my setup. More details with the Mundorf Supreme resistor, especially details in the decays. That is simple as that, my setup is so resolving than each different cables or parts be it active or passive from the source can be heard... or several personn have biass, evenn the non audio ones whom are musicians and heard it at home. Could be the age of your ears as well, go figure; anyway not a biass here or people has collective hallucinations at home (no mushrooms parties)
Last edited:
