because speakers are built that way

there are some transducers that can go above 20kHz very well
ribbon
piezo
Heil drivers

more response in the top end usually means a compromise in the lower end so you will now have trouble with a crossover to the next driver ...
then come the choice to add an extra drive to cover the spectrum

less drivers is better for simplicity and number of components, therefore cost
more drivers , more components, more cost. unless you use cheaper components

a merry go round
and it hasn't really changed since the begining

I have read Mr. Neve talk about audio above 20KHz. I read it as though he believed that the electronic circuits needed to have a wider bandwidth to reduce the distortion components. Not so much that we have to HEAR that high, but that the processing should go that high. Maybe I misunderstood him.
Speaker bandwidth is the result of a lot of physical interactions. Most speaker crossovers do not determine the HF extension of a speaker, but simply protect the driver from program material that is too low in frequency to be properly handled by the driver. The assumption that a loudspeaker need not reproduce material above 20KHz has been prevalent in the industry for many years. Even today, there are many who say, "Why bother?"
Anyway, attempting to "make" a speakers' voice coil vibrate faster than 20KHz. can cause it to overheat and burn up. This is because its' movement is faster than its' ability to dissipate the heat generated by that movement. If this issue is a concern to you, maybe you should contact the folks at Peerless (or Eminence, JBL, etc.) and get their feedback.

this is a bit of the interview the whole thing is here (dead link removed)

START QUOTE.And following on from this, I was visiting Japan and was invited to the laboratories of Professor Oohashi He had discovered that when filteres were applied to an audio signal cutting off frequencies of 20 kHz, the brain started to emit electric signals which can be measured and quantified

These signals were at the frequencies and of the pattern which are associated with frustration and anger. Clearly we discussed this at some length and I also would forward the idea that any frequncies which were not part of the original music, such as quantisizing noise produced by compact discs and other digital sources, also produced similar brain waves.
Fletcher: What about frequencies below 20Hz (theoretically, the low end of the human range of hearing)... how do they affect the tone? END QUOTE

So improving the crossover will not "open" the top and soft dome tweeter cant take the heat. so ribbons can.
So this means the soft dome CAN but will burn up! what if you improve the heat dissipation with that paste thingy?!

Thank you

what if you improve the heat dissipation with that paste thingy?!

Yes it's called ' WALTERS DELIGHT '.

You should try it some time. 8-)

after a quick look at the d-s-t.com homepage (peerless/vifa/scanspeak) i found that about half of their tweeters have usable output beyond 20k

the vifa xt range easily reaches 40k.... (but they're ring radiators... not domes)... and several domes reach 30k+

it seems that the ferrofluid domes roll off sharply above 20k.... i don't know if it has some thing to do with it...

but to say " soft dome tweeter cant take the heat. so ribbons can. " is so not true...... the excursion length decreases with frequency..... but the problem is that if eg. your amp goes into self oscillation above 20k you won't hear it....

and kev

"more response in the top end usually means a compromise in the lower end so you will now have trouble with a crossover to the next driver ... "

why??? my tweeters are linear (+-1dB) from 750hz to 20k+ and 40k+-3dB.... how is that a compromise in the low end??

ok
so I didn't say the whole thing

if you were trying to get out to the Rupert 100K
I think the typical build of tweeters would suffer in the low end

if you want the excursion for 750 I doubt you can get upwards of 40K

and
as far as I am concerned, I've never tested a tweeter that can handle 750Hz at the sorts of powers that would be typical in a practical system.
only one octave higher would take the crossover to 1.5k and increase the power handling
if you wanted much higher power handling ... then 3.0K and it is likely you would be heading for a three way as most large cones are crap at 3.0K.

speaker building is a compromise right from the kick off.

:roll:
however
I don't think Ruperts 100K needed to include the speaker system and was more about correctly handling an audio signal for processing and mixing.

ok.... now i see....

btw my tweet's are cut at 2.5k... i guess for the reason you specified....

Hi,

I was happy to see your post, as I am a speaker engineer working with Vifa, Peerless and Scan-Speak (Tymphany).

FF can have a slight effect on the HF response of a tweeter. However, I think the rolloff you are noting is related to specific models that are designed for the highest power handling. They use FF, heavier coils, wider diaphragms, etc.

Incidentally, it isn't very difficult to get a 20mm dome to go out to 50 KHz, on axis, but the coil and former will be so light that power handling closer to resonance will be prohibitive.

good power handling close to resonance is not an easy thing

an eng at Peerless ... :)
cool 8)
I think Peerless have a phase plugged mid/woof.
looks interesting and one day I may attempt a design with something like that.

or I may just buy a pair
http://www.jlmaudio.com/Active%20Monitor.htm
I think these are Peerless drivers ... Focal ?? tweeter.

I think you got it right with the jlm drivers.

Let me know if I can ever be of assistance. I love DIY!

Kev wrote: good power handling close to resonance is not an easy thing

an eng at Peerless ... :)
cool 8)
I think Peerless have a phase plugged mid/woof.
looks interesting and one day I may attempt a design with something like that.

or I may just buy a pair

I think these are Peerless drivers ... Focal ?? tweeter.

someone posted a link to Wilson Audio in another thread

they have a similar passive unit
different drivers I think ... but close