Why the tweeter needs its own full bandwidth amplifier

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I guess the comment aims at the pressure waves created by the woofer hitting the tweeter cone and generating back EMF by moving it.

I have not measured anything, but as a first guess I wouldn't expect much movement of the small, rigidly mounted tweeter cone by the air pressure of bass notes.

Anyone ever measured the effect?

Regards,
Rundmaus
 
Not many speakers have the back of the tweeter cone accessible to energy from within the cabinet. Many are mounted within their own mini-box or are shielded at the rear.

Having said that I cannot imagine that many tri-amped set ups have the amp deliberately bandwidth limited for treble use - the bandwidth limit has been done at the input by the active crossover so few designers would then go on to choose/ design/ further limit the power amp itself.
 
What tweeter are you referring to that would literally have a cone? Dome tweeters don't and ribbon tweeters don't and usually come already sealed (other than ribbons intended to be used open-backed). And, no tweeter would need to have a full-bandwidth amplifier driving them in order to "cope" with bass notes, whether actively or passively crossed.
Paul

Once you decide to triamp in an active crossover setup, it is important to remember to outfit the tweeter with its own full bandwidth negative feedback amplifier so that it can cope with the bass notes hitting the tweeter cone
 
I think there is some confusion I'd like to clarify.

Imagine the outside of a simple 2 way system. At low frequencies the mid-woofer will be omnidirectional, but the wavefront will encounter the baffle for it's full width and height.

So a question I've often had is, what happens when this wavefront encounters other drivers, such as a tweeter dome or ribbon?

Most of us have been taught that a microphone is just a speaker driver in reverse! That is, they are identical in concept except whether the coil is used to create an electrical signal (microphone) or receive it (speaker). Obviously I'm simplifying a great deal, but it's true. Drivers are subject to the effects of the wavefronts put upon them as much as they are the creators.

This has some interesting connotations that I have not seen researched. For instance, as the OP asked, what happens to the tweeter?

What about in stereo systems? How does feedback, or lack thereof affect the 2 channel reproduction?

Fun stuff. :)

Best,


E
 
tweeters usually have electro-mechanical "high pass" below fs

the tweeter combined back cavity and suspension spring is stiff compared to the woofer and the tweeter area is small

at sufficiently low frequency below tweeter fs the back cavity should resistively vent to the front to equalize atmospheric pressure changes < fs/100?


I think the Synergy guys have seen problems but they are combining in waveguides with compression factors in play
 
This is a rather interesting question from OnAudio. When and amplifier input is band limited and a LF impulse moves the tweeter cone and coupling it back to the amplifier inverting input via the feed-back circuit, it becomes positive output and probably dynamically damping the tweeter cone movement.

How am I doing so far Harrison?
 
Last edited:
Hi Nico!

Yes, but the input has nothing to do with the output in this case. :)

Any amplifier that relies on voltage feedback will behave this way, regardless of input.

Now, the voltage feedback itself is usually bandwidth limited, with a low-pass filter to prevent oscillation. Don't want the amp to try to damp 100kHz noise. :) If memory serves, most feedback designs are NOT high passed, so they can also be used to adjust for DC offset. But now we're getting into my very very hazy memory and I'm sure a more experienced DIYer will correct me.

Best,

E
 
Last edited:
Although I'm a big fan of multi-amping (for past 50 yrs), some risk to tweeters of turn-on very low frequency transients somewhere upstream. I don't know how you'd protect a tweeter (while maintaining sound quality) except by more sophisticated electronics (such as delayed turn-on) upstream.

One good practice is to have tweeters capable of high power (such as the wonderful 1-⅛ inch silk dome, 50 watt, low crossover, from Parts Express). I also believe systems rarely have enough tweeter power headroom, odd as that may sound to some. Just play a girls chorus doing Brahms' Lullaby.

Ben
 
Last edited:
I once read that pressure from the midranges hitting the compression driver is a problem with unity style horns. A solution (by Tom Danley) is to use a passive crossover for which the last element is a coil, parallel to the compression driver.

The Paraline as drawn on the left exposes the HF driver to a much higher pressure zone created by the mid drivers than the off set driver approach on the right.
Normally the exit would be only 1 inch wide (not 4 or 5 as drawn) to allow for decent HF dispersion, if drawn normally it would be more easy to see the difference.

As posted previously, the throat SPL in a driver is quite high, for instance at the horn mouth, 26” from the HF driver throat screen was 126.3 dBA, 143.7 dBA at the HF driver screen. Just 6.5" inches from the screen the level dropped to 136.8 dBA.

As an anecdotal indication of the potential problem, Dave Rat (Rat Sound, Red Hot Chili Peppers FOH engineer) found that he was shattering unpowered berrilium TAD HF diaphragms during testing of the front loaded mid cones which were tightly packed around the HF horn. Removing the series "protection" capacitors let amp damping keep the HF diaphragms in place. That said, the throat arrangement in your Paraline would be at least an order of magnitude more problematic than Dave's "Rat Trap" arrangement.
Hi Art
The TAD drivers are a bit unusual in that they generally have a very low (for a compression driver) resonant frequency and so the traditional series cap leaves them fully un-damped and easy for the diaphragm to move around.

What works much better is a driver with a higher Fs and more excursion capability and then using a parallel choke as the final element across the driver. This way instead of seeing an open circuit as the frequency falls, it sees a short and is highly damped and very hard for external pressure to move around.
I try to use that L as the majority of the compensation slope involved so they are typically a fairly small value as well. In some cases, the L is small enough to make the compression drivers impedance a non-issue as well as it dominates the crossover load.
 
...
So a question I've often had is, what happens when this wavefront encounters other drivers, such as a tweeter dome or ribbon?

Most of us have been taught that a microphone is just a speaker driver in reverse! That is, they are identical in concept except whether the coil is used to create an electrical signal (microphone) or receive it (speaker). Obviously I'm simplifying a great deal, but it's true. Drivers are subject to the effects of the wavefronts put upon them as much as they are the creators.

This has some interesting connotations that I have not seen researched. For instance, as the OP asked, what happens to the tweeter?

What about in stereo systems? How does feedback, or lack thereof affect the 2 channel reproduction? ...

Bob Carver had some comments to that end:

Bob Carver: Here's one of the magic. If we listen to a vacuum tube amplifier, when a vacuum tube amplifier produces a signal it makes the loudspeaker move. The loudspeaker sends sound waves into the room. Sound waves bounce off the wall and they come back to the loudspeaker. Like a microphone they make the loudspeaker move. The loudspeaker makes a little voltage just like the microphone would. That voltage is fed back around the amplifier's feedback loop back to its input. All amplifiers have two inputs. One is the audio signal that comes from the outside world and there is a audio signal which is the feedback signal. Both of these signals are the same. They are the same amplitude, same frequency response, everything. What happens on a vacuum tube amplifier, the amplifier makes another sound that's related to the sound that it heard. In other words, the amplifier is able to listen to the room because it's hearing reverberations, echoes, time delays, all the components associated with the venue. So the loudspeaker speaks and the room speaks back to the loudspeaker. The amplifier hears it by the signal going back to the feedback loop and out it comes again.

Host: It is delayed by some of the time..right?

Bob Carver: It's not delayed by much. The real delay is the acoustic delay (Host: exactly...that's what I'm talking about). Yes...it's delayed..exactly, exactly. That delay makes it sound spacious and big to our brain system. We love sounds that have ambiance and echoes and stuff like that.

Host: But don't solid state amps do the same thing?

Bob Carver: No, for a solid state amplifier, it's output impedance is so low that when it tells the speaker to move, the speaker sends the wave out which bounces off the walls and comes back. A solid state amplifier will not allow the speaker to move in response as in the sound wave coming back and hitting it.

Not the ultimate answer, but some thoughts from an amp designer. Seems simply using a high damping factor solid state amplifier avoids that source of distortion.

I have, however, wondered about this in terms of open baffle speakers. There's a design I've been contemplating building, but I've worried how much the woofers would interfere with the midrange (and dipole planar tweeter if that's what I use) sharing a baffle without separate boxes like that.
 
Why the tweeter needs its own full bandwidth amplifier
Once you decide to triamp in an active crossover setup, it is important to remember to outfit the tweeter with its own full bandwidth negative feedback amplifier so that it can cope with the bass notes hitting the tweeter cone

And? Is there any point to the OP? Who has a tweeter "cone" these days anyway?

Another for the list.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.