My question is in regards to both mid-woofers and pure midrange drivers that are presented with a transmission line enclosure.
It's my understanding that for frequencies with a wavelength larger than that dictated by the largest cross-sectional area of the line (a.k.a bass frequencies), doing things like adding corner braces or trying to maintain a 'smooth flow' of sound pressure throughout the line is pointless.
But what about higher frequencies where the wavelengths have room to bounce around and back out through the cone? Does doing something like B&W does with their 'super snail' speaker help with midrange performance?
To word that another way; Is midrange performance affected by how sharp the turns of the transmission line are? Is there a positive affect on sound by utilizing a benign (if you will) route of travel for the sound as opposed to presenting the driver with an immediate 90 degree turn followed by a several 180 sharp turns, as is often seen in traditional designs?
Without any experience in testing this, or being able to find much about it, I would guess that it's a significant variable for sound quality.
I've been mulling over transmission line designs in my head almost obsessively lately and am trying to gain more clarity on the subject.
Can anyone with some more knowledge chime in on this?
It's my understanding that for frequencies with a wavelength larger than that dictated by the largest cross-sectional area of the line (a.k.a bass frequencies), doing things like adding corner braces or trying to maintain a 'smooth flow' of sound pressure throughout the line is pointless.
But what about higher frequencies where the wavelengths have room to bounce around and back out through the cone? Does doing something like B&W does with their 'super snail' speaker help with midrange performance?
To word that another way; Is midrange performance affected by how sharp the turns of the transmission line are? Is there a positive affect on sound by utilizing a benign (if you will) route of travel for the sound as opposed to presenting the driver with an immediate 90 degree turn followed by a several 180 sharp turns, as is often seen in traditional designs?
Without any experience in testing this, or being able to find much about it, I would guess that it's a significant variable for sound quality.
I've been mulling over transmission line designs in my head almost obsessively lately and am trying to gain more clarity on the subject.
Can anyone with some more knowledge chime in on this?
I have probably the same level of knowledge as you.
I think in terms of microstrip design, as that's an area where I'm more comfortable (RF engineer).
If you've got a microstrip line terminated in a short circuit (closed end for a speaker), then the reflections will be antiphase to the incident energy. If the termination is open, then the reflections are in-phase.
So for a 1/4 wavelength at a given frequency, reflections from the open end of the line are in-phase at the driver, so the driver sees a high impedance. If we organise the 1/4 wavelength at the spot where we want to boost the bass response, that'll do it. There'll also be a boost for 3/4, 5/4, 7/4. Conversely we'll get dips at 1/2, 2/2, 3,2 wavelengths.
To my mind sharp corners in a transmission line are exactly the same as sharp corners in microstrip, with an impedance mismatch (decrease) at the corner and a resultant reflection out of phase. For a frequency where that's a 1/4 wave, we'll see a dip in the response.
In microstrip, we either bevel or radius the corners to reduce the impedance step. Same appears to be true for a speaker.
In a speaker like the Ariel, the reflections are used to create attenuation at higher frequencies. There's multiple hard corners down the line with a variety of short lengths, which will mean that HF information gets attenuated in the line. These features are <<1/4 wave for the main resonance, so don't affect it.
Also you can put wadding and felt in, which make the line lossy. reflections go through the attenuator twice, so get attenuated more than stuff which goes out of the line.
I think in terms of microstrip design, as that's an area where I'm more comfortable (RF engineer).
If you've got a microstrip line terminated in a short circuit (closed end for a speaker), then the reflections will be antiphase to the incident energy. If the termination is open, then the reflections are in-phase.
So for a 1/4 wavelength at a given frequency, reflections from the open end of the line are in-phase at the driver, so the driver sees a high impedance. If we organise the 1/4 wavelength at the spot where we want to boost the bass response, that'll do it. There'll also be a boost for 3/4, 5/4, 7/4. Conversely we'll get dips at 1/2, 2/2, 3,2 wavelengths.
To my mind sharp corners in a transmission line are exactly the same as sharp corners in microstrip, with an impedance mismatch (decrease) at the corner and a resultant reflection out of phase. For a frequency where that's a 1/4 wave, we'll see a dip in the response.
In microstrip, we either bevel or radius the corners to reduce the impedance step. Same appears to be true for a speaker.
In a speaker like the Ariel, the reflections are used to create attenuation at higher frequencies. There's multiple hard corners down the line with a variety of short lengths, which will mean that HF information gets attenuated in the line. These features are <<1/4 wave for the main resonance, so don't affect it.
Also you can put wadding and felt in, which make the line lossy. reflections go through the attenuator twice, so get attenuated more than stuff which goes out of the line.
You could always design a transmission line based on this styling .................
Nautilus Loudspeaker - High End Speakers | Bowers & Wilkins Australia
C.M
Suzy, I'm afraid I'm not familiar with microstrip design. I would never have thought to connect the two in my head. It's pretty neat that they both seem to follow the same principles though, if I recall Martin King's papers correctly.
Tweet, yes the Nautilus speaker is the one I was referring to when I said "super snail". I"ve heard that it sounds really good. And in fact the designs I'm thinking of are similar, but not quite the same, and not so ghastly looking, in my opinion.
Tweet, yes the Nautilus speaker is the one I was referring to when I said "super snail". I"ve heard that it sounds really good. And in fact the designs I'm thinking of are similar, but not quite the same, and not so ghastly looking, in my opinion.
Whether it's a midwoofer in a TL or a pure midrange driver in a TL, both will have stuffing in the line, and the frequencies you're concerned about bouncing around should (will?) be completely absorbed by that stuffing and won't cause any detrimental effects. That's also the reason that using angled or curved deflectors in corners to "steer" the sound are useless; the wavelengths of the frequencies supposedly being steered are too long to even "see" those deflectors, and all of the higher frequencies that could are being absorbed by the stuffing. Others may think otherwise, of course.
Paul
Paul
What Paul said. For the TLs we use for midrange (usually midTweeters) we are trying to absorb the entire rear wave an flatten the impedance.
A post talking about the midTL in the 1st Tysen and a measured impedance curve.
http://www.diyaudio.com/forums/full-range/144099-thread-tysen-variations-fast.html#post1827952
dave
A post talking about the midTL in the 1st Tysen and a measured impedance curve.
http://www.diyaudio.com/forums/full-range/144099-thread-tysen-variations-fast.html#post1827952
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.