The Peter Comeau designed WD 25T EX was similar in concept to the Dynaco A35.
It's a dual chamber design where an upper chamber 'communicates' with a lower chamber via an aperiodic vent.
The key to keeping the bass tightly controlled and free of resonant booming was to adjust the aperiodic foam loading in the vent to reduce the bass resonance peak as much as possible, and use just enough fibre filling in each chamber to absorb internal reflections.
It's a dual chamber design where an upper chamber 'communicates' with a lower chamber via an aperiodic vent.
The key to keeping the bass tightly controlled and free of resonant booming was to adjust the aperiodic foam loading in the vent to reduce the bass resonance peak as much as possible, and use just enough fibre filling in each chamber to absorb internal reflections.
From garage sales i scored similar used as eavestrough screen. Used it in a couple builds. I probably still have some kicking around.
This is essentially what dyna did, and whata variovent is… squish some fiberglass insulation between two pieces...
dave
George Short, of the no-longer-existing North Creek Music Loudspeakers, researched/tested an aperiodic dual chamber system. He titled the system loading as Multichamber Aperiodic Progressive Damping (MAPD). See Post #7 in the thread below:
https://www.diyaudio.com/community/threads/dual-aperiodic-box-idea.54402/
https://www.diyaudio.com/community/threads/dual-aperiodic-box-idea.54402/
Did anyone try to put a driver into a small aperiodic damped enclosure and EQed it?
I did it with 20cm driver in 8 liters with drone cone - worked very well if DSPed to the point.
The drone cone minimizes distortion by helping being the air less compressed and certainly due to work in reflex mode.
But could a Variovent also help the air being less compressed in a small box?
Did anybody measure this for a fixed frequency response like Closed box VS aperiodic VS drone cone all EQed to be linear and same fs and measured distortion?
I did it with 20cm driver in 8 liters with drone cone - worked very well if DSPed to the point.
The drone cone minimizes distortion by helping being the air less compressed and certainly due to work in reflex mode.
But could a Variovent also help the air being less compressed in a small box?
Did anybody measure this for a fixed frequency response like Closed box VS aperiodic VS drone cone all EQed to be linear and same fs and measured distortion?
Interesting experiment.
Did you compare it to closed cabinet?
From my experience I think it was not easy to get the right damping in the vent. Maybe bad luck. But I guess if the went is not made right one end up with cabinet that is just leaking. And I guess horrible resonances at certain Frq.
Did you compare it to closed cabinet?
From my experience I think it was not easy to get the right damping in the vent. Maybe bad luck. But I guess if the went is not made right one end up with cabinet that is just leaking. And I guess horrible resonances at certain Frq.
Hello.
I read somewhere that according to the loudspeaker design, the base (low frequencies) - or rather the frequency range that is "radiated" by the midrange, low-midrange or woofer, may "lag behind". Unfortunately, I don't remember which YT video or homepage it was. Specifically, it's about comparing a transmission line speaker with a closed speaker. Due to the principle, the frequency component that is emitted by the mid-woofer or woofer of the transmission line speaker through the tapered horn channel of the transmission line speaker (is it a horn channel?) is slower than the frequencies that is only emitted forward by the tweeter. At least that's how I understood it.
Does this time delay also apply to loudspeaker designs such as TQWT (or Voights or tapered horns), back-loaded horns or other similar ones, when comparing a closed box to a transmission line?
Cheers, Christian
I read somewhere that according to the loudspeaker design, the base (low frequencies) - or rather the frequency range that is "radiated" by the midrange, low-midrange or woofer, may "lag behind". Unfortunately, I don't remember which YT video or homepage it was. Specifically, it's about comparing a transmission line speaker with a closed speaker. Due to the principle, the frequency component that is emitted by the mid-woofer or woofer of the transmission line speaker through the tapered horn channel of the transmission line speaker (is it a horn channel?) is slower than the frequencies that is only emitted forward by the tweeter. At least that's how I understood it.
Does this time delay also apply to loudspeaker designs such as TQWT (or Voights or tapered horns), back-loaded horns or other similar ones, when comparing a closed box to a transmission line?
Cheers, Christian
Hi, see Group Delay and Excess Group Delay.
Basically all loudspeakers have group delay because low frequency response does not extend all the way to 0Hz, rarely even down to 20Hz, which makes phase shift at lows at least. Both electrical filters, and acoustic filters would introduce this, such as natural high pass of any type of driver / speaker. Most crossovers would make group delay as well even though the system frequency response stays ~flat.
One can get rid of some group delay with FIR type DSP filters, that can be utilized to manipulate phase, simplistic thinking is that it cheat time by delaying the rest of the bandwidth to the longest "group delay". For stereo listening music this is fine, but too much delay in a live music application is distracting, so there could be some limitation how much you can fix with FIR filters. Also not all FIR capable processors have enough "taps" to fix everything, but at least some of group delay can be delt with it. I think it's also possible to introduce audible "ringing" with FIR filters if operator is npt carefull how tge filters are implemented, so it's not automatic silver bullet fix but something one could use.
That said, there are numerous studies that pretty much conclude group delay is not very audible, and provide some thresholds what is audible and what is not. I think audibility might depend on where you listen and how careful you listen to. Differences between more or less group delay are quite subtle fornhome hifi in the big picture, and can be completely overwhelmed by other phenomena such as room modes. There seems to be quite a big difference in sound between various crossover topologies, but group delay is not the only thing that differs so hard to say. Steeper slopes mean more group delay. Although I think I can perceive better dynamics overall with less group delay, kinda more lively sound vs. more boring one, still I would think it's very hard to notice unless lots of listening experience, or capability to AB test it. Still the difference is quite subtle, so might be just expectation bias as well. Logically it could actually change how sound is perceived, because harmonics of sounds are affected which ought to reduce amplitude peaks they make superimposing every fundamental cycle, which hearing system apparently uses for pitch detection and localization for example.
Basically all loudspeakers have group delay because low frequency response does not extend all the way to 0Hz, rarely even down to 20Hz, which makes phase shift at lows at least. Both electrical filters, and acoustic filters would introduce this, such as natural high pass of any type of driver / speaker. Most crossovers would make group delay as well even though the system frequency response stays ~flat.
One can get rid of some group delay with FIR type DSP filters, that can be utilized to manipulate phase, simplistic thinking is that it cheat time by delaying the rest of the bandwidth to the longest "group delay". For stereo listening music this is fine, but too much delay in a live music application is distracting, so there could be some limitation how much you can fix with FIR filters. Also not all FIR capable processors have enough "taps" to fix everything, but at least some of group delay can be delt with it. I think it's also possible to introduce audible "ringing" with FIR filters if operator is npt carefull how tge filters are implemented, so it's not automatic silver bullet fix but something one could use.
That said, there are numerous studies that pretty much conclude group delay is not very audible, and provide some thresholds what is audible and what is not. I think audibility might depend on where you listen and how careful you listen to. Differences between more or less group delay are quite subtle fornhome hifi in the big picture, and can be completely overwhelmed by other phenomena such as room modes. There seems to be quite a big difference in sound between various crossover topologies, but group delay is not the only thing that differs so hard to say. Steeper slopes mean more group delay. Although I think I can perceive better dynamics overall with less group delay, kinda more lively sound vs. more boring one, still I would think it's very hard to notice unless lots of listening experience, or capability to AB test it. Still the difference is quite subtle, so might be just expectation bias as well. Logically it could actually change how sound is perceived, because harmonics of sounds are affected which ought to reduce amplitude peaks they make superimposing every fundamental cycle, which hearing system apparently uses for pitch detection and localization for example.
Last edited:
A loudspeaker with a “hole" in it uses the 180° out of phase output from the back of the driver to reinforce the bass.
In a reflex, it is a helmholz resonator, in a TL the back output is delayed until it is in-phase with the direct output. In a horn (same TL math), that output at the transition from direct to what is coming out the horn shoulkd be exactly an odd number of even wavelengths behind.
So the bass isn’t slower, but what is coming out the ”hole” is at least a ½ cycle behind in time.
dave
In a reflex, it is a helmholz resonator, in a TL the back output is delayed until it is in-phase with the direct output. In a horn (same TL math), that output at the transition from direct to what is coming out the horn shoulkd be exactly an odd number of even wavelengths behind.
So the bass isn’t slower, but what is coming out the ”hole” is at least a ½ cycle behind in time.
dave
For a non-Helmholtz TL/back-horn, the line delay is constructive for wavelength of 1.2X-6X the line-length (so 1.5m-line, 190hz to 38hz), due to simple vector math on phase offset. The upper-frequency range of that is likely low-passed away due to folds and/or stuffing. The lower-frequency range is delayed very little relative to wavelength, but even 5-10ms of delay (1.7-3.4m line plus port-to-baffle) can be perceived as adding venue spaciousness, rather than smearing the direct driver sound. Quarterwave/other resonances (full-line and partials) need to be calculated per MJK/hornresp/akabak/Scottmoose given the actual line geometry. How these resonances are transient and phase delayed exactly, I'm not sure.
Last edited: