Daygloworange said:
If one defines response as "accelerate":
Fc = BL * i - the force on the voice coil related to Bl and the current through the coil.
Since F=Ma:
a = F/M = (BL * i )/ M
Acceleration is inversely proportional to mass, but directly proportional to Bl and i (the current through the coil). You can accelerate a 100 gram woofer as fast as a 5 gram tweeter, it just takes more force.
SPL is proportional to acceleration and cone area. - in fact SPL for a given input is related to: (SD*Bl/Mms*Re)^2
A larger cone doesn't _have to_ accelerate as fast, but at very high frequencies, the radiating area may be smaller, and the acceleration higher as a result. A 5" cone with a 1" whizzer may be behaving approximately as a 1" tweeter at, say, 6kHz, with the result that if a 1" tweeter and our 5" woofer are making the same SPL at 6kHz, they will each be accelerating as fast as the other.
In answer to the OP question.
Jeff Bagby has a good description. I skimmed it and am not sure if he mentioned how important it is that the speakers match closely (fractions of a dB) in frequency response between each other. This is true in two channel as well as multichannel systems.
Theoretically the image must lie on a line between the speakers. Depth, height and beyond speaker images are illusions, yet we have all experienced them when listening to music.
There are anomalies that can sometimes make speakers image "better". crossover gaps (sharp nulls or inverted tweeter polarity) can sometimes give a false sense of depth. Strange sidelobe behavior directed toward floor or ceiling can make for a speaker with more or less height to the image. The room (strength and timing of reflections) and your own HRTF is a big variable here. Some say LEDE rooms are very good for imaging, but I have never experienced one.
Jens Blauert, et al, have written on spatial hearing, mostly on HRTF, but I think there is general talk of psychoacoustics in his book as well.
Jeff Bagby has a good description. I skimmed it and am not sure if he mentioned how important it is that the speakers match closely (fractions of a dB) in frequency response between each other. This is true in two channel as well as multichannel systems.
Theoretically the image must lie on a line between the speakers. Depth, height and beyond speaker images are illusions, yet we have all experienced them when listening to music.
There are anomalies that can sometimes make speakers image "better". crossover gaps (sharp nulls or inverted tweeter polarity) can sometimes give a false sense of depth. Strange sidelobe behavior directed toward floor or ceiling can make for a speaker with more or less height to the image. The room (strength and timing of reflections) and your own HRTF is a big variable here. Some say LEDE rooms are very good for imaging, but I have never experienced one.
Jens Blauert, et al, have written on spatial hearing, mostly on HRTF, but I think there is general talk of psychoacoustics in his book as well.
Matched driver sets go a long way as well. A +/-3db widow can play havoc if you end up with a 6db difference in the same band.
Rob 🙂
Rob 🙂
Loudspeaker Design Cookbook advocates narrow front baffle for imaging, at least for the mid/tweeter.
SY said:
They also had a BBC guy in the team that knew live orchestral music like the back of his hand, for subjective evaluation tests. He voiced it.
Also the size of the box has about the same bulk as a human head. Voices emanate persuasively, especially if you ever heard it in a radio studio from master tape for speech.
Still, a small FR speaker focuses and images better even. If (and only if) measures are taken to have correct tonal balance too. First things first. Its wrong to rely on point source and good phase with a rising FR or severe breakup. Much like designing a tube pre with a shunt noiseless PSU, but using the wrong bias point for the 6DJ8. Useless.
Ron E said:
Acceleration is just one aspect of driver response.
A couple of things you must factor in to the discussion of driver dynamics.
You are taking an example of a tweeter and mid range driver both playing a single frequency, and then building an entire thesis around this.
Your science is correct, however, not entirely comprehensive of what is in fact happening in a real world scenario.
If in fact the theory worked as neatly as this, we would have an ideal speaker with a 18" woofer coupled to a 4" full range driver and be done with it.
The other thing that must be factored in the model you present is not only acceleration, but deceleration of the driver.
Feed both those drivers (1" tweeter and 5" cone driver) a full bandwith signal in their operating range, and chop off the input, then take a look at the spectral decay at the 6k frequency range given in you scenario.
Which driver will have a better spectral decay?
Tying the sides, front, back of the enclosure with horizontal bracing yield the best gains in stiffening of the cabinet, and minimize internal wave disturbances compared to vertical bracing.
There are numerous ways to dampen enclosures. Adding mass is a popular way. There is also contrained layer damping of the enclosure panels, which gets more complex.
There are plenty of threads if one were to search around on the topic.
Cheers
Daygloworange said:
I just defined "response" as acceleration and showed a way a 5" "woofer" could "respond" as fast as a 1" tweeter. Just a simple exercise in "simple physics" 😉 I did not attempt to build some general theory out of it, you took care of extending it into that....Perhaps a better example would be between a 20gram 1" tweeter vs. a 5 gram 1" tweeter. Which would have better spectral decay? Which would likely have smoother frequency response?
Define "better spectral decay" 😉 Then explain how well we hear this, and the ramifications. Ever see the spectral decay of an electrostat?
Ron E said:
Which one would continue to ring longer in the 6k region after the signal was chopped off in the scenario given?
The one that rings longer has greater non linearity, in that frequency range it will be quite audible.
The other thing I forgot to mention was the dispersion characteristics of a smaller driver vs a larger one as frequency increases. The smaller driver wins here as well. This is also important for a speakers ability to image well.
You mean of an electrostatic panel? No I haven't. I am curious what a large surface like that would perform like at different frequencies, in CSD testing.
Cheers
Daygloworange said:
How do you use Black hole as damping material inside the cabinet, lining the material, at the Back(top to bottom or just behind drivers), top only and one side(top to bottom) or both sides? if not how?
You would want to add damping material (in this case Blackhole 5) to any enclosure panel area. Sides, back, top and bottom if you can.
If you envision the speaker enclosure as a drum, then you would want to dampen the drum as much as possible, by damping as much surface area as effectively as possible.
There isn't (typically) much surface area on the back of the front baffle to apply a damping material (such as Blackhole 5), and you would want to avoid shrouding the back portion of the driver with the foam that is part of the Blackhole 5 composite.
The added benefit of Blackhole 5 is that the foam is designed to minimize the effects of the backwave of the driver coming back to the driver and causing non-linearities by re-animating the cone.
Cheers
If you envision the speaker enclosure as a drum, then you would want to dampen the drum as much as possible, by damping as much surface area as effectively as possible.
There isn't (typically) much surface area on the back of the front baffle to apply a damping material (such as Blackhole 5), and you would want to avoid shrouding the back portion of the driver with the foam that is part of the Blackhole 5 composite.
The added benefit of Blackhole 5 is that the foam is designed to minimize the effects of the backwave of the driver coming back to the driver and causing non-linearities by re-animating the cone.
Cheers
Daveis said:
And wide baffles ofer poor imaging? I can't imagine why this might be so. What is the science behind this?
I think narrow baffles give more precise, pinpoint imaging - localization of instruments, while wide baffles present a larger sound stage, which is more natural, and closer to the live event.
Not so easy to have both qualities on the same speaker.
Not so easy to have both qualities on the same speaker.
The best possible soundstage would be created by two point sources without reflections, for example headphones or speakers in free space. When you put speakers into a room, you get reflections from boundaries in the room. This messes up the image, as it provides additional information, which was not in the original. Driver to driver interference also provides additional information like comb filtering. The speaker baffle introduces diffraction effects, which also provide information, which was not in the original. To get your soundstage exact, you have to deal with the things that get your speakers further away from the ideal.
For example, you can build a constant directivity device, which has an equal response on and off axis and which has a narrow dispersion, to minimize room interaction. You can also make room treatments, which minimize the room influence.
For example, you can build a constant directivity device, which has an equal response on and off axis and which has a narrow dispersion, to minimize room interaction. You can also make room treatments, which minimize the room influence.
Daygloworange said:
Looks rather like grass with paths in it....Those big stators are somewhat like NXT panels.
Daygloworange said:
My RD75's/ATC neoplanars in dipole configuration with notch filter @ 5.5 Khz have minimal audible decay problems, low latency in the pass band
Daygloworange said:
xover @ 600 Hz to mid woofers, 6 Khz to tweeters, 8Khz to supertweeter planars run free from 600 Hz up, no upper xover
Daygloworange said:
Dipole dispersion from linesource w/o vertical lobing, low floor bounce with broad horizontal pattern
An externally hosted image should be here but it was not working when we last tested it.
Daygloworange said:
minimal cabinet used, dipole baffle ~ 18" wide with swept wings provides necessary lf reinforcement w/o significant resonance issues
Daygloworange said:
Poly caps (multiple stacks to reduce esr), no bypass used (didn't notice any difference), open core large guage inductors
An externally hosted image should be here but it was not working when we last tested it.
Daygloworange said:
Dipole radiation minimizes room reflections, improves imaging dramatically
Daygloworange said:
No need for bass traps, subwoofer positioning as well as active room correction resolves room modes. Only downside is the size of these speakers (if that matters... 😀 😉
John L.
Cabinet edges cause defraction. The defracted signal is delayed. This reduces precise image. The wider the baffle is, the more delay the defracted signal becomes. This is possibly one of the reasons that narrow baffle produces better image while larger baffle produces larger image due to the longer delay. If practical, I would choose a narrow baffle.
Boundary reflections less than about 5ms are perceived by human ears as part of the original signal and this cause damage to the original image. Reflections longer than 10ms are perceived as separated from the original sound to be "good" air. Place your speakers 2m away from the front wall and use acoustic damping materials (foam block, carpet) to damp all early reflections less than 5ms on the floor (thick wool carpet), ceiling, side walls, and hear the image of your speakers dramatically improved. I did just that. Sound travels 344m per second, so a speaker placed 2m away from the front wall has the front wall reflection delayed by (2+2)/344=0.01163, i.e. about 12ms.
My 2 cents.
Regards,
Bill
Boundary reflections less than about 5ms are perceived by human ears as part of the original signal and this cause damage to the original image. Reflections longer than 10ms are perceived as separated from the original sound to be "good" air. Place your speakers 2m away from the front wall and use acoustic damping materials (foam block, carpet) to damp all early reflections less than 5ms on the floor (thick wool carpet), ceiling, side walls, and hear the image of your speakers dramatically improved. I did just that. Sound travels 344m per second, so a speaker placed 2m away from the front wall has the front wall reflection delayed by (2+2)/344=0.01163, i.e. about 12ms.
My 2 cents.
Regards,
Bill
Here is a link to some responses taken by Dennis Murphy of a small 2 way speaker with, and without roundovers on the top, bottom and vertical edges of the front baffle .
http://murphyblaster.com/content.php?f=cabinets.html
Cheers
http://murphyblaster.com/content.php?f=cabinets.html
Cheers
Ron E said:
They don't have to be..
http://stereophile.com/floorloudspeakers/1204innersound/index4.html
http://stereophile.com/floorloudspeakers/819/index6.html
But they often are:
http://stereophile.com/floorloudspeakers/145/index12.html
Thanks for the links.
What I find interesting are how the cabinet resonances affect the woofer's impedance response on the Innersound Kaya panel hybrid.
I think it goes to show that you must factor a lot of things into the proper implementation of drivers and enclosures.
Cheers
What I find interesting are how the cabinet resonances affect the woofer's impedance response on the Innersound Kaya panel hybrid.
I think it goes to show that you must factor a lot of things into the proper implementation of drivers and enclosures.
Cheers
HiFiNutNut said:
i had built the narrow baffle about 5 years. i never got the image.
i cannot agree with you. you say that it's easy to get the image.
no....no
hey everybody talk a long time. i think we stop a few minutes. and drink a water(like a football match) and talk agian. But at now i'm to be tired. i will stop to talk a few minutes.
😀 😀 😀 😀 😀
- Status
- Not open for further replies.