If you are talking about a MTM using a pair of 18" and a 15" coax, especially knowing that you will be pushing the LF 10-20 dB louder below 100 Hz mixing hip-hop, I'd suggest 100 Hz for the 15" crossover- still allows 114dB at one meter with only 2.5mm excursion, sealed cabinet.
Push a pair of sealed 18" LF up to 114dB at 30 Hz, they are moving almost 10mm, at 300 Hz vocals will sound trashy/gargly through them, not to mention it just plain sounds lousy to have the fundamental baritone vocal range (100-400Hz) separated by almost a meter center to center when listening at a meter or two.
You are right....I would have to add more woofers to cover and lower excursion in the low end. One iteration would be to put the coaxial on top of the 15mid woofer which sits on top of the pairs of 18"s covering the bottom. This would allow proper ctc spacing. Looking at a sim of a normal 15" it would seems a cross over 100hz or higher makes sense....likely higher. Keeping excursion low it would seem that the woofer would not disturb the tweeter so much.
"My 2c take on coaxes is that a vibrating cone doesn't make the worlds best waveguide. And that is why the smaller coaxes shine the brightest."
My first thoughts are, "stop running the coaxial so low" and "large woofers, have less excursion"
I have not looked into why 8' is a good coaxial size but excursion is not the answer. It must be the lesser amount of width, interacting with the tweeter?
I believe you are speaking of the sweet spot. Very good point. I can't wait to compare and contrast a large horn with a nice coaxial.
"My 2c take on coaxes is that a vibrating cone doesn't make the worlds best waveguide. And that is why the smaller coaxes shine the brightest."
My first thoughts are, "stop running the coaxial so low" and "large woofers, have less excursion"
I have not looked into why 8' is a good coaxial size but excursion is not the answer. It must be the lesser amount of width, interacting with the tweeter?
I believe you are speaking of the sweet spot. Very good point. I can't wait to compare and contrast a large horn with a nice coaxial.
Large woofers may have a relatively lower excursion for a given low frequency output, but they also suffer more from beaming effects. You’ll notice many of the pro audio coaxials from 12” upward offer units with dedicated horns for this reason. That brings its own issues, of course.
If you’re dead set on a coaxial solution, have you considered the BMS tri-axial?
If you’re dead set on a coaxial solution, have you considered the BMS tri-axial?
No, I am just experimenting. The multiway seems to be infallible within the sweet spot. I just want to test that theory against a coaxial. Reading the comments here, it seems that phase correction with a multi-way will give better results than a coaxial, as long as, within the sweet spot. Its tuff to beat a large horn in terms of amount of directivity....but at close range and with high output potential it would seem the right coaxial would give good results....but my question is, is there something there that the multiway doesn't have. I found a cheap pair of vx15hp to play with to compare and contrast with the horn
I had a guy tell me that the altec 604 isn't really so great, its just that it was the best at the time. The vx15hp seems a little more contemporary, higher sensitivity and +\- 3db FR to start with. Experience is the best teacher, plus Ive been able to break even selling unused audio things, later on
I had a guy tell me that the altec 604 isn't really so great, its just that it was the best at the time. The vx15hp seems a little more contemporary, higher sensitivity and +\- 3db FR to start with. Experience is the best teacher, plus Ive been able to break even selling unused audio things, later on
Last edited:
Mark100 said:
I have a theory...lol...I bet that sensitivity has a direct effect on.....the start up transient =)
No ones thought this before, tested or proven this point one way or another? If sensitivity reflects a more efficient transfer of electrical energy to mechanical....then it's unlikely that it does not have any measurable affects on the signal.
Last edited:
Voltage sensitivity is a more complex topic than meets the eye....dispersion plays a factor.
With two drivers of the same type and diameter, say 15"...it would seem that being more sensitive to voltage would result in better a better start up transient....the lower sensitivity driver obviously had resistance in the line and I highly doubt its linear.
With two drivers of the same type and diameter, say 15"...it would seem that being more sensitive to voltage would result in better a better start up transient....the lower sensitivity driver obviously had resistance in the line and I highly doubt its linear.
Dispersion is quite a large factor... Sensitivity is literally defined by the directivity index of a loudspeaker. It is a function of efficiency and directivity.
The former is a measure of sound power, and the latter sound pressure. They are measured differently, but can be converted if you know the spherical dispersion of the device under test:
Efficiency and sensitivity conversion - loudspeaker percent and dB per watt and meter loudspeaker efficiency versus sensitivity vs speaker sensitivity 1 watt = 2,83 volt box chart - sengpielaudio Sengpiel Berlin
Sound power does not change with distance or angle.
That link has a section at the bottom on "Loudness" that may be of interest.
Are you talking at an equivalent output sound pressure level, or with equivalent drive voltage? Do both drivers have the same dispersion or not?
The former is a measure of sound power, and the latter sound pressure. They are measured differently, but can be converted if you know the spherical dispersion of the device under test:
Efficiency and sensitivity conversion - loudspeaker percent and dB per watt and meter loudspeaker efficiency versus sensitivity vs speaker sensitivity 1 watt = 2,83 volt box chart - sengpielaudio Sengpiel Berlin
Sound power does not change with distance or angle.
That link has a section at the bottom on "Loudness" that may be of interest.
Are you talking at an equivalent output sound pressure level, or with equivalent drive voltage? Do both drivers have the same dispersion or not?
Last edited:
Ain't this the good old frequency response again, no need to invent it again!?
The "problems transferring the electrical energy to mechanical energy" manifests itself in the frequency response as well as impulse response, which are the same data plotted in different perspective. For most part you can compensate with EQ within the capabilities of the driver in question, hence the use of crossovers because the drivers can't do full bandwidth well alone. Combined multiway system will have nice impulse and frequency response and nice transients.
If you measure at the verge of the system capability, very loud, you'll see the response is poor because now the various parts have non-linear effects on them. Lower the SPL where the system is linear and all is golden. Here is the relation to sensitivity and efficiency to the transients and response in general. I suspect if you measure any system with 1W amplifier power, there is not much difference between high or low sensitivity systems per se, most should be comfortable with 1W.
Another nugget to chew: Listen off-axis and the transients seem poor compared to on-axis because the frequency response has a slope due to directivity. It is exact same system only listened from different perspective. If it was true constant directivity system from bottom to top the transients would sound exactly the same off-axis, only quieter!
If you wanna get back to the group delay stuff, just use FIR filter to linearize the phase to be sure there is no audible group delay.
Last edited:
edit time over. I gotta add disclaimer I'm not too comfortable writing this stuff, but this is how I've understood it. High risk triggering another month long debate as well. maybe I should not have written the post. Well, anyway, the motivation is to learn so what the heck, no better place to post it than this thread
Last edited:
Visuals to aid. Here is step and impulse response of some of the previous examples I posted. The first is from the simulated listening spot, ~90cm tweeter and ear height sitting down 3 meters away. The second one is otherwise exactly the same system but the simulator coordinates are tweaked to show response if standing up at listening position, 90cm above the tweeter axis 3 meters away. Third and fourth are the respective six-packs showing the other responses for reference.
Either "position" could be made perfect with changing the crossovers, relative delays, EQ and linearizing the phase with FIR but not both because the drivers are at separate physical locations. On a coaxial (constant directivity) system both would be more similar to start with. Of course the room would affect in any of these cases etc., the main point here is to illustrate the transient performance depends on where you are at and the response can be tweaked to be better at that position. Only thing one has to care about is the frequency response / phase, and the transients are as good as they are. It would be really difficult to EQ a system looking at the impulse/step response, luckily it can be done by looking at the frequency response and phase!
Edit added GIF of the impulses aligned. There is time of flight difference as well as difference in the shape.
ps. I haven't looked into how to make FIR filters, otherwise would have added that here as well
Either "position" could be made perfect with changing the crossovers, relative delays, EQ and linearizing the phase with FIR but not both because the drivers are at separate physical locations. On a coaxial (constant directivity) system both would be more similar to start with. Of course the room would affect in any of these cases etc., the main point here is to illustrate the transient performance depends on where you are at and the response can be tweaked to be better at that position. Only thing one has to care about is the frequency response / phase, and the transients are as good as they are. It would be really difficult to EQ a system looking at the impulse/step response, luckily it can be done by looking at the frequency response and phase!
Edit added GIF of the impulses aligned. There is time of flight difference as well as difference in the shape.
ps. I haven't looked into how to make FIR filters, otherwise would have added that here as well
Attachments
Last edited:
A thought experiment:
If one has a beaming system the sweetspot whose frequency response (AND "transient response") is tweaked to perfection is small. Approach a constant directivity system and the sweet spot would get bigger. Wider in case of stacked non-coincident sources. Make the system coincident and the response would be great to any direction. But there is still the fact the frequency and "transient" response can be tweaked to a sweet spot. If the standing up position was used as reference in the example the graphs would pretty much flip around, sitting down would not be so nice as standing up. Anyway, you get the point.
It is responsibility of the designer to design a system that has the best sound to a sweet spot. It is then the users duty to arrange the listening situation so that the sweet spot happens where designed, otherwise the response is not as good as intended. Doing It (a system) Yourself for a particular application will allow very good sound but it requires understanding how that can be achieved for the application, including effects of the room.
If one has a beaming system the sweetspot whose frequency response (AND "transient response") is tweaked to perfection is small. Approach a constant directivity system and the sweet spot would get bigger. Wider in case of stacked non-coincident sources. Make the system coincident and the response would be great to any direction. But there is still the fact the frequency and "transient" response can be tweaked to a sweet spot. If the standing up position was used as reference in the example the graphs would pretty much flip around, sitting down would not be so nice as standing up. Anyway, you get the point.
It is responsibility of the designer to design a system that has the best sound to a sweet spot. It is then the users duty to arrange the listening situation so that the sweet spot happens where designed, otherwise the response is not as good as intended. Doing It (a system) Yourself for a particular application will allow very good sound but it requires understanding how that can be achieved for the application, including effects of the room.
Last edited:
- This would be the take away for me. You still left out reverberant field but I get what you are saying. I fell asleep reading this article last night, There are many good points in it.
What Is Microphone Transient Response & Why Is It Important? – My New Microphone
- Not to go back to group delay, unless there is a connection. We might have to confine this argument to dynamic woofer and compression drivers. Electro stat and ribbon obviously have something else going on.
Within high vs low sensitivity drivers, is there not a perceived coloration difference, between the two groups?
Last edited:
Thought experiment continued: Why a coaxial system would sound better given that a coaxial and the non-coincident system both could be tuned for same response at a particular spot? Reasoning from the examples there would be at least two reasons it would. Other is effect of the room in a far field listening situation and the other is sensitivity to the sweet spot in near field listening situation, head moves on the vertical plane relatively much.
In a far field listening situation the room sound (power response) would dominate over direct sound and could be more uniform with a coaxial system but earlier simulated demonstration showed this could be achieved with some non-coincident systems as well. There should not be much difference between the two with far field listening and minimized effect of the room so a coaxial system would not benefit much or at all. Near field I think coaxial could possibly always win, more stable sound, directivity further off-axis doesn't matter too much.
In a far field listening situation the room sound (power response) would dominate over direct sound and could be more uniform with a coaxial system but earlier simulated demonstration showed this could be achieved with some non-coincident systems as well. There should not be much difference between the two with far field listening and minimized effect of the room so a coaxial system would not benefit much or at all. Near field I think coaxial could possibly always win, more stable sound, directivity further off-axis doesn't matter too much.
Last edited:
^^ yeah I don't want to comment on audibility of various cone materials etc. I find more interesting to think the stuff at the root level I've read the forums quite much and different people seem to love different materials and I suspect most of the difference is actually in the implementation. Either could work better on some other use case. After one has found good system design there is possibility to try various different driver models, why not. I suspect there would be better and worse sounding specimen and a large grey mass between that all sound equally good or bad what the design allows. Same for electrostats and ribbons, might be the material or just the directivity or extended frequency response or what ever. I've never heard either so can't comment If building top notch system, the design should be spot on for the application and the drivers should be top shelf stuff.
I've read the forums quite much and different people seem to love different materials and I suspect most of the difference is actually in the implementation. Either could work better on some other use case. After one has found good system design there is possibility to try various different driver models, why not. I suspect there would be better and worse sounding specimen and a large grey mass between that all sound equally good or bad what the design allows. Same for electrostats and ribbons, might be the material or just the directivity or extended frequency response or what ever. I've never heard either so can't comment If building top notch system, the design should be spot on for the application and the drivers should be top shelf stuff.
Last edited:
As long as you can stay away from all the resonances, distortion and other effects of certain materials, every cone becomes exactly the same thing.
There's a fun paper from my university on the topic of off-axis response and listener preference. It's done using simulated impulse responses of a pro audio array system, in an anechoic environment, but the results are still very interesting in light of the objective 'badness' of the off-axis responses versus the on-axis ones:
(PDF) Temporal Distortion Audibility of Directional Loudspeaker Arrays
The full paper is free access.