Hello there!
I’ve been reading along on the forum for a while now and until now I didn’t really have the guts to ask any questions as I would hate to be one of those newbies who asks all the silly questions.
At the moment I should actually be focusing on my exams, but designing a point-source speaker has been an idea of mine that I can’t seem to let go. Unfortunately I don’t have the time to reinvent the wheel, do all the measurements myself and draw my conclusions from those.
Because of all this I have decided to start a thread, hoping that there is somebody out there who can help me out, answer some of my questions or maybe one of you guys has tried any of the ideas below. This way I can hopefully satisfy my curiosity without having to spend precious time doing measurements of different speaker designs.
The easiest solution for a point source speaker is a small full-range driver, of course. I have heard some 3’’ FR’s and I have kind of fallen in love with the pro’s that they offer. I spend a lot of time at my desk because of school etc. and the rumors are true. The imaging and sound stage seem so much better! Especially at such a short distance a lack of time issues, lobbing and reflections really improves the experience in my opinion. The only downside to me is that they kind of lose their sparkle as soon as you move your head out of the ‘sweet spot’. Therefore a tweeter seems mandatory to me and this brings me to coaxial speakers.
Here’s what I know:
Coaxially mounting a tweeter has its obvious pro’s, ranging from no lobbing issues to an (arguable) acoustic center alignment between tweeter and mid-woofer in case the woofer is designed like the Seas T18REX/XFC - H1353-08/06 or the more affordable SB Acoustics SB13PFC25-4-COAX.
Although, in these cases diffraction from the surround of the woofer and the transition from tweeter to woofer cone are an issue. I feel like I could live with some diffraction issues as they seem to be reduced when listened to on the right angle. Also, some say your ears are not sensitive enough to actually perceive the dips in the response of, say, a whizzer cone or the diffracted response of a coaxially mounted tweeter. This is sub-optimal but hey, you can’t have everything in life. The biggest issue would be the IMD caused by the moving cone. Introducing the need for a woofer playing everything below, say, 500Hz or maybe even higher. This means you’ll need to design a proper phase-coherent three-way crossover. I wouldn’t consider the need for an extra woofer an issue as I would often use a subwoofer for the good stuff below 100Hz.
A woofer like the Oberton 12H4CX72 seems to be the solution to the issues mentioned above because the tweeter has its own horn and doesn’t use the cone of the woofer as its moving waveguide. Unfortunately it also introduces some other issues. One issue is the fact that the tweeter is set back a good 160-170mm (6-7’’) as the compression driver is screwed onto the back of the woofer’s magnet. A clever crossover design might be able to fix that (I suppose?) although I am not sure. Please let me know whether it is possible to fix a slight/major time-alignment difference through a passive crossover 😉
In case of the 12H4CX72, it seems like there are considerable cancellations between the cone of the woofer and the protruding horn. I have read that as long as the horn is smaller in diameter than the wavelength of the frequencies that the woofer has to play, this shouldn’t be an issue. The distance between the edge of the horn and the woofer below it also plays a major role in these cancellations I presume. Meaning that in this case the woofer and the CD need to be crossed at roughly 1,2kHz or below. This might explain why the horn is so long, it needs to amplify the CD all the way down to 1000Hz.
This 12’’ woofer is not really an option because of its size but the Beyma 6cx200fe might be a good option. This woofer also doesn’t seem to have nearly as many reflection issues as the Oberton (obviously) and for ‘poor’ students / ’cheapies’ there is even a P-Audio knockoff available, the sn6-150cx, which also measures pretty well (presuming that P-Audio’s graphs are somewhat trustworthy). I guess you’re going to hand in some sound quality (higher distortion) when choosing a knockoff but then again, you can’t have everything life.
Then I remembered that those car audio coaxials, which everybody seems to hate, still exist and I wondered what the downsides were from simply mounting a decent small neodymium tweeter in front of a proper hi-fi woofer. I guess that if you use a small tweeter and cross both tweeter and woofer low enough, you shouldn’t run into any reflection issues caused by the tweeter obstructing the center of the woofer where the remaining few beaming high frequencies would normally come from. It still wouldn’t be ‘time aligned’ like a Seas T18REX/XFC - H1353-08/06 would be but normal two-ways aren’t physically time aligned either.
I know that the lack of a baffle would influence the sound of the tweeter compared to the infinite baffle measurements supplied by the manufacturer. Running a tweeter with a small bump in the 2kHz region, like the Peerless BC25SC55-04 through the VituixCAD Baflle step simulator on a 70*55mm baffle (roughly the tweeter’s own flange) doesn’t seem to influence its response too much in a negative way, some might even argue it looks better than it used to. I would post a picture of the simulated response but I can't figure out how to simply upload a picture in a thread like this...🙁
I am afraid that using a regular dome tweeter would still result in the tweeter getting horn loaded by the woofer so I feel like a tweeter with its own small waveguide might be a better solution. The Peerless BC25SC06-04 for example.
The back of those tweeters doesn’t look particularly aerodynamic so maybe you should mount them in a more aerodynamic sphere-thingy to improve airflow around it.
I have stumbled upon the Fluid Audio FX8. This speaker is a good representation of what I had in mind. It does make some bold claims like a frequency response of 35 Hz – 22 kHz (+/-3dB) but based on some measurements I found online those numbers are a little exaggerated: Test des Fluid Audio FX8 : Concentre de fluide - Audiofanzine. I’m not sure how accurate those measurements are though because the overlayed Presonus speaker seems to exhibit the similar peaks and valleys so my guess is that both of those measurements are influenced by the room.
Some other coaxial speakers that use ideas similar to the ones described above are those made by Geithain GmbH: musikelectronic geithain gmbh - Active loudspeaker. Their measurements look a little more trustworthy to me. They should be, considering the price these monitors go for!
So, after reading all of this, what are your thoughts on the issue? Is a point-source even worth pursuing or should I just stick to designing a good two-way, minimizing the lobbing issues predicted by Xdir or maybe even build an MTM with minimal ctc spacing. Those look like they have very little lobbing issues when done right but they still aren’t a true point source, which most people prefer at shorter distances (desktop and other near-field applications).
Or should I just accept the IMD caused by cone travel? KEF can get away with a 5’’ woofer doing a good 6-8mm peak to peak in the LS50 without anybody complaining about audible HF distortion and the SEAS Loki also doesn’t include a woofer to play everything below 500Hz either.
Maybe a good compromise would be crossing a regular coaxial woofer like the one in the SEAS Loki to a subwoofer at 200Hz and simply building a floor standing speaker. In other words, a tall subwoofer with a separate compartment for the coaxial woofer. Keeping the distance between subwoofer and coax in mind to preserve the point-source experience. This minimizes cone travel of the coax and because the crossover point is below the crucial 300-5000Hz range, where the human ear is said to be the most sensitive, I can simply use the digital crossover from my AV Receiver.
Some might even say that desktop listening is such a terrible idea that I should just use the cheap P-Audio coaxials, accept their distortion levels and just use a proper subwoofer for anything below 100Hz
Isn’t the concept of mounting a small neodymium tweeter in a pod in front of a woofer a very good idea, despite the fact that everybody looks down on car audio coaxials? My own thoughts are that when the right components are used (a good waveguided tweeter and a proper woofer) and the crossover is designed with the right things in mind, a coaxial like the FX8 should be superior to the alternatives mentioned above, right?
Feel free to tell me where my thinking is wrong and guide me into the right direction. There must be a reason speakers like the Fluid Audio FX8 are not more popular…
I’ve been reading along on the forum for a while now and until now I didn’t really have the guts to ask any questions as I would hate to be one of those newbies who asks all the silly questions.
At the moment I should actually be focusing on my exams, but designing a point-source speaker has been an idea of mine that I can’t seem to let go. Unfortunately I don’t have the time to reinvent the wheel, do all the measurements myself and draw my conclusions from those.
Because of all this I have decided to start a thread, hoping that there is somebody out there who can help me out, answer some of my questions or maybe one of you guys has tried any of the ideas below. This way I can hopefully satisfy my curiosity without having to spend precious time doing measurements of different speaker designs.
The easiest solution for a point source speaker is a small full-range driver, of course. I have heard some 3’’ FR’s and I have kind of fallen in love with the pro’s that they offer. I spend a lot of time at my desk because of school etc. and the rumors are true. The imaging and sound stage seem so much better! Especially at such a short distance a lack of time issues, lobbing and reflections really improves the experience in my opinion. The only downside to me is that they kind of lose their sparkle as soon as you move your head out of the ‘sweet spot’. Therefore a tweeter seems mandatory to me and this brings me to coaxial speakers.
Here’s what I know:
Coaxially mounting a tweeter has its obvious pro’s, ranging from no lobbing issues to an (arguable) acoustic center alignment between tweeter and mid-woofer in case the woofer is designed like the Seas T18REX/XFC - H1353-08/06 or the more affordable SB Acoustics SB13PFC25-4-COAX.
Although, in these cases diffraction from the surround of the woofer and the transition from tweeter to woofer cone are an issue. I feel like I could live with some diffraction issues as they seem to be reduced when listened to on the right angle. Also, some say your ears are not sensitive enough to actually perceive the dips in the response of, say, a whizzer cone or the diffracted response of a coaxially mounted tweeter. This is sub-optimal but hey, you can’t have everything in life. The biggest issue would be the IMD caused by the moving cone. Introducing the need for a woofer playing everything below, say, 500Hz or maybe even higher. This means you’ll need to design a proper phase-coherent three-way crossover. I wouldn’t consider the need for an extra woofer an issue as I would often use a subwoofer for the good stuff below 100Hz.
A woofer like the Oberton 12H4CX72 seems to be the solution to the issues mentioned above because the tweeter has its own horn and doesn’t use the cone of the woofer as its moving waveguide. Unfortunately it also introduces some other issues. One issue is the fact that the tweeter is set back a good 160-170mm (6-7’’) as the compression driver is screwed onto the back of the woofer’s magnet. A clever crossover design might be able to fix that (I suppose?) although I am not sure. Please let me know whether it is possible to fix a slight/major time-alignment difference through a passive crossover 😉
In case of the 12H4CX72, it seems like there are considerable cancellations between the cone of the woofer and the protruding horn. I have read that as long as the horn is smaller in diameter than the wavelength of the frequencies that the woofer has to play, this shouldn’t be an issue. The distance between the edge of the horn and the woofer below it also plays a major role in these cancellations I presume. Meaning that in this case the woofer and the CD need to be crossed at roughly 1,2kHz or below. This might explain why the horn is so long, it needs to amplify the CD all the way down to 1000Hz.
This 12’’ woofer is not really an option because of its size but the Beyma 6cx200fe might be a good option. This woofer also doesn’t seem to have nearly as many reflection issues as the Oberton (obviously) and for ‘poor’ students / ’cheapies’ there is even a P-Audio knockoff available, the sn6-150cx, which also measures pretty well (presuming that P-Audio’s graphs are somewhat trustworthy). I guess you’re going to hand in some sound quality (higher distortion) when choosing a knockoff but then again, you can’t have everything life.
Then I remembered that those car audio coaxials, which everybody seems to hate, still exist and I wondered what the downsides were from simply mounting a decent small neodymium tweeter in front of a proper hi-fi woofer. I guess that if you use a small tweeter and cross both tweeter and woofer low enough, you shouldn’t run into any reflection issues caused by the tweeter obstructing the center of the woofer where the remaining few beaming high frequencies would normally come from. It still wouldn’t be ‘time aligned’ like a Seas T18REX/XFC - H1353-08/06 would be but normal two-ways aren’t physically time aligned either.
I know that the lack of a baffle would influence the sound of the tweeter compared to the infinite baffle measurements supplied by the manufacturer. Running a tweeter with a small bump in the 2kHz region, like the Peerless BC25SC55-04 through the VituixCAD Baflle step simulator on a 70*55mm baffle (roughly the tweeter’s own flange) doesn’t seem to influence its response too much in a negative way, some might even argue it looks better than it used to. I would post a picture of the simulated response but I can't figure out how to simply upload a picture in a thread like this...🙁
I am afraid that using a regular dome tweeter would still result in the tweeter getting horn loaded by the woofer so I feel like a tweeter with its own small waveguide might be a better solution. The Peerless BC25SC06-04 for example.
The back of those tweeters doesn’t look particularly aerodynamic so maybe you should mount them in a more aerodynamic sphere-thingy to improve airflow around it.
I have stumbled upon the Fluid Audio FX8. This speaker is a good representation of what I had in mind. It does make some bold claims like a frequency response of 35 Hz – 22 kHz (+/-3dB) but based on some measurements I found online those numbers are a little exaggerated: Test des Fluid Audio FX8 : Concentre de fluide - Audiofanzine. I’m not sure how accurate those measurements are though because the overlayed Presonus speaker seems to exhibit the similar peaks and valleys so my guess is that both of those measurements are influenced by the room.
Some other coaxial speakers that use ideas similar to the ones described above are those made by Geithain GmbH: musikelectronic geithain gmbh - Active loudspeaker. Their measurements look a little more trustworthy to me. They should be, considering the price these monitors go for!
So, after reading all of this, what are your thoughts on the issue? Is a point-source even worth pursuing or should I just stick to designing a good two-way, minimizing the lobbing issues predicted by Xdir or maybe even build an MTM with minimal ctc spacing. Those look like they have very little lobbing issues when done right but they still aren’t a true point source, which most people prefer at shorter distances (desktop and other near-field applications).
Or should I just accept the IMD caused by cone travel? KEF can get away with a 5’’ woofer doing a good 6-8mm peak to peak in the LS50 without anybody complaining about audible HF distortion and the SEAS Loki also doesn’t include a woofer to play everything below 500Hz either.
Maybe a good compromise would be crossing a regular coaxial woofer like the one in the SEAS Loki to a subwoofer at 200Hz and simply building a floor standing speaker. In other words, a tall subwoofer with a separate compartment for the coaxial woofer. Keeping the distance between subwoofer and coax in mind to preserve the point-source experience. This minimizes cone travel of the coax and because the crossover point is below the crucial 300-5000Hz range, where the human ear is said to be the most sensitive, I can simply use the digital crossover from my AV Receiver.
Some might even say that desktop listening is such a terrible idea that I should just use the cheap P-Audio coaxials, accept their distortion levels and just use a proper subwoofer for anything below 100Hz
Isn’t the concept of mounting a small neodymium tweeter in a pod in front of a woofer a very good idea, despite the fact that everybody looks down on car audio coaxials? My own thoughts are that when the right components are used (a good waveguided tweeter and a proper woofer) and the crossover is designed with the right things in mind, a coaxial like the FX8 should be superior to the alternatives mentioned above, right?
Feel free to tell me where my thinking is wrong and guide me into the right direction. There must be a reason speakers like the Fluid Audio FX8 are not more popular…
This design from Seas looks like it would satisfy your needs.
SEAS KingRO4Y Mk II
This is a route I would persue if I was looking for a nearfield solution. I don't like the idea of discontinuities caused by mounting stuff in front of the cone. Limiting excursion of the coaxial driver is the best compromise to me.
SEAS KingRO4Y Mk II
This is a route I would persue if I was looking for a nearfield solution. I don't like the idea of discontinuities caused by mounting stuff in front of the cone. Limiting excursion of the coaxial driver is the best compromise to me.
Geithain is the company which does nearfield monitors in the way you described here. I doubt this approach has more pro's than normal, forethought and carefully crafted multi-driver design. When looking for well-designed coaxial driver obtainable by DIY'ers there is not so many choices left. Making good coaxial driver and then, good system using coaxial driver isn't easy and obvious task at all. Car audio 'coaxials' aren't suitable way how the things should be done in domestic audio and it is out of question.
Which things are commonly overlooked in coaxial drivers and systems desings?
- Uninterrupted waveguiding surface smoothly leading high-frequency sound from tweeter surround until baffle edges (better: roundovers). It excludes whizzer cones, tweeter horns etc from correct designs.
- Little to no waveguiding cone displacement allowable, it excludes designs trying to reproduce bass using coaxial driver at all. Frequency modulation distortion type leads to muddy, non-inteliglible sound. Very common fault across many coaxial desings, almost total design flaw.
- Flat surrounds without sudden geometry changes, especially positive (protruding into waveguide geometry). It excludes 90% of known suspension geometries - single roll, M-roll etc from being good coaxial designs.
- Careful physical centering tweeter dome inside of acoustical center of midrange cone. Shifting tweeter ahead of midrange driver destroys wavefront coherence, trying to correct time-domain response axially leads to severe phase coherence deteriorations off-axis. This is in case of car audio 'coaxials' but also applies for whizzer cones, additional horns isnide mid cone etc.
- Weak tweeter power and stroke loading capabilities, It excludes every driver which is not possible to operate its tweeter below 2kHz where midrange cone still is not already beaming severely. Sensitivity of domestic coaxials also is quite disappointing.
- Trying to evaluate coaxial driver's performance by looking at its on-axis diffraction dips and on-axis hi-frequency roll-off is a general mistake. Most of coaxial designs are plagued with diffraction issues which creates known dips on the frequencies 3-10kHz related to surround's bumps. These dips occures only on axis, but overall power response thrown into room usually is smooth, smoother than in discrete multi-driver designs. What is more, waveguiding of upper midrange-low trebles critical area lessens power output at these frequencies so hi-frequency roll-off becomes compensated and final tonal balance is close to optimal. Good practice is to listen to them a little bit off-axis.
I don't know it what I said is clear enough or complete, it is late here. I'm sorry for poor language.
Which things are commonly overlooked in coaxial drivers and systems desings?
- Uninterrupted waveguiding surface smoothly leading high-frequency sound from tweeter surround until baffle edges (better: roundovers). It excludes whizzer cones, tweeter horns etc from correct designs.
- Little to no waveguiding cone displacement allowable, it excludes designs trying to reproduce bass using coaxial driver at all. Frequency modulation distortion type leads to muddy, non-inteliglible sound. Very common fault across many coaxial desings, almost total design flaw.
- Flat surrounds without sudden geometry changes, especially positive (protruding into waveguide geometry). It excludes 90% of known suspension geometries - single roll, M-roll etc from being good coaxial designs.
- Careful physical centering tweeter dome inside of acoustical center of midrange cone. Shifting tweeter ahead of midrange driver destroys wavefront coherence, trying to correct time-domain response axially leads to severe phase coherence deteriorations off-axis. This is in case of car audio 'coaxials' but also applies for whizzer cones, additional horns isnide mid cone etc.
- Weak tweeter power and stroke loading capabilities, It excludes every driver which is not possible to operate its tweeter below 2kHz where midrange cone still is not already beaming severely. Sensitivity of domestic coaxials also is quite disappointing.
- Trying to evaluate coaxial driver's performance by looking at its on-axis diffraction dips and on-axis hi-frequency roll-off is a general mistake. Most of coaxial designs are plagued with diffraction issues which creates known dips on the frequencies 3-10kHz related to surround's bumps. These dips occures only on axis, but overall power response thrown into room usually is smooth, smoother than in discrete multi-driver designs. What is more, waveguiding of upper midrange-low trebles critical area lessens power output at these frequencies so hi-frequency roll-off becomes compensated and final tonal balance is close to optimal. Good practice is to listen to them a little bit off-axis.
I don't know it what I said is clear enough or complete, it is late here. I'm sorry for poor language.
Last edited:
Hello Vincneus: First, may I compliment you on your excellent written English skill, probably better than most Americans. 22% of young adults here are too stupid to even join our military 😀
You have a good background on the pro and con of coaxials. Active crossover or DSP can help some of the crossover issues. For smaller speakers the limitation is always going to be in the bass. You can get insane levels out of (for example) a BMS coax that has a HF compression driver, but getting a 8" or smaller woofer to similar dB outputs remains difficult 🙂
F
You have a good background on the pro and con of coaxials. Active crossover or DSP can help some of the crossover issues. For smaller speakers the limitation is always going to be in the bass. You can get insane levels out of (for example) a BMS coax that has a HF compression driver, but getting a 8" or smaller woofer to similar dB outputs remains difficult 🙂
F
Thank you, Soldermizer for the compliments! Also many thanks to everybody who has shared their thoughts on the matter. You guys did a great job at explaining why a some designs in particular wouldn't be desirable. I noticed that the tweeter response graphs of the Seas Excel C18EN002/A - E0060-08/06S (the super expensive one) looks a lot less diffracted off-axis than the ones of for example the Dayton CX150, which is on the other end of the spectrum with regards to the price point. If this lack of off-axis diffraction is all due to an inverted surround I understand that a 'flat' surround would be preferable.
If I do decide to use a coaxial woofers with a regular surround, would it be a complete waste of time and money or do you simply get a bit more diffraction and that's it? I have heard some pretty good things about the 5’’ SB Acoustics coaxial and considering the price, it’s very tempting. The graphs looks very acceptable to me at least. Obviously, the super expensive Seas coaxials have tons of other pro’s such as lower distortion rating etc. but they are kind of out of my budget as for now.
I like the idea of the higher sensitivity PA coaxials, partially because a higher sensitivity means less cone movement at the same volume level but I don’t have access to active filtering (yet) so I assume that choosing a PA coaxial with a CD on the magnet wouldn’t be a good choice, right?
I don’t mind building larger (floorstanding) speakers that play down to the 30’s. I’ll just put them on the side of my desk and, as windforce justly recommended, listen to them at a slight angle.
I’m curious what you guys think of the SB acoustics. I have gotten the chance to examine the woofers pretty closely and they look very well put together to me. Little to no gap between tweeter and voicecoil former (in contrary to the Dayton CX woofers!) Some nice foamy stuff around the tweeter and a smoothe transition from cone to surround. I’m not a huge fan of the looks, partially because it forces you to pair it with a woofer from the exact same series in order to keep aesthetics okay. It looks pretty hideous when paired with any other speaker….
Edit: You could rearmount the subwoofer to keep it out of sight. This opens some doors when it comes to woofer choice!
If I do decide to use a coaxial woofers with a regular surround, would it be a complete waste of time and money or do you simply get a bit more diffraction and that's it? I have heard some pretty good things about the 5’’ SB Acoustics coaxial and considering the price, it’s very tempting. The graphs looks very acceptable to me at least. Obviously, the super expensive Seas coaxials have tons of other pro’s such as lower distortion rating etc. but they are kind of out of my budget as for now.
I like the idea of the higher sensitivity PA coaxials, partially because a higher sensitivity means less cone movement at the same volume level but I don’t have access to active filtering (yet) so I assume that choosing a PA coaxial with a CD on the magnet wouldn’t be a good choice, right?
I don’t mind building larger (floorstanding) speakers that play down to the 30’s. I’ll just put them on the side of my desk and, as windforce justly recommended, listen to them at a slight angle.
I’m curious what you guys think of the SB acoustics. I have gotten the chance to examine the woofers pretty closely and they look very well put together to me. Little to no gap between tweeter and voicecoil former (in contrary to the Dayton CX woofers!) Some nice foamy stuff around the tweeter and a smoothe transition from cone to surround. I’m not a huge fan of the looks, partially because it forces you to pair it with a woofer from the exact same series in order to keep aesthetics okay. It looks pretty hideous when paired with any other speaker….
Edit: You could rearmount the subwoofer to keep it out of sight. This opens some doors when it comes to woofer choice!
If this is the only downside, "fixing" anything else is just a distraction.
The IMD of your 3" fullrange drivers doesn't bother you. A larger coaxial will have less IMD.
I'm biased: I have some old P.Audio coaxials, and I like them. Mine are Altec 604 knockoffs. In some respects (measured off axis response), they seem to be better than the originals.
The P.Audio graphs are trustworthy in that they aren't simply made up. The graphs for the sn6-150cx indicate a 2kHz crossover would work well, and that's probably right. I think the crossover region is the most important part of the FR plot. The fact that there is a lot of smooth overlap between the HF and LF drivers is a very good thing.
The P.Audio graphs are, however, over-smoothed. This might hide some problems. The tweeter will probably have bigger peaks ~15kHz than what the graph shows, and the woofer may be rougher 4-5kHz.
I like the idea too, but I think this point is incorrect. Higher sensitivity just means it achieves that movement with less power. That is, a 93dB 6" speaker and a 83dB 6" speaker will need the same cone movement to hit the same SPL. The less sensitive speaker will just waste more power and the voice coil will run slightly warmer while doing do.
Consider: the air in your room doesn't "know" what the driver parameters are, or how hard the amp is working. The air only "knows" how much cone movement is happening.
Active crossovers + filtering are great, but for simplicity and a student budget, consider using a simple passive crossover with active EQ at the source (e.g. use equalizer APO for level adjustment and notch filters).
A minor bonus is that you'll have fewer boxes & cables to manage on your desktop.
You mean because of delay? This might not matter. Minutiae of phase and timing don't matter to all people / all types of music.
Maybe you could find some files to ABX test yourself e.g.
Audibility of allpass phase distortion (test)
Blind testing a 1 ms Timing Difference (3-way)
Note that this is a very basic test, and 1ms delay is roughly a 34cm delay.
...or download some software to add a 0.25ms delay to the treble of a few of your favourite songs. See if it matters to you.
Good info here:
Hammer Dynamics Super 12 Loudspeaker Kit
"The bumps in the frequency response above 7 kHz are due to manner in which the tweeter is mounted on the front baffle. Treble energy diffracts around the narrow tweeter baffle and reflects off the woofer cone, which looks and acts like a parabolic reflector. The reflections give rise to an interference pattern, much like that of a comb filter. The impact on tonal color is slight but noticeable."
Thanks for your insights Hollowboy!
The narrow dispersion was indeed the only problem I experienced, probably because I am not an audiophile and merely an audio enthusiast, but I was afraid that if I were to solve the dispersion problem by simply jamming a tweeter with some crossover components into the box I would create more problems than I was solving lobbing, worse imaging, time alignment issues and so forth.
I never really thought about the 3’’ FR’s producing IMD too. They were obviously crossed to a subwoofer at I believe 150Hz. I thought that because the provided ‘waveguide’ (the cone) doesn’t move separately from the ‘tweeter’ (center of the cone) there wouldn’t be as much IMD as when the waveguide moves around the tweeter, as is the case with a coaxial woofer. I could be wrong though!
I’m glad there are other people with P-Audio stuff around. I really like the looks of your ‘knockoffs’ too! You’re probably right about the smoothing, I’ll keep that in mind when choosing a driver in the future.
If I simulate two 8’’ woofers in the same size sealed box in WinISD, having them both play to 100dB’s takes 5W and 7.5W respectively because the latter has a slightly lower sensitivity. The cone excursion graph shows me that the more efficient moves only 1.2mm whereas the other moves 2.4mm at the same frequency. This leads me to believe that a more efficient woofer of the same size, for some reason needs less cone excursion to produce the same SPL levels.
About the whole horn-in-woofer thing (and the tweeter mounted in front of a woofer): Do you think it is true that the cone movement doesn’t influence the tweeter’s response as much, because it mostly has its own waveguide? That would allow me to cross the coaxial a little lower (100Hz) and make life a bit easier.
I wonder what the measurement of the Hammer dynamics super 12 would look like if it was done on pink noise. The cone would be moving air to a 40Hz sine wave simultaneously with the tweeter producing 8kHz, which is more likely the case when listening to music. Who listens to sweeps? I wonder whether the mentioned peaks and dips above 7kHz would be more of a visible problem. What do you think?
I’ll definitely give those links you sent me a try to see how much it bothers me personally. People often talk about time alignment as crucial factor in a design, but in a two-way system it will always change with respect to your own position, no matter how good you align it on axis. I feel like this not as drastic of a change when the tweeter is mounted in the center of the woofer, even when it is mounted a little behind the woofers acoustic center (on the back of the magnet, for example).
But all in all, I don’t expect my untrained ears to hear a difference… As long as audible distortion is low and dispersion is good enough for the tonal balance to be decent throughout the room, I’m happy 😉
The narrow dispersion was indeed the only problem I experienced, probably because I am not an audiophile and merely an audio enthusiast, but I was afraid that if I were to solve the dispersion problem by simply jamming a tweeter with some crossover components into the box I would create more problems than I was solving lobbing, worse imaging, time alignment issues and so forth.
I never really thought about the 3’’ FR’s producing IMD too. They were obviously crossed to a subwoofer at I believe 150Hz. I thought that because the provided ‘waveguide’ (the cone) doesn’t move separately from the ‘tweeter’ (center of the cone) there wouldn’t be as much IMD as when the waveguide moves around the tweeter, as is the case with a coaxial woofer. I could be wrong though!
I’m glad there are other people with P-Audio stuff around. I really like the looks of your ‘knockoffs’ too! You’re probably right about the smoothing, I’ll keep that in mind when choosing a driver in the future.
If I simulate two 8’’ woofers in the same size sealed box in WinISD, having them both play to 100dB’s takes 5W and 7.5W respectively because the latter has a slightly lower sensitivity. The cone excursion graph shows me that the more efficient moves only 1.2mm whereas the other moves 2.4mm at the same frequency. This leads me to believe that a more efficient woofer of the same size, for some reason needs less cone excursion to produce the same SPL levels.
About the whole horn-in-woofer thing (and the tweeter mounted in front of a woofer): Do you think it is true that the cone movement doesn’t influence the tweeter’s response as much, because it mostly has its own waveguide? That would allow me to cross the coaxial a little lower (100Hz) and make life a bit easier.
I wonder what the measurement of the Hammer dynamics super 12 would look like if it was done on pink noise. The cone would be moving air to a 40Hz sine wave simultaneously with the tweeter producing 8kHz, which is more likely the case when listening to music. Who listens to sweeps? I wonder whether the mentioned peaks and dips above 7kHz would be more of a visible problem. What do you think?
I’ll definitely give those links you sent me a try to see how much it bothers me personally. People often talk about time alignment as crucial factor in a design, but in a two-way system it will always change with respect to your own position, no matter how good you align it on axis. I feel like this not as drastic of a change when the tweeter is mounted in the center of the woofer, even when it is mounted a little behind the woofers acoustic center (on the back of the magnet, for example).
But all in all, I don’t expect my untrained ears to hear a difference… As long as audible distortion is low and dispersion is good enough for the tonal balance to be decent throughout the room, I’m happy 😉
- Status
- Not open for further replies.