Well the phase shuffler should only work with good impulses 
.
It only works on the frequencies above 1000 Hz, shifting the left and right time slightly to avoid listening to the same cross talk pattern at both ears. In the process it widens the phantom image a bit my a few cm.
I'm not seeing that many differences considering the difference in layout, the impulse I showed is left and right channel firing together measured in the exact sweet spot. The APL_TDA plot is also from both speakers firing simultaneously.
The only thing I do see is cleaner result after the impulse which comes as no surprise.
The lack of reflections in your setup is obvious. I didn't have the freedom to place my speakers anywhere I'd like. My metallic ceiling isn't helping me either.
The main thing is a clear and defined peak and no clear reflections after that peak for the first ~15 to 20 ms.
The IR's can't really be compared if we don't know what they show, meaning it's a choice within the measurement program itself what it represents. Usually its mainly showing high frequency information, no surprise your focussed array does better there, right?
The ETC shows an amazing drop to more than -30 dB after the main peak. That's studio level... Your setup behaves as a true near field solution.
Care to run the demo version of APL_TDA? That will plot the room.
I bet it will be all deep blue after the main ridge. I do see a group delay problem at 120 Hz, APL will show that too.
That's where the array hands over to the Bass? Or is it the array itself, mine had a similar
curve before I changed my processing.
Nice plots, do you have a falling FR (not flat but higher in bass than high frequencies) like I have? The IR seems to indicate that. A flat FR wouldn't have a negative spike like that after the main pulse. It could also be caused by that 120 Hz group delay.
If I were in your shoes I'd try and fix that 120 Hz delay, I promise it'll be worth it in percussion.
I'd be very curious to see you do an APL_TDA plot.
.It only works on the frequencies above 1000 Hz, shifting the left and right time slightly to avoid listening to the same cross talk pattern at both ears. In the process it widens the phantom image a bit my a few cm.
I'm not seeing that many differences considering the difference in layout, the impulse I showed is left and right channel firing together measured in the exact sweet spot. The APL_TDA plot is also from both speakers firing simultaneously.
The only thing I do see is cleaner result after the impulse which comes as no surprise.
The lack of reflections in your setup is obvious. I didn't have the freedom to place my speakers anywhere I'd like
. My metallic ceiling isn't helping me either.The main thing is a clear and defined peak and no clear reflections after that peak for the first ~15 to 20 ms.
The IR's can't really be compared if we don't know what they show, meaning it's a choice within the measurement program itself what it represents. Usually its mainly showing high frequency information, no surprise your focussed array does better there, right?
The ETC shows an amazing drop to more than -30 dB after the main peak. That's studio level... Your setup behaves as a true near field solution.
Care to run the demo version of APL_TDA? That will plot the room.
I bet it will be all deep blue after the main ridge. I do see a group delay problem at 120 Hz, APL will show that too.
That's where the array hands over to the Bass? Or is it the array itself, mine had a similar
curve before I changed my processing.
Nice plots, do you have a falling FR (not flat but higher in bass than high frequencies) like I have? The IR seems to indicate that. A flat FR wouldn't have a negative spike like that after the main pulse. It could also be caused by that 120 Hz group delay.
If I were in your shoes I'd try and fix that 120 Hz delay, I promise it'll be worth it in percussion.
I'd be very curious to see you do an APL_TDA plot.
Artsy,
I tried myself to link to the page and as you found the link doesn't work correctly. I purchased this material more than 20 years ago and things in their catalog do change over time. I really don't remember there being a Kraft paper layer on either side of the glass, I know for a fact it was not covered with a foil layer which I would suspect highly would be very reflective at specific high frequencies.
Anytime you change from one material to another such as glass to air to glass you will have a wider bandwidth of absorption but how significant would depend on material changes and volumes. The real difference between one 2" layer and double that thickness would that the thicker section would just have a slightly higher absorption over what I would suspect would be a narrow band at the lower frequency limit of the materials properties, I wouldn't expect an across the band increase in absorption,
I would never consider that any of these materials would be great at the lower frequency range, the bass range as the wavelengths are just to long for any type of material to really have a great affect. Most of the commercial bass traps are really not bass traps but truly I would call them mid bass traps unless you would want to look at the properties of stuffing a long tube and the length of that tube, it is not the transverse distance that would absorb bass frequencies, they would basically pass with little effect.
I agree with osterhammer that a combination of absorption and diffusion is the best solution to use. Add some uneven reflective surfaces with panels and I don't think you can do much better to help a room. A bookcase with books of varying depth could be used as that reflective surface or anything else that has a varying depth to scatter the sound.
I tried myself to link to the page and as you found the link doesn't work correctly. I purchased this material more than 20 years ago and things in their catalog do change over time. I really don't remember there being a Kraft paper layer on either side of the glass, I know for a fact it was not covered with a foil layer which I would suspect highly would be very reflective at specific high frequencies.
Anytime you change from one material to another such as glass to air to glass you will have a wider bandwidth of absorption but how significant would depend on material changes and volumes. The real difference between one 2" layer and double that thickness would that the thicker section would just have a slightly higher absorption over what I would suspect would be a narrow band at the lower frequency limit of the materials properties, I wouldn't expect an across the band increase in absorption,
I would never consider that any of these materials would be great at the lower frequency range, the bass range as the wavelengths are just to long for any type of material to really have a great affect. Most of the commercial bass traps are really not bass traps but truly I would call them mid bass traps unless you would want to look at the properties of stuffing a long tube and the length of that tube, it is not the transverse distance that would absorb bass frequencies, they would basically pass with little effect.
I agree with osterhammer that a combination of absorption and diffusion is the best solution to use. Add some uneven reflective surfaces with panels and I don't think you can do much better to help a room. A bookcase with books of varying depth could be used as that reflective surface or anything else that has a varying depth to scatter the sound.
For me it would largely depend on the goals you want to reach in the room you have available and what other variables you have to work with. (like a spouse or a family , but also things like making the choice to have a single or multi listener environment)
I only added absorption, on first reflection points. My room is still very much "live".
If you look at Martin's room, where he had more freedom to place the speakers away from reflective surfaces and uses a focussed array, the advantages in the sweet spot are apparent. Even though the room is down by more than 30 dB (only in the frequency range the IR or ETC shows, filtered IR's would tell us what it does at other frequencies) there is still a lot of diffuse sound present, though at a very low SPL level. This is a good thing. It's similar to what is done in a lot of studio's that replaced the LEDE concept. It also reminds me of Pano's Cave.
I tried to get my room's ETC graph to sort of resemble a LEDE room. Not entirely though, but have that first 20ms free of reflections as much as possible. That's no easy thing to do in a relative small living room that's not supposed to start looking like a studio. The back wall was way to close to use diffusion panels, so I only absorb the strong reflections, there's enough scattered diffusive energy left in the room. (too much actually)
To get a diffusive tail (one that's not present in small rooms) I use my ambient speakers with software reverb. It has advantages over real rooms as it won't sound the same on every song as a real room would do. As I said before, I use that to hide my room. Making it feel larger than it is.
I wouldn't want to use diffusion panels or otherwise in my room for one simple reason: it's just too small! But only removing the strong reflections leaves enough diffusive energy in the room. The ambient channels have the added benefit to help to get the RT30 plots in REW look more evenly spread across all frequencies which helps stabilize tonality.
Just look at the room jim1961 build where he uses diffusion to re-route the energy, letting it arrive later in time at the listening spot. At the present he is experimenting with ambient speakers and a Lexicon device to simulate the parts you just can't have in a small room.
To get clean/clear imaging it helps to have that clean first wave front.
For a sense of energy it helps to have a diffusive tail arriving from a lateral angle at the listening spot later in time. It also helps battle the effects of cross talk happening at the ear from a stereo setup. Meaning a more 3D sound stage without having to give up the clean/clear imaging.
Plan the room, know what you want from it. That would tell you what to use.
Even though I want the best stereo sound in my sweet spot, I want it to sound very good next to that sweet spot to accommodate my family.
That means my set of compromises is very different from Martin's.
For me, this works, even if I have to give up the sweet spot to my Son during movies, who has experienced the added benefit of that spot. Boy am I glad it sounds more than decent right next to him!
, but also things like making the choice to have a single or multi listener environment)I only added absorption, on first reflection points. My room is still very much "live".
If you look at Martin's room, where he had more freedom to place the speakers away from reflective surfaces and uses a focussed array, the advantages in the sweet spot are apparent. Even though the room is down by more than 30 dB (only in the frequency range the IR or ETC shows, filtered IR's would tell us what it does at other frequencies) there is still a lot of diffuse sound present, though at a very low SPL level. This is a good thing. It's similar to what is done in a lot of studio's that replaced the LEDE concept. It also reminds me of Pano's Cave.
I tried to get my room's ETC graph to sort of resemble a LEDE room. Not entirely though, but have that first 20ms free of reflections as much as possible. That's no easy thing to do in a relative small living room that's not supposed to start looking like a studio. The back wall was way to close to use diffusion panels, so I only absorb the strong reflections, there's enough scattered diffusive energy left in the room. (too much actually)
To get a diffusive tail (one that's not present in small rooms) I use my ambient speakers with software reverb. It has advantages over real rooms as it won't sound the same on every song as a real room would do. As I said before, I use that to hide my room. Making it feel larger than it is.
I wouldn't want to use diffusion panels or otherwise in my room for one simple reason: it's just too small! But only removing the strong reflections leaves enough diffusive energy in the room. The ambient channels have the added benefit to help to get the RT30 plots in REW look more evenly spread across all frequencies which helps stabilize tonality.
Just look at the room jim1961 build where he uses diffusion to re-route the energy, letting it arrive later in time at the listening spot. At the present he is experimenting with ambient speakers and a Lexicon device to simulate the parts you just can't have in a small room.
To get clean/clear imaging it helps to have that clean first wave front.
For a sense of energy it helps to have a diffusive tail arriving from a lateral angle at the listening spot later in time. It also helps battle the effects of cross talk happening at the ear from a stereo setup. Meaning a more 3D sound stage without having to give up the clean/clear imaging.
Plan the room, know what you want from it. That would tell you what to use.
Even though I want the best stereo sound in my sweet spot, I want it to sound very good next to that sweet spot to accommodate my family.
That means my set of compromises is very different from Martin's.
For me, this works, even if I have to give up the sweet spot to my Son during movies, who has experienced the added benefit of that spot. Boy am I glad it sounds more than decent right next to him!
Last edited:
Thanks Everyone for your input...
Martin, that room you linked is awesome! Right up my alley with some of my wood art! I do not know if my room is big enough to incorporate diffusion on the side walls, but that would be cool in the front.
So just like with R-value, there is a point of diminishing returns with thicker panels. I had a feeling bass would be hard to absorb, even in large rooms, it seems uneven to me.
At the moment, I do not share my home with a wife, but I do want a large listening area for my daughter and guests to enjoy. This is the main reason I am seeking a better speaker solution then my Avebury, because they are a one seat sweet spot listen. Yet, there are many unanswered questions and explorations to under go, before I start the next speaker build. My recent upgrades on DAC and power amp has changed the way I listen. I can dig a lot deeper now.
From the Impulses You all have posted, I am seeing this is indeed a very worthy pursuit.
I would like to find some other rooms and see how they used absorption and diffusion.
I included a drawing of my room. It is 10 by 14 feet. Small, but as You can see, the bay window and archway into the next room helps a lot. My initial thoughts would be to either:
Do the front half of the room all diffusion and the back half all absorption.
Or do the front wall diffusion, the front side walls absorption; and the back wall absorption and the back side walls diffusion.
Bear in mind, the front wall will have a flat TV in the front, and eventually I would like a 70" when the money is available. Need to get my daughter thru college first.
Yet, this room IS my sound room, so the sky is the limit - well, maybe size and $$$, but you get the idea. But really, what can be done with a room this size? A lot of people have smaller rooms, I think we all could learn a lot.
Of course, I am thinking I will be needing to move this exploration to my thread(s).
Martin, that room you linked is awesome! Right up my alley with some of my wood art! I do not know if my room is big enough to incorporate diffusion on the side walls, but that would be cool in the front.
So just like with R-value, there is a point of diminishing returns with thicker panels. I had a feeling bass would be hard to absorb, even in large rooms, it seems uneven to me.
At the moment, I do not share my home with a wife, but I do want a large listening area for my daughter and guests to enjoy. This is the main reason I am seeking a better speaker solution then my Avebury, because they are a one seat sweet spot listen. Yet, there are many unanswered questions and explorations to under go, before I start the next speaker build. My recent upgrades on DAC and power amp has changed the way I listen. I can dig a lot deeper now.
From the Impulses You all have posted, I am seeing this is indeed a very worthy pursuit.
I would like to find some other rooms and see how they used absorption and diffusion.
I included a drawing of my room. It is 10 by 14 feet. Small, but as You can see, the bay window and archway into the next room helps a lot. My initial thoughts would be to either:
Do the front half of the room all diffusion and the back half all absorption.
Or do the front wall diffusion, the front side walls absorption; and the back wall absorption and the back side walls diffusion.
Bear in mind, the front wall will have a flat TV in the front
, and eventually I would like a 70" when the money is available. Need to get my daughter thru college first. Yet, this room IS my sound room, so the sky is the limit - well, maybe size and $$$, but you get the idea.
But really, what can be done with a room this size? A lot of people have smaller rooms, I think we all could learn a lot.Of course, I am thinking I will be needing to move this exploration to my thread(s).
Attachments
Been trying to figure what that is...
Okay, Ronald, this is the part I am learning with the measurements. When I use Holm, I get an Impulse like the one you posted last night, and I do understand what I am looking at and what improvements are. REW gives the ETC like above. But I really do not understand what I am looking at Are we trying to improve this graph by leveling off the vertical lines that peak above the curve (like at 49.5 ms), or are we trying to lower the overall curve faster over time? Maybe both? Any links to reading materials would help.
Okay, Ronald, this is the part I am learning with the measurements. When I use Holm, I get an Impulse like the one you posted last night, and I do understand what I am looking at and what improvements are. REW gives the ETC like above. But I really do not understand what I am looking at
Are we trying to improve this graph by leveling off the vertical lines that peak above the curve (like at 49.5 ms), or are we trying to lower the overall curve faster over time? Maybe both? Any links to reading materials would help.
Seeing that room I'd know what I would do . We could continue this on your thread though.
You don't have enough room to place the speakers away from walls. So use them to strengthen the bottom end, but at the same time avoid as many reflections as possible.
I wouldn't use diffusion on the front wall or back wall in a room of this size. Its actually quite similar to my room.
. We could continue this on your thread though.You don't have enough room to place the speakers away from walls. So use them to strengthen the bottom end, but at the same time avoid as many reflections as possible.
I wouldn't use diffusion on the front wall or back wall in a room of this size. Its actually quite similar to my room.
Okay, Ronald, this is the part I am learning with the measurements. When I use Holm, I get an Impulse like the one you posted last night, and I do understand what I am looking at and what improvements are. REW gives the ETC like above. But I really do not understand what I am looking at
Both an IR and ETC are a different way of looking at the same measurement. As are waterfall graphs and FR curves.
What I look for in the ETC is a steep drop after the initial peak and a gradual decline afterwards. The first 20 ms being more important to me.
The IR shows that same peak in a different way. The reflections you see are bumps in that flat tail, which should correspond to peaks in the ETC graph. On the ETC graph you get to see the SPL level in reference to the main peak.
What you'd want is an entirely different matter though. I studied a lot of threads over on Gearslutz about studio rooms to gain some knowledge. But even over there not all people agree with each other on what's right. It did help me get a grip and understanding of what we are looking at.
Threads like: https://www.gearslutz.com/board/studio-building-acoustics/511073-civil-etc-discussion.html are a start to get some reading done. I've looked up all I could find on LEDE concepts and Haas Kickers...
Be aware that an ETC does not show all frequencies. You can actually apply band pass filters in REW to look at a "slice" of the frequency measurement.
That's what's fun about the APL_TDA measurement. It shows you a lot about the room in a nice 3D picture. Showing all reflections, but it's still up to you to determine their cause.
A thought about time alignment...
With a more time aligned speaker a "Boom" sounds like: Boom
With less strickt time alignment that would change to Booomm.
The "B" being the high frequency, the "oo" are the mids, the "m" being the low end.
It smears in time. It's much more complicated of course (pfew) as musical instruments have harmonies spread over a wide range of frequencies. But this is what I get out of time alignment. More force in mid notes as the lower harmonic parts don't lag. The room will still add it's own tail to whatever "Boom" we play though. But the "Boom" remains a "Boom".
Not having the time alignment right doesn't sound wrong to us, but having it right gives more power to a lot of instruments. Like they are happening right there.
Not having the time alignment is like slowing down the lower impact. Can you hear it?
I could tell you, but it's better to try yourself.
With a more time aligned speaker a "Boom" sounds like: Boom
With less strickt time alignment that would change to Booomm.
The "B" being the high frequency, the "oo" are the mids, the "m" being the low end.
It smears in time. It's much more complicated of course (pfew) as musical instruments have harmonies spread over a wide range of frequencies. But this is what I get out of time alignment. More force in mid notes as the lower harmonic parts don't lag. The room will still add it's own tail to whatever "Boom" we play though. But the "Boom" remains a "Boom".
Not having the time alignment right doesn't sound wrong to us, but having it right gives more power to a lot of instruments. Like they are happening right there.
Not having the time alignment is like slowing down the lower impact. Can you hear it?
I could tell you, but it's better to try yourself.
Thanks, I had a feeling it was something like that... I just have to learn how to use it now.
I can hear and feel it - at the same time. Well, almost. There are a couple of ms between "OO" and "M". On my non-time aligned rig (Arrays and Subs) I hear the "BOO" with a delayed felt "M" More like "BOOuMMM" The "u" might be an artifact from my system. I can definitely hear the two distinct characters of the different drivers (NSB verses Daytons). That extra "u" gets masked out when time aligned. You are right, the "M" lags on a bit, if it is not time aligned. I have found that the bass delay gives me the impression there is too much low, even if the measurements suggest otherwise. Might be better for the neighbors thou.
The Avebury is interesting in this matter, as the bass is derived from the rear horns bouncing off my room. It is in alignment, but it lingers a bit also. It would be more "BOOMM" I still get the impression of impact, but it is not as instant as having everything time aligned in the frontal plane. I would take this compromise over the non-aligned subs. Right now, it is the situation I am in only having one DAC. I can not align the subs to the arrays, and I am finding I really do not care for the subs as much in this situation. But I have gained far more in other areas with these source improvements, that I really do not miss the "M" too much with just the arrays. Of course I have Avebury for now...
We've discussed this in the past, Wesayso, and I have a slightly different opinion. If you have a perfectly flat frequency response and not gross timing errors, a BOOM will sound exactly like a BOOM. I'm sorry, but our ears are just not that discerning. What is a gross timing error? I don't know. But a standard speaker will regular crossovers and a perfectly flat frequency response will sound extremely good, and I would be willing to bet money that it would be indistinguishable from a speaker that had the same frequency response but the phase corrected.
Our hearing mechanism has evolved for better survival. Loudness and direction is all we care about. If the relative (small) timing difference between two frequencies does not alter loudness or direction perception, there is no need to determine that there is a difference.
I'm going to try and design a text for this. Hope to put it up soon.
Definitely not my experience. Above 1000 Hz it will be very hard to hear any timing errors, I'll admit to that. They will be too small to notice. But the lower we go in frequency the more you will be able to truly hear these differences.
You should try it sometime. Its not that easy to get it right though. Once you do have time coherency down to, say, 30 Hz, play an action movie with some bullets flying and explosions.
It will scare the pants off of you just listening to it. It's way more immediate and certainly addresses parts of our survival skills. it will get your heart beating.
I also like the movie "Rush" for that test. Once those F1 engines fire up real close to you it will almost make you jump up and hide for cover. Play it at a realistic SPL level (lol).
You sense it with your body as well as your ears, that will definitely tell you the source is close. Do you think that part didn't matter in our survival?
Just think about how you hear thunder. Does it not scare you more if it gets closer? You feel, hear and see it instantaneously. That does bring up the adrenaline.
Now how would that thunder sound on a non time coherent speaker compared to a time coherent one. Believe me, done right you will hear the difference. The lower in frequency you go, the more it will get the juices flowing.
Don't make the mistake of over correcting with a simple phase alignment done quickly. That will probably sound different. You've got to make sure the timing at the listening position is right, in frequency and phase.
You want more excitement and dynamics? Just give it a shot. You don't want to know how many times these arrays made me jump up. Especially during good movie tracks. It's way more than your ears alone that listens...
Last edited:
I could hear the difference between these two graphs:
and
The FR being close enough to say they were the same:
Both is with left and right speaker playing, 1/6 oct FDW. (not actual measurements but the prediction is shown here)
I've listened to both versions extensively. I created the second one (following minimum phase) because the sound was "off".
That second one is what I listen to even today. Never found a reason to mess with it again.
I expected the first one to be the one I would like best. It wasn't. The second graph, closer to minimum phase is what sounded natural. The other one just sounded "off" in comparison.
The GD difference is really small, the difference in sound wasn't. Or should I say the difference in feel. That's bordering on the other side, getting the bass to fire early. Much easier to pick up than having it arrive late. One will get your foot tapping, the other will make you go, "what was that?"
and
The FR being close enough to say they were the same:
Both is with left and right speaker playing, 1/6 oct FDW. (not actual measurements but the prediction is shown here)
I've listened to both versions extensively. I created the second one (following minimum phase) because the sound was "off".
That second one is what I listen to even today. Never found a reason to mess with it again.
I expected the first one to be the one I would like best. It wasn't. The second graph, closer to minimum phase is what sounded natural. The other one just sounded "off" in comparison.
The GD difference is really small, the difference in sound wasn't. Or should I say the difference in feel. That's bordering on the other side, getting the bass to fire early. Much easier to pick up than having it arrive late. One will get your foot tapping, the other will make you go, "what was that?"
Last edited:
Hi Wesayso
I have just discovered your thread, which has obviously been in progress for some time. Your build quality and I'm sure sound quality is jaw dropping. I too am convinced of the virtues of line array design. I have built 84 inch tall cabinets with 16 drivers each, 5 & 1/4 inch diameter. I choose 16 drivers to present an exact 8 ohm load to the power amplifier. The diameter was selected to extend the low frequency limit.
I am surprised that your drivers are providing full range coverage. My design philosophy was to provide HI FI performance at Rock concert levels. I have built 2 (sub) woofer cabinets that are driven by a modified 1000 wpc Crest CA-18 amplifier.
I wrestled with the positioning of the Plannar Tweeter Array. See Attached Photo. I believe that there could be some interference issue between the MF and HF arrays, But I am very pleased with the sound. I am toying with the idea of a (32 db/octave slope) active cross-over to replace the 24 db/octave unit I am using. Your thoughts? My project appears somewhat crude, in comparison to your line arrays, but I was very interested in getting started quickly!
I have just discovered your thread, which has obviously been in progress for some time. Your build quality and I'm sure sound quality is jaw dropping. I too am convinced of the virtues of line array design. I have built 84 inch tall cabinets with 16 drivers each, 5 & 1/4 inch diameter. I choose 16 drivers to present an exact 8 ohm load to the power amplifier. The diameter was selected to extend the low frequency limit.
I am surprised that your drivers are providing full range coverage. My design philosophy was to provide HI FI performance at Rock concert levels. I have built 2 (sub) woofer cabinets that are driven by a modified 1000 wpc Crest CA-18 amplifier.
I wrestled with the positioning of the Plannar Tweeter Array. See Attached Photo. I believe that there could be some interference issue between the MF and HF arrays, But I am very pleased with the sound. I am toying with the idea of a (32 db/octave slope) active cross-over to replace the 24 db/octave unit I am using. Your thoughts? My project appears somewhat crude, in comparison to your line arrays, but I was very interested in getting started quickly!
Attachments
Nice project! I couldn't tell you if moving to higher slopes would help you without seeing a lot more data. Where do you cross over to the planar tweeters? How far are they apart?
One of my goals being time coherency made me choose a setup without crossovers. Maybe one day I'll add subs to reinforce the low end. I could use it for movies, right now I have to dial back the volume on movies to keep it sane and protect my drivers. In the mid frequencies they can easily do their job up to insane levels of SPL.
I'm guessing it would be hard to integrate subs and keep the current timing and more important: even bass the arrays project in the room.
In music there isn't much they cannot handle.
One of the most challenging tracks I found was A Perfect Circle - Lullaby. You feel the pressure in that song while your low end gets a work-out. Makes me want a bigger/stronger amplifier as that's my current bottleneck.
Keep in mind I have 17 Hz set to -3dB, but in reality I cannot really play that low at higher SPL levels. Movies do really use that low end but most music doesn't.
That could be right. I have learned to listen, that's for sure. Most of the time I listen with a big grin on my face.
It's still the moments the music just takes over that I enjoy the most. Being breathless from listening to a marvellous piece of music, or moved to tears.
That was the goal and in that regard I made it to the finish line. I still have an urge to learn more though.
I expected time coherency to have an effect on imaging. I cannot say that is true. As long as FR and Phase is in agreement in the left and right channels you can have great imaging without it being time coherent. So why am I so strongly defending that topic? Because in my humble opinion it makes the music sound alive. It is fun and exiting! Scary at times (with SPL up).
theaudiopath,
It seems you are very concerned with absolute time alignment of your main arrays and the tweeter arrays. One detail I noticed when looking closer was that you have alternating distances for the 5 1/2" speakers and then you have the tweeter array on a plane with these speakers. I don't think there are any easy answers for exact time alignment in a physical sense here. You've created a nice looking array with a few physical constraints that you now have to deal with that is very different than the Wesayso arrays. While Ron has chosen no crossovers you have to have them and he has absolute physical alignment in his arrays. If absolute alignment is your goal might I suggest a simple dsp time delay for the two arrays and a modification to your main array electrical wiring. Since you have stepped the two sets of 5 1/2" speakers why not separate the two levels of drivers and have a slight time correction between the two layers and drive each layer from a different channel. Yes this will require an extra amplifier to do and a second time delay but that seems to be the only way I see that you can get absolute time alignment without some sort of smear. Physically moving the tweeter array to achieve time alignment would seem to introduce some other problems with a reflection off the main array if you did that so you are simply exchanging one problem, time alignment, for another, diffraction between arrays.
If I had the time to do it right now I would build Ron some new drivers with much more excursion so he could have his cake and eat it to on the low end.
It seems you are very concerned with absolute time alignment of your main arrays and the tweeter arrays. One detail I noticed when looking closer was that you have alternating distances for the 5 1/2" speakers and then you have the tweeter array on a plane with these speakers. I don't think there are any easy answers for exact time alignment in a physical sense here. You've created a nice looking array with a few physical constraints that you now have to deal with that is very different than the Wesayso arrays. While Ron has chosen no crossovers you have to have them and he has absolute physical alignment in his arrays. If absolute alignment is your goal might I suggest a simple dsp time delay for the two arrays and a modification to your main array electrical wiring. Since you have stepped the two sets of 5 1/2" speakers why not separate the two levels of drivers and have a slight time correction between the two layers and drive each layer from a different channel. Yes this will require an extra amplifier to do and a second time delay but that seems to be the only way I see that you can get absolute time alignment without some sort of smear. Physically moving the tweeter array to achieve time alignment would seem to introduce some other problems with a reflection off the main array if you did that so you are simply exchanging one problem, time alignment, for another, diffraction between arrays.
If I had the time to do it right now I would build Ron some new drivers with much more excursion so he could have his cake and eat it to on the low end.
Last edited:
Hi Kindhornman;
I believe your were referring to my line arrays. Although not the prettiest solution, I thought it was clever to overlap the drivers, for two reasons. First; to reduce the centre to centre distance of the drivers. My research told me that when the driver spacing exceeded half the wavelength ( C to C ), the drivers act as individual sources, with the corresponding interference patterns. I believe this to be far more of an issue than the 3/8 of an inch that the drivers had to be staggered, to reduce spacing. At 2 kHz, 3/8 inch is a small fraction of a wave length.
I have no formal audio training, but have spent considerable time researching, before the first panel was cut. I have seen line arrays that are convex, from top to bottom, presumably for coverage reasons in larger (outdoor) systems. I have also seen concave designs, I believe to address distance issues from the listener to the middle drivers, verses drivers at the top and bottom. I concluded that a straight column was the best (and simplest) compromise.
I also over-lapped my drivers because I found it impossible to get sufficient air displacement from any driver smaller than 5 and 1/4 inches, for a reasonable bottom end response. I hope to 'tweak' my design to come up with a speaker system that could be built at a reasonable cost, produce 115 db (plus, yes I am a nutter) at 12 feet (not 1 meter!) and avoid the horrors of compression drivers. My listening area has a very high ceiling, but the design had to fit in a standard ceiling height house (8 feet), if it was to have any mass appeal.
I believe that the beaming effect of many cone drivers would limit the line effect at higher frequencies, so I have the planars crossed over at 2 kHz, and again, spaced as close as possible (one half wavelength) at 2 k
I believe your were referring to my line arrays. Although not the prettiest solution, I thought it was clever to overlap the drivers, for two reasons. First; to reduce the centre to centre distance of the drivers. My research told me that when the driver spacing exceeded half the wavelength ( C to C ), the drivers act as individual sources, with the corresponding interference patterns. I believe this to be far more of an issue than the 3/8 of an inch that the drivers had to be staggered, to reduce spacing. At 2 kHz, 3/8 inch is a small fraction of a wave length.
I have no formal audio training, but have spent considerable time researching, before the first panel was cut. I have seen line arrays that are convex, from top to bottom, presumably for coverage reasons in larger (outdoor) systems. I have also seen concave designs, I believe to address distance issues from the listener to the middle drivers, verses drivers at the top and bottom. I concluded that a straight column was the best (and simplest) compromise.
I also over-lapped my drivers because I found it impossible to get sufficient air displacement from any driver smaller than 5 and 1/4 inches, for a reasonable bottom end response. I hope to 'tweak' my design to come up with a speaker system that could be built at a reasonable cost, produce 115 db (plus, yes I am a nutter) at 12 feet (not 1 meter!) and avoid the horrors of compression drivers. My listening area has a very high ceiling, but the design had to fit in a standard ceiling height house (8 feet), if it was to have any mass appeal.
I believe that the beaming effect of many cone drivers would limit the line effect at higher frequencies, so I have the planars crossed over at 2 kHz, and again, spaced as close as possible (one half wavelength) at 2 k
