I can see something like this working well:
Originally made as microphone windshield: Foam Conversion - Custom Audio Microphone Windshields & Microphone Covers
What do you guys think...
Originally made as microphone windshield: Foam Conversion - Custom Audio Microphone Windshields & Microphone Covers
What do you guys think...
Attachments
I'm not sure about any possible merits in using those foam balls.
They will most likely start to reflect effectively at frequencies way too high for what we need, since they're made to transmit wide band sound while blocking a draft.
Even a hard and rigid tube will practically be invisible for wavelengths larger than half the circumference of the tube (f < ~8,75 kHz @ 25 mm tube diameter). Tell me, if I messed up my calculations here...
I'd also be aware of reflections other than in the horizontal plane, since those might add additional destructive interference patterns, counteracting the line source effect in the already compromised high frequencies.
Tell me about your thoughts...
They will most likely start to reflect effectively at frequencies way too high for what we need, since they're made to transmit wide band sound while blocking a draft.
Even a hard and rigid tube will practically be invisible for wavelengths larger than half the circumference of the tube (f < ~8,75 kHz @ 25 mm tube diameter). Tell me, if I messed up my calculations here...
I'd also be aware of reflections other than in the horizontal plane, since those might add additional destructive interference patterns, counteracting the line source effect in the already compromised high frequencies.
Tell me about your thoughts...
We are told from a physics standpoint that all drivers start to beam at some point, the measurements below prove otherwise. I did on and off axis measurements of my Arrays and Avebury. line source (on left) -vs- true point source (on right). My degree placement is not exact, but close enough to validate this point. I am using a Dayton "Omni" measurement mic, noting some of the same issues discussed in the recent posts. The mic was pointing directly at the center of the drivers for all these measurements. I included the distortion readings just to reaffirm the possibilities of what a large quantity of cheap drivers can do working together.
Allen
I think you graphs proves excatly what line sources do compared to point sources.When linesources get directive they only do it in the horizontal plane wheras pointsources do it in horizontal as well as vertical plane, so they loose 6 dB/oct vs. array 3 dB/oct as far as i know and can see from your graphs.
By the way my linesources perform the same dissapearing act as wesayso´s
Koldby
I'm a bit double in my thoughts on this. One side of me wants to believe the tale of center to center spacing. But the other side of me looks at actual distance from the driver to one's ear.
Why would an arced array work, I'm not talking about Keele's CBT, but the other way around. Every driver at the same distance to one's ear. Very focused sound. But at a very small sweet spot. Now why does the CBT array work? It is bent the other way around. Granted it also has shading applied trying to "recreate" a spherical wave.
What I wanted from a reticulated foam ball in front of a driver is trying to slow down waves passing straight through. If we assume high frequency is being produced by the center cap. We would then create a wider wave front of high frequency forming a flatter total wave front. Sort of like letting a ribbon play that frequency.
Even a ribbon would have time arrival problems in distance to one's ear if it's long enough.
So shaping the wave front might be a key here. A reduction in comb filtering.
There's no way around that comb filtering is happening from reading my REW graphs. (there, I said it
). Anyway I could reduce that would help give the array an even better response.A ribbon has the same distance problems from the center to the outer limits, but it is seamless. Not a point source every (in my case) 85 mm. The foam ball could stretch the total square area of high frequency reproduction to a bigger area than that point source it originally is.
You're absolutely right, longer wave lengths would travel around it. But the shorter, thus higher, waves would slow down in the middle path trough the foam (longest distance trough the foam). yet the sides would have less obstruction, creating something resembling more of a flat wave front instead of its natural spherical one.
I've studied your STEP response and my own. There is something happening with that tube in front of the drivers but I'm not sure it wins from pointing the drivers more on axis yet.
One of the reasons I haven't tried any of these ideas yet is that I can't really convince myself it would work.
I still believe the driver to ear distance is the most important factor at play concerning comb filtering. In that case only attenuating high frequencies from the outer drivers would work. Even that could be done with foam. But it wouldn't be pretty. It would look like a very large version of this:
Once I get my thoughts together I'll be ready to run some experiments (lol).
Every one tells me it's center to center spacing. Yet I can't forget the distance from the center to one's ear for every separate driver.
The STEP from your tube solution still shows the wavy shape of the separate time arrivals at the microphone. I haven't seen a measurement of a long ribbon tweeter but I bet it is smoother.
(Disclaimer: this entire text was typed while I was intoxicated. No harm was intended while typing this message but no coherency can be claimed. High levels of Alcohol were at play)
Attachments
As to the mention of concave arrays, back in '91 built some 3way towers that used a concave short shaded line (that's a mouthful). There were two 5" full rangers above and three below a surface mounted jbl pro titanium tweeter. The decision was made to acoustically time align for an optimum listening distance of 3.5m. For a bit I struggled with just how much nearfield correction to apply when it occured to me that the small distance difference between it and a straight array was akin to depth of field in photography and made the line near sighted. Typical listening distance would be around 2.5m at the time, ever hoping for a larger room, but not neccessarily a big room where they would spread their wings and remain tightly focused. Superb imagining and depth. Soundstage was far wider than the spacing would suggest, which I partly attributed to the high offset low diffraction baffle design.
They were quite popular at our clubhouse pool parties too
They were quite popular at our clubhouse pool parties too
I think we have different objectives.
Me, trying to widen the hf dispersion, since I'm basically (critically) listening 60° off axis. You, rather the opposite?
The step's raggedness shows more than just different driver's arrivals, I'm sure you're aware of that.
There are diffractions from the various baffle structure features, as well as reflections from the tube and internal chassis and box structure.
You'll see an even worse step (and of course frequency) response taken closer to the array due to comb filter issues from the individual drivers and all of the above mentioned additional reflections being magnified by the closeness.
Also, I'm not sure the foam effectively slows the wave propagation through it, since it's mainly made up of air bubbles with very thin membranes in between. And the speed of sound is the same in those bubbles ...
Let us know how that goes... I wouldn't want to shade drivers with passive components, but padding down the high frequency output of the outer most drivers (gradually increasing padding to the outer drivers) with reticulated foam is an interesting way. It has been on my mind ever since I saw the reduction in high frequency from Dr. Geddes waveguide fill with reticulated foam to reduce the effect of hom's in the waveguide.
But as said, I'm afraid it would take quite a thick layer to have the right effect for huge line arrays? I have no idea how to make that look pretty.
Yes indeed, different objectives. I'm looking for ways to soften the effect of comb filtering. Also trying to determine if it is needed. I would love to see the step of a narrow long ribbon. Of coarse the step contains more information than I suggested but the initial rise in the first ~ 0.3 ms is showing the different arrival times or time-smear if you will.
I was surprised to see a rise in high frequency on your FR plots, way off axis with that tube in front of the drivers. Is that rise mainly the tube or is it accompanied by a boost.
Last edited:
Of course you're right, I do indeed boost 1,5dB@12,5kHz 2dB@16kHz 6dB@20kHz to get some sparkle (Behringer GEQ set to true response mode).
Without the tubes though, the measured response at the off-axis listening position drops about 12dB from 10 to 20kHz - can only recall from memory, since I didn't save any graphs.
Quest for the line source high
Interesting, the highs would be "shaded" and the mids and lows would not. You would have to EQ your highs for a certain listening distance like a point source, but it would not be as extreme as the 6 dB/doubling of distance, I can only speculate, maybe 4.5 dB/doubling of distance? I think you could get smoother highs, due to less comb filtering, but for me personally, I am not willing to compromise the large listening area the straight arrays provide...
I would like to see a STEP of a singular large 72" tall ribbon. What does "time arrival" problems sound like without comb filtering? They say our ear "locks on" to the closest point, but I can not validate this point.
Wesayso, I am glad You are at least thinking productively when You are intoxicated, Thanks for your honesty.
The Scaena is still an array of tweeters. I think the comb filtering is minimized with an array of ribbons (or are they planers) like that. I am not sure about the driver spacing of the two arrays is the best for a 5500 Hz crossover frequency. 13740 (speed of sound in inches) divided by 5500 (Hz) is about 2 1/2 inches or about 64 mm. Wouldn't there be comb filtering between the tweeter and main arrays? Sure You can "toe" each array in on your listening area, but again, maybe giving up the large listening area.
It seems there are many things we like about our full-range arrays, if we could only have that "ribbon" type of sound on the highs. There are a couple of other threads of people building "full-range" ribbon drivers, covering 200 -300 Hz. But as Marin says... (next post)...
Interesting, the highs would be "shaded" and the mids and lows would not. You would have to EQ your highs for a certain listening distance like a point source, but it would not be as extreme as the 6 dB/doubling of distance, I can only speculate, maybe 4.5 dB/doubling of distance? I think you could get smoother highs, due to less comb filtering, but for me personally, I am not willing to compromise the large listening area the straight arrays provide...
I would like to see a STEP of a singular large 72" tall ribbon. What does "time arrival" problems sound like without comb filtering? They say our ear "locks on" to the closest point, but I can not validate this point.
Wesayso, I am glad You are at least thinking productively when You are intoxicated, Thanks for your honesty.
The Scaena is still an array of tweeters. I think the comb filtering is minimized with an array of ribbons (or are they planers) like that. I am not sure about the driver spacing of the two arrays is the best for a 5500 Hz crossover frequency. 13740 (speed of sound in inches) divided by 5500 (Hz) is about 2 1/2 inches or about 64 mm. Wouldn't there be comb filtering between the tweeter and main arrays? Sure You can "toe" each array in on your listening area, but again, maybe giving up the large listening area.
It seems there are many things we like about our full-range arrays, if we could only have that "ribbon" type of sound on the highs. There are a couple of other threads of people building "full-range" ribbon drivers, covering 200 -300 Hz. But as Marin says... (next post)...
Marin's observation...
So maybe the full-range ribbon is not all it is cracked up to be... But maybe building a 72" tall ribbon tweeter to cross in at 3000 hz? Can we get the driver spacing between the arrays down to 4.58 inches or 115 mm center to center? The magnets and steel bars needed for the structure of ribbon drivers makes that a challenge, but maybe designing it for 3000 Hz and above is possible... Then can we get it to integrate? Quite a journey none the less... Any takers?
Allen
(Disclaimer: I am writing this after three 12 hour night shifts of work and slightly lacking sleep)
So maybe the full-range ribbon is not all it is cracked up to be... But maybe building a 72" tall ribbon tweeter to cross in at 3000 hz? Can we get the driver spacing between the arrays down to 4.58 inches or 115 mm center to center? The magnets and steel bars needed for the structure of ribbon drivers makes that a challenge, but maybe designing it for 3000 Hz and above is possible... Then can we get it to integrate? Quite a journey none the less... Any takers?
Allen
(Disclaimer: I am writing this after three 12 hour night shifts of work and slightly lacking sleep
)
It would be more like an expanding line array I guess... But you'd have to choose an ideal listening height for that driver wouldn't be shaded.
But from my experience you do hear the closest driver first and the tail of the high frequency output isn't heard as echo, but added to that first arrival.
One of the fun things I heard with the array laying down and testing with a battery if all drivers were connected properly. As I moved along the array I only heard what seemed like one driver reacting to my battery pulse.
I should state all these ideas have been in my head while working on this project before I even had sound. Just curiosity on my part. Trying to figure out if we can have even more perfection.
I'd love to try a long ribbon tweeter but wouldn't be able too without heavy funding
. I've never heard the Scaena's and as such I really have no clue what they are able off. That doesn't make me less curious though. I've asked people with first hand listening experience about their impressions. They have impressed quite a few listeners over the years.Of coarse ideas of a single tweeter in the middle has also been on my mind, but like you Allen, I'm a bit convinced the listening experience is due to the line behavior and I wouldn't want to compromise that either.
Right now I have the speakers hooked up to my Asus card, with the ground loop isolator back in the chain. I'll have to redo my measurements and check if my conclusion about the ground loop isolator was true. (It seemed to bring in extra distortion)
One thing I do know, the up sampling DAC does sound different from my Asus card. We'll see which one I like most after another measurement session.
Yes.
And it's the same between platforms as well.
That's one of the reason I went from PC to Mac, as the drivers from Presonus for my audio interface are so much better on the Mac than on the PC side.
So, I do believe that sound is handled differently between sound cards. You can even hear it between playing software. On the Mac, the way Audirvana handles sound is way cleaner than Amarra. Audirvana is all about clarity and detail... which works well for some music, but add in electronic and it sounds... dull. Amarra sounds fuller on electronic music, but muddied on clean acoustic tracks.
Sorry, I derailed a bit here.
It's all about choices. Hardware and software will have a huge impact, after the speakers themselves. Software is easy to try, lots of demos out there. Hardware... not so much. Better buy from a store that allows returns!
Are both of these software packages claiming and/or achieving bit-perfect output?
I don't know why the Asus sounds different (yet) but there are physical differences beside the normal hardware choices. I have the Asus connected with a ground loop isolator. That could be a factor. I run the Asus at native 24 bit 44.100 signal for now. The DAC is always up sampling to 24/192.
I'm always a bit careful as to why something could sound different. I need more time (hopefully happening soon) to delve in to the "why" question.
I do realise if you listen for differences you will hear them. Sometimes even fooling myself into hearing differences, I'll admit to that. Like when I changed a parameter in the MSED plugin and was sure to hear a difference, only to find out the box to activate the plugin wasn't even 'ticked'.
I don't know why the Asus sounds different (yet) but there are physical differences beside the normal hardware choices. I have the Asus connected with a ground loop isolator. That could be a factor. I run the Asus at native 24 bit 44.100 signal for now. The DAC is always up sampling to 24/192.
I'm always a bit careful as to why something could sound different. I need more time (hopefully happening soon) to delve in to the "why" question.
I do realise if you listen for differences you will hear them. Sometimes even fooling myself into hearing differences, I'll admit to that. Like when I changed a parameter in the MSED plugin and was sure to hear a difference, only to find out the box to activate the plugin wasn't even 'ticked'.
Yes, both software are supposed to be bit-perfect, but there are differences in the output qualities. Amarra is the great party app, sounding much fuller, but annoying when you are sitting alone on the couch wanting to hear the little details. Audirvana provides that experience, but sounds a bit dull when you have people in the room and want to have some engaging music.
Just as with speaker parts and crossover parts, there are levels of quality in DACs, audio interfaces and their respective drivers. Drivers account for a lot in the mix. It was so obvious with my interface listening to the Windows vs Mac drivers (which were written by the same company for the same product!). Even my wife, who doesn't really care about this, thought the sound, after I switched to the Mac, sounded nicer... but she didn't know I had switched to the Mac. She just thought I had some different settings.
Some of the best products and drivers come from a company called RME. Their quality and support is amazing... which also means their products are more expensive to buy. I wish we had a RME distributor around here. But, at the time, I found a Presonus distributor, so I went with that.
My trick when I want to A/B a setting that is easy to do with a check/uncheck at the click of a mouse. I place the pointer on the check box, close my eyes, and click a bunch of times, not paying attention, until I have no idea if it is checked or unchecked. Without opening my eyes, I A/B the settings and decide which I like best... then I open my eyes to see which I preferred. Eyes can indeed trick you in "hearing" something you want to hear, even if it's not really there.
Just as with speaker parts and crossover parts, there are levels of quality in DACs, audio interfaces and their respective drivers. Drivers account for a lot in the mix. It was so obvious with my interface listening to the Windows vs Mac drivers (which were written by the same company for the same product!). Even my wife, who doesn't really care about this, thought the sound, after I switched to the Mac, sounded nicer... but she didn't know I had switched to the Mac. She just thought I had some different settings.
Some of the best products and drivers come from a company called RME. Their quality and support is amazing... which also means their products are more expensive to buy. I wish we had a RME distributor around here. But, at the time, I found a Presonus distributor, so I went with that.
My trick when I want to A/B a setting that is easy to do with a check/uncheck at the click of a mouse. I place the pointer on the check box, close my eyes, and click a bunch of times, not paying attention, until I have no idea if it is checked or unchecked. Without opening my eyes, I A/B the settings and decide which I like best... then I open my eyes to see which I preferred. Eyes can indeed trick you in "hearing" something you want to hear, even if it's not really there.
JRiver "Surround" effect with point source...
I had limited play time on my night off, but I did get the Avebury set-up and ready. I tried them with the "surround" effect in JRiver. Oddly enough, I really did not care for the effect with this system. The effect does the same thing on Avebury as it does with the arrays, but the effect even in "subtle" mode sounds over done through the Avebury. It is as if Avebury does not throw out enough sound to make the effect sound "convincing". The Avebury does not have the encompassing sound of a line source. As I have said before, the impression I feel when I switch from the Arrays to the Avebury is that I am going from being in the concert, to watching it on TV. You can add all the ambiance You want to the show You are watching, but it just does not have the true to life "scale" of the real event. I think our line arrays bring us closer to that. I feel like my head is in a vice when I listen to point sources.
The funny thing, with all the cool things we can do with EQ and DSP, You still can not beat the physical aspect of moving a lot of air over a large area. I had more fun playing with the DSP settings with the arrays, where as the more I altered with the Avebury, the more "false" it sounded (except EQ to flatten the response). I preferred the straight, dry, and focused sound with the Avebury, even though it is not very engaging. Point sources just do not cut it for me anymore. But having said all of that, I enjoy using the Avebury for my front channels when I set-up my 4.2 system for movies. In that case, the distant, focused sound in the front works, as it has the 3D ambiance added in by the back two speakers, in my case the arrays.
I would still like to try the tweeter with the Avebury, see if the full disappearing act changes my perception.
We will have to see what I learn on the next night off...
Allen
I had limited play time on my night off, but I did get the Avebury set-up and ready. I tried them with the "surround" effect in JRiver. Oddly enough, I really did not care for the effect with this system. The effect does the same thing on Avebury as it does with the arrays, but the effect even in "subtle" mode sounds over done through the Avebury. It is as if Avebury does not throw out enough sound to make the effect sound "convincing". The Avebury does not have the encompassing sound of a line source. As I have said before, the impression I feel when I switch from the Arrays to the Avebury is that I am going from being in the concert, to watching it on TV. You can add all the ambiance You want to the show You are watching, but it just does not have the true to life "scale" of the real event. I think our line arrays bring us closer to that. I feel like my head is in a vice when I listen to point sources.
The funny thing, with all the cool things we can do with EQ and DSP, You still can not beat the physical aspect of moving a lot of air over a large area. I had more fun playing with the DSP settings with the arrays, where as the more I altered with the Avebury, the more "false" it sounded (except EQ to flatten the response). I preferred the straight, dry, and focused sound with the Avebury, even though it is not very engaging. Point sources just do not cut it for me anymore.
But having said all of that, I enjoy using the Avebury for my front channels when I set-up my 4.2 system for movies. In that case, the distant, focused sound in the front works, as it has the 3D ambiance added in by the back two speakers, in my case the arrays. I would still like to try the tweeter with the Avebury, see if the full disappearing act changes my perception.
We will have to see what I learn on the next night off...
Allen
Last edited:
Thanks for testing this Allen, I am not surprised with the results.
I've listened to the arrays with the effect on and off and every time I prefer to have it turned on. But in testing with Headphones I want it off.
I love the arrays for movies though, It all comes alive when watching with the arrays. But sometimes I wish I had a bigger TV screen.
I've listened to the arrays with the effect on and off and every time I prefer to have it turned on. But in testing with Headphones I want it off.
I love the arrays for movies though, It all comes alive when watching with the arrays. But sometimes I wish I had a bigger TV screen
.