torgeirs,
Don't get your worry for low end
admit have no experience with line arrays myself other than had visiting wesayso's beautiful living room.
These could be a typical inside house use.
One response below is 1x TC9 feeded 12,5 watts and the other is 25x TC9 feeded 12,5 watts, look the lower red curves which show excursion fall from max excursion to about 0,5mm with 25x TC9 and now lets tap response out at 25x TC9 curve four meter away (-6dB) from 1x TC9 curve which is close to PRO world recommend 84-86dB SPL at listening position.
Grey is sum of room gain times 25x TC9 response into REW.
Don't get your worry for low end
admit have no experience with line arrays myself other than had visiting wesayso's beautiful living room.These could be a typical inside house use.
One response below is 1x TC9 feeded 12,5 watts and the other is 25x TC9 feeded 12,5 watts, look the lower red curves which show excursion fall from max excursion to about 0,5mm with 25x TC9 and now lets tap response out at 25x TC9 curve four meter away (-6dB) from 1x TC9 curve which is close to PRO world recommend 84-86dB SPL at listening position.
Grey is sum of room gain times 25x TC9 response into REW.
Attachments
Last edited:
There is that level of change in the frequency response but it is because there is EQ cut above 150Hz due to the line array gain in that region. Seems like you keep missing that part. If you add in the room gain the boost would likely drop to 12dB or so but I will show the in room plots when I get there.
REW has a dual tone function with selectable frequencies so I can try to run that test at some point.
I don't consider this to be a system for ear damaging levels. 90 to 100dB is my target range.
Strange to compare a system like that with so much more volume displacement capability. Of course a line array of small full range drivers cannot compete.
If I wanted to play at 110dB + then I would be building something different.
I don't know either REW says 6% at 50Hz in my measurement.
Seems like there might have been some confusion over the distortion graph being EQ'd vs No EQ. What is interesting is that the EQ did not raise the distortion very much in the low frequencies outdoors.
Lets only compare or discuss above 50Hz because below that environmental noise could well have affected the measurement.
If we were to add 10 to 15dB to the signal from room gain like Byrtt showed the distortion would be very good to quite a low frequency.
Attachments
The graph in:
http://www.diyaudio.com/forums/atta...rray-cnc-cabinet-distortion-outdoor-no-eq.jpg
says it all and don't defy physics
Again the needed execurtion at 30 vs 250Hz sees the difference in dB not the absolute values. I have seen enough room curves to know that a EQ value at 30 Hz is impossible to see from a general room gain curve. Maybe you are lucky and has a room peak node at 30 Hz as my living room system. There I need only 2 * 12 inch low execurion units.
Remeber that there are RMS value and peak value. So if crest is 20 dB and the peak is at 30 Hz the speaker has to cope with 87 + 20 dB = 107 dB.
But, at 87 dB RMS and normal roomsize (roomgain) and RT60 values I believe you in saying this construction can play very undistorted.
http://www.diyaudio.com/forums/atta...rray-cnc-cabinet-distortion-outdoor-no-eq.jpg
says it all and don't defy physics
Again the needed execurtion at 30 vs 250Hz sees the difference in dB not the absolute values. I have seen enough room curves to know that a EQ value at 30 Hz is impossible to see from a general room gain curve. Maybe you are lucky and has a room peak node at 30 Hz as my living room system. There I need only 2 * 12 inch low execurion units.
Remeber that there are RMS value and peak value. So if crest is 20 dB and the peak is at 30 Hz the speaker has to cope with 87 + 20 dB = 107 dB.
But, at 87 dB RMS and normal roomsize (roomgain) and RT60 values I believe you in saying this construction can play very undistorted.
To make any useful bass with an array of small drivers you need the room. This system is meant to be close to a wall (I am 0.5 m from the wall with the baffle for a reason)
Don't underestimate the sheer size of the array as it has advantages compared to a single 12". But we can't defy physics and it's still only comparable to a 12" with about 7mm x-max. It suffers less from room modes because of the many separate drivers all interact a little different with boundaries. They average out most ill effects.
I swap energy between my arrays in the bottom end to get the most out of it. I have one array close to a corner, which helps to get 30 Hz without much boost, though it suffers from it's position at 70 Hz. The other array isn't in a corner and struggles more to hit 30 Hz, but does not have a problem at 70 Hz and together they sum real well over a wide listening area. They "kind of" work like DSP-ed distributed subs. Though lacking the power of multiple real subs for home theatre.
Don't underestimate the sheer size of the array as it has advantages compared to a single 12". But we can't defy physics and it's still only comparable to a 12" with about 7mm x-max. It suffers less from room modes because of the many separate drivers all interact a little different with boundaries. They average out most ill effects.
I swap energy between my arrays in the bottom end to get the most out of it. I have one array close to a corner, which helps to get 30 Hz without much boost, though it suffers from it's position at 70 Hz. The other array isn't in a corner and struggles more to hit 30 Hz, but does not have a problem at 70 Hz and together they sum real well over a wide listening area. They "kind of" work like DSP-ed distributed subs. Though lacking the power of multiple real subs for home theatre.
Home theatre or rockconcert SPLs (and EQing)
So for it's intended use it should be super.
Guess you measure both arrays seperately indoors. And then both as mono 50 speaker sub.
And adjust both speakers for best mono 50 speaker low bass? (Under about 70 Hz depending on room size).
I would even make a mono sub filter and sum it with a stereo HP filter if I had this system with DSP before the stereo amp.
I would be careful to not adjust one of the speakers more than 3 dB over the other, that is double cone displacement, at lowest bass. The reason is not to have displacement induced mid and HF distortion "dominated" by one of the speakers.
Anyway, congrats with very nice build and excelent price/performance ratio!
Just an fun fact. I really like the sound of the Seas fa22rcz
H1597-08 FA22RCZ
It has a rising instead of a falling bass to treble response, but the same unlinear treble. My ears are at least quite insensitive to such HF comb filtering
So for it's intended use it should be super.
Guess you measure both arrays seperately indoors. And then both as mono 50 speaker sub.
And adjust both speakers for best mono 50 speaker low bass? (Under about 70 Hz depending on room size).
I would even make a mono sub filter and sum it with a stereo HP filter if I had this system with DSP before the stereo amp.
I would be careful to not adjust one of the speakers more than 3 dB over the other, that is double cone displacement, at lowest bass. The reason is not to have displacement induced mid and HF distortion "dominated" by one of the speakers.
Anyway, congrats with very nice build and excelent price/performance ratio!
Just an fun fact. I really like the sound of the Seas fa22rcz
H1597-08 FA22RCZ
It has a rising instead of a falling bass to treble response, but the same unlinear treble. My ears are at least quite insensitive to such HF comb filtering
The EQ cuts from 150Hz to 1.5K which reduces distortion in that range as you see in the EQ vs no EQ plots above.
What you might not appreciate is that the drivers are barely moving to produce the un EQ'd response at 80dB. The rolloff of the sealed box dictates the low end response and that is basically the same as a single driver in a 2 litre box. But because there are 25 drivers when the low end boost is applied it still doesn't cause the drivers to have to move a huge amount which is why the distortion doesn't rise hardly at all with the EQ boost at that level. Cutting sharply below 20Hz helps a lot too.
The downside like with all EQ boosted sealed speakers is that to achieve the boost the overall signal level has to be attenuated and then more power applied back with the amplifier limiting the maximum spl.
In this system it is much more likely to run out of amp power before the drivers run out of linear excursion.
https://www.parts-express.com/pedocs/specs/264-1062--tymphany-tc9fd18-08-spec-sheet.pdf
So 85 dB at mid and 60 dB at 30Hz
32 drivers is 2^5 so max 6dB*5 = 30 dB rise with 32 drivers summing perfectly. That is 90 dB sensitivity total max at 30 Hz. That is 2,83 volt applied to each driver.
Interesting to model the cone displacement at that frequency and volts. (or is it 3mm at 9 volts => 1mm at 3 volts.)
And the x-max is tiny 2.6mm so very little cone displacement is needed to push the cone into unlinear area.
So 85 dB at mid and 60 dB at 30Hz
32 drivers is 2^5 so max 6dB*5 = 30 dB rise with 32 drivers summing perfectly. That is 90 dB sensitivity total max at 30 Hz. That is 2,83 volt applied to each driver.
Interesting to model the cone displacement at that frequency and volts. (or is it 3mm at 9 volts => 1mm at 3 volts.)
And the x-max is tiny 2.6mm so very little cone displacement is needed to push the cone into unlinear area.
Last edited:
You can't use a manufacturers spec sheet (that was taken presumably on an IEC baffle) and compare like that. The enclosure that I have used is completely different and loads the driver very differently.
XRK has made a model in akabak very similar to wesayso's tower which is probably the best to use to answer these sorts of questions.
The other alternative is to look at the graph Byrtt showed of a standard box model of 25 combined drivers and see the output at a given frequency and draw a line from there to give maximum output.
x-max is relative and there are 25 of those drivers at 2.6mm, like I have said before it takes a significant amount of power to get to x-max even at 30 Hz with significant EQ (300 to 400W at 8 Ohms). More power than I have available in an amplifier so for me it is a theoretical concern.
As wesayso has said before it is equivalent to a 12 inch driver in cone area with x-max of 7mm. That is a fairly decent driver that could be EQ'd in a sealed box to produce reasonable output.
There are quite a few arrays like these that have been built and they have all been shown to work as intended. They are not for rock concert level spl's unless subwoofers are used to get the low end or you just go without it.
I built this as a standalone full range speaker ~20Hz at what I consider to be loud enough levels.
I have had mine since 2014 (after being inspired by wesayso's Two Towers) and I can attest that low end is indeed well taken care of.
I have placed a few unknowing friends in front of mine and there are three main characteristics they become very aware of / surprised by:
1 - Clarity
2 - Field of view (sound stage, wall of sound)
3 - Very uniform low frequency response, and how low it goes with such apparent ease.
But do they suit the listening criterias of all? Hardly
They work for mine
Last edited:
For info my visit at wesayso's home was close to 24 hours and not one moment it was possible to visual to see any cone movement and it was not because we didn't listen kick drum from Eagles track Hotel California from live album hell freezes over
Well at 30 Hz the baffle don't matter unless it is comparable to backwall size.
Enclose simulation from scan-speaks excel sim.
Share links to zaph's measurements of similar vifa driver.
http://www.zaphaudio.com/temp/Vifa-TC9FD18-08-TS.gif
http://www.zaphaudio.com/temp/Vifa-TC9FD18-08-FR.gif
http://www.zaphaudio.com/temp/Vifa-TC9FD18-08-HD.gif
Scan speak estimates 61 dB at 30Hz for single driver at 9 volts.
+6 dB each doubling of number of drivers gives max 5x6dB = 30 dB
So drivers combined can produce max 91dB at 30Hz when compared to other single drivers.
The drivers are actually a 15" with Fs at about 120 Hz , 3mm x max and 125dB/m SPL at 2,83V (At high frequency they only add 3dB each doubling and betwwen 6 and 3dB in between)
As you see from the zaph links the distortion is fixed under the frequency where displasement is fixed. The ratio changes because the SPL is falling.
Just to inform other readers that like roule of tumb estimates that can be done without a computer at hand.
How it then interacts with the room is pure luck (or unluck) in my opinion.
But from your statements I think a solution with many small subs is a workable solution especially with low (70Hz) LP filter. So that Sub cone movement don't distort the rest of the sound
Enclose simulation from scan-speaks excel sim.
Share links to zaph's measurements of similar vifa driver.
http://www.zaphaudio.com/temp/Vifa-TC9FD18-08-TS.gif
http://www.zaphaudio.com/temp/Vifa-TC9FD18-08-FR.gif
http://www.zaphaudio.com/temp/Vifa-TC9FD18-08-HD.gif
Scan speak estimates 61 dB at 30Hz for single driver at 9 volts.
+6 dB each doubling of number of drivers gives max 5x6dB = 30 dB
So drivers combined can produce max 91dB at 30Hz when compared to other single drivers.
The drivers are actually a 15" with Fs at about 120 Hz , 3mm x max and 125dB/m SPL at 2,83V (At high frequency they only add 3dB each doubling and betwwen 6 and 3dB in between)
As you see from the zaph links the distortion is fixed under the frequency where displasement is fixed. The ratio changes because the SPL is falling.
Just to inform other readers that like roule of tumb estimates that can be done without a computer at hand.
How it then interacts with the room is pure luck (or unluck) in my opinion.
But from your statements I think a solution with many small subs is a workable solution especially with low (70Hz) LP filter. So that Sub cone movement don't distort the rest of the sound
Attachments
torgeirs you continue want low end to be a problem, then okay fine for me and probably for some rooms.
Can say 80-90% of speakers i heard that Eagles live track reproduced is too exaggerated for that kick drum part which tells me this is not what sound engineer was listening when he finalized that mix, and a prove is into headphones it sounds absolute natural and fine, same natural lows was heard on wesayso's system interaction including room and actual remember and noticed how xrk971 when kick drum for that track set in he got a big smile on his lips, this is telling me how clean deep and natural plus loud that happening sounded.
Can say 80-90% of speakers i heard that Eagles live track reproduced is too exaggerated for that kick drum part which tells me this is not what sound engineer was listening when he finalized that mix, and a prove is into headphones it sounds absolute natural and fine, same natural lows was heard on wesayso's system interaction including room and actual remember and noticed how xrk971 when kick drum for that track set in he got a big smile on his lips, this is telling me how clean deep and natural plus loud that happening sounded.
Disclaimer, torgiers, I don't think anything will sway your opinion. But to answer a few points...
Speaking purely for myself, I can't say my arrays were cheap to build. I've spend quite a bit of money and even more time on mine and don't regret it in any way.
My bass is still stereo and the swapping of energy is to balance what each speaker does (leveling the cone movement), as it evens out the amount of EQ needed on the bottom end between the left and right speaker, so it doesn't make it worse, it makes it more even. These decisions are driven by looking at actual FIR correction signals (the amount of boost needed) and distortion levels.
I agree, at 30 Hz it's close enough to the wall boundary to consider that to be the baffle. Just look how long a 30 Hz wave is in meters.
Did you look at ra7's data on this doubling of distance? You may be surprised.
Why not use one of the direct measured distortion numbers from actual measurements of arrays with DSP applied instead of using single driver data? More than one chart and array available, right here on this forum.
That would give you the listening position view on this (using the room), not a 1m measurement and then figuring you'd be ok in your room. How few people on her actually show listening spot measurements...
Personally, I'm wired to verify each step, I hate making assumptions. It probably comes from my formal training as an engineer long ago. Neh, I even did that well before I "was trained". If I thought I was going off into a deep end I would never have dared to go all out like I did. It takes a lot of patience, blood sweat and tears to do it like this:
In all honesty, I had backup plans ready with tweeters and subs (at least "on or in" my mind). It just turned out so well (even better than expected) that I did not need to reach for such a plan at all. At least, not for music.
Home theatre is a little different though. They really do use that low end, hardly any song I've come by has a full power bottom end. One of the strongest bottom ends in music I have encountered are on songs like "Lullaby" by "A Perfect Circle" or some RAP music by "E40".
If we look at that "Lullaby" track, analyzed by frequency level:
It shows a peak at 25 Hz, only 6 dB lower than the biggest peak (at ~34 Hz) in the song.
I bet many haven't "felt" what this song brings to the table, I love it. But with a song like that, I'm pushing the boundaries of my amp, not my speakers (yet). This is a song that could get me in trouble if I up the level above my "set" reference.
I'd say sim it with an actual wall nearby or better with the actual room (as said, it needs the room) and that will tell you your numbers are not complete here, if you leave the room out. It has not been a secret ever that you do need the room (as in: the wall behind the array).
Simmed without help from the room the X-max figure will get you to about 94 dB at 30 Hz, which luckily isn't x-mech on these drivers, they have a little extra movement to spare. This is before any room gain is accounted for.Free standing outside you'd be wise to cut them off at ~70 Hz to hit an honest 110 dB.
BYRTT, do you remember where the torture tests of the TC9 drivers are at?
They can take quite a bit of punishment. Not that easy to blow up at all! Of course it can be done.
I would not have spend close to 2 years full time on these speakers if I had any doubt or if I thought I was depending on luck to make it work. The data available on line array and their behavior across the net was good enough validation (together with many sims) that this can work very well. For that to happen the in room position is key to make it work. It's actually a lot more forgiving than that 12" with 7mm x-max or that 15" (it's closer to 16" actually) with 2.55 mm x-max which are equal in displacement. (both as used in a sealed enclosure with DSP)
No luck involved at all, just pure data that steered my decision making to go all out, one year of building the "damn things" and another year spend to learn to get the maximum out of them (with DSP). (And another year of thinking it trough and going over existing arrays and sims time and again before having the balls to do it My original plans date back to September 2011)
In hindsight, time well spend for me. Many more months were spend later on, to optimize everything even further, with many a fun experiment thrown in.
APL Timing chart:
This might explain the experience BYRTT describes. No weird modes in the room after the big bangs. No overhang on notes
Frequency and phase:
Pictured is the stereo measurement with a 6 cycle FDW. Again, in the listening spot, no stand outs or weird room effects...
Without any (FDW) window applied, this time smoothed at 1/12 octave here's the left/right channel balance:
This graph also shows where the left takes over from the right speaker and vice versa. Together they produce the graphs shown above.
Impulse of a stereo measurement at the exact sweet spot:
You've already seen one of many distortion charts available in my thread.
I guess we cannot take away the worries you have about the cones flapping to produce the low end, while also having to play the higher notes.
What I mostly worried about was the havoc the room (any room) usually adds to any speaker you place in there. I'm comfortable with the ~88 dB average although I did play at louder levels without noticing it instantly because my girl dusted off my amplifier. I usually notice something like that due to hearing my room protesting on deep bass notes. Something will rattle sooner or later.
It's clear to me you play at different levels. You'd need subs. Personally I'd do line array subs (lol). If you're up for it, download the free APL_TDA demo and show me/us what that bass part of yours looks like at the listening spot.
(together with a distortion plot and FR)
I cannot imagine it staying clean after the main pulse. I'd expect many things to start rattling sympathetically. But then again, I don't know your room (nor it's size).
In short: You probably wouldn't be satisfied with the SPL levels these arrays can bring. They do bring a lot more than most assume they are capable of. And I'm not talking about bass levels alone. High fun factor, many smiles per minute and lots of chills down the spine. In short: they (re)produce the emotion I want to hear/feel that's embedded in the music I like. They actually made me enjoy many more genre's than I used to listen to, I even like listening to Dance now! (lol)
That's pretty weird for an old Rocker like me!
Home theatre or rockconcert SPLs (and EQing)
So for it's intended use it should be super.
Guess you measure both arrays seperately indoors. And then both as mono 50 speaker sub.
And adjust both speakers for best mono 50 speaker low bass? (Under about 70 Hz depending on room size).
I would even make a mono sub filter and sum it with a stereo HP filter if I had this system with DSP before the stereo amp.
I would be careful to not adjust one of the speakers more than 3 dB over the other, that is double cone displacement, at lowest bass. The reason is not to have displacement induced mid and HF distortion "dominated" by one of the speakers.
Anyway, congrats with very nice build and excelent price/performance ratio!
Speaking purely for myself, I can't say my arrays were cheap to build. I've spend quite a bit of money and even more time on mine and don't regret it in any way.
My bass is still stereo and the swapping of energy is to balance what each speaker does (leveling the cone movement), as it evens out the amount of EQ needed on the bottom end between the left and right speaker, so it doesn't make it worse, it makes it more even. These decisions are driven by looking at actual FIR correction signals (the amount of boost needed) and distortion levels.
I agree, at 30 Hz it's close enough to the wall boundary to consider that to be the baffle. Just look how long a 30 Hz wave is in meters.
Did you look at ra7's data on this doubling of distance? You may be surprised.
Why not use one of the direct measured distortion numbers from actual measurements of arrays with DSP applied instead of using single driver data? More than one chart and array available, right here on this forum.
That would give you the listening position view on this (using the room), not a 1m measurement and then figuring you'd be ok in your room. How few people on her actually show listening spot measurements...
Personally, I'm wired to verify each step, I hate making assumptions. It probably comes from my formal training as an engineer long ago. Neh, I even did that well before I "was trained". If I thought I was going off into a deep end I would never have dared to go all out like I did. It takes a lot of patience, blood sweat and tears to do it like this:
In all honesty, I had backup plans ready with tweeters and subs (at least "on or in" my mind). It just turned out so well (even better than expected) that I did not need to reach for such a plan at all. At least, not for music.
Home theatre is a little different though. They really do use that low end, hardly any song I've come by has a full power bottom end. One of the strongest bottom ends in music I have encountered are on songs like "Lullaby" by "A Perfect Circle" or some RAP music by "E40"
.If we look at that "Lullaby" track, analyzed by frequency level:
It shows a peak at 25 Hz, only 6 dB lower than the biggest peak (at ~34 Hz) in the song.
I bet many haven't "felt" what this song brings to the table, I love it. But with a song like that, I'm pushing the boundaries of my amp, not my speakers (yet). This is a song that could get me in trouble if I up the level above my "set" reference.
I'd say sim it with an actual wall nearby or better with the actual room (as said, it needs the room) and that will tell you your numbers are not complete here, if you leave the room out. It has not been a secret ever that you do need the room (as in: the wall behind the array).
Simmed without help from the room the X-max figure will get you to about 94 dB at 30 Hz, which luckily isn't x-mech on these drivers, they have a little extra movement to spare. This is before any room gain is accounted for.Free standing outside you'd be wise to cut them off at ~70 Hz to hit an honest 110 dB.
BYRTT, do you remember where the torture tests of the TC9 drivers are at?
They can take quite a bit of punishment. Not that easy to blow up at all! Of course it can be done.
I would not have spend close to 2 years full time on these speakers if I had any doubt or if I thought I was depending on luck to make it work. The data available on line array and their behavior across the net was good enough validation (together with many sims) that this can work very well. For that to happen the in room position is key to make it work. It's actually a lot more forgiving than that 12" with 7mm x-max or that 15" (it's closer to 16" actually) with 2.55 mm x-max which are equal in displacement. (both as used in a sealed enclosure with DSP)
No luck involved at all, just pure data that steered my decision making to go all out, one year of building the "damn things" and another year spend to learn to get the maximum out of them (with DSP). (And another year of thinking it trough and going over existing arrays and sims time and again before having the balls to do it
My original plans date back to September 2011) In hindsight, time well spend for me.
Many more months were spend later on, to optimize everything even further, with many a fun experiment thrown in.APL Timing chart:
This might explain the experience BYRTT describes. No weird modes in the room after the big bangs. No overhang on notes
Frequency and phase:
Pictured is the stereo measurement with a 6 cycle FDW. Again, in the listening spot, no stand outs or weird room effects...
Without any (FDW) window applied, this time smoothed at 1/12 octave here's the left/right channel balance:
This graph also shows where the left takes over from the right speaker and vice versa. Together they produce the graphs shown above.
Impulse of a stereo measurement at the exact sweet spot:
You've already seen one of many distortion charts available in my thread.
I guess we cannot take away the worries you have about the cones flapping to produce the low end, while also having to play the higher notes.
What I mostly worried about was the havoc the room (any room) usually adds to any speaker you place in there. I'm comfortable with the ~88 dB average although I did play at louder levels without noticing it instantly because my girl dusted off my amplifier. I usually notice something like that due to hearing my room protesting on deep bass notes. Something will rattle sooner or later.
It's clear to me you play at different levels.
You'd need subs. Personally I'd do line array subs (lol). If you're up for it, download the free APL_TDA demo and show me/us what that bass part of yours looks like at the listening spot. (together with a distortion plot and FR)
I cannot imagine it staying clean after the main pulse. I'd expect many things to start rattling sympathetically. But then again, I don't know your room (nor it's size).
In short: You probably wouldn't be satisfied with the SPL levels these arrays can bring. They do bring a lot more than most assume they are capable of. And I'm not talking about bass levels alone. High fun factor, many smiles per minute and lots of chills down the spine. In short: they (re)produce the emotion I want to hear/feel that's embedded in the music I like. They actually made me enjoy many more genre's than I used to listen to, I even like listening to Dance now! (lol)
That's pretty weird for an old Rocker like me!
Thanks a nice overview there wesayso you makes me want build arrays if day had more hours : )
Could it be the XO less design you have and its snappy good timing down to lows that call for less head room at lows that builders got used to normal will pay back with traditional XO designs, i mean honestly after had half an hour Beolap 90 demo would have swear it was using linear phase XO guided by signs from how DC thumbs was presentated, then later when stereophile released measurement could see myself was cheated by the very hot low end house curve it perform in narrow mode (https://www.stereophile.com/content/bang-olufsen-beolab-90-loudspeaker-measurements).
Synthetic example below is 17Hz-22kHz pass band with a downward tilt as frq goes up, using traditionally minimum phase XO at HT recommended 80Hz frq point both LR 4th and 8th order shown will introduce some delays, do we have tendency mask these delays asking for a more hot house curve at lows and thereby demand for sub section has more head room as the nice performing ones torgeirs talked about.
Could it be the XO less design you have and its snappy good timing down to lows that call for less head room at lows that builders got used to normal will pay back with traditional XO designs, i mean honestly after had half an hour Beolap 90 demo would have swear it was using linear phase XO guided by signs from how DC thumbs was presentated, then later when stereophile released measurement could see myself was cheated by the very hot low end house curve it perform in narrow mode (https://www.stereophile.com/content/bang-olufsen-beolab-90-loudspeaker-measurements).
Synthetic example below is 17Hz-22kHz pass band with a downward tilt as frq goes up, using traditionally minimum phase XO at HT recommended 80Hz frq point both LR 4th and 8th order shown will introduce some delays, do we have tendency mask these delays asking for a more hot house curve at lows and thereby demand for sub section has more head room as the nice performing ones torgeirs talked about.
Attachments
Solution suggests that there is a problem that needs fixing. If you want to ignore the data and opinions of actual line array builders and listeners in favour of your own knowledge or rules of thumb then go ahead.
I read all of the line array threads many times before I built these. On seeing others data I was convinced enough to build my own. And like wesayso my version is not cheap, good value yes but not cheap to me.
When I get back home I will take some measurements in room and I will see if I can get the dual 30Hz and 1KHz test to work. If I can I will see how loud it gets before distortion becomes an issue. At that point we can discuss the data.
Until then can we just agree to disagree?
In my humble opinion a very good question, BYRTT. In nature, all sounds we hear are as is... meaning they don't have the time lag of the reproduction chain that's quite common due to crossovers or bass loading schemes (read ports).
We may be able to compensate that sense by exaggerating the bottom end. I've read a couple of papers where it states our brain uses masking if a bottom end becomes too loud. Basically it means it (the brain) will block out parts of the bottom end to be able to focus on what comes above.
We also know our brain is powerful enough to come up with ground tones if it only hears it's harmonics, meaning a speaker with a lacking low end can still be perceived as having those ground tones as long as it's harmonics are there.
Personally what I find particularly convincing is feeling the (ground) tones as much as hearing them. It's why I enjoy listening to my speakers even better than my headphones. I will admit I'm not playing loud enough to get the huge kick in the chest I've felt at live performances. While that kick mainly comes from the 100-200 Hz area, it needs to be loud to really give that sensation.
The bottom end moves you in the belly, more than on the chest.
I think most people also underestimate the role of their room in their stereo enjoyment. A smaller room does not have real reverb tones like a larger space has.
If one adds these tones synthetically the whole experience changes. With eyes closed you no longer feel like you're in a small room, the walls disappear (as long as the first arriving reflections that give away the room size are absorbed/diffracted enough). It changes the whole experience. I've had a lot of fun with that part of my experiments. It gives a completely different sense of energy, one that fills the room, without detracting from imaging queues.
For people to relate measurements to actual listening experiences we cannot separate the room from the speakers. What I've tried to do is get that combination to work as seamlessly as possible (well, as possible for me anyway).
You tell us how that Beolab 90 compares to what you've heard at my place. The Beolab did have a more potent bottom end with the twin Scan Speak 32W's.
Did you miss anything at my place? Did it allow you to drift off into the music?
Anyway, we're drifting off topic again. I could talk about this stuff for ages, out of pure enthusiasm, but let's wait for fluid to show us some in house measurements. I've had a visitor that stated: The energy and power of these lines is a lot like a good PA system. I thought of that as a compliment, even though a lot of PA systems do not extend that bottom end as far as people think. Maybe the timing does help making it sound dynamic.
Last edited:
Yes, sorry to make a stirr. Just to clarify. To compare the 25 units with one 16 inch, the about 120 hz Fs has to be taken into the equation.
I think it is interesting to read that a "25 cone unit" that is driven below fs from 150 hz can produce undistorted sound at normal levels.
I look forward to the 30 hz + 1k test. How loud can the 30 hz be before distortion can be heard/ measured at 1k
Think the bass stimuli could also be pink noise from 30 to 150 hz. (Region where displacement is "constant" with Volts and frequency.)
I think it is interesting to read that a "25 cone unit" that is driven below fs from 150 hz can produce undistorted sound at normal levels.
I look forward to the 30 hz + 1k test. How loud can the 30 hz be before distortion can be heard/ measured at 1k
Think the bass stimuli could also be pink noise from 30 to 150 hz. (Region where displacement is "constant" with Volts and frequency.)
Pretty sure you've heard about a Linkwitz transform where Mr. Linkwitz EQ's the low end of a subwoofer to get what he needs? It's the same principles as he's working under the (in box) Fs to do so. That sub doesn't stop playing at Fs. Though it wont be asked to do high frequencies.
Look at the Murphey Corner Line Array papers to see what it does with examples of another driver if you can't wait. He didn't do all tests you ask for but some examples are there.
I hope you do realize you're in the full range part of the forum. Many here play with one 3.5 to 4" driver that does it all
. Clearly you're trying to make a point for using a multi way solution while the point of using these arrays is to not use any crossover. Different strokes for different folks...