Just ordered 32 of these Foster 00463 midbass drivers as well as 32 of these DAYTON ND20FA neo dome tweets to build some open back lines. On the surface the response graphs look quite respectable seeing these will be under $300.00 in material costs. Should be high sensitivity and pretty low distortion/linear.
I'm hoping the line will get down below 100 cycles and plan on crossing them at 2500-3000 cycles. I will bi-amp these.
What do you guys think?
Any problems with series induction interaction (rolloff) from partial series wiring of the line array Fosters?
Neo tweeter response
Foster response 00463 - lower of the two
I'm hoping the line will get down below 100 cycles and plan on crossing them at 2500-3000 cycles. I will bi-amp these.
What do you guys think?
Any problems with series induction interaction (rolloff) from partial series wiring of the line array Fosters?
Neo tweeter response
Foster response 00463 - lower of the two
An externally hosted image should be here but it was not working when we last tested it.
Notice the loss in spl here in this array (and this is with 2 inch drivers):
http://www.partsexpress.com/projectshowcase/Kuze3201/Kuze3201noeq.jpg
Now some of that sp-loss won't be as exagerated because you aren't using as many drivers as that design had. However, because of the larger diameter of the drivers for yours vs. the 2 inch drivers in this design - sp-loss will likely begin a bit low in freq.
Effectivly what this means is that you'll need to either shelve the midband or add some upper freq. gain to the woofers for a fairly flat response at crossover.
As for the tweeters.. Zaph has them tested and recommends about 3.5 kHz for the crossover point. However you'll be using far more of them, AND you'll likely be using a steeper filter - so 2.5 kHz *may* be OK.
http://www.zaphaudio.com/tweetermishmash/
The tweeter line length at 2.5 kHz for aprox. 24 inches will net you a maximum of 2 meters listening distance to maintain linesource character. If you crossover at 3.5 kHz then add a meter to that figure (..but you'll also have to add even more gain to the woofer's top-end). (..the woofer line should be long enough to net an "infinite" line length.)
I think the thing that will be critical to good sound (not suggested above), will be making sure that there is very little airflow resistance behind the midbass drives.
http://www.partsexpress.com/projectshowcase/Kuze3201/Kuze3201noeq.jpg
Now some of that sp-loss won't be as exagerated because you aren't using as many drivers as that design had. However, because of the larger diameter of the drivers for yours vs. the 2 inch drivers in this design - sp-loss will likely begin a bit low in freq.
Effectivly what this means is that you'll need to either shelve the midband or add some upper freq. gain to the woofers for a fairly flat response at crossover.
As for the tweeters.. Zaph has them tested and recommends about 3.5 kHz for the crossover point. However you'll be using far more of them, AND you'll likely be using a steeper filter - so 2.5 kHz *may* be OK.
http://www.zaphaudio.com/tweetermishmash/
The tweeter line length at 2.5 kHz for aprox. 24 inches will net you a maximum of 2 meters listening distance to maintain linesource character. If you crossover at 3.5 kHz then add a meter to that figure (..but you'll also have to add even more gain to the woofer's top-end). (..the woofer line should be long enough to net an "infinite" line length.)
I think the thing that will be critical to good sound (not suggested above), will be making sure that there is very little airflow resistance behind the midbass drives.
Great info ScottG - Thanks
What I'm thinking - ( I won't have the drivers mounted till next week) - is starting with a narrow baffle say 7-8" wide and it may do some of the shelving on it's own and then adjust the response with with various size slanted wings to try and achieve a wide range linear response.
I have lots of extra wood here to expiriment with.
I'll post more on the woofer/mid line after I get some hands on with these things. I've never built a long line like this before - The hole saw will really be getting a work out next week.
After the Foster line is figured out I'll add the upper treble and try 16 of the domes a side. Another meter of them a side will surely add some cost - maybe it will be worth it..
What I'm thinking - ( I won't have the drivers mounted till next week) - is starting with a narrow baffle say 7-8" wide and it may do some of the shelving on it's own and then adjust the response with with various size slanted wings to try and achieve a wide range linear response.
I have lots of extra wood here to expiriment with.
I'll post more on the woofer/mid line after I get some hands on with these things. I've never built a long line like this before - The hole saw will really be getting a work out next week.
After the Foster line is figured out I'll add the upper treble and try 16 of the domes a side. Another meter of them a side will surely add some cost - maybe it will be worth it..
Interesting.. I did some modeling of the baffle and discovered some stuff I *should* have known.. but didn't.
I went through the Edge - which only allowed me to place 7 midbass 5.25 inch drivers on the panel (..but did allow me to make, but not see all of it, a panel that was 85 inches high - strange). Anyway - just bear this in mind (..and note that the exageration in sp loss at higher freq.s are likely greater than my 7 driver version).
The modled baffle was 85 inches tall and 8 inches wide. NOT calculated as an open baffle. The drivers were all set to the outer edge of one side of the baffle.
.. the sp loss IS significant at 1 meter distance for the midbass line. Its only down 1 db at 1 kHz. At 1.5 khz its down 3 db. At 2 khz its down about 5.5 db. And at 3 kHz its down about 6 db. From there its is NOT pretty (..and is pretty much unusable).
Now here is the interesting thing..
At 2 meters + it isn't so bad up to 2 kHz. Pretty much flat to 1.3 kHz. Down 3 db at 2 kHz and down 6 db at 3 kHz.
At 3 meters sp is actually UP about .5 db at 1.3 kHz. Down 1.5 db at 2 kHz and down 3 db at 3 kHz.
So this tells me that unless line array is specifically engineered for set listening distance, that tonal balances will occur with alternate listening distances. (..of course room reflections and total power response will also play a part. Note that at 10 + meters its pretty much the "farfield" and the thing is pretty flat overall - even at much higher freq.s)
As for the low-end - the 8 inch baffle and the midbasses you have selected will be more than sufficient for baffle-loss compensation. Even at 10 meters at 100 Hz the the line is only down about 4.5 db. At 1 meter the line down only 3 db at 100 Hz. So with these drivers you have the "thumbs-up" for the "lower"-end response.
For an Open baffle configuration..
You are 6 db down at 150 Hz for an 8 inch baffle with spl's starting to roll-off at around 500 Hz. This means you'll be -2 db with these midbass drivers at 150 Hz (..assuming the driver curves are an accurate representation). At 100 Hz figure about -6 to -7 db down for these midbass drivers. So for open baffle and a lower freq. response - these drivers just don't "cut it" without some eq.
HOWEVER - with such a high qts they effectivly have "eq", so they might make it open baffle 8 inches.
(note that considering the high qts of the midbass drivers - IF you go with an enclosed backing to them, (i.e. non-open baffle), then seriously consider an aperiodic enclosure.)
I went through the Edge - which only allowed me to place 7 midbass 5.25 inch drivers on the panel (..but did allow me to make, but not see all of it, a panel that was 85 inches high - strange). Anyway - just bear this in mind (..and note that the exageration in sp loss at higher freq.s are likely greater than my 7 driver version).
The modled baffle was 85 inches tall and 8 inches wide. NOT calculated as an open baffle. The drivers were all set to the outer edge of one side of the baffle.
.. the sp loss IS significant at 1 meter distance for the midbass line. Its only down 1 db at 1 kHz. At 1.5 khz its down 3 db. At 2 khz its down about 5.5 db. And at 3 kHz its down about 6 db. From there its is NOT pretty (..and is pretty much unusable).
Now here is the interesting thing..
At 2 meters + it isn't so bad up to 2 kHz. Pretty much flat to 1.3 kHz. Down 3 db at 2 kHz and down 6 db at 3 kHz.
At 3 meters sp is actually UP about .5 db at 1.3 kHz. Down 1.5 db at 2 kHz and down 3 db at 3 kHz.
So this tells me that unless line array is specifically engineered for set listening distance, that tonal balances will occur with alternate listening distances. (..of course room reflections and total power response will also play a part. Note that at 10 + meters its pretty much the "farfield" and the thing is pretty flat overall - even at much higher freq.s)
As for the low-end - the 8 inch baffle and the midbasses you have selected will be more than sufficient for baffle-loss compensation. Even at 10 meters at 100 Hz the the line is only down about 4.5 db. At 1 meter the line down only 3 db at 100 Hz. So with these drivers you have the "thumbs-up" for the "lower"-end response.
For an Open baffle configuration..
You are 6 db down at 150 Hz for an 8 inch baffle with spl's starting to roll-off at around 500 Hz. This means you'll be -2 db with these midbass drivers at 150 Hz (..assuming the driver curves are an accurate representation). At 100 Hz figure about -6 to -7 db down for these midbass drivers. So for open baffle and a lower freq. response - these drivers just don't "cut it" without some eq.
HOWEVER - with such a high qts they effectivly have "eq", so they might make it open baffle 8 inches.
(note that considering the high qts of the midbass drivers - IF you go with an enclosed backing to them, (i.e. non-open baffle), then seriously consider an aperiodic enclosure.)
Well let me build some baffles and see what really happens before you make me change my mind!
What inspired me to do this line array thing was hearing a 96 driver line using 30 cent one by three inch TV speakers at fellows house a couple of weeks ago.
The tiny drivers were in 3' wide sealed baffles with no tweeters and some little 6.5" isobaric bass drivers built into sub enclosures(biamped) in the back of the baffles for below 120 cycles. Other than the size (bigger then my horn systems at over 8 feet tall and 3 feet wide and aaround a fooot and a half deep!) and lack of high frequency extension I was pretty impressed with the sound. Mostly In the way the lines drove the room and the huge sweet spot. The upper bass and mids were very coherent and his line arrays projected a nice meaty, 3D image that was quite enveloping and full size. They also would play at serious levels with little strain like a good horn.
He used no eq at with this and up to 6k or so, I feel this 96 driver system is superior to my Acoustat Spectra 3 electrostats. His line was also easier to drive, no beaming, and walked all over the Acoustats in dynamics and pure FUN - The Acoustat is a pretty good sounding panel speaker with it's segmented panels.
So what I'm hoping to do is get these $ 3.00 drivers to sound as good as his 30 cent drivers with some response above 10K and a smaller footprint.
What inspired me to do this line array thing was hearing a 96 driver line using 30 cent one by three inch TV speakers at fellows house a couple of weeks ago.
The tiny drivers were in 3' wide sealed baffles with no tweeters and some little 6.5" isobaric bass drivers built into sub enclosures(biamped) in the back of the baffles for below 120 cycles. Other than the size (bigger then my horn systems at over 8 feet tall and 3 feet wide and aaround a fooot and a half deep!) and lack of high frequency extension I was pretty impressed with the sound. Mostly In the way the lines drove the room and the huge sweet spot. The upper bass and mids were very coherent and his line arrays projected a nice meaty, 3D image that was quite enveloping and full size. They also would play at serious levels with little strain like a good horn.
He used no eq at with this and up to 6k or so, I feel this 96 driver system is superior to my Acoustat Spectra 3 electrostats. His line was also easier to drive, no beaming, and walked all over the Acoustats in dynamics and pure FUN - The Acoustat is a pretty good sounding panel speaker with it's segmented panels.
So what I'm hoping to do is get these $ 3.00 drivers to sound as good as his 30 cent drivers with some response above 10K and a smaller footprint.
Naw.. those still comb.
- these are the drivers.. notice that they have the right rising response to compensate for the loss of sp at higher freq.s. (..at least in the near field.)
http://www.tymphany.com/datasheet/printview.php?id=116
Notice the combing here (but it has been eq'ed) for the tall 2 inch driver array. Subjectivly from about 3.5 kHz to 12 kHz it will have the subjective effect of a bit of white-noise overlayed on the soundstage and the imaging slightly more diffuse.
http://www.partsexpress.com/projectshowcase/Kuze3201/Kuze3201weqnosmooth.jpg
The vifa drivers will exhibit this even more and it can't be eq.ed away. It however doesn't really become objectionable until 12 kHz - and even then its pretty benign. The best that eq. can do in this circumstance is to smooth out some of the dips and peaks where its the worst (rather like "averaging").
parts value-wise.. lets see 50 drivers with a manufacturer price less than 30 US each and a little eq., that with all the normal associated business costs and no dealer to deal with *some how* manages to warrent a price tag of 19K US and is most certainly NOT a fullrange design. Sometimes I just want to beat the living cr@p out of "high-end" manufacturers.
- these are the drivers.. notice that they have the right rising response to compensate for the loss of sp at higher freq.s. (..at least in the near field.)
http://www.tymphany.com/datasheet/printview.php?id=116
Notice the combing here (but it has been eq'ed) for the tall 2 inch driver array. Subjectivly from about 3.5 kHz to 12 kHz it will have the subjective effect of a bit of white-noise overlayed on the soundstage and the imaging slightly more diffuse.
http://www.partsexpress.com/projectshowcase/Kuze3201/Kuze3201weqnosmooth.jpg
The vifa drivers will exhibit this even more and it can't be eq.ed away. It however doesn't really become objectionable until 12 kHz - and even then its pretty benign. The best that eq. can do in this circumstance is to smooth out some of the dips and peaks where its the worst (rather like "averaging").
parts value-wise.. lets see 50 drivers with a manufacturer price less than 30 US each and a little eq., that with all the normal associated business costs and no dealer to deal with *some how* manages to warrent a price tag of 19K US and is most certainly NOT a fullrange design. Sometimes I just want to beat the living cr@p out of "high-end" manufacturers.
based on that I need to see if Madisound shipped the drivers I ordered yesterday and try and switch them to the 'full range' version with the rising response - !
Top response is 'Full Range' - same price
Top response is 'Full Range' - same price
I actually started off writing just that (in my first post). Then I read it a little closer and had seen that you already had made the purchase - so I figured "eh, work with what you've got".
Note that JA (stereophile) measures at 1.25 meters and this seems to correlate well with an in-room response at an average listening distance of 3-4 meters. (..he also averages over a horizontal "window" at the listening position for the higher freq. response.)
Then I read it a little closer and had seen that you already had made the purchase - so I figured "eh, work with what you've got". Note that JA (stereophile) measures at 1.25 meters and this seems to correlate well with an in-room response at an average listening distance of 3-4 meters. (..he also averages over a horizontal "window" at the listening position for the higher freq. response.)
Magnetar,
A big driver high dynamics guy impressed by a bunch of 1"x3" TV speakers says alot. I wanted to try the ovals, but never found one that sounded good individually.
I don't envy all those holes. That's the only thing I really hate about building arrays. I'm sure you have a high-eff compression tweet around. You may want to try one in place of the center mid at first. It'll need an Lpad to adjust the level to match the array for listening distance, but it solves a number of other issues, not to mention less holes and wires.
Hint- Solder the wires to drivers before installation or you're looking at a back breaking nightmare. Also, the undiscussed secret to good vs great OB's is addressing edge diffraction, not baffle step but actual diffraction at the edges. It's far more important with OB's than boxes due to much greater velocity at the sides (think in terms of rushing to fill the rear wave rarefaction and vice versa). This is even more important with narrow fronts, since you're talking about higher frequencies bending around the baffle. I believe wings increase the effect because the travel distance is even greater (more velocity).
Unless your buddy got some magical 1x3's, yours should be even better. Have fun and keep us posted.
A big driver high dynamics guy impressed by a bunch of 1"x3" TV speakers says alot. I wanted to try the ovals, but never found one that sounded good individually.
I don't envy all those holes. That's the only thing I really hate about building arrays. I'm sure you have a high-eff compression tweet around. You may want to try one in place of the center mid at first. It'll need an Lpad to adjust the level to match the array for listening distance, but it solves a number of other issues, not to mention less holes and wires.
Hint- Solder the wires to drivers before installation or you're looking at a back breaking nightmare. Also, the undiscussed secret to good vs great OB's is addressing edge diffraction, not baffle step but actual diffraction at the edges. It's far more important with OB's than boxes due to much greater velocity at the sides (think in terms of rushing to fill the rear wave rarefaction and vice versa). This is even more important with narrow fronts, since you're talking about higher frequencies bending around the baffle. I believe wings increase the effect because the travel distance is even greater (more velocity).
Unless your buddy got some magical 1x3's, yours should be even better. Have fun and keep us posted.
ScottG said:
I actually started off writing just that (in my first post).
Note that JA (stereophile) measures at 1.25 meters and this seems to correlate well with an in-room response at an average listening distance of 3-4 meters. (..he also averages over a horizontal "window" at the listening position for the higher freq. response.)
I'm not so smart sometimes - It took several posts by you for me to absorb your point, You did good - I actually get it now!
I'll order the new woofers and send these back if they already shipped em. Shouldn't be a problem if I don't open the box - just shipping cost hopefully.
I normally measure at 1, 2 and 3 meters from 200 cycles up. And 0, 30 and 45 degrees off axis at 1 meter
Your advice is greatly appreciated!
johninCR said:
No doubt cutting that many oval holes was a nightmare! I just shook my head in awe when I saw these speaks and imagined the labor he had in them- they really were finished very well. He even took 6" PVC pipe, split it perfectly, painted it piano black and used it on the edge of the baffles. The speakeres themselves had mahogany veneer.
His website is here. No picture of the line arrays yet.
Older home theater project:
!!
You can contact him about what drivers he used at his website.
I have plenty if compression drivers but am trying to do something different - I'm bored, and like I said was impressed by his line enough to build some myself.
Thanks for the tips!
Foster/Neo line array
I have a line array with 32 of the neos per channel. You will need to cut their flanges if you want to avoid combing distortion starting around 9khz. I cut mine to the edge of the steel casing. The fixes combing to about 17320(13541/0.78 c-ti-c distance).
ADDITIONally, the fosters my be too big to avoid horizontal combing between the tweeters and the mid ranges.
Do you plan to add a sub woofer? You will need something to cover the low end.
ZarATHU
I have a line array with 32 of the neos per channel. You will need to cut their flanges if you want to avoid combing distortion starting around 9khz. I cut mine to the edge of the steel casing. The fixes combing to about 17320(13541/0.78 c-ti-c distance).
ADDITIONally, the fosters my be too big to avoid horizontal combing between the tweeters and the mid ranges.
Do you plan to add a sub woofer? You will need something to cover the low end.
ZarATHU
Re: Foster/Neo line array
Hello,
I have the drivers now and was contemplating triming the tweeters down - I was thinking it may not make an audible difference above 10K. Have you measured it both trimmed and un-trimmed and compared? How far away? What did you hear/measure?
Also where are you crossing them over at with what crossover?
Does it sound good?
What midwoofers are you using? Seems to me there will always be some horiz comb problems unless you mount the mid/tweeter within half a wavelength at the crossover point. How did you get around it? Did you measure it if you tried it further than a half wavelength?
What happened to the response of your midwoofers when you measured your array compared to one individual driver? Did the drivers have a rising response to begin with?
How wide is your baffle? Do you have baffle step rollof? How tall is your array?
I have several different subs to use. What are you using?
I have 32 of both of the 5.25 fosters and haven't sent the first 'woofers' back to madisound yet I was thinking maybe I'd extend the main (wideband foster) array's bass with another pair of the foster 'woofer' arrays - maybe with a little eq - IOW four lines for a stereo pair using 64 of the fosters. 32 bass and 32 wideband/tweeters. What do you think? I think it would be 'fast' and look cool.
Thanks for your reply.
Hello,
I have the drivers now and was contemplating triming the tweeters down - I was thinking it may not make an audible difference above 10K. Have you measured it both trimmed and un-trimmed and compared? How far away? What did you hear/measure?
Also where are you crossing them over at with what crossover?
Does it sound good?
What midwoofers are you using? Seems to me there will always be some horiz comb problems unless you mount the mid/tweeter within half a wavelength at the crossover point. How did you get around it? Did you measure it if you tried it further than a half wavelength?
What happened to the response of your midwoofers when you measured your array compared to one individual driver? Did the drivers have a rising response to begin with?
How wide is your baffle? Do you have baffle step rollof? How tall is your array?
I have several different subs to use. What are you using?
I have 32 of both of the 5.25 fosters and haven't sent the first 'woofers' back to madisound yet I was thinking maybe I'd extend the main (wideband foster) array's bass with another pair of the foster 'woofer' arrays - maybe with a little eq - IOW four lines for a stereo pair using 64 of the fosters. 32 bass and 32 wideband/tweeters. What do you think? I think it would be 'fast' and look cool.
Thanks for your reply.
Hello,
>I have the drivers now and was contemplating triming the tweeters down - I was >thinking it may not make an audible difference above 10K. Have you measured it both >trimmed and un-trimmed and compared? How far away? What did you hear/measure?
I followed Jim Griffin's paper exclusively and didn't take any chances since it would be a huge effort to recut a baffle that already had 17 circular holes in it. However I spoke with someone who did not cut them and was commenting that his tweeter sounded vage and warbly, which is eactly what you would expect to hear about combing in the 10-17 K range.
>Also where are you crossing them over at with what crossover?
Rane AC23, at about 2750hz(with 32 of them per channel)
>Does it sound good?
I like it.
>What midwoofers are you using?
>Seems to me there will always be some horiz comb >problems unless you mount the >mid/tweeter within half a wavelength at the crossover point.
>How did you get around it? Did you measure it if you tried it further than a half >wavelength?
When you use domes you can't use a mid woofer. You have to use a real midrange, even if its a widerange model. To avoid horizontal combing you can exceed about 4 inches in the mid and this is too small to be called a woofer. I used 3.5 inch Sammis which were a dynamite little buyout speaker for PE, but are no longer available.
>What happened to the response of your midwoofers when you measured your array >compared to one individual driver? Did the drivers have a rising response to begin with?
Youy can't make that comparison. When you get morfe than about 6 or seven of them in a line the mechaincs of lowered distortion and lower percentage of total amplitude take the combination spekaers way down into the linear range. Bumps and drops in the speakers disappear. A speaker with a typical kcik at 7K individually, has no suck kick in a group of 6. I have a group of 17 mids and 32 tweets.
>How wide is your baffle?
8 inches
>Do you have baffle step rolloff?
Be more specific in how you are using those terms
>How tall is your array?
90 inches
>I have several different subs to use. What are you using?
I'm not actually using a sub, its more of a woofer. It crosses at about 165, and goes down to about 25hz. I'm using two, one for each channel, and they are Goldsound 12 inch DVC models with a 15 mm Xmax.
>I have 32 of both of the 5.25 fosters and haven't sent the first 'woofers' back to >madisound yet I was thinking maybe I'd extend the main (wideband foster) array's bass >with another pair of the foster 'woofer' arrays - maybe with a little eq - IOW four lines >for a stereo pair using 64 of the fosters. 32 bass and 32 wideband/tweeters. What do >you think? I think it would be 'fast' and look cool.
I think they are too big if you are going to use the little neo's, but who knows, they might work fine. They are certainly a good price. If i was going to use them, though i would be forced to buy the B&G Neo ribbon's at 60 bucks a piece.
Zarathu
>I have the drivers now and was contemplating triming the tweeters down - I was >thinking it may not make an audible difference above 10K. Have you measured it both >trimmed and un-trimmed and compared? How far away? What did you hear/measure?
I followed Jim Griffin's paper exclusively and didn't take any chances since it would be a huge effort to recut a baffle that already had 17 circular holes in it. However I spoke with someone who did not cut them and was commenting that his tweeter sounded vage and warbly, which is eactly what you would expect to hear about combing in the 10-17 K range.
>Also where are you crossing them over at with what crossover?
Rane AC23, at about 2750hz(with 32 of them per channel)
>Does it sound good?
I like it.
>What midwoofers are you using?
>Seems to me there will always be some horiz comb >problems unless you mount the >mid/tweeter within half a wavelength at the crossover point.
>How did you get around it? Did you measure it if you tried it further than a half >wavelength?
When you use domes you can't use a mid woofer. You have to use a real midrange, even if its a widerange model. To avoid horizontal combing you can exceed about 4 inches in the mid and this is too small to be called a woofer. I used 3.5 inch Sammis which were a dynamite little buyout speaker for PE, but are no longer available.
>What happened to the response of your midwoofers when you measured your array >compared to one individual driver? Did the drivers have a rising response to begin with?
Youy can't make that comparison. When you get morfe than about 6 or seven of them in a line the mechaincs of lowered distortion and lower percentage of total amplitude take the combination spekaers way down into the linear range. Bumps and drops in the speakers disappear. A speaker with a typical kcik at 7K individually, has no suck kick in a group of 6. I have a group of 17 mids and 32 tweets.
>How wide is your baffle?
8 inches
>Do you have baffle step rolloff?
Be more specific in how you are using those terms
>How tall is your array?
90 inches
>I have several different subs to use. What are you using?
I'm not actually using a sub, its more of a woofer. It crosses at about 165, and goes down to about 25hz. I'm using two, one for each channel, and they are Goldsound 12 inch DVC models with a 15 mm Xmax.
>I have 32 of both of the 5.25 fosters and haven't sent the first 'woofers' back to >madisound yet I was thinking maybe I'd extend the main (wideband foster) array's bass >with another pair of the foster 'woofer' arrays - maybe with a little eq - IOW four lines >for a stereo pair using 64 of the fosters. 32 bass and 32 wideband/tweeters. What do >you think? I think it would be 'fast' and look cool.
I think they are too big if you are going to use the little neo's, but who knows, they might work fine. They are certainly a good price. If i was going to use them, though i would be forced to buy the B&G Neo ribbon's at 60 bucks a piece.
Zarathu
Thanks
>What happened to the response of your midwoofers when you measured your array >compared to one individual driver? Did the drivers have a rising response to begin with?
>>Youy can't make that comparison. When you get morfe than about 6 or seven of them in a line the mechaincs of lowered distortion and lower percentage of total amplitude take the combination spekaers way down into the linear range. Bumps and drops in the speakers disappear. A speaker with a typical kcik at 7K individually, has no suck kick in a group of 6. I have a group of 17 mids and 32 tweets.>>>
When I said comparison I meant measurment. Like measuring the frequency response of an individual driver and then measuring the response of several in a line and comparing them. This will show the net effect of the multiple drivers in the line. From what I can tell the interferance between the individual units causes an alteration of the response. What I was looking for was what effect your line had iIn the upper midrange and high range of the cone specificaly.
>Do you have baffle step rolloff?
>>Be more specific in how you are using those terms>>
In the real world a little 8' baffle will only support a linear response down to a specific frequency. Then the baffle becomes too narrow or small and you have a net loss of those frequencies. This causes a a non linear range that needs to corrected if you want an accurate midrange to low midrange and bass transition. The low frequencies 'roll off' because a little baffle won't support them like it does in the upper range- the baffle is just too small.
I was looking for the net effect of this in your line compared to using one of your 3" drivers on the same size baffle. It should have an effect starting around 1.5 k so.
Have you measured your lines?
Thanks again
>What happened to the response of your midwoofers when you measured your array >compared to one individual driver? Did the drivers have a rising response to begin with?
>>Youy can't make that comparison. When you get morfe than about 6 or seven of them in a line the mechaincs of lowered distortion and lower percentage of total amplitude take the combination spekaers way down into the linear range. Bumps and drops in the speakers disappear. A speaker with a typical kcik at 7K individually, has no suck kick in a group of 6. I have a group of 17 mids and 32 tweets.>>>
When I said comparison I meant measurment. Like measuring the frequency response of an individual driver and then measuring the response of several in a line and comparing them. This will show the net effect of the multiple drivers in the line. From what I can tell the interferance between the individual units causes an alteration of the response. What I was looking for was what effect your line had iIn the upper midrange and high range of the cone specificaly.
>Do you have baffle step rolloff?
>>Be more specific in how you are using those terms>>
In the real world a little 8' baffle will only support a linear response down to a specific frequency. Then the baffle becomes too narrow or small and you have a net loss of those frequencies. This causes a a non linear range that needs to corrected if you want an accurate midrange to low midrange and bass transition. The low frequencies 'roll off' because a little baffle won't support them like it does in the upper range- the baffle is just too small.
I was looking for the net effect of this in your line compared to using one of your 3" drivers on the same size baffle. It should have an effect starting around 1.5 k so.
Have you measured your lines?
Thanks again
>Have you measured your lines?
OK now I understand what you mean. That concept is too integrated in my whole system to be worried about all by itself.
Yes...but not outside or in an anechoic chamber. I think its impossible to do a real line array without doing frequency response measurements(or sound pressure measurements), since to get the tweeters, mids and woofers balanced to amplitude with the electronic crossovers, I have to be able to amplify or equalize one part of the array over others. Each part has a different level of sensitivity. The tweeters are up around 106, and the mids are around 95, and the woofers are around 92. Obviously in a passive design I would have to pad the tweeters and the mid ranges, but in an electronic design, i just boost the mids and woofers a little bit, and lower the volume on the tweeters a little bit.
However, i have to admit varying the boost on certain occasions to augment certain kinds of music. I do it with the equalizer rather that with the crossover, since that will modify the crossover points.
So my choice is that I either use a sound level meter or a freq. response graph; I have both, and for quick and dirty set up, the Sound level meter is easier. But I did it in the room that its in. If I moved the speakers, I would have to do it all over again.
I have no idea what the specific baffle roll-off might be since I didn't do the measurements just on the speakers but actually on the speakers, the speakers in a group, and the room together.
At the moment I'm not terribly sure where I put the original graphs, and my measurement system is not set up since I'm in the middle of renovating the room itself. Renovation: new drywall in some places, new dripped ceiling, wall treatments, and a separation partial wall to separate it from the wood-stove part of the room. The speakers are covered with plastic(which tends to screw up the frequency response ;-), and my laptop computer measurement system got moved upstairs out of the drywall dust!
I'm sorry I'm not as helpful as I might be, especially if I had the measurement setup ready to go. I hope I'm making sense, since I don't have an EE background.
Zarathu
OK now I understand what you mean. That concept is too integrated in my whole system to be worried about all by itself.
Yes...but not outside or in an anechoic chamber. I think its impossible to do a real line array without doing frequency response measurements(or sound pressure measurements), since to get the tweeters, mids and woofers balanced to amplitude with the electronic crossovers, I have to be able to amplify or equalize one part of the array over others. Each part has a different level of sensitivity. The tweeters are up around 106, and the mids are around 95, and the woofers are around 92. Obviously in a passive design I would have to pad the tweeters and the mid ranges, but in an electronic design, i just boost the mids and woofers a little bit, and lower the volume on the tweeters a little bit.
However, i have to admit varying the boost on certain occasions to augment certain kinds of music. I do it with the equalizer rather that with the crossover, since that will modify the crossover points.
So my choice is that I either use a sound level meter or a freq. response graph; I have both, and for quick and dirty set up, the Sound level meter is easier. But I did it in the room that its in. If I moved the speakers, I would have to do it all over again.
I have no idea what the specific baffle roll-off might be since I didn't do the measurements just on the speakers but actually on the speakers, the speakers in a group, and the room together.
At the moment I'm not terribly sure where I put the original graphs, and my measurement system is not set up since I'm in the middle of renovating the room itself. Renovation: new drywall in some places, new dripped ceiling, wall treatments, and a separation partial wall to separate it from the wood-stove part of the room. The speakers are covered with plastic(which tends to screw up the frequency response ;-), and my laptop computer measurement system got moved upstairs out of the drywall dust!
I'm sorry I'm not as helpful as I might be, especially if I had the measurement setup ready to go. I hope I'm making sense, since I don't have an EE background.
Zarathu
I think that your numbers may be wrong if I'm understanding the concept:
"Baffle diffraction loss is a function of wavelength. The wavelength (l) of a given frequency at room temperature and standard atmospheric pressure may be found by the following equation:
"The baffle diffraction step may be most accurately described as a loss of bass, and not a gain in treble. However, if we view a graph of baffle diffraction, the amplitude "rises" as we move from left to right across the frequency axis. As an approximation, the rise begins at the frequency whose wavelength is 1/8 the smallest dimension of the baffle. This dimension is typically the width of the loudspeaker since most are tall and narrow. Using the same 18" baffle as in the previous example, the response would begin to rise at [1/8 * (13560/18)], or 94 Hz. Also, the maximum amplitude is attained at a frequency whose wavelength is twice the smallest dimension of the baffle; in this case [2(13560/18)], or 1.5 kHz. The -3dB point has been empirically determined to be:
The f3 for our arbitrary baffle width of 18" would be 253 Hz."
http://www.t-linespeakers.org/tech/bafflestep/index.html
Again.....most of these issues are much bigger when one has to use passive crossovers to accomplish the 6 db loss due to the baffle---by having to pad the tweeter even more than just because of differences in sensitivity. With electronic crossovers(and amplifiers within the electronic crossovers), and separate amps for each tweeter/mid/woofer line, and a 2/3 octave constant Q equalizer, the baffle step rolloff issues become essentially non-issues to be dealt with part in parcel with the various frequency response of the speakers and to room as a whole.
In my opinion, they are only important if you are selling the speakers as a stand-alone item, with internal passive crossovers.
Zarathu
"Baffle diffraction loss is a function of wavelength. The wavelength (l) of a given frequency at room temperature and standard atmospheric pressure may be found by the following equation:
"The baffle diffraction step may be most accurately described as a loss of bass, and not a gain in treble. However, if we view a graph of baffle diffraction, the amplitude "rises" as we move from left to right across the frequency axis. As an approximation, the rise begins at the frequency whose wavelength is 1/8 the smallest dimension of the baffle. This dimension is typically the width of the loudspeaker since most are tall and narrow. Using the same 18" baffle as in the previous example, the response would begin to rise at [1/8 * (13560/18)], or 94 Hz. Also, the maximum amplitude is attained at a frequency whose wavelength is twice the smallest dimension of the baffle; in this case [2(13560/18)], or 1.5 kHz. The -3dB point has been empirically determined to be:
The f3 for our arbitrary baffle width of 18" would be 253 Hz."
http://www.t-linespeakers.org/tech/bafflestep/index.html
Again.....most of these issues are much bigger when one has to use passive crossovers to accomplish the 6 db loss due to the baffle---by having to pad the tweeter even more than just because of differences in sensitivity. With electronic crossovers(and amplifiers within the electronic crossovers), and separate amps for each tweeter/mid/woofer line, and a 2/3 octave constant Q equalizer, the baffle step rolloff issues become essentially non-issues to be dealt with part in parcel with the various frequency response of the speakers and to room as a whole.
In my opinion, they are only important if you are selling the speakers as a stand-alone item, with internal passive crossovers.
Zarathu
Thanks for your help
- I'm going to try and have a line of the Fosters built today - I will make these comparisons (single driver versus a line of same) and pass along the info in a post in this thread. I think it may be helpful for others considering build a line array.
It may help in determining what type of response one should seek in a driver for a line array. My guess, based on Scotts help earlier, is you would want a driver with a rising response starting pretty low to overcome some of the interferance of the drivers within the line, and not use the tradional 'flat response' driver most folks seem to use. The same thing can be said of a horn driver - a rising response is going to give you a wider net bandwith.
I'm also interested in the net response on the low end of the scale using a narrow baffle in a line array. With a single driver you loose sensitivity big time, with a line? I'll find out soon enough.
- I'm going to try and have a line of the Fosters built today - I will make these comparisons (single driver versus a line of same) and pass along the info in a post in this thread. I think it may be helpful for others considering build a line array.
It may help in determining what type of response one should seek in a driver for a line array. My guess, based on Scotts help earlier, is you would want a driver with a rising response starting pretty low to overcome some of the interferance of the drivers within the line, and not use the tradional 'flat response' driver most folks seem to use. The same thing can be said of a horn driver - a rising response is going to give you a wider net bandwith.
I'm also interested in the net response on the low end of the scale using a narrow baffle in a line array. With a single driver you loose sensitivity big time, with a line? I'll find out soon enough.
You could always start with higher sensitivity to begin with, and the Fosters are only 86.7 SPL to begin with. I'm not sure that I'd want to deliberately use speakers with rising FR's in the range. You could manipulate the bass rolloff by using a vented box and fooling around with the tuning and thebox size to give your sysem the appropriate boost where you want it.
But again, measurements of the baffle rolloff might be different for a baffle consisting of 17 of the midranges floor to ceiling. What about the addition of a waveguide on the midrange?
I take it that you are NOT using electronic crossovers and Tri-amping with equalization?
Zarathu
But again, measurements of the baffle rolloff might be different for a baffle consisting of 17 of the midranges floor to ceiling. What about the addition of a waveguide on the midrange?
I take it that you are NOT using electronic crossovers and Tri-amping with equalization?
Zarathu
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.