So I am continuing to mull options for my future listening/HT room.
How would I go about modelling something like a line array of four 2"ish full-range drivers (e.g. MA CHN40) mated to an 8" woofer, crossed over at say 300Hz.
I'm wondering if this kind of design, across 8 surround/overheads might
a) give me enough low frequency SPL to avoid subwoofers altogether, and
b) allow me to eliminate (or massively suppress) "sweet spots" by EQing between them and whatever the main LCR are
How would I go about modelling something like a line array of four 2"ish full-range drivers (e.g. MA CHN40) mated to an 8" woofer, crossed over at say 300Hz.
I'm wondering if this kind of design, across 8 surround/overheads might
a) give me enough low frequency SPL to avoid subwoofers altogether, and
b) allow me to eliminate (or massively suppress) "sweet spots" by EQing between them and whatever the main LCR are
You tell us. 😉 20*log(8) = ~ +18 dB to whatever one speaker's peak SPL is if close together, so probably less depending on room acoustics.
🙂
Putting the woofer SPL to one side then, the line arrays - are there any design principles for array cabinets? How would one then model them? I've had a poke around, and while there are certainly a few array build logs on here (not least Two Towers), I didn't find anything on initial design or modelling
Putting the woofer SPL to one side then, the line arrays - are there any design principles for array cabinets? How would one then model them? I've had a poke around, and while there are certainly a few array build logs on here (not least Two Towers), I didn't find anything on initial design or modelling
GM,
You are too generous for your calculations of system SPL values.
In my Near Field Line Array White Paper I used 10*log (number of drivers) to compute the efficiency gain for a line array. It is also modified by the sensitivity gain/loss which accounts for the connection of the various drivers. Thus the system sensitivity becomes:
System Sensitivity = SPL + Efficiency Gain + Sensitivity Gain/Loss
See page 17 of the NFLAWP for a sample calculation.
http://www.audioroundtable.com/misc/nflawp.pdf
Colinb4987,
A line array is a stack of the volume for the individual drivers in the array. For example a stack of 4 drivers is the volume for the 4 individual drivers. The NFLAWP discusses the spacing between the drivers and how many drivers you need.
Jim
You are too generous for your calculations of system SPL values.
In my Near Field Line Array White Paper I used 10*log (number of drivers) to compute the efficiency gain for a line array. It is also modified by the sensitivity gain/loss which accounts for the connection of the various drivers. Thus the system sensitivity becomes:
System Sensitivity = SPL + Efficiency Gain + Sensitivity Gain/Loss
See page 17 of the NFLAWP for a sample calculation.
http://www.audioroundtable.com/misc/nflawp.pdf
Colinb4987,
A line array is a stack of the volume for the individual drivers in the array. For example a stack of 4 drivers is the volume for the 4 individual drivers. The NFLAWP discusses the spacing between the drivers and how many drivers you need.
Jim
Hi Colin,
When I started my array build I only had done some crude modeling (in WinISD Pro) and a promise of success by following in other people's footsteps. I did have back-up plans available but luckily never needed them.
These days we have some awesome tools at our disposal. Just look here: https://www.diyaudio.com/community/threads/cloning-ids-25s.203356/
for a complete (starting) model of a 25 driver line array consisting of 25x Vifa TC9//fD18-08. (you can throw out the floor reflection model and use what VituixCAD models in it's room settings)
You could use it "as is" or strip it to serve your own purpose. VituixCAD is a powerful tool but may take a some time to get to know it. It can model your proposal from scratch, even up to making a basic prediction of what a certain size driver's directivity would be like, but the more accurate your input is, the more accurate the predictions. Ideally one would input a measurement of a mock-up driver in a cabinet with on and off axis data and o from there. If you don't have that data, the next best thing is using the Diffraction modeler inside VituixCAD with the real driver's on axis response to model its off-axis behavior. That is what was used in the linked model above. I can tell you, the results of that model were quite accurate when compared to my measurements in-room of my actual 25 driver array. I can highly recommend that piece of software as a valuable tool to predict what you're going to get.
I've used it extensively myself, at first to learn more about the array behavior in my room and after that to optimize it further by shading it.
Personally, I'd never use a short unshaded array, especially for HT, as it gives a narrow listening window in the vertical plane. Easy enough to model.
I do use my tall arrays for home theater together with a pair of 12" subs. No center channel so it's 'only' a 4.2 setup but quite satisfying 😉.
For it to work like that I use a lot of trickery, (like EQ and further DSP) to get the results to my liking.
When I started my array build I only had done some crude modeling (in WinISD Pro) and a promise of success by following in other people's footsteps. I did have back-up plans available but luckily never needed them.
These days we have some awesome tools at our disposal. Just look here: https://www.diyaudio.com/community/threads/cloning-ids-25s.203356/
for a complete (starting) model of a 25 driver line array consisting of 25x Vifa TC9//fD18-08. (you can throw out the floor reflection model and use what VituixCAD models in it's room settings)
You could use it "as is" or strip it to serve your own purpose. VituixCAD is a powerful tool but may take a some time to get to know it. It can model your proposal from scratch, even up to making a basic prediction of what a certain size driver's directivity would be like, but the more accurate your input is, the more accurate the predictions. Ideally one would input a measurement of a mock-up driver in a cabinet with on and off axis data and o from there. If you don't have that data, the next best thing is using the Diffraction modeler inside VituixCAD with the real driver's on axis response to model its off-axis behavior. That is what was used in the linked model above. I can tell you, the results of that model were quite accurate when compared to my measurements in-room of my actual 25 driver array. I can highly recommend that piece of software as a valuable tool to predict what you're going to get.
I've used it extensively myself, at first to learn more about the array behavior in my room and after that to optimize it further by shading it.
Personally, I'd never use a short unshaded array, especially for HT, as it gives a narrow listening window in the vertical plane. Easy enough to model.
I do use my tall arrays for home theater together with a pair of 12" subs. No center channel so it's 'only' a 4.2 setup but quite satisfying 😉.
For it to work like that I use a lot of trickery, (like EQ and further DSP) to get the results to my liking.
Thanks. I started playing with VituixCAD, but couldn't make head nor tail of it! Bits are buried where you don't expect them. I suspect trying to do it on a 12" screen didn't help - I'll try again when I'm at my desk/monitor!
I'm rapidly realising the limitations of short arrays on the vertical axis, so beginning to think again about my options. Most likely landing on one of
I.e. some form of WAW/FAST
Much more to learn and ponder!
I'm rapidly realising the limitations of short arrays on the vertical axis, so beginning to think again about my options. Most likely landing on one of
- single FR driver in sealed cabinet built into the wall, with front horn
- single FR driver in MLTL (smaller Woden Poplar) cabinet built into the wall
I.e. some form of WAW/FAST
Much more to learn and ponder!
Last edited:
I still love arrays, but only the tall ones 😀. They work well for movies and have the desired SPL capabilities.
If I hadn't done arrays, I would have used some type of Unity/Synergy speakers. I bet they are a blast in a Home Theater setup too..
If I hadn't done arrays, I would have used some type of Unity/Synergy speakers. I bet they are a blast in a Home Theater setup too..
5 drivers wired in a Bessel array lined up vertically..............no discernable combing or lobing with the response and directivity of a single driver.