Hmm, be interesting to see where that lands. I've not looked at passive crossovers at all in this situation but you're probably looking at a being 15-20dB down on the Supravox at its low end. I would guess that'd turn it into the equivalent of a 75 or 80dB efficient driver once it's stuck behind a passive crossover which renders it flat, maybe 85 or so if you're willing to accept the phase consequencies of using LC tanks to boost the response. In comparison the Neo3's "flat" to 700Hz after dipole conversion. I'm whistling in the dark here, but I wouldn't be surprised if you found you had to Lpad a Neo3 to get it line up with the Supravox.
I also wonder how good of a tube amp you could build for what the passive crossover parts would cost. Put the good SET on the Supravox and use the other on the tweeters. Off the top of my head my guess would be taking the Supravoxes to 200Hz would cost enough equalization power the system would be in reasonable power balance with a couple watts on the tweeters. I'm a sand guy, but I have the (possibly incorrect) impression fairly decent results can be obtained at those power levels at what's probably not too much of an incremental cost riser over designing and tuning passive crossovers.
I also wonder how good of a tube amp you could build for what the passive crossover parts would cost. Put the good SET on the Supravox and use the other on the tweeters. Off the top of my head my guess would be taking the Supravoxes to 200Hz would cost enough equalization power the system would be in reasonable power balance with a couple watts on the tweeters. I'm a sand guy, but I have the (possibly incorrect) impression fairly decent results can be obtained at those power levels at what's probably not too much of an incremental cost riser over designing and tuning passive crossovers.
The way to avoid being down on the Supravox at its low end, is as you know cross it before the dipole roll off cuts in too much:
If I’m using SL’s spreadsheet spl_max1 correctly, the excursion limited SPL of the Supravox at say 250 Hz, is 120 dB.
At 200 Hz it’s still 114
100 Hz - 96
50 Hz - 78
The Supravox has a healthy Xmax for a sensitive (96 dB) full range driver, of 7 mm
That’s assuming that on a baffle of 250 mm, ie barely wider than the driver, that that D = 240. Is that a good estimate of D?
Fequal (where the SPL of the dipole = the SPL of a monopole) is 293 Hz
Have I missed something?
The cost of this tube amp is an order of magnitude over passive or active crossover parts. No other tube amp of this quality is affordable to me!
(Other than saving a good amount of money on decent large inductors)
If I cross actively at say 200 Hz to a SS amp, would I be better off?
If I’m using SL’s spreadsheet spl_max1 correctly, the excursion limited SPL of the Supravox at say 250 Hz, is 120 dB.
At 200 Hz it’s still 114
100 Hz - 96
50 Hz - 78
The Supravox has a healthy Xmax for a sensitive (96 dB) full range driver, of 7 mm
That’s assuming that on a baffle of 250 mm, ie barely wider than the driver, that that D = 240. Is that a good estimate of D?
Fequal (where the SPL of the dipole = the SPL of a monopole) is 293 Hz
Have I missed something?
The cost of this tube amp is an order of magnitude over passive or active crossover parts. No other tube amp of this quality is affordable to me!
(Other than saving a good amount of money on decent large inductors)
If I cross actively at say 200 Hz to a SS amp, would I be better off?
I've tried a floor mirror simulation for the woofer baffle, including some correction for the wings:
The dipole peak would be around 200 Hz. It looks like you can cross over to the Supravox anywhere from 200-300 Hz, having a margin of one octave to the rapid SPL fall-off from 600 Hz upward.
Adding the floor bounce for the Supravox for 3 (pink) and 5 m (red) listening distance:
If you are listening at no farther than 3 m distance you can try to circumvent the floor bounce by taking the woofer up to 400 Hz.
Rudolf
The dipole peak would be around 200 Hz. It looks like you can cross over to the Supravox anywhere from 200-300 Hz, having a margin of one octave to the rapid SPL fall-off from 600 Hz upward.
Adding the floor bounce for the Supravox for 3 (pink) and 5 m (red) listening distance:
If you are listening at no farther than 3 m distance you can try to circumvent the floor bounce by taking the woofer up to 400 Hz.
Rudolf
Attachments
Rudolf
Thank you kindly for trying to model my design.
Yes the initial design is to have the 400 mm woofer near the bottom of a rectangular 500 * 800 mm baffle.
However it has the 240 mm Supravox on an extension of this. (I attach the drawing again).
I downloaded the Edge but it looks like it only models baffles which are rectangles - not surprising.
I do wonder if the extra “corners” in my shape would cause extra undesirable diffraction points. If I know up-front that would happen, I could implement it and surround the unusual shape of my Supravox baffle extension on top with eg acrylic baffle extensions.
Being transparent, the extensions could have whatever shape is the most desirable acoustically. Eg like (if I recall correctly) Nuuk’s dome/ oval profile; or more easily, a simple rectangle.
I’ve read some of the threads that CLS posted.
http://www.diyaudio.com/forums/multi-way/142015-my-open-baffle-dipole-beyma-tpl-150-a-36.html posts 357 - 364
Some things that were posted by Michael (mige0):
While the top end polar responses of a driver without a baffle are way better, the penalty for the nude speaker is roughly a 6 dB SPL loss.
For me, that rules that out.
Some measurements showed that a square baffle the width of the basket, improves smoothness above 1 kHz, however a severe peak arises at ~ 4 kHz
The peak could probably be notched.
If you want the widest bandwidth of off axis regularity, the less the baffle the better.
I do want wide bandwidth of off axis regularity.
The sum of the above suggests to me for the wide range Supravox, a modest size square baffle, which will need a notch filter; but to avoid loss of sensitivity at the low end - needs to be crossed to the woofer higher than otherwise.
The unusual shape of my ply Supravox baffle shown in my drawing, would be surrounded by acrylic extensions to provide a square baffle.
Should it then be separate from the woofer baffle by some distance?
And preferably a suspended driver.
One thing going against these reported experiences, is that Linkwitz said Electro-acoustic models
“ . . sharp interference nulls when D is a multiple of a wavelength. This behavior can be considerably smoothed by making the baffle f rectangular which gives a variation to the length of D”
And what might be done above the dipole peak frequency, where the response “falls apart”?
Thank you kindly for trying to model my design.
Yes the initial design is to have the 400 mm woofer near the bottom of a rectangular 500 * 800 mm baffle.
However it has the 240 mm Supravox on an extension of this. (I attach the drawing again).
I downloaded the Edge but it looks like it only models baffles which are rectangles - not surprising.
I do wonder if the extra “corners” in my shape would cause extra undesirable diffraction points. If I know up-front that would happen, I could implement it and surround the unusual shape of my Supravox baffle extension on top with eg acrylic baffle extensions.
Being transparent, the extensions could have whatever shape is the most desirable acoustically. Eg like (if I recall correctly) Nuuk’s dome/ oval profile; or more easily, a simple rectangle.
I’ve read some of the threads that CLS posted.
http://www.diyaudio.com/forums/multi-way/142015-my-open-baffle-dipole-beyma-tpl-150-a-36.html posts 357 - 364
Some things that were posted by Michael (mige0):
While the top end polar responses of a driver without a baffle are way better, the penalty for the nude speaker is roughly a 6 dB SPL loss.
For me, that rules that out.
Some measurements showed that a square baffle the width of the basket, improves smoothness above 1 kHz, however a severe peak arises at ~ 4 kHz
The peak could probably be notched.
If you want the widest bandwidth of off axis regularity, the less the baffle the better.
I do want wide bandwidth of off axis regularity.
The sum of the above suggests to me for the wide range Supravox, a modest size square baffle, which will need a notch filter; but to avoid loss of sensitivity at the low end - needs to be crossed to the woofer higher than otherwise.
The unusual shape of my ply Supravox baffle shown in my drawing, would be surrounded by acrylic extensions to provide a square baffle.
Should it then be separate from the woofer baffle by some distance?
And preferably a suspended driver.
One thing going against these reported experiences, is that Linkwitz said Electro-acoustic models
“ . . sharp interference nulls when D is a multiple of a wavelength. This behavior can be considerably smoothed by making the baffle f rectangular which gives a variation to the length of D”
And what might be done above the dipole peak frequency, where the response “falls apart”?
“If I cross actively at say 200 Hz to a SS amp, would I be better off?”
Thinking it over and reading Linkwitz, I now think the answer is a definite yes.
My pendulum of thought had swung from (a few years ago) all active is best, to believing that one amp to give the same sonic signature, from about 100 Hz up.
But the bottom end, esp on a dipole, is calling for EQ and more watts . .
Thinking it over and reading Linkwitz, I now think the answer is a definite yes.
My pendulum of thought had swung from (a few years ago) all active is best, to believing that one amp to give the same sonic signature, from about 100 Hz up.
But the bottom end, esp on a dipole, is calling for EQ and more watts . .
It does flat baffles, but it can handle non-rectangular shapes. There's a place up top where you enter the number of corners/edges in the baffle shape. Also, in the graph, you can drag the corners around. I've modeled a baffle similar to yours - wider near the bottom, then narrowing in stages as it gets higher.
It won't directly do wings, so you need to 'open them up', i.e. add the effective additional width to the flat baffle.
Attachments
See that -6dB/octave slope below the dipole peak in Rudolph's plots? I'm assuming you'll want to EQ that flat. As I mentioned earlier, the Supravox will peak around 1k with baffle and probably around 1.3k without. 125-200Hz is two and some to three and some octaves below the peak; hence 15-20dB of equalization. If you're concerned about preserving watts for peak SPL that's a fair amount of power to burn off in the crossover.
As a starting point, don't worry about it. Real dipoles aren't point sources. So there aren't nulls to zero. I'm not working with drivers as good as the Supravox and TD15M but you can find some of my results here. None of the parametric equalization bands are set to more than 5dB or are of notch type Q. Except for one I'm using to squelch a cone breakup at 3.1kHz. 😛
This remark'll probably bring someone out of the woodwork, but I don't know anyone who feels passive crossovers produce better sound quality than active. There are folks who assert that with an order of magnitude or two more effort you can as good results with passive as with active analog. For an open baffle they might be right, depending on how significant the back wave reflections from the spider are. (But, since you're out to build the best system you can, I'd strongly encourage you to try out linear phase crossover and equalization.)
Personally I'd be cautious of padding down the Supravox with ~10x its impedance at the dipole peak as the amp'll mostly be controlling resistors and inductors instead of the driver. To the extent the system's linear it doesn't matter but, roughly speaking, the amp's ability to supppress nonlinearities is reduced by the amount of attenuation in the EQ. That's just a ballpark figure; there are a couple different ways to approach modeling the system and the source impdeance of the tube amp is part of it. But I would expect some loss of definition as a result. Whether or not it's enough to matter depends on your listening preferences and how good the amp and drivers are. Comparative THD sweeps with active versus passive cross would be a good place to start on the measurement side. Be interesting data to see.
Hopefully if John K or someone else with relevant experience is reading this thread they'll chime in and let us know if my guesstimates are in the right direction or not. All the dipole builds I'm recalling on DIY with detailed measurements are fully active.
Last edited:
This thread has opened up a *lot more than to consider than I’d imagined
The system mix of baffle size, baffle aesthetics, drivers, SPL, system cost, amp sound, coherence, smoothness, crossover do-ability. .
Thanks Saurav, I’ll have to do that at the first good opportunity
(Though right now they’re replaying the last time my team in the local football game won the annual Grand Final, a tightly contested game, but one in which in the space of 10 mins, my team kicked 5! goals .. )
“And what might be done above the dipole peak frequency, where the response falls apart?”
Latest thoughts: however extended and good the Supravox is, above the woofer if the XOs are passive, rather than going up to a super tweeter, crossing to a good tweeter - on a * smaller baffle crossed at more like – now seems a much smoother response is achievable.
This project might end up in stages or iterations . .
Hi TW20,
I’ve spent a lot of time researching drivers, now I realize not enough on crossovers - can you suggest a good link to a webpage on linear phase XO & EQ?
I’m not looking for *peak SPL, but capability of at least 105 dB uncompressed and “flat”.
Whatever combination of EQ and amp output “best” achieves that.
To that goal with minimal and simpler EQ, maybe:
- a tweeter on a small baffle - crossed much lower than a super tweeter
- active filters are easier
So now considering crossing to say the modestly priced Peerless HDS tweeter 810921.
Specs http://www.tymphany.com/datasheet/printview.php?id=170
An impressive comparison against very good, far more expensive tweeters:
http://www.customanalogue.com/elsinore/elsinore_15.htm
It’s 3 db less sensitive than the Supravox at 93 dB, but with higher power handling, as an active tweeter driven by eg a class A amp, which IME have excellent treble (it also would not limit the Supravox in SPL).
Coherence would be lost, but smoothness gained .
The system mix of baffle size, baffle aesthetics, drivers, SPL, system cost, amp sound, coherence, smoothness, crossover do-ability. .
Thanks Saurav, I’ll have to do that at the first good opportunity
(Though right now they’re replaying the last time my team in the local football game won the annual Grand Final, a tightly contested game, but one in which in the space of 10 mins, my team kicked 5! goals .. )
“And what might be done above the dipole peak frequency, where the response falls apart?”
Latest thoughts: however extended and good the Supravox is, above the woofer if the XOs are passive, rather than going up to a super tweeter, crossing to a good tweeter - on a * smaller baffle crossed at more like – now seems a much smoother response is achievable.
This project might end up in stages or iterations . .
Hi TW20,
I’ve spent a lot of time researching drivers, now I realize not enough on crossovers - can you suggest a good link to a webpage on linear phase XO & EQ?
I’m not looking for *peak SPL, but capability of at least 105 dB uncompressed and “flat”.
Whatever combination of EQ and amp output “best” achieves that.
To that goal with minimal and simpler EQ, maybe:
- a tweeter on a small baffle - crossed much lower than a super tweeter
- active filters are easier
So now considering crossing to say the modestly priced Peerless HDS tweeter 810921.
Specs http://www.tymphany.com/datasheet/printview.php?id=170
An impressive comparison against very good, far more expensive tweeters:
http://www.customanalogue.com/elsinore/elsinore_15.htm
It’s 3 db less sensitive than the Supravox at 93 dB, but with higher power handling, as an active tweeter driven by eg a class A amp, which IME have excellent treble (it also would not limit the Supravox in SPL).
Coherence would be lost, but smoothness gained .
Last edited:
a good tweeter on a *small baffle, crossed between say 3 - 5 kHz:
This approach would not make use of an octave or two of the Supravox's very smooth top end - albeit on a conventional baffle,
but with an 8 incher's directionality in the treble etc not its key strength; give smoother response with less EQ, and have much of the treble played by an excellent (giant killing?) dedicated tweeter.
The reason I got the Supravox wasn't because it was full range/ I wanted to extract maximum bandwidth from it, but is sufficiently high sensitivity for 16 watts without dipole roll-off, good Qts for OB, was meant to be great if not asked to excurse too much (and tunable), and was flexible in XO points.
This approach would not make use of an octave or two of the Supravox's very smooth top end - albeit on a conventional baffle,
but with an 8 incher's directionality in the treble etc not its key strength; give smoother response with less EQ, and have much of the treble played by an excellent (giant killing?) dedicated tweeter.
The reason I got the Supravox wasn't because it was full range/ I wanted to extract maximum bandwidth from it, but is sufficiently high sensitivity for 16 watts without dipole roll-off, good Qts for OB, was meant to be great if not asked to excurse too much (and tunable), and was flexible in XO points.
Maybe consider using something like a DCX2496 at least in the initial stages. I'm pretty sure the first iteration you build won't be your last 🙂 And for me, the DCX has been quite handy in trying out different XO frequencies / slopes and seeing how they sound. Sometimes things behave in non-obvious ways, and I end up preferring a configuration that I didn't expect, which then sends me on an investigation into the possible causes behind that.
I have a 97dB midrange, used to have a horn tweeter with a passive XO between them and a DIY 2A3 SET driving those 2. I've gone to a 'nude' Neo3, the DCX, and solid state for all 6 channels. I miss the sound of the 2A3 sometimes, but my crossovers are now much better aligned than anything I had before, so overall the sound is better integrated.
I have a 97dB midrange, used to have a horn tweeter with a passive XO between them and a DIY 2A3 SET driving those 2. I've gone to a 'nude' Neo3, the DCX, and solid state for all 6 channels. I miss the sound of the 2A3 sometimes, but my crossovers are now much better aligned than anything I had before, so overall the sound is better integrated.
I haven't written it yet. 😀 In the meantime, see my previous suggestion to look at cuibono's Violet DSP thread. John K also has some discussion on his website.
So define your target power spectral density for 105dB (flat below 20Hz and 1/f above is a reasonable default) and do the integration over whatever baffles, drivers, and crossovers are of interest. I suspect you'll find you need active EQ on the Supravox.
If you want easy crossover and EQ setup digital's hard to beat. The obvious quick fix in your case would be to use the DCX up as a two way between the TD15Ms and Supravox+tweeter. Do all EQ in the DCX and implement only the crossover between the Supravox and tweeter in passive. Doesn't provide the ability to equalize the Supravox and tweeter independently---which is handy but unlikely to make or break the design---but it's the best option I can see within the current constraints. You can always mod the DCX or swap in a better digital system later.
If you want good directivity avoid a tweeter baffle. Some sims in Edge will show you why and help you analyze the dipole peak.
Saurav
“Maybe consider using something like a DCX2496 at least in the initial stages. I'm pretty sure the first iteration you build won't be your last
And for me, the DCX has been quite handy in trying out different XO frequencies / slopes and seeing how they sound”
I hadn’t even thought of it a couple of day ago, but now think that way too.
I have a 97dB midrange, used to have a horn tweeter with a passive XO between them and a DIY 2A3 SET driving those 2.
An earlier idea of mine was a 45 SE driving BMS coaxial compression drivers on JBL 500 Hz horns, crossing actively straight to the Lambdas driven by the 845 SE.
“I've gone to a 'nude' Neo3, the DCX, and solid state for all 6 channels”
Nude Neo3 with SS. I tend to think you need to be more careful with an amp’s resolution with good ribbons: What amps do you and others find work well with them?
“Maybe consider using something like a DCX2496 at least in the initial stages. I'm pretty sure the first iteration you build won't be your last
And for me, the DCX has been quite handy in trying out different XO frequencies / slopes and seeing how they sound”
I hadn’t even thought of it a couple of day ago, but now think that way too.
I have a 97dB midrange, used to have a horn tweeter with a passive XO between them and a DIY 2A3 SET driving those 2.
An earlier idea of mine was a 45 SE driving BMS coaxial compression drivers on JBL 500 Hz horns, crossing actively straight to the Lambdas driven by the 845 SE.
“I've gone to a 'nude' Neo3, the DCX, and solid state for all 6 channels”
Nude Neo3 with SS. I tend to think you need to be more careful with an amp’s resolution with good ribbons: What amps do you and others find work well with them?
Last edited:
TW,
I will read all the links, including cuibono's Violet DSP thread and JohnK’s discussion Stored Energy and Time Domain Distortion. Hope I can understand it sufficiently.
Your approach sounds very good. I need to take time off work and social life to do this well.
Has anyone used JohnK’s ABC Dipole?
I will read all the links, including cuibono's Violet DSP thread and JohnK’s discussion Stored Energy and Time Domain Distortion. Hope I can understand it sufficiently.
Your approach sounds very good. I need to take time off work and social life to do this well.
Has anyone used JohnK’s ABC Dipole?
The basics are pretty simple. It's the implementation details that get complicated. The most straightforward way I've found to think about it is the 360 degree phase rotation of a warped phase LR4 crossover causes the sound at the crossover point to be the opposite of what it should be because of the 180 degree phase rotation at that frequency. A sharp edged transient like a drumbeat has an approximately constant power spectral density (flat SPL in the frequency domain, if you will) but once you run it through the LR4 crossover the energy around the crossover frequency acts to reduce the transient's height rather than increasing it as it should. If you can see how that would reduce the drumbeat's attack then you've got the core concept behind why linear phase produces a cleaner, more detailed sound.
It's not for everyone---some listeners prefer warped phase---but I'll let cuibono's reaction and StigErik's crossover choices stand for themselves.
A solid LME49811 build will run you about USD 125 in parts per channel and beat most amps on the market. See any of the several 49811 build threads in the chipamp forum; I'd recommend tomchr's Yet Another LME49811+Sanken STD03 thread and looking at Panson Audio's kits. If you want something off the shelf I'd suggest Parasound Halo A23s. They've been discovered so used ones are rather more expensive than they once were but I've put 'em toe to toe with various amps costing six to ten times as much and the A23s have consistently sounded as good or better. They're nothing fancy, just a solid cap multiplier 2SC3519/2SA1386 discrete design from John Curl without the excessive markup so common in audio.
Full disclosure: I stand to make a tidy profit by replacing my pair of A23s with two four channel 49811+STD03 based amps and selling the A23s. If I wasn't enjoying replying on this thread I'd probably have started laying out the boards by now. That copy of SoundEasy will just about pay for the transformers, heat sinks, and the sheet metal and bar stock for the enclosures. So thanks. 😛
I went fairly inexpensive to try out the concept, so for now the midrange and tweeters are off an Adcom GFA-2535 (basically a dual 535), and the woofers off a Behringer A500. I'm pretty sure my next 2 changes will be trying a non-PDR Neo3 (I picked the PDR based on sims, but since then a few others have measured the non-PDR, and I think that may be better suited to my goals), and getting better amplification. I don't really know much about SS amplification. Emotiva is one of the brands I'm considering.
Yes. It's more than worth the money, even if you do nothing but read the help documents. I don't play with it too much these days, since I'm mostly playing with XO changes and not baffle changes, but in the initial design stages, I relied on that and EDGE a lot.
Last edited:
There is no Supravox driver in that "two drivers" simulation. Maybe you need to better understand the concept of "mirror sources" WRT calculating floor influences. 😉
If you give a sufficient number in the "# of corners" field in EDGE, the programm will automatically form those corners into a circle or ellipse - depending on the values for x and y in the "Baffle size" entry. You only need to push the "Apply!" button
I used the lower four "corners" of the circle to make them into the attached rectangular baffle:
Attachments
“the 360 degree phase rotation of a warped phase LR4 crossover causes the sound at the crossover point to be the opposite of what it should be because of the 180 degree phase rotation at that frequency. A sharp edged transient like a drumbeat has an approximately constant power spectral density (flat SPL in the frequency domain, if you will) but once you run it through the LR4 crossover the energy around the crossover frequency acts to reduce the transient's height rather than increasing it as it should. If you can see how that would reduce the drumbeat's attack then you've got the core concept behind why linear phase produces a cleaner, more detailed sound.”
Yes, at least the concept easily makes sense.
I’ll think about LME49811 based amps. What have you compared them to?
Thanks Saurav
I'll have to spend some time with ABC Dipole
Yes, at least the concept easily makes sense.
I’ll think about LME49811 based amps. What have you compared them to?
Thanks Saurav
I'll have to spend some time with ABC Dipole
Last edited:
I'll refresh myself on "mirror sources"
In EDGE, I've created the baffle with simmed wings as suggested.
in post #23 Rudolf you said
“It looks like you can cross over to the Supravox anywhere from 200-300 Hz, having a margin of one octave to the rapid SPL fall-off from 600 Hz upward”
When I work out how to position and size them differently from the 1st driver, without changing the size of the 1st driver, I'll do the 2nd and 3rd drivers (ie Supravox and tweeter).
In EDGE, I've created the baffle with simmed wings as suggested.
in post #23 Rudolf you said
“It looks like you can cross over to the Supravox anywhere from 200-300 Hz, having a margin of one octave to the rapid SPL fall-off from 600 Hz upward”
When I work out how to position and size them differently from the 1st driver, without changing the size of the 1st driver, I'll do the 2nd and 3rd drivers (ie Supravox and tweeter).
If you check the chip amp and solid state forums you'll find plenty of measurements of lots of different amps. Douglas Self and Rod Elliott have both published a fair amount of data from their amps and Panson's 49811 kits are well speced. Also, pretty much any decent chipamp's datasheet has a fair amount of data. I haven't looked, but I'd imagine one could find the same for tubes. I've also listened to dozens of commercial amps and a few DIYs over the years.
Unfortunately most amp measurements don't go down to typical tweeter listening levels of 10-100uW but for BJTs in class AB the measurements I do have support Douglas Self's rule of thumb crossover distortion increases by a factor of three for every order of magnitude reduction in power. So one can likely project from available THD as well as SnR data with a fair amount of accuracy. Main things I look at are Avol, GBP, SnR, slew rate, and THD (IMD if it's available). You may also want to consider the LME49830, though I'm not aware of any thermal track MOSFETs.
I believe Edge only simulates one driver size at a time.
'Mirror source' - if you driver is close to the ground, the floor reflection reinforces the main drier's output (or, it plays in half-space, depending on how you want to look at it). I think that's what Rudolf was trying to model with the 2 drivers.
You probably need to model it one driver at a time, and for the woofer, include the floor reflection in the model.
You probably need to model it one driver at a time, and for the woofer, include the floor reflection in the model.
- Status
- Not open for further replies.