Well, I certainly seems as if I have started a war, or so it would seem, but I do thank everyone who had useful input and weren't arguing over what qualifications they did or didn't possess 
Just for your information, when I said I planned to cross the .5 woofer over at 200 Hz, that was merely a ballpark figure to better explain the project. I have since learnt that it would be best to cross this woofer over at the baffle step frequency, and I plan on making the cabinets approximately 20 cm wide or so, and I think this would put the baffle step freq. around the 550 Hz mark (I am not absolutely sure on this, someone please correct me if I am wrong), and this should 'close the gap' a bit more between the two drivers. I may even just run the two woofers in parallel if need be, in which case all the arguing would be for nothing other than some amusing reading.
I can appreciate what Alvis is saying, but for the sake of 500 Hz difference between the drivers, in the upper range, I really don't think it will make much of a perceivable difference.
And to those of you wondering why I'm using a dome mid between the woofer and tweeter, the woofers have a small break-up resonance between 1 and 2 K, and then drop very sharply. To go straight to a tweeter would mean crossing over at 2K which is rather low, whereas using the dome mid and ribbon tweeter, going by the manufacturers recommended crossover points, all the drivers match nearly perfectly so should integrate well.
Crossover design will be fun, but I have have a few different computer programs to aid with design such as Xover pro (not the most advanced program but pretty handy).
And to conclude, a few pages back someone asked 'why'. Well, why not? I want to try something new, that will be more exciting and have more to think about than a two way bookshelf speaker. It may not be everyone's idea of an ideal system, but once its all tweaked and adjustments are made etc, I think it could work very well. I am just waiting for the mid drivers to arrive and the wood to be delivered, and then I will make a new thread for the actual build of the speakers themselves.
Just for your information, when I said I planned to cross the .5 woofer over at 200 Hz, that was merely a ballpark figure to better explain the project. I have since learnt that it would be best to cross this woofer over at the baffle step frequency, and I plan on making the cabinets approximately 20 cm wide or so, and I think this would put the baffle step freq. around the 550 Hz mark (I am not absolutely sure on this, someone please correct me if I am wrong), and this should 'close the gap' a bit more between the two drivers. I may even just run the two woofers in parallel if need be, in which case all the arguing would be for nothing other than some amusing reading.
I can appreciate what Alvis is saying, but for the sake of 500 Hz difference between the drivers, in the upper range, I really don't think it will make much of a perceivable difference.
And to those of you wondering why I'm using a dome mid between the woofer and tweeter, the woofers have a small break-up resonance between 1 and 2 K, and then drop very sharply. To go straight to a tweeter would mean crossing over at 2K which is rather low, whereas using the dome mid and ribbon tweeter, going by the manufacturers recommended crossover points, all the drivers match nearly perfectly so should integrate well.
Crossover design will be fun, but I have have a few different computer programs to aid with design such as Xover pro (not the most advanced program but pretty handy).
And to conclude, a few pages back someone asked 'why'. Well, why not? I want to try something new, that will be more exciting and have more to think about than a two way bookshelf speaker. It may not be everyone's idea of an ideal system, but once its all tweaked and adjustments are made etc, I think it could work very well. I am just waiting for the mid drivers to arrive and the wood to be delivered, and then I will make a new thread for the actual build of the speakers themselves.
Right, can you sum this up for me.. because its interesting, but I am a bit confused.
2 woofers in one box. Each woofer crossed to the mid/tweeter at differing frequencys...
If a sine wave was sent to this completed speaker at a frequency that only the bassmid was playing, lets say 500hz.
At the same time both drivers are reproducing a bassline, for example
Would not the bass/mid dutied driver, still vibrate the bass dutied woofer. at the 500hz, example, frequency ?
Or if just a series of notes of melody that where within the 400-800hz range, where being played, for example, at the start of a piece of piano music. Would the bass/mid driver vibrations, vibrate the inactive bass cone?
Thanks. Just curious, as usual
2 woofers in one box. Each woofer crossed to the mid/tweeter at differing frequencys...
If a sine wave was sent to this completed speaker at a frequency that only the bassmid was playing, lets say 500hz.
At the same time both drivers are reproducing a bassline, for example
Would not the bass/mid dutied driver, still vibrate the bass dutied woofer. at the 500hz, example, frequency ?
Or if just a series of notes of melody that where within the 400-800hz range, where being played, for example, at the start of a piece of piano music. Would the bass/mid driver vibrations, vibrate the inactive bass cone?
Thanks. Just curious, as usual
Last edited:
Right, can you sum this up for me.. because its interesting, but I am a bit confused.
2 woofers in one box. Each woofer crossed to the mid/tweeter at differing frequencys...
If a sine wave was sent to this completed speaker at a frequency that only the bassmid was playing, lets say 500hz.
At the same time both drivers are reproducing a bassline, for example
Would not the bass/mid dutied driver, still vibrate the bass dutied woofer. at the 500hz, example, frequency ?
Or if just a series of notes of melody that where within the 400-800hz range, where being played, for example, at the start of a piece of piano music. Would the bass/mid driver vibrations, vibrate the inactive bass cone?
Thanks. Just curious, as usual
Hi,
Basically there is no real effect because the acoustic coupling between the
drivers is very low at midrange frequencies compared to bass frequencies.
Some midrange will bleed through the bass only driver at insignificant levels,
compared to the level it reproduces anyway due to the low slope roll-off.
rgds, sreten.
Mixed signals...
BOTH drivers are active and remain active throughout the range. With a lower frequency filter on one of the drivers the PROPORTION is altered, and the transition is GRADUAL not abrupt.
Let's look at an example with 2 identical mid-woofers: MW#1 running full range up to 3KHz, and we opt for a first-order 200Hz low-pass for MW#2
Here's what happens without accounting for internal coupling or driver spacing effects (we'll address that after this example).
At 200 Hz, the drivers are substantially in phase. Totally in phase would give a + 6db boost. But electrically, MW2 is at 0.707 (3dB down) with a phase difference of 45 degrees relative to MW1.
Because they sum vectorially, the combined amplitude is 1.847 equating to +5.33dB over the output of MW1 alone at 200Hz.
At 400 Hz, MW2 is now at 0.448 (-7dB) with a phase angle of 63 deg. Vectorially this will sum with MW1 for a combined amplitude of 1.265 or +2dB over MW1 alone.
At 800 Hz, MW2 is 0.242 (that's -12dB) and 76 degrees out of phase w. MW1 The vectorially summed response is now 1.086 for net increase of +0.7dB over MW1 alone.
Note:At 100Hz it's nearly +6dB with very little phase shift.
So as can be seen it's GRADUAL change, not abrupt. The drivers (and the internal pressure waves!!) remain substantially in-phase throughout the operating range. There is virtually no "cancellation" from "out of phase" internal pressure waves.
The comb-filtering from having two drivers covering the same range COULD be a problem if they are spaced far apart, but if they are close (say 35cm on center, using 17cm drivers, that's 7" in the US), this problem is relatively minor. At 800 Hz the wavelength is still over 43 cm, and by then the response of MW2 is 12 dB down - resulting in relatively minor comb-filtering.
OK so what about box "lift"and Baffle Step Compensation (BSC)?
If the box is about 9" wide, (and with an ideal "egg-shaped" box without peaks and valleys) box lift is about +6dB at 500Hz and above, +4.7dB at 400Hz and about +2dB at 200Hz, and +0.5dB at 100Hz, the combined output is +7.33 dB at 200Hz, +6.7dB at 400 Hz and +6.7dB at 800Hz.
When "normalized" to the midrange at 1KHz (which is lifted +6dB by the box), these responses would then be + 0.7dB at 100Hz, +1.33dB at 200 Hz, +0.7 dB at 400Hz, +0.7dB at 800 Hz
So, the 200 Hz roll-off on MW2 would yield almost completely flat on-axis when measured anechoically.
On top of this, the small phase shift at low frequencies operates to lower the in-box resonant frequency of the drivers, extending the 3dB down point by 2 or 3 Hz when using a sealed box.
But we don't listen in anechoic environments, we have room gain, and the box lift effect disappears off axis - so the net in room acoustic response may get a little bass-heavy by maybe +3 to +6dB. Many would view this as a problem, but because most recordings are mastered at very high SPL and played back at lower levels in the home, some of this excess gets scrubbed off due to the "Fletcher-Munson" effect... it becomes a bit of built-in loudness compensation and will pleasing in most rooms.
Now as for the internal pressure waves from MW2 "modulating" the midrange on MW1, as the examples above show, the relative phase difference between drivers is almost nil in the bass transitioning to only 76 degrees at 800Hz - 76 degrees is not even yet in quadrature, and so is still additive ...... and remember, by then the output of MW2 is down by 12 dB... so MW2 not going to add too much to the mids and upper mids.
And what it does add is still close enough in phase so it will not muddy or color the sound in a very noticeable way.
Yes it's true, there will be some MINOR time-smear and combing effects , especially if the drivers are spaced far apart, and yes it is audible as compared to the single driver - but barely so - and for most listeners, I suspect we'd gladly have that extra bass extension, extra efficiency and balance.
When crossing over to the tweeter, we are then mostly concerned with the mid and upper mid response of MW1 alone, without any need for BSC as the bass levels remain balanced.
As for the "evanescent fluids coupling inside the box" theory, well, maybe, but it's certainly not a dominant or even very noticeable effect!
Sreten's response is right on target, but short. I'll elaborate a little.Right, can you sum this up for me.. because its interesting, but I am a bit confused.
2 woofers in one box. Each woofer crossed to the mid/tweeter at differing frequencys...
If a sine wave was sent to this completed speaker at a frequency that only the bassmid was playing, lets say 500hz.
At the same time both drivers are reproducing a bassline, for example
Would not the bass/mid dutied driver, still vibrate the bass dutied woofer. at the 500hz, example, frequency ?
Or if just a series of notes of melody that where within the 400-800hz range, where being played, for example, at the start of a piece of piano music. Would the bass/mid driver vibrations, vibrate the inactive bass cone?
Thanks. Just curious, as usual
BOTH drivers are active and remain active throughout the range. With a lower frequency filter on one of the drivers the PROPORTION is altered, and the transition is GRADUAL not abrupt.
Let's look at an example with 2 identical mid-woofers: MW#1 running full range up to 3KHz, and we opt for a first-order 200Hz low-pass for MW#2
Here's what happens without accounting for internal coupling or driver spacing effects (we'll address that after this example).
At 200 Hz, the drivers are substantially in phase. Totally in phase would give a + 6db boost. But electrically, MW2 is at 0.707 (3dB down) with a phase difference of 45 degrees relative to MW1.
Because they sum vectorially, the combined amplitude is 1.847 equating to +5.33dB over the output of MW1 alone at 200Hz.
At 400 Hz, MW2 is now at 0.448 (-7dB) with a phase angle of 63 deg. Vectorially this will sum with MW1 for a combined amplitude of 1.265 or +2dB over MW1 alone.
At 800 Hz, MW2 is 0.242 (that's -12dB) and 76 degrees out of phase w. MW1 The vectorially summed response is now 1.086 for net increase of +0.7dB over MW1 alone.
Note:At 100Hz it's nearly +6dB with very little phase shift.
So as can be seen it's GRADUAL change, not abrupt. The drivers (and the internal pressure waves!!) remain substantially in-phase throughout the operating range. There is virtually no "cancellation" from "out of phase" internal pressure waves.
The comb-filtering from having two drivers covering the same range COULD be a problem if they are spaced far apart, but if they are close (say 35cm on center, using 17cm drivers, that's 7" in the US), this problem is relatively minor. At 800 Hz the wavelength is still over 43 cm, and by then the response of MW2 is 12 dB down - resulting in relatively minor comb-filtering.
OK so what about box "lift"and Baffle Step Compensation (BSC)?
If the box is about 9" wide, (and with an ideal "egg-shaped" box without peaks and valleys) box lift is about +6dB at 500Hz and above, +4.7dB at 400Hz and about +2dB at 200Hz, and +0.5dB at 100Hz, the combined output is +7.33 dB at 200Hz, +6.7dB at 400 Hz and +6.7dB at 800Hz.
When "normalized" to the midrange at 1KHz (which is lifted +6dB by the box), these responses would then be + 0.7dB at 100Hz, +1.33dB at 200 Hz, +0.7 dB at 400Hz, +0.7dB at 800 Hz
So, the 200 Hz roll-off on MW2 would yield almost completely flat on-axis when measured anechoically.
On top of this, the small phase shift at low frequencies operates to lower the in-box resonant frequency of the drivers, extending the 3dB down point by 2 or 3 Hz when using a sealed box.
But we don't listen in anechoic environments, we have room gain, and the box lift effect disappears off axis - so the net in room acoustic response may get a little bass-heavy by maybe +3 to +6dB. Many would view this as a problem, but because most recordings are mastered at very high SPL and played back at lower levels in the home, some of this excess gets scrubbed off due to the "Fletcher-Munson" effect... it becomes a bit of built-in loudness compensation and will pleasing in most rooms.
Now as for the internal pressure waves from MW2 "modulating" the midrange on MW1, as the examples above show, the relative phase difference between drivers is almost nil in the bass transitioning to only 76 degrees at 800Hz - 76 degrees is not even yet in quadrature, and so is still additive ...... and remember, by then the output of MW2 is down by 12 dB... so MW2 not going to add too much to the mids and upper mids.
And what it does add is still close enough in phase so it will not muddy or color the sound in a very noticeable way.
Yes it's true, there will be some MINOR time-smear and combing effects , especially if the drivers are spaced far apart, and yes it is audible as compared to the single driver - but barely so - and for most listeners, I suspect we'd gladly have that extra bass extension, extra efficiency and balance.
When crossing over to the tweeter, we are then mostly concerned with the mid and upper mid response of MW1 alone, without any need for BSC as the bass levels remain balanced.
As for the "evanescent fluids coupling inside the box" theory, well, maybe, but it's certainly not a dominant or even very noticeable effect!
More clarification
Hi again! Just in case the previous example wasn't clear enough in de-bunking the notion of midrange modulation by "evanescent acoustic interference pressure waves" let's look at the following considerations:
1) In the bass frequencies below 200Hz MidWoofer1 and MidWoofer 2 are operating almost totally in phase and almost exactly the same amplitude. Which means they will have total output of 6dB more than the equivalent MW1 alone.
So if we operate them at the same matching SPL, say 90dB at 1m, the "2.5" version with the added MW2 will have SUBSTANTIALLY lower harmonic distortion, perhaps by a factor of 3 to 10 times lower, due to exponentially decreased distortion with less cone excursion. The smaller the driver, the more this becomes a factor.
It will also have SUBSTANTIALLY less Doppler distortion, because the cone from MW1 will have drastically reduced excursion at the same bass SPL
the above holds true whether using a single-volume box or a split-volumes box with the same total internal volume for each driver.
2) The idea that the back-pressure from MW2 will somehow modulate the cone on MW1 would be true, IF and ONLY if MW2 were not being actively driven with a bass signal that is essentially identical to that which is driving MW1.
The bass back-pressure from MW1 is identical to that of MW2.
To put it simply, if you push one hand against the other with 1lb of force on each, neither of them moves. Now increase the force by 10 times... do they move? No, because the forces are still equal. Same with the cones from MW1 and MW2
3) Introducing a small phase shift, as with a 1st order lowpass will not create substantial cancellation unless the inter-driver distances are conducive to creating comb-filtering by being too far apart.
Note: A first order filter is asymptotically limited to a max phase shift of 90 degrees. Meaning: even at the asymptotic extreme, this filter cannot "cancel" any output because it has no vectorial component which is going backwards.
IN SUM:
"Midrange modulation" from the 2nd woofer? Due to the decreases in cone movement vs the single driver variant, the "midrange modulation" in the 2.5 configuration is likely to be much LESS.... not more.
I hope these last points will help put the "evanescent acoustic bass waves interfering with the mid-range" theory to rest. Pshaw!! What bunkus!
Then again, some things, like half-baked, yet rigidly held notions, SHOULD be evanescent. Let's hope these clarifications help to further "evanesce" the issue!
Special Note: Deliberately introducing a large phase shift on MW2 vs MW1 as in Modafferi's patents does create some interaction between the drivers. It clearly would indeed affect the mids, especially when using his extreme version where the phase-shift goes all the way to 180 degrees by 500Hz, creating nearly total cancellation of output.
Hi again! Just in case the previous example wasn't clear enough in de-bunking the notion of midrange modulation by "evanescent acoustic interference pressure waves" let's look at the following considerations:
1) In the bass frequencies below 200Hz MidWoofer1 and MidWoofer 2 are operating almost totally in phase and almost exactly the same amplitude. Which means they will have total output of 6dB more than the equivalent MW1 alone.
So if we operate them at the same matching SPL, say 90dB at 1m, the "2.5" version with the added MW2 will have SUBSTANTIALLY lower harmonic distortion, perhaps by a factor of 3 to 10 times lower, due to exponentially decreased distortion with less cone excursion. The smaller the driver, the more this becomes a factor.
It will also have SUBSTANTIALLY less Doppler distortion, because the cone from MW1 will have drastically reduced excursion at the same bass SPL
the above holds true whether using a single-volume box or a split-volumes box with the same total internal volume for each driver.
2) The idea that the back-pressure from MW2 will somehow modulate the cone on MW1 would be true, IF and ONLY if MW2 were not being actively driven with a bass signal that is essentially identical to that which is driving MW1.
The bass back-pressure from MW1 is identical to that of MW2.
To put it simply, if you push one hand against the other with 1lb of force on each, neither of them moves. Now increase the force by 10 times... do they move? No, because the forces are still equal. Same with the cones from MW1 and MW2
3) Introducing a small phase shift, as with a 1st order lowpass will not create substantial cancellation unless the inter-driver distances are conducive to creating comb-filtering by being too far apart.
Note: A first order filter is asymptotically limited to a max phase shift of 90 degrees. Meaning: even at the asymptotic extreme, this filter cannot "cancel" any output because it has no vectorial component which is going backwards.
IN SUM:
"Midrange modulation" from the 2nd woofer? Due to the decreases in cone movement vs the single driver variant, the "midrange modulation" in the 2.5 configuration is likely to be much LESS.... not more.
I hope these last points will help put the "evanescent acoustic bass waves interfering with the mid-range" theory to rest. Pshaw!! What bunkus!
Then again, some things, like half-baked, yet rigidly held notions, SHOULD be evanescent. Let's hope these clarifications help to further "evanesce" the issue!
Special Note: Deliberately introducing a large phase shift on MW2 vs MW1 as in Modafferi's patents does create some interaction between the drivers. It clearly would indeed affect the mids, especially when using his extreme version where the phase-shift goes all the way to 180 degrees by 500Hz, creating nearly total cancellation of output.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.