Hi,
I am just wondering, has anyone made/used multi way subwoofers for PA?
Hypothetically, one could have two 4th order bandpass subs with one up to 60hz or so, and another from 60-120hz. I was thinking this could have some advantages:
With 4th order BP only having one output, this could allow a cleaner sound if the mid and low subs are time aligned, instead of having 4 different outputs if bass reflex were used to do this (mid driver, mid port, sub driver, sub port)
Perhaps this could overcome some of the limitations of BP enclosures?
I am just wondering, has anyone made/used multi way subwoofers for PA?
Hypothetically, one could have two 4th order bandpass subs with one up to 60hz or so, and another from 60-120hz. I was thinking this could have some advantages:
- Easier design of high efficiency bandpass boxes with smaller range.
- The 'mid' sub, having a higher tuning frequency, would have less group delay than a low tuned BP or BR, for more punch and (maybe?) transient response overall.
- The low tuned sub could be positioned further ahead of the stage to time align its group delay.
With 4th order BP only having one output, this could allow a cleaner sound if the mid and low subs are time aligned, instead of having 4 different outputs if bass reflex were used to do this (mid driver, mid port, sub driver, sub port)
Perhaps this could overcome some of the limitations of BP enclosures?
Group delay is inherent in low frequency reproduction regardless of cabinet design.. those long wavelengths take longer to produce that is just the physics of it, and that also means it's not really possible to "time allign" subbass.. exactly how do you line up wavelengths that are many meters long with others that are only centimeters or milimeters long? I think the best results occur when everything is aligned for best impulse response, after that it is what it is.
And bandpass designs are the worst of all for adding even more delay and mud so you rarely see anyone using them in pro audio.
The theoretical best speaker system is a single point source for all frequencies but that isn't possible given our current technology, so we are forced to breakup the spectrum into several frequency bands. In general the best results occur with the fewest number of divisions but that doesn't stop people from using many, but I'm definitely in the less is more camp.
And bandpass designs are the worst of all for adding even more delay and mud so you rarely see anyone using them in pro audio.
The theoretical best speaker system is a single point source for all frequencies but that isn't possible given our current technology, so we are forced to breakup the spectrum into several frequency bands. In general the best results occur with the fewest number of divisions but that doesn't stop people from using many, but I'm definitely in the less is more camp.
Last edited:
Hi re the delay caused by low frequency reproduction, what I'm thinking about is:
Given the 90 degree phase lag (in port output) at the tuning frequency of a bass reflex or the output of a BP due to the port and enclosure interaction, and if the enclosure is tuned to say 30hz, then this should come to (1/30) /4 =8.33ms which wouldn't cause issues for a 30hz signal but would be more significant for a higher frequency so of course a BP enclosure response would be poor if you ask it to reproduce a wide bandwidth. Plus 100% of the output is going through the port.
90 degrees or 1/4 wavelength of a enclosure tuned to 70 hz would be much shorter delay at 3.5ms, if we split this out.
8.33 - 3.5 = 4.83ms difference. So with dual band subs, we could position the 30hz sub 1.66 meters ahead of the 70hz sub which would theoretically give us a somewhat time coherent output at the bass frequencies. (need to do some calculations to see the average phase/time lag of the port output over the frequency range)
Also a 4th order BP having 100% of the output coming out of the port means 100% of output has the delay VS a bass reflex which has the direct driver output as well.
Given the 90 degree phase lag (in port output) at the tuning frequency of a bass reflex or the output of a BP due to the port and enclosure interaction, and if the enclosure is tuned to say 30hz, then this should come to (1/30) /4 =8.33ms which wouldn't cause issues for a 30hz signal but would be more significant for a higher frequency so of course a BP enclosure response would be poor if you ask it to reproduce a wide bandwidth. Plus 100% of the output is going through the port.
90 degrees or 1/4 wavelength of a enclosure tuned to 70 hz would be much shorter delay at 3.5ms, if we split this out.
8.33 - 3.5 = 4.83ms difference. So with dual band subs, we could position the 30hz sub 1.66 meters ahead of the 70hz sub which would theoretically give us a somewhat time coherent output at the bass frequencies. (need to do some calculations to see the average phase/time lag of the port output over the frequency range)
Also a 4th order BP having 100% of the output coming out of the port means 100% of output has the delay VS a bass reflex which has the direct driver output as well.
I think BP4 for PA is a bad idea.
Remember the excursion of the driver increases rapidly below the vent tuning frequency, and that the vent tuning frequency in a BP4 is in the middle of the passband, not at the bottom of the passband like in BP6 or vented alignments.
That means you'll have to use a much higher HPF frequency for the BP4 than you would a similar vented box, cutting off a lot of low frequency extension.
Next up: re phase & group delay, it's 180deg for a typical vented box, not 90.
However, for BP4 the phase is around 0 degrees at the tuning frequency.
Neither have flat group delay away from the tuning frequency so you can't simply do your time alignment by basing it on calculated group delay at any one frequency.
You'll actually get better alignment by checking the phase at the crossover region between each box and aligning there, otherwise you risk having uneven frequency response if they are out of phase at crossover.
Next, you should really be doing your time alignment in the DSP rather than physically in the venue - this allows you to keep the speakers close together meaning your alignment will be valid over a wider range of listening positions. Remember that if you pull a speaker one or two meters forward of another, that time alignment is only valid on that axis - any of the audience who are off to the side will not get the benefit of it. Also, depending on how big the venues are that you're working, you may not even have space to move your cabinets around like that.
Hope this helps,
David.
Remember the excursion of the driver increases rapidly below the vent tuning frequency, and that the vent tuning frequency in a BP4 is in the middle of the passband, not at the bottom of the passband like in BP6 or vented alignments.
That means you'll have to use a much higher HPF frequency for the BP4 than you would a similar vented box, cutting off a lot of low frequency extension.
Next up: re phase & group delay, it's 180deg for a typical vented box, not 90.
However, for BP4 the phase is around 0 degrees at the tuning frequency.
Neither have flat group delay away from the tuning frequency so you can't simply do your time alignment by basing it on calculated group delay at any one frequency.
You'll actually get better alignment by checking the phase at the crossover region between each box and aligning there, otherwise you risk having uneven frequency response if they are out of phase at crossover.
Next, you should really be doing your time alignment in the DSP rather than physically in the venue - this allows you to keep the speakers close together meaning your alignment will be valid over a wider range of listening positions. Remember that if you pull a speaker one or two meters forward of another, that time alignment is only valid on that axis - any of the audience who are off to the side will not get the benefit of it. Also, depending on how big the venues are that you're working, you may not even have space to move your cabinets around like that.
Hope this helps,
David.
I think BP4 is a good way to make the most of drivers that have a too-high Qts for ported boxes. I built a 2x18" BP4 a while back, and it worked really well for the cheap-ish drivers that I was forced to use.
They had an Xmax of about 10mm, and got to about +/-15mm without sounding awful, so I ran that box without a HPF. It was optimised for 45Hz-90Hz, but would still grunt something out at 30Hz on account of the sealed back chambers.
Chris
They had an Xmax of about 10mm, and got to about +/-15mm without sounding awful, so I ran that box without a HPF. It was optimised for 45Hz-90Hz, but would still grunt something out at 30Hz on account of the sealed back chambers.
Chris
Sure, if cost or availability dictate using a particular driver who's parameters suit BP4 better that's fair enough.
Would you mind sharing what the driver was by any chance?
Cheers,
David.
Sure, it was this one: https://bishopsound.com/products/18...h-faston-terminals-woofer-bdp18-4-8-ohm#specs
Looks like I've mis-remembered the excursion figures, though. Perhaps it was 7.5mm Xmax and 10mm of not-awful.
Despite the promisingly-large magnet, the BL is pretty low and Qts is pretty high. The box I came up with was compact (as 2x18" boxes go), featured force cancellation, and direct ventilation of the magnets. ie, 3x chambers: 2x sealed (outside), middle ported, drivers mounted with magnets in the middle chamber.
I still have the drivers on a shelf somewhere, but I don't have much use for them.
Chris
Looks like I've mis-remembered the excursion figures, though. Perhaps it was 7.5mm Xmax and 10mm of not-awful.
Despite the promisingly-large magnet, the BL is pretty low and Qts is pretty high. The box I came up with was compact (as 2x18" boxes go), featured force cancellation, and direct ventilation of the magnets. ie, 3x chambers: 2x sealed (outside), middle ported, drivers mounted with magnets in the middle chamber.
I still have the drivers on a shelf somewhere, but I don't have much use for them.
Chris
Sometimes a kick bin is used for upper bass, together with a subwoofer deep bass. You could call that multi way.
Punch can be a lack of output at the 100 - 200 Hz band (with respect to the 30 - 60 Hz boom), which is not necessarily handled by the subwoofers.
Punch can be a lack of output at the 100 - 200 Hz band (with respect to the 30 - 60 Hz boom), which is not necessarily handled by the subwoofers.
Last edited:
I am not a fan of separate kick sections. 100Hz is not so low it should be easy enough to ensure that the main speakers are capable of enough output in this range; if there isn't enough 'kick' then you should be able to EQ it in. A box covering only covering such a narrow frequency range is going to be tricky to integrate with the other speakers due to interaction between the LP and HP of the crossover.
Depends on the size of the system. Nexo Alpha is a good example of where a 4-way system is pretty much mandatory - you can't get that output from a 3-way setup.
Anything smaller than several 2" HF drivers per side, though, and I agree that a 3-way system is the way to go.
Chris
Anything smaller than several 2" HF drivers per side, though, and I agree that a 3-way system is the way to go.
Chris