That's not true. The lower cutoff frequency is lowered as more drivers with their corresponding enclosures are placed together due to mutual coupling effects. That's because the relationship between the compliance of the outer air and the compliance of the air inside the enclosures changes when more drivers are radiating together.
Relative cone displacement of each driver is also reduced that way, thus reducing distortion.
Relative cone displacement of each driver is also reduced that way, thus reducing distortion.
No, I can't quantify it exactly and I don't know any enclosure simulator allowing the user to change outer air compliance except "hornresp", but this one is only useful for horns (and the modelled effect is quite dramatic).
However, as the theoretical efficiency increase when two drivers are placed together is 3dB/W, I think that the theoretical lower cutoff displacement could be half an octave.
However, as the theoretical efficiency increase when two drivers are placed together is 3dB/W, I think that the theoretical lower cutoff displacement could be half an octave.
I doubt about what you're saying because everyone would use multiple drivers instead of a single one. Most people still use a single driver.
Multiple drivers could have an effect on the F3, but I think it takes more than 2 or 4, it would take a line array at least IMO.
Again, this statement is not theory based so I could be very wrong too!
I agree with you that the distortion and the excursion decrease with multiple drivers.
Multiple drivers could have an effect on the F3, but I think it takes more than 2 or 4, it would take a line array at least IMO.
Again, this statement is not theory based so I could be very wrong too!
I agree with you that the distortion and the excursion decrease with multiple drivers.
Judging by what I read here everyday, little people knows actually what they are doing while the rest act following strange instincts and fashion trends.
Concerning the title of the thread, there is something that I have already mentioned but I think that it's worth more emphasis. Big drivers of 10" or more diameter are popularly regarded as having very poor high frequency response, but that's not because of the own drivers, it's because people always puts them in the center of some wall of a perfectly square or rectangular and big enclosure, whose resonant modes will ruin the response above 100Hz of any driver. That kind of enclosures are terrble.
In order to take advantage of the ultra-wide frequency bandwidth and the high output provided by most 10" and bigger drivers, enclosures without even a single pair of paralell walls and preferably of asymmetric design are a must. Asymmetric driver placement is also recommended. Clean output up to 1Khz may be achieved this way even with some 18" PA drivers (provided that you also get rid of the typical wood mechanical resonances).
Concerning the title of the thread, there is something that I have already mentioned but I think that it's worth more emphasis. Big drivers of 10" or more diameter are popularly regarded as having very poor high frequency response, but that's not because of the own drivers, it's because people always puts them in the center of some wall of a perfectly square or rectangular and big enclosure, whose resonant modes will ruin the response above 100Hz of any driver. That kind of enclosures are terrble.
In order to take advantage of the ultra-wide frequency bandwidth and the high output provided by most 10" and bigger drivers, enclosures without even a single pair of paralell walls and preferably of asymmetric design are a must. Asymmetric driver placement is also recommended. Clean output up to 1Khz may be achieved this way even with some 18" PA drivers (provided that you also get rid of the typical wood mechanical resonances).
In my experience, the effect is both measurable and listenable quite easily. It's the same phenomena that happens when you compare half-space frequency response and corner frequency response of the same driver/enclosure. Increasing the number of adjacent perpendicular walls behind the speaker lowers bass cutoff point.
Given a sealed enclosure, it has to be understood that bass roll-off is caused by driver suspension compliance and enclosure air compliance acting as springs and limiting cone displacement to a fixed value only proportional to voice coil current, despite frequency.
Since for flat frequency response the required cone displacement is inversely proportional to the frequency squared, but we have the same available force at all frequencies to move the cone, when the frequency is progressively reduced there is a progressive change in the order of cone excursion increase as the suspension takes over the moving mass as a displacement-limiting factor. That change is from 1/freq^2 to 1/freq^0=1, thus causing a 2nd order roll-off in the response (typical sealed box roll-off). Note that moving mass alone does not cause any lower or upper cutoff, because a constant ondulatory force applied on a constant mass produces a movement whose amplitude is 1/freq^2 despite frequency.
In other words, what causes low-frequency rolloff in a sealed enclosure is not the decreasing frequency itself, instead, for each voice coil current level there is a proportional maximum cone excursion level (derived from motor force applied to suspension and enclosure air springs) and the bass roll-off happens after that level is reached despite frequency. Thus, if you manage to get the same or higher SPL with the same voice coil current but less cone displacement, an unavoidable consequence will be lowered bass cutoff, as cone excursion limiting due to suspension and enclosure air compliance will just start to happen at a lower frequency.
BTW: That explains why horns allow to get good 30Hz efficiency of drivers with Fs=40hz or higher. The trick is always increasing useful SPL-producing loading to reduce cone displacement given the same voice coil force and frequency.
Given a sealed enclosure, it has to be understood that bass roll-off is caused by driver suspension compliance and enclosure air compliance acting as springs and limiting cone displacement to a fixed value only proportional to voice coil current, despite frequency.
Since for flat frequency response the required cone displacement is inversely proportional to the frequency squared, but we have the same available force at all frequencies to move the cone, when the frequency is progressively reduced there is a progressive change in the order of cone excursion increase as the suspension takes over the moving mass as a displacement-limiting factor. That change is from 1/freq^2 to 1/freq^0=1, thus causing a 2nd order roll-off in the response (typical sealed box roll-off). Note that moving mass alone does not cause any lower or upper cutoff, because a constant ondulatory force applied on a constant mass produces a movement whose amplitude is 1/freq^2 despite frequency.
In other words, what causes low-frequency rolloff in a sealed enclosure is not the decreasing frequency itself, instead, for each voice coil current level there is a proportional maximum cone excursion level (derived from motor force applied to suspension and enclosure air springs) and the bass roll-off happens after that level is reached despite frequency. Thus, if you manage to get the same or higher SPL with the same voice coil current but less cone displacement, an unavoidable consequence will be lowered bass cutoff, as cone excursion limiting due to suspension and enclosure air compliance will just start to happen at a lower frequency.
BTW: That explains why horns allow to get good 30Hz efficiency of drivers with Fs=40hz or higher. The trick is always increasing useful SPL-producing loading to reduce cone displacement given the same voice coil force and frequency.
OK - I'm in trouble. The 12" Hi-Fi Dayton Reference subs are out of stock, and will be till mid-May. My options are thus to go with the 10" Hi-Fi version, or the 12 HO.
This seriously throws a monkey wrench in my plans, as it was looking good to use a couple of those 12 HF's crossed over high...
This seriously throws a monkey wrench in my plans, as it was looking good to use a couple of those 12 HF's crossed over high...
perhaps this has been said but i believe that at a cetain separation, at a certain frequency and below you begin to move into the near field of the combined pair at whcih point the rolloff changes from six db per octave to three. so this should give you another octave? This effect depends on the two driver's distance from eachother and I think it is either one half or one quarter the wavelength of the frequency in question. I think at ten feet it is something like fifty hertz so a rise might be seen in the efficiency there and the rollof should be slower below.
I hope!
I hope!
Phase correction is not about the overall phase response of the system, but about the particular phase response of each driver/enclosure, that must be tailored for proper summing with the rest of the drivers in the overlapping crossover region.
In other words, phase correction is always required, and poor systems result when it's just ignored.
In other words, phase correction is always required, and poor systems result when it's just ignored.
Hi,
Here's a plot of my corner loaded tempests, with an 80Hz low pass filter. The plot shows 1 playing versus both playing. They are placed directly next to each other. Each driver has its own amp.
I cannot see any change in the response - Shows the 6dB increase that using 2 drivers gives, if the power is supplied.
Rob.
Here's a plot of my corner loaded tempests, with an 80Hz low pass filter. The plot shows 1 playing versus both playing. They are placed directly next to each other. Each driver has its own amp.
I cannot see any change in the response - Shows the 6dB increase that using 2 drivers gives, if the power is supplied.
Rob.
Attachments
You don't see the cut-off point moving lower because there is no cut-off!! The horn-like corner loading together with the 12dB/oct room gain are supressing completely the 12dB/oct roll-off from the sealed enclosures.
If you can, repeat the test outdoors with the enclosures placed together and absolutely no walls near (10 meter clearance should be OK), except the own floor of course.
Also, note that gain is becoming *more* than 6dB below 60Hz while it's *less* than 6dB above 60Hz. This suggests a 5dB gain figure due to placing the two enclosures together and a 2dB additional gain figure below 60Hz due to enclosure zeros being shifted to a lower frequency while room poles being the same.
If you can, repeat the test outdoors with the enclosures placed together and absolutely no walls near (10 meter clearance should be OK), except the own floor of course.
Also, note that gain is becoming *more* than 6dB below 60Hz while it's *less* than 6dB above 60Hz. This suggests a 5dB gain figure due to placing the two enclosures together and a 2dB additional gain figure below 60Hz due to enclosure zeros being shifted to a lower frequency while room poles being the same.
Hi eva,
The tempests have an f-3 of ~35Hz. If using 2 lowered the f-3 then surely you'd notice the top plot change around the 35Hz mark even in room ? According to what you've stated there should be a boost visible between approx 25Hz - 35Hz if the 2 subs had an f3 lowered by 1/2 an octave.
I'm remaking the boxes soon (few weeks) so if the weathers kind I'll do the outside measurements.
The 6dB figure is pretty consistant. The slight +/- 1 could well be to the curve 'rippling' as I don't use the peak hold function, I just pause the rta and save it to memory. (was only a quick measurement)
Rob
The tempests have an f-3 of ~35Hz. If using 2 lowered the f-3 then surely you'd notice the top plot change around the 35Hz mark even in room ? According to what you've stated there should be a boost visible between approx 25Hz - 35Hz if the 2 subs had an f3 lowered by 1/2 an octave.
I'm remaking the boxes soon (few weeks) so if the weathers kind I'll do the outside measurements.
The 6dB figure is pretty consistant. The slight +/- 1 could well be to the curve 'rippling' as I don't use the peak hold function, I just pause the rta and save it to memory. (was only a quick measurement)
Rob
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.