I'm thinking to experiment with (stereo) cardioid subs:
- Pair of 15" drivers
- One at each end of a short 15"-dia tube, both outward-facing
- Delay the rear driver, by the length of the tube.
Maybe this needs two chambers, back-to-back, instead of a single chamber coupling both drivers. But the whole front-back length should be less than 18" to maintain cardioid pattern to a few hundred Hz. So the enclosed volume is really small.
I don't have much knowledge about how to model this sort of thing, so thinking I should just build it and see what happens. For the cost of an afternoon cutting up tubing...
Before I start, does anyone have experience with this sort of build, or other cardioid subs at home? Do they "work" in a smallish room?
- Pair of 15" drivers
- One at each end of a short 15"-dia tube, both outward-facing
- Delay the rear driver, by the length of the tube.
Maybe this needs two chambers, back-to-back, instead of a single chamber coupling both drivers. But the whole front-back length should be less than 18" to maintain cardioid pattern to a few hundred Hz. So the enclosed volume is really small.
I don't have much knowledge about how to model this sort of thing, so thinking I should just build it and see what happens. For the cost of an afternoon cutting up tubing...
Before I start, does anyone have experience with this sort of build, or other cardioid subs at home? Do they "work" in a smallish room?
Separate chambers would be required for cardioid subs.
Cardioid subs are used to reduce level behind the sub, but result in a more ragged frequency response and pattern control of a few octaves centered around a specific frequency.
If by "smallish room" you mean smaller than a theater, no, they don't "work", as low frequency waves are long enough to bounce off opposing walls and fill in the reduced level behind the cardioid subs.
Cardioid subs are used to reduce level behind the sub, but result in a more ragged frequency response and pattern control of a few octaves centered around a specific frequency.
If by "smallish room" you mean smaller than a theater, no, they don't "work", as low frequency waves are long enough to bounce off opposing walls and fill in the reduced level behind the cardioid subs.
He did a lot of work, and as he says:
"Cardioid cannot fight against reflections from back wall".
Oh well.
The Nexo Ray subs use two drivers.
Seems unlikely they would have a single chamber when the two speakers are fed different signal to create different polar patterns.
Have you actually opened a Nexo Ray sub and saw only one chamber?
Pictures or it didn't happen
.Just for discussions sake: the rear driver can be 6db down to produce the desired effect of wave cancellation. No need to use an identical woofer when an 8-10" would do just fine
Typically a ratio of two forward to one rearward speaker (a six dB difference) works well, but the speakers must be matched in phase response, easier to do if using the same speaker type.Just for discussions sake: the rear driver can be 6db down to produce the desired effect of wave cancellation. No need to use an identical woofer when an 8-10" would do just fine
For practical reasons I'd like to lowpass-filter the rear driver; I don't want pattern control (or anything rearward) much above the point where baffle-step naturally makes the front driver directional by itself. Filtering can be done either mechanically (e.g. slot loading) or with DSP (e.g. biquads). Both ways would introduce some phase shifts relative to the front driver.
The ideal might be to measure the on-axis front-driver response, measure from the same position driven only by the rear driver, then adjust phase and delay using FIR filters so that they combine constructively across the band of interest. In other words, optimize for the forward lobe, and let the rest fall where it may.
Having the forward driver physically a few inches higher than the rear will point the cardioid (or hyper, etc) upward. For whatever that's worth.
I'm still not yet absolutely convinced this is worthwhile the effort to build. But it's a good gedankenexperiment
The ideal might be to measure the on-axis front-driver response, measure from the same position driven only by the rear driver, then adjust phase and delay using FIR filters so that they combine constructively across the band of interest. In other words, optimize for the forward lobe, and let the rest fall where it may.
Having the forward driver physically a few inches higher than the rear will point the cardioid (or hyper, etc) upward. For whatever that's worth.
I'm still not yet absolutely convinced this is worthwhile the effort to build. But it's a good gedankenexperiment
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.