Hi All,
I did some reading on cardioid sub from JohnK and Kimmo Saunist, as well as multi-sub from GedLee and Harman. I feel this topic quite interesting and like to share some thoughts here.
1. Wall enforces a boundary condition that air particle velocity has to be zero.
2. In the context of mode, pressure and velocity are orthogonal. Thus, pressure node is velocity anti-node, and vice-versa.
3. Omni (monopole) source is a pressure source, dipole source (figure of 8) is a velocity source, and cardioid is half and half.
4. A mode is maximally excited when a pressure source is placed at a pressure anti-node, or a velocity source is placed at a velocity anti-node.
5. The 0th order mode has uniform pressure distribution across the whole room.
6. The 1st order mode is maximally excited by placing a monopole at the wall, or a dipole at the center of the room.
7. Considering people rarely put a dipole at the center of the room, and dipole cannot excite the 0th order mode, using a dipole bass forms an acoustic high-pass filter from the speaker to the listener.
8. We can minimize the mode excitation by putting the source at the corresponding node, or by putting the same sources (wit the same signal) at out-of-phase segments of a mode. Harman suggests using four monopoles (playing the same signal) at quarter spacing to all walls. In such a manner, the 1st order mode is not excited because the mode only has two segments and they are out of phase. The 2nd order mode is not excited because monopoles are placed at pressure nodes. 3rd order mode is excited quite strongly in this way.
9. Dipole and monopole will maximally excite some modes, when they sit at the corresponding anti-nodes. They will also NOT excite some modes at all, when they sit at corresponding nodes.
10. Cardioid always excites all modes, but never excites them to their full strengths.
11. Interaural time difference (ITD) is the key cue for direction finding at low frequencies. I agree with Earl that the stereo pair should be full-range and not EQed for room modes. Then use multiple subs to control the uneven response due to room modes in low frequencies.
12. Mathematically, room mode and ray-tracing are two different sets of basis functions. They are both sufficient to model a room on their own if sufficient number of terms are used. When limited terms are used, mode is good for low frequency and ray-tracing is good for high frequency.
13. Cardioid mid and high can be used to control the front-wall reflection. This concept is closer to ray-tracing than mode excitation.
I found it quite interesting how a far field pattern property translates to basic source types, then mode excitation. Any discussion is welcome.
---
Correct typo as picowallspeaker pointed out.
I did some reading on cardioid sub from JohnK and Kimmo Saunist, as well as multi-sub from GedLee and Harman. I feel this topic quite interesting and like to share some thoughts here.
1. Wall enforces a boundary condition that air particle velocity has to be zero.
2. In the context of mode, pressure and velocity are orthogonal. Thus, pressure node is velocity anti-node, and vice-versa.
3. Omni (monopole) source is a pressure source, dipole source (figure of 8) is a velocity source, and cardioid is half and half.
4. A mode is maximally excited when a pressure source is placed at a pressure anti-node, or a velocity source is placed at a velocity anti-node.
5. The 0th order mode has uniform pressure distribution across the whole room.
6. The 1st order mode is maximally excited by placing a monopole at the wall, or a dipole at the center of the room.
7. Considering people rarely put a dipole at the center of the room, and dipole cannot excite the 0th order mode, using a dipole bass forms an acoustic high-pass filter from the speaker to the listener.
8. We can minimize the mode excitation by putting the source at the corresponding node, or by putting the same sources (wit the same signal) at out-of-phase segments of a mode. Harman suggests using four monopoles (playing the same signal) at quarter spacing to all walls. In such a manner, the 1st order mode is not excited because the mode only has two segments and they are out of phase. The 2nd order mode is not excited because monopoles are placed at pressure nodes. 3rd order mode is excited quite strongly in this way.
9. Dipole and monopole will maximally excite some modes, when they sit at the corresponding anti-nodes. They will also NOT excite some modes at all, when they sit at corresponding nodes.
10. Cardioid always excites all modes, but never excites them to their full strengths.
11. Interaural time difference (ITD) is the key cue for direction finding at low frequencies. I agree with Earl that the stereo pair should be full-range and not EQed for room modes. Then use multiple subs to control the uneven response due to room modes in low frequencies.
12. Mathematically, room mode and ray-tracing are two different sets of basis functions. They are both sufficient to model a room on their own if sufficient number of terms are used. When limited terms are used, mode is good for low frequency and ray-tracing is good for high frequency.
13. Cardioid mid and high can be used to control the front-wall reflection. This concept is closer to ray-tracing than mode excitation.
I found it quite interesting how a far field pattern property translates to basic source types, then mode excitation. Any discussion is welcome.
---
Correct typo as picowallspeaker pointed out.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.