Thanks GM! Yeah, I'm fixing and oldish house with several bad structure decisions by nowadays standards and things escalate quickly. If I only was older and second time home owner before buying the first.. Or wealthier to build a new one, or poor enough not to have bought any in the first place. If there are any positive sides, a good chance to build the house for stereo system and not the other way around... 😀
What about if there is no low frequencies present to couple the port and the cone, could higher frequencies then exit the port? I mean I'm not sure how much these unwanted port "leakage" even affects the speaker system sound so all this might be academic, but the resonances seem to appear in the measurements at least.
Measurements are most often swept sine (at least that seems to be the default setting in ARTA for example) there isn't any lows at the moment when the sweep hits a box resonance, for example the ~680Hz, and it easily appears in the resulting frequency response? But with music this resonance is not so apparent due to low frequencies keeping the port busy? I guess this could be tested with a noise test signal or with two sine tones, one hitting a internal resonance frequency and another the port tuning frequency toggling the port tone on and off and listen/measure how much it affects.
ironsf, regarding rear wall near midwoofer: You could try! People seem to like results when the backwave is "absorbed" as much as possible, search the forum for "the dagger" for example, or look B&W nautilus. Although there is lots of people happy with seemingly what ever rectangular box they are having as mid back chamber. I tried three different proto boxes for a mid just because the hobby, and one with the back wall as close as possible to get the wavelength resonance just above crossover point and most open space for the speaker structure openings to the sides yielded smoothest looking results. This was with a 8" PA mid with huge magnet and smallish enclosure volume, very difficult to build "a dagger". Still having peaks and dips all around the pass band, need to do few more protos I think and refine the measurement setup and other plans or change the driver. It is acceptable but not looking pretty so, it is part of the hobby to over tinker 🙂
What about if there is no low frequencies present to couple the port and the cone, could higher frequencies then exit the port? I mean I'm not sure how much these unwanted port "leakage" even affects the speaker system sound so all this might be academic, but the resonances seem to appear in the measurements at least.
Measurements are most often swept sine (at least that seems to be the default setting in ARTA for example) there isn't any lows at the moment when the sweep hits a box resonance, for example the ~680Hz, and it easily appears in the resulting frequency response? But with music this resonance is not so apparent due to low frequencies keeping the port busy? I guess this could be tested with a noise test signal or with two sine tones, one hitting a internal resonance frequency and another the port tuning frequency toggling the port tone on and off and listen/measure how much it affects.
ironsf, regarding rear wall near midwoofer: You could try! People seem to like results when the backwave is "absorbed" as much as possible, search the forum for "the dagger" for example, or look B&W nautilus. Although there is lots of people happy with seemingly what ever rectangular box they are having as mid back chamber. I tried three different proto boxes for a mid just because the hobby, and one with the back wall as close as possible to get the wavelength resonance just above crossover point and most open space for the speaker structure openings to the sides yielded smoothest looking results. This was with a 8" PA mid with huge magnet and smallish enclosure volume, very difficult to build "a dagger". Still having peaks and dips all around the pass band, need to do few more protos I think and refine the measurement setup and other plans or change the driver. It is acceptable but not looking pretty so, it is part of the hobby to over tinker 🙂
Last edited:
An interesting article about bass-reflex loudspeakers and port linearity (or non-linearity) can be found here:
How Good Is Your Port | audioXpress
How Good Is Your Port | audioXpress
My experience is only trail and error. I would be very wary of placing the rear wall close to the woofer basket because of the greater risk of reflections through the cone, which is often transparent to them.
From tmuikku;
"What about if there is no low frequencies present to couple the port and the cone, could higher frequencies then exit the port? I mean I'm not sure how much these unwanted port "leakage" even affects the speaker system sound so all this might be academic, but the resonances seem to appear in the measurements at least."
If there was no LF energy, then a port would be pointless.
If anyone doesn't believe that mid and other noises come out of a port go and listen to some even professional systems. Even my ATC SCM100ASLs sent a noise out of the ports.
Not all of these noises are turbulence from port profiles.
"What about if there is no low frequencies present to couple the port and the cone, could higher frequencies then exit the port? I mean I'm not sure how much these unwanted port "leakage" even affects the speaker system sound so all this might be academic, but the resonances seem to appear in the measurements at least."
If there was no LF energy, then a port would be pointless.
If anyone doesn't believe that mid and other noises come out of a port go and listen to some even professional systems. Even my ATC SCM100ASLs sent a noise out of the ports.
Not all of these noises are turbulence from port profiles.
Yeah Pharos, with a song one might listen there could be LF energy or not depending on the genre, composition, era, produced etc. 🙂 Just wondering when the "port being busy, not letting inside sounds out" GM mentioned is actually true and when not. gdan just posted an article where there are some examples when the port is indeed very busy so makes me wonder does this phenomenon GM mentioned only apply to too small ports, or when playing loud, and / or when having low frequency content playing? As you say there is unwanted leaking sound coming out from the port, seen in the graphs in this thread and I think I can hear them through the port as well when listening close. This is of course a bit hard to say for sure (for me at least), since the driver is so close the ear (and mic) hears other stuff than the port as well. So, just making a discussion about the subject, thanks!🙂
Uh oh! Now a SW becomes a reflection!
:rolleyes:
Of course if you relate to the Hyperphysics site there will be that description of the phenomenon - Which you have experienced for sure!
I wandered trough some phisics & music sites but the SW phenomenon is applied only to musical instruments (of course for string instruments and the ones with air columns) and in loudspeaker boxes only the transversal ones that take place eminently on the baffle> that's why it's often repeated to make a solid baffle. The longitudinal waves ..uh...oh, are those chaotical molecules inside a box moving in waves? I don't think so. But let's take the classic example of the tall box and it's often repeated that for d=90 cm you'll see a "stationary wave at 380 Hz", well, applying the wave equation you'll get that lambda BUT the speaker box is not a musical instrument and the transient it's not the same. Ok, I get lost, but I'm still asking who are you and others to give patents ?
The energy coming out at mid frequencies is a separate factor from noises due to its own bass functioning.
Both factors have to be considered in port design. A large area port allows more of the internal mid noise through for obvious reasons, so it is desired from this consideration to minimise the area. But in doing this we increase the air velocity of the reflex action.
Whether to have one or more ports is also in contention, many pro Hi-Fi designers have used two, but it must be borne in mind that more smaller ones increases the frictional are for bass airflow, and turbulence at the port entrances and exits.
Curved smoothing is often used at bot ends of the port, and some dimple the inner surface, or line it with felt.
When I rebuilt my ATC 100s, I used felt, and placed a hoover belt around the inner tube end to make it less sharp. (The impedance to airflow is not the same in each direction, and so it acts as a diode resulting in a diaphragm displacement from centre position.
At LOW level you should be able to 'insert' (place very close to), your ear into the port entrance, hence cutting off the sound from the woofer, of course turbulence will be much reduced at reduced excursions.
Both factors have to be considered in port design. A large area port allows more of the internal mid noise through for obvious reasons, so it is desired from this consideration to minimise the area. But in doing this we increase the air velocity of the reflex action.
Whether to have one or more ports is also in contention, many pro Hi-Fi designers have used two, but it must be borne in mind that more smaller ones increases the frictional are for bass airflow, and turbulence at the port entrances and exits.
Curved smoothing is often used at bot ends of the port, and some dimple the inner surface, or line it with felt.
When I rebuilt my ATC 100s, I used felt, and placed a hoover belt around the inner tube end to make it less sharp. (The impedance to airflow is not the same in each direction, and so it acts as a diode resulting in a diaphragm displacement from centre position.
At LOW level you should be able to 'insert' (place very close to), your ear into the port entrance, hence cutting off the sound from the woofer, of course turbulence will be much reduced at reduced excursions.
I was having the same issue with my Dynaudio speakers with rear port. I could have reduced port noise drastically just adding a back chamber made with a pair of Amazon card box filling with acoustic form without affecting frequency response. This would be a temporary solution, but it certainly works.
