Sure!
As far as I experienced the flared ports have a much more severe turbulent flow separation and "resonator unloading" characteristic. So once the flared port compresses (or, more precisely, flow separates, the resonance is lost, driver excursion is multiplied) most of the output is lost.
Tubular ports start chuffing and (mildly) compressing earlier, but the air flow is not disrupted completely.
My explanation is:
The flow disruption of flared ports happens in the port flare. With tube ports the turbulent flow stays more outside the oscullating air of the port.
The idea seems reasonable, though I'm not sure that is what's happening here.
What you are describing sounds like a breakdown of port function, whereas I just measured a small difference in the amount of port compression. Note that the flared ports still had more output, but the difference was larger at low levels. Hence, the flared ports seemed to be working better even though they compressed more.
Here are the measurements in question.
Brown: unflared 5.66V, Green: flared 5.66 V
Red: unflared 45.28V, Orange: flared 45.28V
Dark green: unflared 87 V, Light green: flared 87V
At 5.66V the flared ports had 1.2 dB more output.
At 87V, the unflared ports had 0.4 db less compression, but the flared ports still had 0.8 dB more output compared to the unflared ports.
The difference could be attributable to a small change in tuning frequency or perhaps temperature/humidity. The sensitivity above ~52 Hz is slightly higher for the port without flares.
What you are describing sounds like a breakdown of port function, whereas I just measured a small difference in the amount of port compression. Note that the flared ports still had more output, but the difference was larger at low levels. Hence, the flared ports seemed to be working better even though they compressed more.
Here are the measurements in question.
Brown: unflared 5.66V, Green: flared 5.66 V
Red: unflared 45.28V, Orange: flared 45.28V
Dark green: unflared 87 V, Light green: flared 87V
At 5.66V the flared ports had 1.2 dB more output.
At 87V, the unflared ports had 0.4 db less compression, but the flared ports still had 0.8 dB more output compared to the unflared ports.
The difference could be attributable to a small change in tuning frequency or perhaps temperature/humidity. The sensitivity above ~52 Hz is slightly higher for the port without flares.
the flared port speaker seems to be tuned lower. definitely!
also, as far as I have observed from response and impedance measurements the flared ports usually have a higher resonator Qb.
That is: higher port output level.
most simulators assume a standard Qb value of 7. My test speakers usually seem to have a Qb of 10-15.
Thanks for the data!
Do you mean the build up of a pressure unbalance that will offset the diaphragm?
Correct ... I'll have to do a pressure measurement for pressure difference in/ouside the enclosure!
Definitely.
I am curious about the actual pressure difference and its influence on diaphragm position.
The pressure unbalance is created because of flow difference in- and outwards at the port.
That happens when the flow is separating. My optimizer tool tries to avoid this flow separation, because it correlates with compression, noise and tuning change.
But of course the flow separation will start very slowly, even before any noticeable compression happens.
I'll do a research on manometers!
EDIT: I think I'll start with a transparent water filled tube ...
Last edited:
Would you mind sharing the port dimensions?
And is the (port) output actually near 116 dB?
The subwoofer has two ports that consists of a straight rectangular section that is about 50 cm long. The "pipe" section cross-section is 13*11.4 cm.
The ports ends have 55mm radii, but only on two sides, e.g the ports are placed in the corners, and two sides of the rectangle are terminated at the enclosure wall. So end cross sectional area would be (13+5.5)*(11.4+5.5)*2 = 625 cm2 and in the straight section it would be: 13*11.4*2 = 296 mm2
The output was measured at 2 meters, ground plane (2 pi) with a factory calibrated UMIK-1, so the SPL readings should be mostly OK.
yes, Klippel has a couple of paper on this (sorry for sending you into the woods, but I have to look them up myself again lol haha)
Obviously this is only at a small frequency band.
I have tested this with just steady laminar (water) flows in the past, and the difference is quite substantial.
The question is how this translates to loudspeaker acoustics, since theory and practice are not always the same thing.
On the other hand, it's not that hard to make it happen.