Hello All,
I'm testing a prototype but have some interesting behaviour.
Would anyone have any idea's why I would be getting this weird dip in the port response at the tuning frequency?
Shared album - Billy Jukes - Google Photos
The speaker was sat on the ground and the microphone was located 5cm from the port opening.
This box uses a rear triangular port for other design reasons.
I have tried varying the length but I still get the same dip at the tuning frequency.
The port is only 12cm long and the largest internal dimension of the box is roughly 42cm so I can't see any issues with modes in the port or box.
The box Vol is roughly 25L and is triangular in shape.
The Fs of the driver is 60Hz so I don't see any issues there.
Here's a measurement at 5cm from the port with the same driver in another box with a front, square port of similar area and length and slightly smaller box volume with a wedge-shaped box. No issues in the port response here.
Shared album - Billy Jukes - Google Photos
Any ideas will be grateful
I'm testing a prototype but have some interesting behaviour.
Would anyone have any idea's why I would be getting this weird dip in the port response at the tuning frequency?
Shared album - Billy Jukes - Google Photos
The speaker was sat on the ground and the microphone was located 5cm from the port opening.
This box uses a rear triangular port for other design reasons.
I have tried varying the length but I still get the same dip at the tuning frequency.
The port is only 12cm long and the largest internal dimension of the box is roughly 42cm so I can't see any issues with modes in the port or box.
The box Vol is roughly 25L and is triangular in shape.
The Fs of the driver is 60Hz so I don't see any issues there.
Here's a measurement at 5cm from the port with the same driver in another box with a front, square port of similar area and length and slightly smaller box volume with a wedge-shaped box. No issues in the port response here.
Shared album - Billy Jukes - Google Photos
Any ideas will be grateful
Last edited:
I had something similar recently and it puzzled me for a while. At first, I thought it was an air leak and resealed all the joints in the box. That didn't work. I tried another driver and the problem was still there. So it was the cabinet.
I eventually traced it to a baffle resonance and fixed it by glueing a brace across the baffle just above the main driver. Try pressing on the baffle and you should see the dip disappear. It's easier to troubleshoot using the impedance curve, which should show a blip at the offending frequency.
I eventually traced it to a baffle resonance and fixed it by glueing a brace across the baffle just above the main driver. Try pressing on the baffle and you should see the dip disappear. It's easier to troubleshoot using the impedance curve, which should show a blip at the offending frequency.
Last edited:
too many variables are unknown so to compare the two is dubious...for all we know the difference in mic distance to adjacent boundaries is causing the dip.
Cheers Steve
I've checked for potential leaks as well because the construction of the rear port is ropey. I haven't checked the bracing yet but I'll try this tonight.
I've checked for potential leaks as well because the construction of the rear port is ropey. I haven't checked the bracing yet but I'll try this tonight.
Last edited:
my point exactly, with no information as to the distance to adjacent boundaries it's meaningless.
if your thinking that the close proximity to the port succeeds in eliminating interference, that's not totally correct. at that frequency even 1/4 and 1/2 wave intervals can mess up the measured response.
Both measurements were taken in the same location, outside with the speaker placed on 25mm felt, on top of shallow gravel. The nearest hard boundary (other than the floor) was the wall of my house which was c.7m away
ok it's apparent that care was taken to eliminate variables, all i can suggest is that the path length difference between a rear and front firing port may be the cause.
and just to insure it's not me misreading the first measurement was taken at .5 cm and the second was at 5 cm
and just to insure it's not me misreading the first measurement was taken at .5 cm and the second was at 5 cm
so in one measurement the mic was behind the box and in the other it was in front (same side as driver)?
if so then i would say the notch is due to phase difference at the measurement locations.
if so then i would say the notch is due to phase difference at the measurement locations.
2 separate measurements on 2 separate occasions of 2 different speakers. Same speaker location. Different mic positions (one 5cm from rear port, one 5cm from front port)
This is the best i can do with word 🙂
Shared album - Billy Jukes - Google Photos
Grey box below the speaker is felt
The mic was a B&K 4189 on a 2250 G4
Shared album - Billy Jukes - Google Photos
Grey box below the speaker is felt
The mic was a B&K 4189 on a 2250 G4
Last edited:
i mentioned in post #8 that path length difference was at play, and thanks for the illustration btw.
how do i go about quickly explaining the rather complex subject of phase cancellation...?
it has to do with instantaneous pressure at two different locations made more complex due to the fact that there's two sources, the driver and the port and that the distance between them is different...in the case of the front facing port when the loudspeaker's acoustic output decreases as it approaches resonance the port takes over but both are producing positive pressure (the other name for bass reflex design is the phase inversion enclosure).
so now relocate the port to the rear (along with our measurement location) and imagine driver/port operation, your two positive sources are now 180 deg apart how does that not result in a null at the tuning frequency?
how do i go about quickly explaining the rather complex subject of phase cancellation...?
it has to do with instantaneous pressure at two different locations made more complex due to the fact that there's two sources, the driver and the port and that the distance between them is different...in the case of the front facing port when the loudspeaker's acoustic output decreases as it approaches resonance the port takes over but both are producing positive pressure (the other name for bass reflex design is the phase inversion enclosure).
so now relocate the port to the rear (along with our measurement location) and imagine driver/port operation, your two positive sources are now 180 deg apart how does that not result in a null at the tuning frequency?
Thanks for the explanation Turk but I understand phase cancellation.
The distance from the centre of the driver to the ground is roughly 22cm.
The distance from the port to the mic is 5cm,
The distance from the centre of the 12 inch driver to the mic is 70cm (around the cabinet)
This results in a path length difference of 65cm which is half the wavelength of 263Hz (343/(0.65*2)).
In theory, I shouldn't get a cancellation dip below this main cancellation without other hard surfaces nearby, and even so, the reflected energy would be considerably lower.
The dip is at 87Hz therefore Im looking for a pathlength difference of roughly 2m to cause this dip. There's nothing that I can think of in this setup that has such a pathlength.
The distance from the centre of the driver to the ground is roughly 22cm.
The distance from the port to the mic is 5cm,
The distance from the centre of the 12 inch driver to the mic is 70cm (around the cabinet)
This results in a path length difference of 65cm which is half the wavelength of 263Hz (343/(0.65*2)).
In theory, I shouldn't get a cancellation dip below this main cancellation without other hard surfaces nearby, and even so, the reflected energy would be considerably lower.
The dip is at 87Hz therefore Im looking for a pathlength difference of roughly 2m to cause this dip. There's nothing that I can think of in this setup that has such a pathlength.
Last edited:
yes a bass reflex port is 180 deg out...and the distances we're talking about corresponds to 1/4 wave @ 87 hz ...give or take 30cm or a foot , no?
i'd venture if you moved away from the ground plane, eliminate that boundary the notch would still be there but the Q would be narrower.
i'd venture if you moved away from the ground plane, eliminate that boundary the notch would still be there but the Q would be narrower.
Last edited: