Port output realive to driver output (unusual question)

[fatmarley]

A while ago someone took a measurement of some speakers, tuned them 90 degrees and took another measurement (It was in a garage IIRC). I'm pretty sure the 90 degree measurement was down around 6db all the way down to the lowest frequencies (It was a long time ago though, so I could be wrong)

Has anyone measured a ported speaker, turned It 90 degrees and taken another measurement? Just wondered If port output would be down relative to the driver output If you put a port In the side?. If It made no difference where you put the port, the port output should stay at the same level when you turn the speaker 90 degrees but the higher frequencies should drop.

Yes I know low frequencies are omnidirectional and It shouldn't make any difference but has anyone actually taken some measurements to prove this fact?

[Sydney]

What Measurement approach was used?
In a Garage could pose issues:
https://www.klippel.de/fileadmin/kli...rts%282%29.pdf

[dlr]

The output at the lowest frequencies won't change from port or driver, but at some point two things will be a factor, the driver directionality and the delay. The location of the port will affect this.

The output target for port design is the listening position in the front hemisphere. If the port is in the back there will be some amount of additional delay relative to the front vs. a front port with much smaller delay. At 90 degrees the delay from the port changes, it's lessened for the rear located port of course. The front driver delay may change if the mic distance from the driver changes, but that will likely be far less of a change that for the rear port case.

The other consideration is below the port tuning. Below it the system response will tend toward a dipole response for a rear port, but since the port opening is small, it will be partial and will depend on typical dipole considerations, primarily the distance from driver to port in the front hemisphere with the side having the added change of distance to mic.

At the lowest frequencies the driver output will be dropping as will the port output due to that since the port's source is the driver. This is the normal low end rolloff of a ported system. At the limit, the port is essentially an acoustic short-circuit and the distance from front to rear becomes inconsequential.

There's no way to know exactly what the difference will be without measuring or having software that can model it.With a rear mounted port you could measure port and driver individually and combine them in software with different delays, but at typical listening distances the impact should be reduced. Measure at 1m front and side vs. something like 8' and you have the same issue that you have with diffraction where the response changes with distance.The floor influence will be a factor as well. I would think that the change in summed and perceived response would be in the area of port tuning since it will be at its maximum in and around that point, mid-bass maybe, not low bass.

dlr

[billfitzmaurice]

A while ago someone took a measurement of some speakers, tuned them 90 degrees and took another measurement (It was in a garage IIRC). I'm pretty sure the 90 degree measurement was down around 6db all the way down to the lowest frequencies

That's quite possible, depending on the distance from the speaker to the mic. At a close distance the inverse square rule would be in full effect. Measured in both instances from a few meters away, not so much.

[Whitneyville1]

I think I remember this discussion before, with the "conclusion" being you might be able to model/measure it, but at real world listening distances, you can't hear it. I won't swear to it because of the room factor(s) and so-on, but the omni-directional effect of low frequencies, this is a logical answer to me. Side mounted passive radiators, I think may support this.

[fatmarley]

Thanks guys. I'll have to get the measuring gear out to see how things look In reality.

[dlr]

Thanks guys. I'll have to get the measuring gear out to see how things look In reality.

Bill's reply is more likely the accurate one, though I think for a rear mounted port there will also some influence as I suggested. The problem is that with measuring, it will be difficult to separate or isolate issues. For most low frequency measurements, you would take a close-mic woofer and port measurement, though front mounted ports are a bit problematic due to proximity to the driver whereas the rear port is much better isolated from the driver for the close-mic situation. But for the rear mounted case adding the two is usually done without adding any delay for back-to-front distance as far as I recall, but then dipoles usually operate across a much broader spectrum. For a dipole, you would have to include this distance since it's critical to dipoles. In using John K's dipole spreadsheet I don't recall if loss with distance was part of it. That distance is used to determine the delay from driver rear output to front output. It's been quite a few years since I worked with it for my dipole system.

dlr

[johnk...]

Bill's reply is more likely the accurate one, though I think for a rear mounted port there will also some influence as I suggested. The problem is that with measuring, it will be difficult to separate or isolate issues. For most low frequency measurements, you would take a close-mic woofer and port measurement, though front mounted ports are a bit problematic due to proximity to the driver whereas the rear port is much better isolated from the driver for the close-mic situation. But for the rear mounted case adding the two is usually done without adding any delay for back-to-front distance as far as I recall, but then dipoles usually operate across a much broader spectrum. For a dipole, you would have to include this distance since it's critical to dipoles. In using John K's dipole spreadsheet I don't recall if loss with distance was part of it. That distance is used to determine the delay from driver rear output to front output. It's been quite a few years since I worked with it for my dipole system.

dlr

Yep, need to include delay for rear port. But it really only matters below the cut off. Port output at resonance is max where as driver output is minimum. Above resonance port output drops off rapidly. Below resonance both port and driver approach the same level so phase and any delay is more important.

[dlr]

Yep, need to include delay for rear port. But it really only matters below the cut off. Port output at resonance is max where as driver output is minimum. Above resonance port output drops off rapidly. Below resonance both port and driver approach the same level so phase and any delay is more important.

Thanks for chiming in. I was thinking that below resonance the port output might be too restricted for it to have much impact, but since the woofer is also rolling off, the volume displacement of both the driver and the port would likely be similar at the lower limit as you point out. Any time I've seen measurements that include a rear mounted port I don't recall that anyone added delay to the port measurement, though it should be to be completely accurate. It would be interesting to see what the impact is. Seems to me that it would actually be most important at the two F3 points of the port since the woofer and port volume displacement might be closest there with more influence at the upper F3 due to the shorter wavelength. If the delay makes a significant difference at the listening position, then at 90 degrees part of the change will be due to reducing the delay to near zero due to equal distance from the test point. So you have change due to distance in both relative magnitude and phase of driver and port.

I will suggest one thing and that is that although the driver output is minimum near port resonance, the driver output is not always maximally reduced, apparently due to oddities of box dimensions and driver/port location. Some port responses have a bit of a "plateau". Stereophile measurements occasional showed this.

dlr