Front vs rear port?

[Swerd]

Strangely enough, his port measurement stops right around the frequency range where a peak could develop. Is there a peak at 700 or 800Hz? How would we know?

Good point. I chose that plot as an example of a properly performing port, but I didn't notice the plot stopped at 500 Hz.

 

[Swerd]

Strangely enough, his port measurement stops right around the frequency range where a peak could develop. Is there a peak at 700 or 800Hz? How would we know?

Beave's sharp eyes noticed the absence of frequency response info above 500 Hz for the port (red trace) on these speakers (B&W 805 D3). Thanks . The more I thought about it the more I wondered why. B&W uses large diameter flared ports covered with highly noticeable dimples, said to "smooth the flow of air". Why would a magazine review of the speaker omit information that could illustrate that? Is it no better than a non-dimpled port? Is it worse?

Any way, I looked for another example of a ported 2-way speaker where the port resonance (or noise) is shown up to 1000 Hz, and it is acceptably low. Here the red trace shows the port noise is more than 25 dB quieter than the speaker.

 

[TLS Guy]

Beave's sharp eyes noticed the absence of frequency response info above 500 Hz for these speakers (B&W 805 D3). Thanks . The more I thought about it the more I wondered why. B&W uses large diameter flared ports covered with highly noticeable dimples, said to "smooth the flow of air". Why would a magazine review of the speaker omit information that could illustrate that? Is it no better than a non-dimpled port? Is it worse?

Any way, I looked for another example of a ported 2-way speaker where the port resonance (or noise) is shown up to 1000 Hz, and it is acceptably low. Here the red trace shows the port noise is more than 25 dB quieter than the speaker.

This should not surprise. A port is an open pipe in organ parlance, and so it will contain a half wavelength at resonance.

It will radiate even harmonics at decreasing intensity.

So the port will resonate at a frequency of the speed of sound divided by the port length X 2

So if we take the speed of sound as 1,100 ft/sec. A 5.5" port will resonate at 1000 Hz.

So if the speaker inside the box is radiating sound at 1000 Hz the port will resonate.

I do put damping near the port and behind the driver. The only way to stop it totally is to have the driver crossed over well before port resonance. That is not going to work for two ways.

So we are talking here about something inevitable and expected.

However it does have to be considered a disadvantage of ported speakers, although not a big one.

 

[Swerd]

This should not surprise. A port is an open pipe in organ parlance, and so it will contain a half wavelength at resonance… …

Thanks for that helpful explanation, especially the illustrations.

However, it makes the absence of port frequency response data for those B&W 805 D3 speakers more puzzling because the crossover frequency of these 2-way speakers, 3300 Hz, is well above the expected port resonance frequency of about 1000 Hz.

Something in John Atkinson's text (Measurements page, below figure 2) does partially address this question:

"The impedance-magnitude trace suggests that the large, flared port on the front baffle is tuned to 50Hz or so. The minimum-motion notch in the woofer's nearfield output (fig.3, blue trace) occurs at 48Hz, and the port's response (red) peaks between 35 and 90Hz in classic fashion. (I haven't plotted the port's output above 500Hz, as the measurement at higher frequencies was contaminated by the woofer's output.)"​

What Atkinson says in print is almost always factually correct, but at times I've found that what he doesn't say can be more important. Maybe I am being too skeptical, but I feel like I need a lawyer to help me read statements such as I quoted above. Perhaps (this is entirely my speculation here) this statement was the product of negotiations between B&W and Stereophile about what to say about an undesirable result of Atkinson's measurements.

 

[TLS Guy]

Well there are three types of sonic radiation from a port, one wanted the others not.

The first is the intended bass radiation to augment the driver.

The next is port resonance determined by the physical dimensions of the port.

The third is just plain radiation of frequencies radiated by the back of the cone above tuning.

This latter can really only be controlled by port position damping and most effective of all turns in the port.

 

[Beave]

Thanks for that helpful explanation, especially the illustrations.

However, it makes the absence of port frequency response data for those B&W 805 D3 speakers more puzzling because the crossover frequency of these 2-way speakers, 3300 Hz, is well above the expected port resonance frequency of about 1000 Hz.

Something in John Atkinson's text (Measurements page, below figure 2) does partially address this question:

"The impedance-magnitude trace suggests that the large, flared port on the front baffle is tuned to 50Hz or so. The minimum-motion notch in the woofer's nearfield output (fig.3, blue trace) occurs at 48Hz, and the port's response (red) peaks between 35 and 90Hz in classic fashion. (I haven't plotted the port's output above 500Hz, as the measurement at higher frequencies was contaminated by the woofer's output.)"​

What Atkinson says in print is almost always factually correct, but at times I've found that what he doesn't say can be more important. Maybe I am being too skeptical, but I feel like I need a lawyer to help me read statements such as I quoted above. Perhaps (this is entirely my speculation here) this statement was the product of negotiations between B&W and Stereophile about what to say about an undesirable result of Atkinson's measurements.

It's strange. I can't figure out why the port output measurement would be contaminated by the woofer's output in this particular speaker; while in several other front-ported speakers with similarly close woofer and port placement, he was able to get a clean measurement to 1kHz.

It would be a good question to ask him.

 

[TLS Guy]

It's strange. I can't figure out why the port output measurement would be contaminated by the woofer's output in this particular speaker; while in several other front-ported speakers with similarly close woofer and port placement, he was able to get a clean measurement to 1kHz.

It would be a good question to ask him.

The answer is simple and obvious. The port is very large in diameter. The bigger the hole the more sound will come out from inside the cabinet.

 

[Beave]

The answer is simple and obvious. The port is very large in diameter. The bigger the hole the more sound will come out from inside the cabinet.

But then why wouldn't he should show that in his measurement, as he does for other similar speakers?

I get the impression that he's saying he cuts the measurement off at 500Hz because his measurement is getting contaminated by the woofer's output firing outward, not inward into the cabinet and then out through the port.

 

[shadyJ]

But then why wouldn't he should show that in his measurement, as he does for other similar speakers?

I get the impression that he's saying he cuts the measurement off at 500Hz because his measurement is getting contaminated by the woofer's output firing outward, not inward into the cabinet and then out through the port.

I believe TLS Guy is right. Depending on where the port is situated and its configuration, you can hear the output of other drivers through the port. Atkinson may have cut off the measurement at 500 hz because the port wasn't actually generating the output being recorded above that; the woofer was. Myself, I would have left it in the measurements- it is energy being generated by the speakers, who cares how that happened. But you have to keep in mind that sound that leaks through ports like that isn't going to be audible, unless you listen to the back of the speaker instead of the front.

Personally I do prefer listening the speakers from the rear, it makes for an extremely warm sound that I prefer. People who listen to speakers that are facing them do not have an appreciation for good sound and are not true music lovers!

 

[Beave]

I believe TLS Guy is right. Depending on where the port is situated and its configuration, you can hear the output of other drivers through the port. Atkinson may have cut off the measurement at 500 hz because the port wasn't actually generating the output being recorded above that; the woofer was. Myself, I would have left it in the measurements- it is energy being generated by the speakers, who cares how that happened. But you have to keep in mind that sound that leaks through ports like that isn't going to be audible, unless you listen to the back of the speaker instead of the front.

Personally I do prefer listening the speakers from the rear, it makes for an extremely warm sound that I prefer. People who listen to speakers that are facing them do not have an appreciation for good sound and are not true music lovers!

But this is a front ported speaker! If this is sound from the woofer going through the cabinet and then coming out the (front!) port, then it could be audible. And it would be delayed slightly versus the direct sound from the woofer.

 

[KEW]

The answer is simple and obvious. The port is very large in diameter. The bigger the hole the more sound will come out from inside the cabinet.

Isn't that all the more reason to measure it? ...to ensure it is not excessive?

 

[TLS Guy]

I'm detecting some surprise about all this.

Unwanted sound emanates from all box speakers.

We have discussed the port but a more major source is direct transmission through the speaker walls.

Now sound travels actually better through liquids and even better through solids than air. The more tightly packed the molecules the better. It also travels faster in water and faster in solids than water. Since frequency is the speed of sound divided by wavelength, if the speed is higher there is an increase in pitch as it traverses the solid.

If you put a speaker totally inside a closed box you would still hear it no matter how perfectly you braced the cabinet.

So what form does sound transmission take through a speaker cabinet. We talk a lot about panel resonance. This is when the sides are at a resonant point, resulting in a single or multiple peak rises in output from the cabinet walls.

However sound also passes though the walls by direct molecular transmission. The tighter the molecules are packed (denser) the more efficient the transmission. This transmission has nothing to do with resonance and is an inherent property of the material of the cabinet walls.

So lets consider a closed box and forget a port at this time. Lets also assume panel resonance is perfectly controlled for this discussion. Some of the sound from the rear radiation will be adsorbed by the damping. Some will undergo multiple reflections and be reduced in strength. But some will excite molecular collisions and pass through the walls. As it passes it increases in speed and pitch. As the molecules on the outside surface set the air molecules vibrating the sound slows and original pitch is restored. Unfortunately all frequencies are not treated equally. So there is a band pass effect. This is audible and a source of coloration in speakers that is difficult to control, and never perfectly controlled.

Unfortunately ported speakers can only be lightly damped otherwise the speaker does not work.
Sealed and TL speakers are much more heavily damped and have less of this coloration.

It gets worse. Reflections from the rear radiation are transmitted through the loudspeaker cone more easily than the cabinet walls, and are a definite source of coloration in speakers. Again this is worse for ported speakers than other designs. In addition this is the reason why different cone materials do have a slightly different timbre. The band pass frequencies of various cone materials are different. All of this direct sound transmission through solids tends to upset mid range clarity preferentially.

All this has been a big push for planar loudspeakers and open backed speakers. The cry of these enthusiasts is "Get rid of the box!'" Well they have a point but in ridding themselves of the box create other ills. I will say this however that this is a major reason for the delicate transparent sound of the best electrostatic speakers.

This is a problem I pondered a lot over the years. So reducing this has a lot to do with the way my speakers are designed.

The mid range line enclosure apart from the front panel is highly isolated. The sides by the angled side panels with heavy damping. This also reduces cabinet width and brakes up the time paths of the reflections.

The top, rear and bottom radiations are reduced by having to traverse the bass line.

So the only significant direct transmission is though the front baffle, which is double thickness. There is no further measures to ameliorate this that I can think of.

This is effective though, and quite a few have thought the mid line to be an electrostatic panel.

 

[Beave]

That's a really long way of not answering a question!

 

[Alex2507]

You gotta hit that rear port if you expect to hear the brown note.

 

[TLS Guy]

That's a really long way of not answering a question!

The point I would have thought was obvious. The leakage from the port is just one of multiple sources of leakage of sound from the rear of the cone out through the enclosure.

So showing the port leakage would be misleading and NOT be a valid comparison by itself to the total leakage of the speaker under test and other similar speakers.

 

[Bucknekked]

I'm detecting some surprise about all this.
Unwanted sound emanates from all box speakers.
We have discussed the port but a more major source is direct transmission through the speaker walls.
Now sound travels actually better through liquids and even better through solids than air. The more tightly packed the molecules the better. It also travels faster in water and faster in solids than water. Since frequency is the speed of sound divided by wavelength, if the speed is higher there is an increase in pitch as it traverses the solid.

If you put a speaker totally inside a closed box you would still hear it no matter how perfectly you braced the cabinet.

OK, maybe others aren't going to admit it, but, I am willing to admit I'm surprised by a number of Mark's observations. I read stuff here on the AH because there are just enough nuggets that are worth learning to make up for some of the boneheaded stuff that goes on. This was a learning post for me.

All speaker boxes emanate unwanted sound. I don't suppose I ever thought about that much because I assume some of these dense, heavy and well made cabinets would be pretty inert for sound transmission.

Sound travels better through liquids and solids than air. I knew sound travels better in liquids because of all the National Geographic whale coverage. Thanks to the whales, I get sound through liquids. But sound through solids being easier than air? That's something new to me I'll have to go study up on a bit. I would not have guessed that.

Put a speaker in a completely enclosed box and you would still hear it. Another phenomenon I would never have thought possible. Again, I would have thought inert materials and bracing and damping would have made it a dead box. Apparently not so. This strikes me as a good DIY study project. If I had a wood shop I would want to give that a try. Sounds like a challenge.

That's 3 things to start the day I didn't know. Pretty good start.

 

[shadyJ]

Cabinets will resonate sound, sure, but the question then becomes one of audibility. On a reasonably well built cabinet, it is simply not an issue. You don't have to go to great lengths to reduce these resonances into insignificance either. For example, B&W's matrix cabinets and Magico's cabinets are overkill and definitely not worth the weight penalty. If cabinet resonances are a serious problem they would show up on the frequency response graph and impedance response.

 

[zieglj01]

Personally I do prefer listening the speakers from the rear, it makes for an extremely warm sound that I prefer. People who listen to speakers that are facing them do not have an appreciation for good sound and are not true music lovers!

With the leakage that tends to come through some front ports, I tend to get annoyed and distracted.

 

[Bucknekked]

With the leakage that tends to come through some front ports, I tend to get annoyed and distracted.

Oh, I understand that. When one of my ports has some leakage, it usually ruins the day

 

[Bucknekked]

Cabinets will resonate sound, sure, but the question then becomes one of audibility. On a reasonably well built cabinet, it is simply not an issue. You don't have to go to great lengths to reduce these resonances into insignificance either. For example, B&W's matrix cabinets and Magico's cabinets are overkill and definitely not worth the weight penalty. If cabinet resonances are a serious problem they would show up on the frequency response graph and impedance response.

A good point. Audible leakage, or just leakage?
Another DIY project for somebody with wood, time and an interest.