monitor port design -Harvey?

[1tonio]

Harvey or anyone?
How does the design of a monitors port design effect the "sound".
As an example:
1. Dynaudio Bm5A has rear port design with a rectangluar opening at the of the cabinet.
2. Adams A7 has a port in front with a round opening.

I realise its design is to match the complete system, but are there any specific general reasons for the different types?

Are there any acoustical response's due to these designs to be considered ?

Inquiring minds want to know.

On a side note, how much do you think the importing fees are included with the final purchase price?

T

[solo2racr]

Port shape has no bearing on the sound coming from the port. It's all about the cubic size of the port (area x length). The only time the area would be a problem is if the port is to small and the port "whistles" from the velocity of the air passing through it. Port size and length are figured from the Qts of the speaker and the enclosure. I would think that, in a monitor, that a Qts of .7 (flat response) would be favorable to a more common Qts of .9 to 1.1 ( increased bass output at the expense of low freq extension ) in a consumer speaker.

The difference in a front port VS. a rear port would be that the rear port is meant to use the wall behind it to help "amplify" the bass freq. coming from it. With the front port, ya gets what ya gets

[1tonio]

Qts? Could you expand on that?

Not sure I follow on this. Wouldn't the amplification have a delayed effect by reflecting off the wall on a rear port>
A front port would seem to be a combination with the direct waveform from the drivers kinda mixed in?

The difference in a front port VS. a rear port would be that the rear port is meant to use the wall behind it to help "amplify" the bass freq. coming from it. With the front port, ya gets what ya gets

[dementedchord]

things to keep in mind....

the central purpose of the port is not to generate sound... but rather allow the cab to "breathe" so the pressure inside the cab doesnt inhibit the woofers movement...

the little bit of sound from the port is out of phase relitive to the woofer....


the distance the sound has to travel to get to the woofer determine what freqs are affected and a rear port is obviously longer and tends to disapate into the room...


if the wall behind is not properly treated there's a booming effect as it reflects off the hard surface...

[dementedchord]

double post...sorry

[Harvey Gerst]

Before I try to answer your question, I feel that it's important that I review some basic laws of physics about how real speakers work in the real world. I'll skip the math and techy stuff and try to present it in easy to understand terms. It all boils down to one basic axiom,and four smaller axioms:

1. The basic axiom: "Everything in designing a speaker is a trade off."

This can also be stated as:

"There no such thing as a free lunch."

or

"You want something; what are you willing to give up to get it?"

These are the four smaller axioms:

1. "Small won't get you big, without some tricks."

2. "To get good bass response, you hafta move the air in the room - a lot of air."

3. "As the frequency goes up, the dispersion narrows."

4. "As the frequency goes down, the bass will roll off dramatically at some point, depending on basic axiom 1."

The problem:

Think of the air in a room as a big bale of loosely packed cotton. You want to move the whole bale of cotton, but all you have is a small stick. (The "stick" is your speaker.) The stick won't move the whole bale; all it will do is poke into the bale and move the cotton near the stick. You need what's called "better coupling". Speakers are like small sticks; they don't do a good job of moving large objects. You need a better way to tie the speaker to the air in the room. You can:

1. Use a bigger stick: You can use a bigger cone, but there are limits to how big you can make it, and drawbacks start to outweigh benefits as the size goes up. if the stick is as big as the room, how do you move the stick? The bigger cone has to be heavier and stiffer, so that it doesn't flex as it moves. Heavier means sluggish. That limits the cone to slower starting notes like pipe organs. Stiffer, yet light cones means more exotic materials (i.e., expensive") like Kevlar, Aluminum, and Graphite Composites.

2. Move the stick more: This was the principle of the acoustic suspension speakers; make the speaker move longer distances to push more air. Unfortunately, it required a lot heavier cone and a long voice coil which dropped the efficiency way down and made the system sluggish as hell.

3. Use a lotta little sticks: The Bose approach, where you use multiple bass drivers to simulate the cone area of a larger driver. The problem is still back to basic physics; even though you equal the area of a large speaker, the cone diameter of each speaker determines one of the low frequency cutoff points, and you pay for the bass boost with phase cancellations and beaming at higher frequencies.

4. Taper the stick from small to large: It's called an acoustic transformer, and that's how horns work. They transform a high energy, large motion, speaker cone to to a lower energy, less motion, signal appearing at the horn mouth that couples better to the air in the room. The problem with low frequency horns is that the mouth of the horn has to be huge and (like every transformer), the throat of the horn (or the transformer's primary) can easily saturate when overdriven.

5. Tie a second stick to the first stick: This is what a Helmholtz resonator does; it can either be done with a tuned hole in the box to move more air in a very small frequency range just below where the speaker starts to roll off (Axiom 5), or by using any tuned mass (like a passive radiator) to move air in that range. The efficiency of this port is tied to a lot of other factors, including cabinet volume.

It's the same principle as blowing across a Coke bottle to produce a note. The air mass in the box and the opening (and duct length) combine to tune the note to a desired frequency, usually just below where the response drops off naturally - around the -3 dB point. A passive radiator serves the same purpose as a tuned port.

6. Use eq to boost the bass and fix problems: That works ok, but only up to a point. You can't fix room nodes electronically, since those are caused by bass buildup over time, and they're different for each room. You can do some slight boost to help a steady dropoff, but you quickly run out of power (or speaker capacity) at very low frequencies.

A brief side trip about Axiom 3: High Frequency Dispersion: This dispersion problem is true of microphones as well as speakers. Even with a perfect omni measurement mic, you have a choice of flat response on-axis, but the high end will drop off as you move off axis, or you can have flat response off-axis, but the high end will increase as you move on axis. With speakers, it's the same thing; as you raise the frequency, the beam narrows and you lose highs as you move further off-axis.

So, what have we learned? Basically, just three things; that (below a certain frequency), a speaker needs help to produce low end, and that bigger is generally better, but not without some compromises. And, we know that above a certain frequency, the high end dispersion begins to narrow as the frequency goes up.

As far as front or back porting, I like front porting myself, but the frequency is probably low enough so that it's non-directional anyway.

[mshilarious]

There needs to be a book on my shelf, "The Stuff Harvey Gerst Knows That You Don't", and everybody needs a copy

[solo2racr]

Originally Posted by 1tonio

Qts? Could you expand on that?

Qts is the total system Q. Qt is the calculated resistance of the driver by mechanical ( Qm ) and electrical ( Qe ) means. Qts is calculated from the cubic space inside the enclosure (air resistance) and the Qt of the woofer.

[solo2racr]

Originally Posted by dementedchord

the central purpose of the port is not to generate sound... but rather allow the cab to "breathe" so the pressure inside the cab doesnt inhibit the woofers movement...

What you are describing is a situation where the woofer is operating below f3. IE: below the freq the port was designed for. Below this freq, the woofer essentially has no support from the air behind it. It is like it is operating in a free air environment. Trying to drive a woofer below f3 is simply asking for the voice coil to fry. With no air support behind the cone, the electro-mechanical action (voice coil and magnet assembly) tries it's best to control the cone. In trying to move the cone in the opposite direction from being over extended, the motor assembly gets hot from all the extra work and expands, The voice coil starts to rub on the magnet and shorts out. Woofer blown!

[Harvey Gerst]

Originally Posted by solo2racr

What you are describing is a situation where the woofer is operating below f3. IE: below the freq the port was designed for. Below this freq, the woofer essentially has no support from the air behind it. It is like it is operating in a free air environment. Trying to driver a woofer below f3 is simply asking for the voice coil to fry. With no air support behind the cone, the electro-mechanical action (voice coil and magnet assembly) tries it's best to control the cone. In trying to move the cone in the opposite direction from being over extended, the motor assembly gets hot from all the extra work and expands, The voice coil starts to rub on the magnet and shorts out. Woofer blown!

Yup, and at the port resonance, the port (and the air mass in the box) actually provides most of the sound output, with the woofer barely moving at all. Below that point, things get really shitty really fast. The speaker is "un-loaded" below port resonance and free to move as far as it can. In a closed system (a sealed acoustic suspension system, with no port, for example), the trapped air in the box provides damping for the cone motion.

[1tonio]

woo doggy !!

Harvey thank you so much for an in depth -mini manual

While I wrap my head around your eloquent explanation, have you listened to either of the models I mentioned? If so ,do you have any opinions on them?

I'll be back. BTW , I'm using Event PS8's(front ports) and with some mids & HF support with Klipsch Heresys' (3way w/ mid horns).

Even though I used the Adam A7 & Dyn BM5A as an example, I've had my trigger button on some Dyn BM6A's(rear rectagular ports also). Though I heard some top dogs says that they are scooped in the mids.

T

[Almy]

Man. Ok Havey, it's high time you wrote a book. "Everything I needed to know about gear I learned from Harvey." I love the way you put things in perspective without losing the 'why' and without only showing the 'what.'

[Harvey Gerst]

Hey, ya gotta remember that my first real job (if you don't count working at a supermarket) was at JBL during its "Golden Age". My first love is, and always will be, speaker design.

I don't listen to many monitors at AES or NAMM these days; I find somethings I don't like in a lot of them. Plus at trade shows, I get people mad at me because I'll point out a problem they didn't hear for most of the show till I point it out.

At this last NAMM, one of the new Tascam monitors they were demoing had a polarity reversal problem. They didn't hear it till I pointed it out - on the last day of the show. And they argued that there was no problem, till I offered to bet them 20 bucks they were wrong. They wouldn't take the bet. I left them arguing over my bet.

At another booth, I listened to some pretty good monitors from Canada, except one side was way too bright. I checked the playback chain and I didn't see any different eq, and they assured me I was wrong. Then I asked if these were powered monitors with separate level controls? They proudly answered, yup. I said, then somebody has been messing with the treble level control, but no, they said that couldn't be. Guess what? Speaker on the left was set to "flat", speaker on the right had 10 dB of tweeter boost!!

How can these people show off their monitors, when they're apparently deafer than I am?

I've been threatening to design a decent monitor for the last 10 years; maybe I'll get around to it this year.

[solo2racr]

Originally Posted by Harvey Gerst

I've been threatening to design a decent monitor for the last 10 years; maybe I'll get around to it this year.

Since I built my first 2 way system 20 years ago, I've been wanting to do the same thing. Getting a good match of drivers is the hard part

[1tonio]

OK, so basically the port is an crutch or assisting mechanism to push the low freq into the enviroment. Based on the design, it could enhance the output of LF or it could totally screw up the response.

If LF's are more omni directional, they shouldn't effect the time phase between itself and higher frequencies? But isn't low freq's slower in nature?
Or is it more about the initial LF only, since LF builds up with time?

This is what I don't understand with front vs rear ports. Even LF have some directionality no? especially freq's close to the crossover point. Or are ports normally designed in a small range under the drivers rolloff point(I'll throw in 100hz) that there would be minimal driectionality.

T

[robin watson]

Originally Posted by mshilarious

There needs to be a book on my shelf, "The Stuff Harvey Gerst Knows That You Don't", and everybody needs a copy

Or as the late, great Roy Scheider might have said:

"You're gonna need a bigger shelf."

[robin watson]

Originally Posted by Almy

Man. Ok Havey, it's high time you wrote a book. "Everything I needed to know about gear I learned from Harvey." I love the way you put things in perspective without losing the 'why' and without only showing the 'what.'

I think you're going to have to change either the proposed title of the book, or the author.

[solo2racr]

Originally Posted by 1tonio

OK, so basically the port is an crutch or assisting mechanism to push the low freq into the environment. Based on the design, it could enhance the output of LF or it could totally screw up the response.

If LF's are more omni directional, they shouldn't effect the time phase between itself and higher frequencies? But isn't low freq's slower in nature?
Or is it more about the initial LF only, since LF builds up with time?

This is what I don't understand with front vs rear ports. Even LF have some directionality no? especially freq's close to the crossover point. Or are ports normally designed in a small range under the drivers rolloff point(I'll throw in 100hz) that there would be minimal directionality.

T

The Thiele & Small parameters of a woofer will pretty much dictate the type of enclosure that is needed for that woofer. (boiling it down to sealed VS. ported and ignoring all the variables of each design) You can't just add a port to a seal enclosure to get more bass and vice versa. The port (or lack of), enclosure size, and woofer are designed as a system. Change one thing and the system no longer works correctly.

Ports are tuned to a freq. that is close to the woofers "rolloff" (free air resonance). Not much of any kind of directional sound at these low frequencies.

As the freq rises, the enclosure design has lesser and lesser effect on the woofer. By the time the woofer gets to the xover point (say 2500hz) the enclosure has no bearing on the woofer.

This part I came up with on my own so if it is in error, please correct me.
It would seem that freq has no bearing on the time a sound arrives from point A to point B. I say this from the standpoint that Mach is a constant (there ARE a few variables like air density) and isn't influenced by freq. An airplane will break the sound barrier at X mph regardless of the freq it is putting out. If it were, aircraft mfgs would build supersonic aircraft that resonated at a freq. that was favorable to mach+ speeds.

Sounds reasonable to me

[Harvey Gerst]

1tonio said:
OK, so basically the port is an crutch or assisting mechanism to push the low freq into the enviroment. Based on the design, it could enhance the output of LF or it could totally screw up the response.

It's an assisting mechanism; every woofer will roll off at some point, usually after the resonant frequency. If you tune the port slightly below that frequency, you can bump the bass and extend the frequency response a little lower. The tradeoff is that below the port resonance, the speaker is unloaded and trying to get more even bottom end can mess up the speaker.

If LF's are more omni directional, they shouldn't effect the time phase between itself and higher frequencies? But isn't low freq's slower in nature?

The port is usually tuned around 50 Hz or lower, so you don't really hear where that very low stuff is coming from.

Or is it more about the initial LF only, since LF builds up with time?

You're talking about room nodes. Standing waves build up over time, and the frequencies are dependent on the room dimensions. It's a separate subject from speaker port design or port placement.

This is what I don't understand with front vs rear ports. Even LF have some directionality no?

Most ports are tuned in the 50 to 30 Hz range. Low frequency directionality is generally considered to end around 80 to 100 Hz. At 80 Hz and below, you can't tell where the sound is coming from.

Especially freq's close to the crossover point. Or are ports normally designed in a small range under the drivers rolloff point(I'll throw in 100hz) that there would be minimal directionality.

The crossover point of most two way nearfield monitors is around 2,500 Hz. Most lower priced nearfield monitors start running into problems below 50 or 60 Hz.

[COOLCAT]

Originally Posted by Harvey Gerst

This is what I don't understand with front vs rear ports. Even LF have some directionality no?

Most ports are tuned in the 50 to 30 Hz range. Low frequency directionality is generally considered to end around 80 to 100 Hz. At 80 Hz and below, you can't tell where the sound is coming from.

great thread btw..

yea, the low freq directionality topic always seemed to be a debate, in my mind anyway, until I came across personally what Harvey mentioned. the directionality ends around 100-80hz. it was the first time the topic was clarified.

I checked this out with a single tone generator and its true, very cool.....

selecting a tone, 150hz or higher, its obvious which speaker the bass was coming from...

then gradually selecting lower test tones, 125, 100, 80, 75, 50.. the sound mysteriously went omnidirectional. right around 80-100hz.

great fun ..lab project thing if your into that stuff.

very cool like when you press the "mono" button and everything sucks to the center...even though the speakers haven't physically moved.

The book.."The Stuff Harvey Gerst Knows That You Don't"....thats good ms....so true.

[1tonio]

AHA!!


Thats what I was looking for !!! Harvey, you have been most gracious with this information.

solo2racr, you've have a great help also.

NOw to save up a few more $$$ for the ultimate set.

T

[thajeremy]

Originally Posted by Harvey Gerst

I've been threatening to design a decent monitor for the last 10 years; maybe I'll get around to it this year.

did you say something about another Group Buy???

[Harvey Gerst]

Might be time for a bump.

[dementedchord]

hey harvey... what's the deal with transmission lines???

[Harvey Gerst]

Originally Posted by dementedchord

hey harvey... what's the deal with transmission lines???

Think of a transmission line speaker as either a bass reflex speaker (with a very long port), or as a stuffed organ pipe. Above its resonance, it acts like a sealed enclosure (which absorbs all back resonances), and at resonance, it puts out a huge signal. With a good driver, they can sound extremely smooth, but they get rather large quickly when you need the resonance to be low.

Typically, transmission line designs are based on 1/4 wavelength of the lowest desired frequency. Oddly enough, even though this design has been around for many years, it wasn't until 1965 when they actually figured out everything it was doing. Basically, the idea is to damp as much of the resonances off the back of the speaker as possible, smoothing out the response, then use the 1/4 wavelength design to damp the main resonance, producing a smooth frequency response and a smoother impedance curve.