Cab designing, Xmax + port velocity wrt harmonics

[AlexanderB]

This is a combination of empirical observations and the science though. As I have understood it the science predicts an air velocity in the port, but it is empirically observed that values of theoretical velocity in excess of a given value result in chuffing. It doesn't matter if the real values of air velocity in the port are way larger or smaller than the theoretical ones, just so long as they are related so the empirical determination works.

Actually the thing that surprises me most about chuffing is the crude aerodynamics of most ports I've seen. If worried about the phenomenon surely both ends of the ports should have well considered surfaces. I presume it's different in high end kit. Haven't examined any in detail.

Agree 100%. It is VERY cheap to add a aerodynamic flare/flange to a round port and it can make great improvements.

 

[AlexanderB]

Actually all you need to do is look at the Fourier Transform of the output signal from an EB. It is predominantly 2nd harmonic.

How is it that you think we hear the fundamental when it's 10 dB or more down due to the superposition of pickup and cabinet response?

1. Distribution of energy across the tonal spectrum varies greatly between basses. My $2500 luthier five stringer has MUCH more sub 50 Hz output than most other basses I have tried, including my EMG loaded Fender Jazz Bass Plus V.

2. Fundamental objection is easy enough to shoot down. BT,DT, still have the T shirt.

For instance every modern, non-custom shop Fender and Squier I have heard and tried suffer from a bit of "dead spot" in the open E. Almost anyone (even our drummer!) can hear the 41 Hz part fade out after a few seconds, leaving the 82 Hz as thr lowest remaining frequence. Very notable.

If you have a processor controlled PA, try playing around with the subs. Mine are fairly flat to 32-33 Hz WITHOUT the processor. Set the crossover at low pass 48 dB/octave at 40 Hz, play the low B and C, and watch the fundamental in action. Then, swap to high pass 48 dB/octave at 50 Hz, and hear your bottom string without fundamental.

If you do not hear this, the issue is with the bass, gear or your ears, IMHO.

My personal bass sound is somewhere in between, definitely fundamental in the mix, but not massively boosted.

I used to run a bi-amped rig, active crossover (-24 dB/octave at 70 Hz) and 2x600W and a sub with -3dB around 32 Hz.
Heavy, limited SPL but lots of fundamental even at B and below. It has been over 10 years since I retired it, as also I was willing to trade performance for convenience.

 

[fdeck]

My design program includes an experimental feature to take the harmonic content of the bass waveform into account. I call it the "waveform based" analysis in the graphs:

Invalid Link Removed

Usual disclaimers: This program has not been vetted by professional engineers, and a single curve can't represent varying conditions.

 

[Rick James]

Actually the thing that surprises me most about chuffing is the crude aerodynamics of most ports I've seen. If worried about the phenomenon surely both ends of the ports should have well considered surfaces.

Where the port output is concerned the wavelengths are so long that there's nothing to be gained by improving the aerodynamics. Even a four inch radius flare is acoustically invisible to a twenty foot wavelength. Besides, you don't hear chuffing at 50 Hz. What you hear is more like 500 Hz, way above the port output frequencies, and the reason you hear it is it's caused by the friction between the air in the port and the port walls. Flaring the port ends reduces the noise caused by turbulence at the junction of the port and baffle, but it doesn't do anything to reduce noise created in the rest of the duct. You can reduce the noise by flaring the port ends, or you can eliminate it by keeping the port velocity low.

 

[seamonkey]

I see this repeated over and over, as being "a truth". If there were any scientific truth to this, please present any peer reviewed paper, scientific investigation, correctly executed blind a/b test or similar...

Perception of missing fundamental pitch by 3- and 4-month-old human infants. - PubMed - NCBI
"Perception of missing fundamental pitch by 3- and 4-month-old human infants."

http://rabbit.eng.miami.edu/students/ddickey/pics/AES_Convention_Paper_May_2008.pdf
"Harmonic and Intermodulation Analysis of Nonlinear Devices Used in Virtual Bass Systems"

There's actually quite a few papers on this.

Or Pick up a demo of Waves MaxxBass and try it yourself. It's actually quite interesting how much you can actually cut, yet still hear. it's natural in human hearing as evident by infant testing.

 

[seamonkey]

This is a combination of empirical observations and the science though. As I have understood it the science predicts an air velocity in the port, but it is empirically observed that values of theoretical velocity in excess of a given value result in chuffing. It doesn't matter if the real values of air velocity in the port are way larger or smaller than the theoretical ones, just so long as they are related so the empirical determination works.

Actually the thing that surprises me most about chuffing is the crude aerodynamics of most ports I've seen. If worried about the phenomenon surely both ends of the ports should have well considered surfaces. I presume it's different in high end kit. Haven't examined any in detail.

Big patent lawsuits by JBL and Bose about port designs. They are all about reducing chuffing and turbulence. Chuffing and turbulence are not "sound" but side artifacts of air flow.
Then again, some people think the pant flapping effect indicates how much bass is being produced. More than one company tries to sell turbulence as a bonus. Simple measurements tell the truth.

 

[agedhorse]

The simplest way to avoid port chuffing is to avoid the conditions that cause it in the first place. By paying attention to (and respecting) the velocity, is can be a non-issue.

 

[AlexanderB]

Perception of missing fundamental pitch by 3- and 4-month-old human infants. - PubMed - NCBI
"Perception of missing fundamental pitch by 3- and 4-month-old human infants."

http://rabbit.eng.miami.edu/students/ddickey/pics/AES_Convention_Paper_May_2008.pdf
"Harmonic and Intermodulation Analysis of Nonlinear Devices Used in Virtual Bass Systems"

There's actually quite a few papers on this.

Or Pick up a demo of Waves MaxxBass and try it yourself. It's actually quite interesting how much you can actually cut, yet still hear. it's natural in human hearing as evident by infant testing.

Well, I have tried a couple of such psychoacoustic devices (rack mount and software), and while it can be a neat trick for music intended to be played back on systems with limited bandwith. I also use "beater tones" from time to time, when I want to make the four stringer sound "deeper" than it actually is.
Still, neither of those things is a 1:1 replacement for the actual fundamental.
Like I said, been there, done that, have the T shirt.

Anyway, I think these later posts are of less interest to the OP.

 

[AlexanderB]

The simplest way to avoid port chuffing is to avoid the conditions that cause it in the first place. By paying attention to (and respecting) the velocity, is can be a non-issue.

Still, it is so often badly executed in low/mid price level commercial MI cabs...
At least, the more intense and deep bass in music the last 15 years or so has somehat increased the awareness of how different designs and priorities can affect bass enclosures.

 

[BadExample]

Perception of missing fundamental pitch by 3- and 4-month-old human infants. - PubMed - NCBI
"Perception of missing fundamental pitch by 3- and 4-month-old human infants."

http://rabbit.eng.miami.edu/students/ddickey/pics/AES_Convention_Paper_May_2008.pdf
"Harmonic and Intermodulation Analysis of Nonlinear Devices Used in Virtual Bass Systems"

There's actually quite a few papers on this.

Or Pick up a demo of Waves MaxxBass and try it yourself. It's actually quite interesting how much you can actually cut, yet still hear. it's natural in human hearing as evident by infant testing.

But did the infants dig the tone?

 

[JimChjones]

Flaring the port ends reduces the noise caused by turbulence at the junction of the port and baffle, but it doesn't do anything to reduce noise created in the rest of the duct. .

Agree with you all the way, it's just that I am bemused that so many ends I've seen are so simplistically treated since that must be a major generator of turbulence. In particular I look at commercial ports and see the outer end flared and inner end with a sharp transition which seems daft. It's way beyond my pay grade to consider how turbulence against port walls might be minimised other than to guess promoting laminar flow would help. It's jet engine designer grade stuff. But the ends must be huge contributors and well worth looking at, as I'm sure the serious designers do.

 

[reddesert]

Cone excursion;
I expect that Winisd (and similar programs I've used) show cone excursion in a beautiful graph at the power level you have entered as if the speaker was tested with a sine sweep gen. It would be very unusual for me to ever produce such a pure tone.
....
An example: winisd says a driver will reach it's Xmax at 41Hz at 100W. If we were to play an open E on a jazz bass we would have the fundamental 41Hz and some equally large 2nd and 3rd harmonics etc. As these frequencies are simultaneous, wouldn't the spectral distribution of power work in our favour to limit the excursion? Port velocity vs frequency is possibly similarly distributed?

fdeck's calculator, which he posted a link to above, shows a nice plot of the excursion as a function of frequency for both the idealized pure frequency case and a "waveform" model. I don't know all the details or how empirical the waveform model is, but it makes a good visual comparison. In the ideal pure sine case, the speaker cone doesn't move at all at the tuned port frequency, but I've never seen that when playing a real bass through an amp. The waveform model has a less extreme variation in excursion. It doesn't buy you extra Xmax at frequencies below the port tuning though.

Re: the missing-but-perceived fundamental issue, there are plenty of studies of it and other types of auditory perception that reflect that our ears aren't dB meters. Missing fundamental - Wikipedia One example I like, because we've all heard it: "Most common telephones cannot reproduce sounds lower than 300 Hz, but a male voice has a fundamental frequency approximately 150 Hz. Because of the missing fundamental effect, the fundamental frequencies of male voices are still perceived as their pitches over the telephone."

 

[fdeck]

I don't know all the details or how empirical the waveform model is, but it makes a good visual comparison. In the ideal pure sine case, the speaker cone doesn't move at all at the tuned port frequency, but I've never seen that when playing a real bass through an amp. The waveform model has a less extreme variation in excursion. It doesn't buy you extra Xmax at frequencies below the port tuning though.

Thanks. My program uses a sum of amplitudes of the first four harmonics. First, I looked at the waveforms from my basses on an oscilloscope, then made up this approximate waveform. It looked like I wasn't seeing a lot of variation when I played with different parameters and additional modeling complexity (e.g., accounting for phase shifts of the harmonics), so I decided that this 4-harmonic model is probably good enough for my own use.

My main goal was to answer the question: Why do bass speakers seem to work pretty well, despite having seemingly inadequate excursion at the lowest frequencies? So, the graphs represent the most pessimistic possible case (sine based) and a more optimistic case (waveform based). It's still a judgment call to decide what you think the practical power handling of your speaker is, under your playing conditions. Amps with "scooped" voicings may be more bass heavy than my model waveform.

Edit: You can do "view page source" in your browser and see all of my calculations, plus they are documented in a more readable form in some of the articles at my web page. I intend to make them even more readable in a new draft that I've been procrastinating on finishing up. My goal was for all of this to be "open source" to the extent possible.

 

[Rick James]

I look at commercial ports and see the outer end flared and inner end with a sharp transition which seems daft. It's way beyond my pay grade to consider how turbulence against port walls might be minimised other than to guess promoting laminar flow would help. It's jet engine designer grade stuff.

It is, another application of Bernoulli's Principle. The slug of air moving back and forth in the port moves faster if the area is small, slower if the area is large. If it moves fast enough it can be heard. It's mostly heard where the port joins the baffle. If the port end is flared the speed of the air mass slows before it hits the baffle, because the area is larger. That's what causes a reduction in noise, not streamlining. Since the flare only lowers the speed of the air mass at the port exit noise in the duct created by the air mass rubbing against the duct walls is unaffected. To eliminate noise throughout the length of the duct just make the area large enough to reduce the speed of the air mass throughout it, and then flaring isn't necessary as it won't have any effect.

Why do bass speakers seem to work pretty well, despite having seemingly inadequate excursion at the lowest frequencies?

Because the power density of the electric bass waveform isn't highest at the lowest frequencies, the first octave of its range. It's highest an octave higher. That's why an electric bass cab should be designed to have maximum port output, and therefore minimum cone excursion, not in the lowest octave but in the second octave.

 

[agedhorse]

Agree with you all the way, it's just that I am bemused that so many ends I've seen are so simplistically treated since that must be a major generator of turbulence. In particular I look at commercial ports and see the outer end flared and inner end with a sharp transition which seems daft. It's way beyond my pay grade to consider how turbulence against port walls might be minimised other than to guess promoting laminar flow would help. It's jet engine designer grade stuff. But the ends must be huge contributors and well worth looking at, as I'm sure the serious designers do.

It's only daft if it's really the cause of a problem.

 

[BadExample]

That's why an electric bass cab should be designed to have maximum port output, and therefore minimum cone excursion, not in the lowest octave but in the second octave.

Hey Rick, maximum port output, are you meaning the tuning, or do you mean the opening size designed to be maximum velocity short of chuffing? or something else?

 

[fdeck]

Here is something I've noticed when playing around with designs. I call it the "small box crisis." I know that my own wishful thinking makes me want to get the most possible sound out of the smallest possible box. Who wouldn't want that? I know there's the iron triangle of small, loud, and low, which applies to box volume. But there's another problem as well, having to do with ports.

As the box gets smaller, the port has to get bigger in order to maintain the same tuning frequency at a decent enough size to avoid turbulence. There's a point where it seems that the port has to grow faster than the box is shrinking. This is where it is no longer useful to reduce the box size. And it seems like more of a problem with smaller diameter drivers like that cool little Faital 8PR200.

For us DIY'ers, compromising on port design to make small boxes work "on paper" may be a pitfall worth avoiding. One thing I've tried is to put the port on the side of the box. This is unconventional, but it aligns the port along a longer dimension, so it can be deeper and correspondingly bigger.

 

[Rick James]

Hey Rick, maximum port output, are you meaning the tuning, or do you mean the opening size designed to be maximum velocity short of chuffing?

The tuning, Fb. Port output is maximum and cone excursion minimum at Fb. I see guys who play low B saying they tune to 30-35Hz to get maximum output on the low B. The flaw in that is there's a lot more power going to the speaker at 62 Hz when playing a low B than at 31 Hz. If the Fb is 31 Hz port output at 62 Hz is down, excursion is up, and you end up with less output capacity playing the low B note, not more. By the same token if you make the port area large enough to take full power at 31 Hz without chuffing the port will be a lot larger than it needs to be. You make the port area large enough to take full power without chuffing at 62 Hz. This is where experience comes in. Speaker modeling software assumes constant signal voltage to the speaker through the entire speaker pass band. That's appropriate if you're designing a stereo speaker, but not a pro-sound speaker. Experience is what allows a good designer to know which corners should be cut and which you can't.

 

[Passinwind]

Here is something I've noticed when playing around with designs. I call it the "small box crisis." I know that my own wishful thinking makes me want to get the most possible sound out of the smallest possible box. Who wouldn't want that?

Lots of people don't want that at all, at least in my experience. Never underestimate the power of music as fashion show.

 

[DukeLeJeune]

It's mostly heard where the port joins the baffle. If the port end is flared the speed of the air mass slows before it hits the baffle, because the area is larger.

Actually the place where turbulence (chuffing) is most likely to set in with a round port is, at the end inside the box. It's a fluid-flow problem, the air being the fluid. The front baffle acts sort of like a crude funnel to facilitate fluid flow into the outer end of the port, but typically the inner end has no such structure to facilitate intake fluid flow, so that is where audible turbulence (chuffing) first sets in. Thus with a flared port, the inner flare is arguably more important than the outer flare.

One way to do a "poor man's inner flare" is to cut the inner end of the port on a 45-degree angle, which makes the inner opening an ellipse of increased cross-sectional area. This increased cross-sectional area facilitates intake fluid flow. Credit to Earl Geddes for the idea. Measure the port length down the centerline, rather than along the long side.