How obvious is the difference between a 20hz capable subwoofer and 25hz, 30hz, etc?

[Howard]

Simulations are not accurate. Engineers knows this, so they figure in the tolerance. Thiele and Small provides some part to help simulate. They treated the speaker driver as a circuit and a simulations follow by adding nested circuits to simulate sealed, vented, bandpass, passive radiator.

And do you do this?

I helped someone build a subwoofer providing what volume and what tuning frequency. I said it may or may not perform well because of the simulations. The person said it worked great, but I could not to a listen test because of distance problems.

So you've never built something yourself. Great.

It will take me over a week to calculate a design, so I did use Unibox. I also use WinISD Pro Alpha. I do not have to prove that.

lol

I will not upload the images to my server, but prefer to have my name to be anonymous. It is not because of here or anywhere else. For personal stuff, I prefer to be anonymous. I prefer not to have a facebook and myspace account.

Did you know that on the Internet, you can upload images to a multitude of file hosts without any personal data attached?

I did prove it and you just do not like how I am saying it.

No, you didn't. You're talking out of your ass and you know it.

To explain the difference about difference percentage and error percentage to someone like that should know how to read equations. If you insist on me to lecture on this simple equation, the following is word for word from wikipedia.

""""One final note to make is that a lot of confusion lies in mistakenly assuming that percent difference is the same as percent error. The difference is that percent difference is comparing two experimental values, whereas percent error compares one experimental value with the actual/accepted value.""""

Error Percentage Example:
Let us say you work in a lab and you got a thermometer with unheard of accuracy. You think the manufactures accuracy is bull, so you come up with an experiment. You do this by figuring out known controls to check the accuracy. Some known controls that could be used for the experiment is the temperature of boiling water or the temperature when water freezes. Let us say that you use boiling water. After some readings, you use the error percentage.

Difference Percentage Example:
Let us say that you are comparing two different CPU from a benchmark on anandtech. Since the benchmarks provides experimental values, but not as controls. You have to treat them as theoretical values. The only controls is the configuration in each program and the process how they did the benchmark.

Fine and dandy, but if I do it your way with an example... values A and B are 200 and 100, respectively.

(200-100)/(200+100) * 100% = 33.3%

Does this mean that 200 is 33.3% more than 100?

Using a GR Research SW-12B 12 inch woofer and a 500 watt Bash amplifier. A 3 cubic feet tuned at around 23 Hz. Could use three 3 inch ports at a length of about 31 inches. To make it at this length, a 3 inch flexible pipe can be used with out causing the air to be restricted compared to using elbows. It will cost about $330. Sure it is more expensive, but it is smaller.

Pros of your design:

It's smaller

Cons of your design:

Has lower maximum SPL
Has lower sensitivity
It's more expensive
More difficult to build
Has more distortion (the SW-12B does not have shorting rings or any special features to reduce distortion relative to the RSS390, and it also requires more excursion for the same SPL which almost always increases distortion as well)

Distortions are anything that does not make up the signal. The materials adds distortion to the signal even though it is below the loudness of the signal coming from the subwoofer or speaker driver.

My point is, you are calling all distortion THD when THD means harmonic distortion only. Spurious noise (driver self noise or port chuffing) is not harmonically related to the input signal.

It is from the Linkwitz Transform spreadsheet from Linkwitz site.

I assume you referring to this:

http://www.linkwitzlab.com/images/graphics/air-distortn.gif

To requote you:

Also there is distortions cause by the trapped sound waves created on the rear of the woofer inside a box. To calculate the distortion for the box use the following.

0.014*Sd*Xmax/Vb = THD %

There is a fundamental misunderstanding here that you happen to be cursed by. His equations are for finding the distortion of the COMPLIANCE of the internal air volume (as if we didn't already know that when you compress a gas spring, it pushes harder), NOT the THD. Good going.

Hard to say because you just quote and bicker even though they are just opinions.

For context, you are referring to my question "Which one of my points are you referring to?".

Just another piece of evidence that shows you're talking out of your ass.

Sure you can use that formula, but it will not be near as accurate as the more reputable equations. They are the following.

n0 = 9.64*10^(-10)*Fs^3*Vas/Qes
SPL = 112 + 10*LOG(n0)

These are posted at the following.

http://www.diysubwoofers.org/sld/sealed3.htm

They are also in other books that have those equations, but mainly it is in Ray Alden. If you have a problem with it, complain to Ray Alden.

Those are valid equations (did not confirm personally), but more importantly, the equations that I have directly show the factors of efficiency. The equations you have, though equivalent, do not tell you immediately what will change the efficiency. For example, looking at my 2nd equation:

n0 = rho*(Sd*BL)^2/(Mms^2*2pi*c*Re);

You can immediately see that by efficiency is proportional to the square of radiating surface area and to the square of the force-current ratio (B*L), and inversely proportional to the square of the moving mass and inversely linearly proportional to the DC resistance of the voice coil.

Therefore, by increasing the Sd can generally increase efficiency as surface area tends to grow faster than the mass, considering the entire mechanical assembly.

Hard to explain in a short time frame. I do not have time write up to prove that I am right.

Sorry to hear that.

I already know compression. Compression is caused by the air pressure being greater than the energy of the materials used in the cone. At this point the Sd decreases because the cone is deforming.

Really? The Sd decreases because the cone gets smaller? Interesting effect.

By the way, you'll note that I said "power compression":

And, as reducing Sd tends to decrease efficiency, the higher power required increases power compression (linear distortion rather than non-linear distortion)

Which has nothing to do with any mechanical deformation - rather, it is the reduction in speaker output due to the increase in impedance of the driver. The impedance increases because of the rise in temperature in the voice coil due to Joule heating when high amounts of power are sent to the driver. This effect is very commonly noticeable in pro audio applications when high amounts of power are used on a regular basis, and ONLY in voltage-drive systems (i.e. where the damping factor is high). As the damping factor decreases, as is the case for most vacuum amplifiers, voltage drive tends towards current drive and instead of maintaining a proportional output voltage, the amp maintains a proportional output current (which tends to increase the power delivered to a hotter coil, which is the opposite). P = I^2R, for constant current and rising R, P increases, while for voltage drive P = V^2/R, where for constant voltage and rising R, P decreases.

If you don't believe me, first of all you need to do research, and second of all you need to do more research. JBL has many white papers on power compression, I believe.

You are only going to make it worst for your self by designing something to linear limits. If you doing any business, your customers will come back and in force you to fix the subwoofer. If you provide head room, customers will come back for more good business.

What I did was get the best for jlee that I could given his budget and space constraints. if you don't like the fact that I prefer to utilize the maximum amount of Xmax possible given all other factors (and it is a manufacturer-supplied Xmax generally rated at 70% of neutral BL), tough shit. By the way, with 300W in my design (as I specified), I don't even get close to using up all of the Xmax until some stupidly low frequency. I think it's around 15Hz but I can't be sure, don't have the program here.

Can not count for everything in design like a subwoofer. Telling anybody to be careful the volume control is not advice. Anybody will happily turn it to max. Any engineer does count the maximum volume and a little bit above the max.

Did you just conveniently ignore the fact that I said gain can be user-restricted? And do you know that gain is different from volume control? That it can be set with fixed resistors?

All woofers wattage limit is stated in free air or no box. When the woofer is put into a box, the wattage limit changes. It can be lower or higher.

Yes, when a driver is put into an enclosure, the maximum power can change, but it is NEVER higher than in free-air unless there is something in the enclosure that sucks heat away. Why? Let's break "wattage limit" into its various parts:

1) Thermal power handling - this is the wattage limit for the voice coil and is basically how much power you can send without melting anything
2) Mechanical excursion limits - this is the point at which the moving assembly hits the motor on the inward stroke, and the amount of power required for this to happen depends on the frequency of the signal and the enclosure tuning
3) Distortion limits - this is the wattage limit that is determined by the acceptable amount of distortion, but is not generally considered as a "hard limit"

Case number 1 is the dominant case except where case 2 exists; therefore, if Xmech (mechanical xmax) is not being achieved, the driver's power limit is that which will not cause the glues to melt, or the plastic voice coil former to deform, or the magnet wire insulation to burn off, or whatever. Now, putting the driver into an enclosure will generally reduce its excursion for any given wattage, but since in many cases the driver is thermally limited (everywhere in the passband except low bass, depending on tuning), the enclosure will NOT help suck out heat and will actually serve to insulate the driver.

It is best to set a desire goal for the wattage handling and then come up with a design that suits that goal.

What kind of designer are you? As a designer, your top priority should be to meet your customer's requests, or to work with your customer to find out what he wants. Why the fuck should he care how much wattage the system can handle? The customer generally has SPL, frequency response, specific polar radiation, size, weight, budget, etc. as requirements but when the hell does anybody specifically ask for 1000W (as an example)? Fucking stupid.

Did you not realize that jlee specified several constraints such as budget, size, intended purpose etc. but specifically DID NOT care about the wattage of the system? If I give him more wattage than you do, do you think he cares?

I will say you are confused and contradicting your self. Bracing does not make it sound better and not any more than using $1000 for speaker cable. Bracing just provides strength to the walls, so they do not flex. Sure bracing can help control the vibrations, so they are minimal.

Let me ask you this - even if bracing reduces the flex of the walls, why do it in the first place? There is no way a well-made plain box will break apart even from stupid amounts of displacement, so why? And bracing doesn't control vibration in and of itself. It might make the walls flex less, but the box as a whole will still experience the same reaction forces from the driver.

Now let me tell you something - the flex of the walls is considered to be, by many reputable people, AUDIBLE.

I am done speaking to you because you do not how to quit. You can post your improbable corrections, but I am not going to post back.

Good fucking riddance. One warning sign we should've all noticed was that you claim to be a designer, but yet don't want to reveal yourself. Either you're a really shitty designer, or you aren't a designer at all.

 

[s44]

Good news: looks like Howard chased the always wrong guy off this forum.

Bad news: he's invaded the Gadgets forum.

 

[Howard]

 

[Viperoni]

I skimmed through the thread quickly Howard, but well done!

 

[electroju]

I skimmed through the thread quickly Howard, but well done!

You just skimmed, but you did not think. Most Howard posts are wrong. Howard does not apply what he has learned.

 

[electroju]

Good news: looks like Howard chased the always wrong guy off this forum.

Bad news: he's invaded the Gadgets forum.

Nope still here. Howard is still wrong on most what he said. I tried stating what he said is wrong prefers to be big baby about it.

 

[Howard]

Nope still here. Howard is still wrong on most what he said. I tried stating what he said is wrong prefers to be big baby about it.

http://www.youtube.com/watch?v=dISKvZetkQI

 

[electroju]

(200-100)/(200+100) * 100% = 33.3%

Does this mean that 200 is 33.3% more than 100?

It should be the following.

(200-100)/((200+100) / 2) = 0.66 X 100 = 66%

At this point you should think about what you read.

Pros of your design:

It's smaller

Cons of your design:

Has lower maximum SPL
Has lower sensitivity
It's more expensive
More difficult to build
Has more distortion (the SW-12B does not have shorting rings or any special features to reduce distortion relative to the RSS390, and it also requires more excursion for the same SPL which almost always increases distortion as well)

Who saids that shorten rings are require for subwoofers. They are not. It is good to have low inductance, but do not need it for subwoofers. The only times when low inductance is required is trying produce the full frequency response up to 20000 Hz.

My design has the same output as yours. The complexity is the same since it is a vented box.

My point is, you are calling all distortion THD when THD means harmonic distortion only. Spurious noise (driver self noise or port chuffing) is not harmonically related to the input signal.

I assume you referring to this:

http://www.linkwitzlab.com/images/graphics/air-distortn.gif

To requote you:


There is a fundamental misunderstanding here that you happen to be cursed by. His equations are for finding the distortion of the COMPLIANCE of the internal air volume (as if we didn't already know that when you compress a gas spring, it pushes harder), NOT the THD. Good going.

THD is Total Harmonic Distortion. It is anything that is not the signal. It does not have to be the amplifier. It can be mechanical too. The following proves my point even though it is from wikipedia.

http://en.wikipedia.org/wiki/Total_harmonic_distortion

Those are valid equations (did not confirm personally), but more importantly, the equations that I have directly show the factors of efficiency. The equations you have, though equivalent, do not tell you immediately what will change the efficiency. For example, looking at my 2nd equation:

n0 = rho*(Sd*BL)^2/(Mms^2*2pi*c*Re);

You can immediately see that by efficiency is proportional to the square of radiating surface area and to the square of the force-current ratio (B*L), and inversely proportional to the square of the moving mass and inversely linearly proportional to the DC resistance of the voice coil.

Therefore, by increasing the Sd can generally increase efficiency as surface area tends to grow faster than the mass, considering the entire mechanical assembly.

You forgot Fs. Mms is part of Fs, but Fs also includes the resonance of the materials being used.

My equations do paint a better picture. Qes is the main factor. Then it is Fs and then it is Vas. The Qes which is Re and Bl. Adding a resister in series increases Qes which reduces SPL. Vas is a combination of Sd and Cms. Fs is part of Mms and the materials used.

You do not define the variables for equations. For example rho and c has no relationship to anything.

Really? The Sd decreases because the cone gets smaller? Interesting effect.

By the way, you'll note that I said "power compression":

Which has nothing to do with any mechanical deformation - rather, it is the reduction in speaker output due to the increase in impedance of the driver. The impedance increases because of the rise in temperature in the voice coil due to Joule heating when high amounts of power are sent to the driver. This effect is very commonly noticeable in pro audio applications when high amounts of power are used on a regular basis, and ONLY in voltage-drive systems (i.e. where the damping factor is high). As the damping factor decreases, as is the case for most vacuum amplifiers, voltage drive tends towards current drive and instead of maintaining a proportional output voltage, the amp maintains a proportional output current (which tends to increase the power delivered to a hotter coil, which is the opposite). P = I^2R, for constant current and rising R, P increases, while for voltage drive P = V^2/R, where for constant voltage and rising R, P decreases.

If you don't believe me, first of all you need to do research, and second of all you need to do more research. JBL has many white papers on power compression, I believe.

Those formulas are ohms law, but they have no relationship to AC. AC is technically less powerful than DC. Dampening factor has nothing to do with power compression. Power compression relates to cone deform. Yes Re changes and there will be a limit, but this only happens when the voice coil is not moving. When a speaker driver is put in a box, the statements you say is not as easy as you say about compression. If you want to know the exact wattage limit of the speaker driver, you have to put it through a test until it breaks. Manufactures do this to know the limits of the speakers.

Dampening factor has a affect on a transient response, but it is very, very, very small. It is not a variable that should be included. High dampening factor keeps the Q of the speaker driver consistent to what you measured. Low dampening factor changes or increases the Q, but it is small. Dampening factor is a good marketing statement, but is not a good spec to bring up.

Power compression comes after Sd is deformed from high output. More Sd means a higher chance of compression. If you want more output with minimal compression, use multiple smaller speaker drivers.

What I did was get the best for jlee that I could given his budget and space constraints. if you don't like the fact that I prefer to utilize the maximum amount of Xmax possible given all other factors (and it is a manufacturer-supplied Xmax generally rated at 70% of neutral BL), tough shit. By the way, with 300W in my design (as I specified), I don't even get close to using up all of the Xmax until some stupidly low frequency. I think it's around 15Hz but I can't be sure, don't have the program here.

You have big problem thinking that Xmax should be reached. The stated "70% said neutral Bl" means no control after this. Distortions increase and your equations that you stated bite you in your ass. My design is more giving while your design is not. My design will have less distortion. Do not want to use all the stated Xmax because the mechanical limit is close. Though not all manufactures state "70% Bl", so it is a fictional spec that you are stating. I rather be cautious instead of using all the rated Xmax. To know the mechanical limits, you have get two drivers and abuse one by pressing down on the cone until it reaches the back plate or until the assembly starts to coming apart.

It is a give and take. You have to give up SPL for quality or you have to give up quality for SPL.

Did you just conveniently ignore the fact that I said gain can be user-restricted? And do you know that gain is different from volume control? That it can be set with fixed resistors?

Yes, when a driver is put into an enclosure, the maximum power can change, but it is NEVER higher than in free-air unless there is something in the enclosure that sucks heat away. Why? Let's break "wattage limit" into its various parts:

1) Thermal power handling - this is the wattage limit for the voice coil and is basically how much power you can send without melting anything
2) Mechanical excursion limits - this is the point at which the moving assembly hits the motor on the inward stroke, and the amount of power required for this to happen depends on the frequency of the signal and the enclosure tuning
3) Distortion limits - this is the wattage limit that is determined by the acceptable amount of distortion, but is not generally considered as a "hard limit"

Case number 1 is the dominant case except where case 2 exists; therefore, if Xmech (mechanical xmax) is not being achieved, the driver's power limit is that which will not cause the glues to melt, or the plastic voice coil to deform, or the magnet wire insulation to burn off, or whatever. Now, putting the driver into an enclosure will generally reduce its excursion for any given wattage, but since in many cases the driver is thermally limited (everywhere in the passband except low bass, depending on tuning), the enclosure will NOT help suck out heat and will actually serve to insulate the driver.

The knob that control the loudness of the subwoofer is not gain. It is volume. Look at the schematic. Manufactures provides this confusion until schematic verifies it as it being the volume. Enforcing anybody to go easy on the volume will eventually go the opposite way and test your design at its limits. It is best to say turn it loud and I designed for that purpose too.

There are plenty of ways the voice coil can be cooled. They are the former, basket, and just the voice coil moving. If the voice coil is not moving usually at the tuning frequency or resonant frequency, the voice coil will heat up more. In a bandpass box, the voice coil has a higher chances of over heating. The former and the basket can absorb the heat and dissipate the heat. Density of the materials used for the former and basket helps it further.

Ever heard of the Shiva back during the Adire Audio days. Some people have used it with a 1200 watts (professional amplifier) when it is designed to handle 600 watts. Your statements are moot because you miss my point about it depends on the design.

What kind of fucking designer are you? As a designer, your top priority should be to meet your customer's requests, or to work with your customer to find out what he wants. Why the fuck should he care how much wattage the system can handle? The customer generally has SPL, frequency response, specific polar radiation, size, weight, budget, etc. as requirements but when the hell does anybody specifically ask for 1000W (as an example)? Fucking stupid.

Did you not realize that jlee specified several constraints such as budget, size, intended purpose etc. but specifically DID NOT care about the wattage of the system? If I give him more wattage than you do, do you think he cares?

Wattage is not always proves that one subwoofer is better than the other if it can not handle it. Your design does not handle 300 watts. It handles 150 easily, but 200 watts is pushing it. When I said that my design can handle 1000 watts, I do not mean you should use 1000 watt amplifier. I mean peak. During music, there is always musical peaks. These peaks can reach very high. Amplifiers can handle the peaks for short time before they clip. Luckily for the amplifier the musical peaks are short, but the woofer is not so lucky. It have to handle it. If it can not, damage will occur. Subwoofers are not meant for music because of the heavy mass. The heavy mass will cause more damage to the subwoofer than woofers of the same diameter that have a lighter mass. Counting for the peak protects the subwoofer somewhat, but not completely. If the customer wants output, they have to be told to fork over more money or the design have to changed to a different driver because there is a limitation. I rather push for the Exodus Tempest X2 or the Maelstorm (18-inch).

Any designer or engineer will count for the peaks. Civil engineer counts for extreme loads on bridge or building that may not happen at one time. They make it, so it can handle it.

Let me ask you this - even if bracing reduces the flex of the walls, why do it in the first place? There is no way a well-made plain box will break apart even from stupid amounts of displacement, so why? And bracing doesn't control vibration in and of itself. It might make the walls flex less, but the box as a whole will still experience the same reaction forces from the driver.

Now let me tell you something - the flex of the walls is considered to be, by many reputable people, AUDIBLE.

You are thinking about this too much. Bracing helps by minimizing flexing, but only if the sides are flimsy. You can build a box with composites and it does not need bracing. The only times when bracing is required is using 0.75 inch MDF in a very big box or the sides are resonating in the same spectrum the speaker driver or subwoofer is trying to produce.

Good fucking riddance. One warning sign we should've all noticed was that you claim to be a designer, but yet don't want to reveal yourself. Either you're a really shitty designer, or you aren't a designer at all.

I am designer, but you do too much research and you do not apply what you learn to what you have said here.

 

[Howard]

OK, does anybody need me to refute this or are we all agreed that his posts are worthless?

 

[s44]

If you keep him too busy to spread his nonsense into other threads, it would be a service.

 

[Howard]

It should be the following.

(200-100)/((200+100) / 2) = 0.66 X 100 = 66%

At this point you should think about what you read.

Don't give me that attitude when I did exactly what you did in your earlier post. Looks like I "forgot" to divide by 2 as well. Even so, 200 isn't 66% more than 100... or if it is, you'll have to explain that to me.

Who saids that shorten rings are require for subwoofers. They are not. It is good to have low inductance, but do not need it for subwoofers. The only times when low inductance is required is trying produce the full frequency response up to 20000 Hz.

Nobody said they're required, but they are preferred because they reduce flux modulation - a primary component in the variation of BL(i), which is a cause of distortion the same way that BL(x) is.

My design has the same output as yours. The complexity is the same since it is a vented box.

Way to ignore everything I said like you usually do, fool.

THD is Total Harmonic Distortion. It is anything that is not the signal. It does not have to be the amplifier. It can be mechanical too. The following proves my point even though it is from wikipedia.

http://en.wikipedia.org/wiki/Total_harmonic_distortion

Really, THD is everything that is not the signal? Then why does your link describe THD+N? What exactly is that?

You forgot Fs. Mms is part of Fs, but Fs also includes the resonance of the materials being used.

My equations do paint a better picture. Qes is the main factor. Then it is Fs and then it is Vas. The Qes which is Re and Bl. Adding a resister in series increases Qes which reduces SPL. Vas is a combination of Sd and Cms. Fs is part of Mms and the materials used.

You do not define the variables for equations. For example rho and c has no relationship to anything.

I didn't forget Fs. Fs is the resonance frequency of the spring-mass system, which is determined by the mass (Mms) and compliance (Cms, inverse of Kms). Fs does not include the resonance of the materials being used, and that matters only in the frequency response (and distortion, to some extent).

Why would you want to vary Qes directly when it is a derivative of Re and BL? Why wouldn't you say that, by adding a resistor in series, you increase Re and therefore decrease n0?

Again, Vas is a derivative factor which is probably used only because it is easier to measure than Cms. The materials of the driver do not affect the Mms other than because the materials may vary in density.

rho and c are the commonly used symbols for the density of air at standard conditions and speed of sound at standard conditions, respectively.

http://en.wikipedia.org/wiki/Density_of_air (shows both rho and c being used)

Never heard of a designer that didn't know those symbols...

Those formulas are ohms law, but they have no relationship to AC.

What are the AC power equations?

AC is technically less powerful than DC.

I think you need to put some qualifiers in that ridiculous statement.

Dampening factor has nothing to do with power compression.

There's a fairly strong relationship. If damping factor changes, the effect of Joule heating changes. If Re is high and damping factor is high, less power goes to the speaker. If Re is low and damping factor is high, more power goes to the speaker. If Re is high and damping factor is low, more power goes to the speaker. If Re is low and damping factor is low, less power goes to the speaker.

Power compression relates to cone deform.

No it doesn't.

Yes Re changes and there will be a limit, but this only happens when the voice coil is not moving.

Absolutely wrong.

When a speaker driver is put in a box, the statements you say is not as easy as you say about compression. If you want to know the exact wattage limit of the speaker driver, you have to put it through a test until it breaks. Manufactures do this to know the limits of the speakers.

So? What does this have to do with what I said?

Dampening factor has a affect on a transient response, but it is very, very, very small. It is not a variable that should be included. High dampening factor keeps the Q of the speaker driver consistent to what you measured. Low dampening factor changes or increases the Q, but it is small. Dampening factor is a good marketing statement, but is not a good spec to bring up.

Why are you talking about this? I never even mentioned transient response in relation to damping factor.

Yes, higher electrical damping reduces Qes and reduces total Q... so?

Power compression comes after Sd is deformed from high output. More Sd means a higher chance of compression. If you want more output with minimal compression, use multiple smaller speaker drivers.

What the fuck are you smoking? Prove to me that deformation of the diaphragm during even high output is enough to significantly reduce output (e.g. a reduction of more than 0.5dB).

There are several ways to reduce power compression and that's only one way (and not always the best way).

You have big problem thinking that Xmax should be reached.

Shut your mouth. That design doesn't hit Xmax at max power in the practical passband.

The stated "70% said neutral Bl" means no control after this.

Which is why calculated maximum output doesn't rely on any more excursion than this.

Distortions increase and your equations that you stated bite you in your ass.

What are you smoking? Which of my equations will bite me in my ass?

My design is more giving while your design is not.

Maybe, but you certainly have not convinced anyone of that.

My design will have less distortion.

How?

Do not want to use all the stated Xmax because the mechanical limit is close. Though not all manufactures state "70% Bl", so it is a fictional spec that you are stating. I rather be cautious instead of using all the rated Xmax. To know the mechanical limits, you have get two drivers and abuse one by pressing down on the cone until it reaches the back plate or until the assembly starts to coming apart.

What you are looking for there is called Xmech and is an available spec for better drivers.

It is a give and take. You have to give up SPL for quality or you have to give up quality for SPL.

Maybe for your designs.

The knob that control the loudness of the subwoofer is not gain. It is volume. Look at the schematic. Manufactures provides this confusion until schematic verifies it as it being the volume. Enforcing anybody to go easy on the volume will eventually go the opposite way and test your design at its limits. It is best to say turn it loud and I designed for that purpose too.

I know what a volume knob is. Gain is set inside the amp.

There are plenty of ways the voice coil can be cooled. They are the former, basket, and just the voice coil moving. If the voice coil is not moving usually at the tuning frequency or resonant frequency, the voice coil will heat up more. In a bandpass box, the voice coil has a higher chances of over heating. The former and the basket can absorb the heat and dissipate the heat. Density of the materials used for the former and basket helps it further.

Yes, the voice coil former heats up first (the voice coil first, actually), and from there heats the diaphragm. If the diaphragm is metal, or another material with good thermal conductivity, significant heat can be transferred this way. The motor is the next to heat up, and then the basket. Neither the motor or the basket do anything but heat up the inside of the box, though, so when the motor warms up the amount of heat being pulled away from the back is significantly reduced.

Ever heard of the Shiva back during the Adire Audio days.

Yes, I used to follow AA.

Some people have used it with a 1200 watts (professional amplifier) when it is designed to handle 600 watts. Your statements are moot because you miss my point about it depends on the design.

That's nice. I, too, can put 1200W into a driver that's rated for 600W. What was your point, anyway?

Stupid paragraph

Come back when you're able to properly transcribe your ideas into words.

Any designer or engineer will count for the peaks. Civil engineer counts for extreme loads on bridge or building that may not happen at one time. They make it, so it can handle it.

Sure, I'll account for the worst case, but we aren't given a worst case in this situation. jlee did not ask for a certain amount of SPL. Did he?

You are thinking about this too much. Bracing helps by minimizing flexing, but only if the sides are flimsy. You can build a box with composites and it does not need bracing. The only times when bracing is required is using 0.75 inch MDF in a very big box or the sides are resonating in the same spectrum the speaker driver or subwoofer is trying to produce.

Nope. No matter how strong the walls may be, they will still flex, unless you have found some wonder material that has an infinite flexural modulus.

I am designer

No you're not. Shut up, fool.

 

[velillen]

electroju if your a designer can we see some pics of your designs? Seriously a picture of a sub isnt giving anyway anything so dont pull you want to be anonymous as an answer

And you guys are really adding bracing to the argument? Ask any carpenter and they will tell you that is so just up to the box size more than anything. Small ass box isnt going to need bracing. Big ass box will. Medium size is up to the user. Sure it would be fine without but it will flex with any material to a small degree. obviously 1.5 inch oak or something isnt gonna move much at all vs .75 inch mdf. As for how much bracing again thats so personal and you can find that just by looking at people's DIY subs. Some have 1 bracing. Some have 1 for/aft one right/left. Some have 3 or 4 bracing. Lets not add bracing to the mix

 

[Howard]

electroju if your a designer can we see some pics of your designs? Seriously a picture of a sub isnt giving anyway anything so dont pull you want to be anonymous as an answer

His "customers" are probably not gonna want to reveal anything, either.

 

[electroju]

Don't give me that attitude when I did exactly what you did in your earlier post. Looks like I "forgot" to divide by 2 as well. Even so, 200 isn't 66% more than 100... or if it is, you'll have to explain that to me.

Fine I wrong again, so the following equation is the right one.

(x - y) / y = percent difference

Another percentage difference comparing two different results of the same experiment.

(x - y) / ( (x + y) / 2) = percent difference

Other statements that you said is wrong.

Nobody said they're required, but they are preferred because they reduce flux modulation - a primary component in the variation of BL(i), which is a cause of distortion the same way that BL(x) is.

The problem with shortening rings is they are costly. Subwoofers still do not need shortening rings to achieve low distortions. It is better to design a subwoofer so it uses less Xmax of the desire woofer. You do not think this way in the design, so I can see why you stating this way. You prefer to use the full Xmax, but only thing you are hurting is the distortion. Shortening rings are not the only way to reduce magnetic distortions. There are different types of motor structures like XBL^2 that helps reduce magnetic distortions at Xmax. Woofers using underhung will not be able to handle its full rated Xmax. It gets worst with overhung.

Again low inductances are not required for low distortion for subwoofers until the woofer is used in a 2-way or 3-way design. In a multi-way loudspeaker low inductance woofers are better. Low inductance helps with the speed of the woofer, but I am not going to get in that one.

Way to ignore everything I said like you usually do, fool.

I do not ignore, but most of your statements are bull. Like the dampening factor, how to use Xmax, volume and gain control, power compression,

Really, THD is everything that is not the signal? Then why does your link describe THD+N? What exactly is that?

THD+N is THD that includes noise. IMHO, noise is still distortions.

Why would you want to vary Qes directly when it is a derivative of Re and BL? Why wouldn't you say that, by adding a resistor in series, you increase Re and therefore decrease n0?

Changing the Qes can help the a desire woofer for an infinite baffle setup. If the Qts is 0.7, you can take advantage of the Fs.

rho and c are the commonly used symbols for the density of air at standard conditions and speed of sound at standard conditions, respectively.

http://en.wikipedia.org/wiki/Density_of_air (shows both rho and c being used)

Never heard of a designer that didn't know those symbols...

c also represents speed of light. The rho can also be Spearman Rho.

What are the AC power equations?

I think you need to put some qualifiers in that ridiculous statement.

You are the engineer. Why should I look it up.

The following is what it is.
http://hyperphysics.phy-astr.gsu.edu/hbase/electric/powerac.html

There's a fairly strong relationship. If damping factor changes, the effect of Joule heating changes. If Re is high and damping factor is high, less power goes to the speaker. If Re is low and damping factor is high, more power goes to the speaker. If Re is high and damping factor is low, more power goes to the speaker. If Re is low and damping factor is low, less power goes to the speaker.

There is no "strongly relationship" that dampening factor has a huge impact on Re. The dampening factor can change or increase the Q of the speaker, but it does not change the loudness of the speaker. You have to prove it because you are stating bull shit. The reason why it is bull shit is because the resistance that make the dampening factor makes it too insignificant to change Re in any way. If you want to get dirty, there are cables and other passive components. We can also relate the speaker driver as a circuit to make the circuit ring with out causing Re to rise in any way.

No it doesn't.

Power compression is part of Sd deforming. Yes, it is also cause by the voice coil heating up. The following proves you are wrong on your statements on power compression.

http://www.harman.com/EN-US/OurComp...ip/Documents/Scientific Publications/7059.pdf

That is an old article because these days woofers includes vents to help cool it off. One thing you miss is the difference between music and movie content. Movie content is not as intensive as music. Music is constant and peak outputs. Movies have a idle for audio. Subwoofer can easily dissipate the heat during movies, but not during music. Power compression does not matter at this time, so what does it matter to you. I think it matters to you because you feel you are more educated in the power compression area than others and you want to have a big head about it.

So? What does this have to do with what I said?

You said that any design can handle the power that the manufacture stated for its speaker driver. This is not true. It depends on the design. Then you said "of course it is the design". You are reversing your statements you are feel that you know better. The design that you suggested does not quite handle the wattage handling of what the manufacture stated. Also my does not too, but handle more power.

Why are you talking about this? I never even mentioned transient response in relation to damping factor.

Yes, higher electrical damping reduces Qes and reduces total Q... so?

High dampening factor does not lower or reduce Qes. High dampening factor keeps the specs of the speaker driver consistent to what it has been measured. Also it helps keep the design consistent to simulations. Low dampening factor increase the Q of the speaker. If the speaker driver is related as a circuit, the dampening factor will make it ring or oscillate causing its Q to be higher than what has been measure before. Qes changes or increases because it does not have enough feedback to fight against the ringing caused by the other circuit the amplifier. The Re will not change because that is the speaker driver spec.

What the fuck are you smoking? Prove to me that deformation of the diaphragm during even high output is enough to significantly reduce output (e.g. a reduction of more than 0.5dB).

There are several ways to reduce power compression and that's only one way (and not always the best way).

What you are smoking that power compression have to be brought up when the woofer is not even close to thermal damage. Also you do not know the difference between movie and music content.

Shut your mouth. That design doesn't hit Xmax at max power in the practical passband.

You first because again it about the design and your design is out of the passband. My design is in the passband.

Which is why calculated maximum output doesn't rely on any more excursion than this.

There is thing called mechanical Xmax after "70% said neutral Bl". The "70% of neutral Bl" is a manufactures opinions that the speaker driver can not go any further. It can go to mechanical Xmax, but this is then relying on the suspension for control instead of the voice coil.

What are you smoking? Which of my equations will bite me in my ass?

It proves my point that any equation you stated in this thread does not matter to you in your design. For example distortion VS Xmax.

Maybe, but you certainly have not convinced anyone of that.

Not to you, but others. I specify a design using GR Research and someone acknowledge it as being good. All you said that design is too complex to build. Vented or bass reflex are complex. Have to use a stethoscope or the woofer tester to help tuning the box to the desire frequency. If you plug up one or two ports of my design, it can provide different tuning for different applications.

How?

My designs uses less Xmax and has more control during its operation. Your design uses near Xmax. Xmax is the linear limits and it is also where it loses control.

What you are looking for there is called Xmech and is an available spec for better drivers.

Xmech or mechanical excursions is available for any driver. Xmech is usually not stated. You have to contact the manufacture or use the abuse technique to find out the Xmech. Xmech is not available to the so called "better drivers". In some cases Xmax is not stated, so you have to rely on something else like the voice coil height and magnetic gap.

Maybe for your designs.

Engineers always give and take in their designs. When I build or design a computer, I decide on a fast processors or a fast hard drive that suits my budget and what I need.

I know what a volume knob is. Gain is set inside the amp.

Your point is...

Yes, the voice coil former heats up first (the voice coil first, actually), and from there heats the diaphragm. If the diaphragm is metal, or another material with good thermal conductivity, significant heat can be transferred this way. The motor is the next to heat up, and then the basket. Neither the motor or the basket do anything but heat up the inside of the box, though, so when the motor warms up the amount of heat being pulled away from the back is significantly reduced.

Your point is...that your Mr Know it all prefers to change issues to power compression because you are more familiar with that than designing a subwoofer.

That's nice. I, too, can put 1200W into a driver that's rated for 600W. What was your point, anyway?

My point is about the design. If it is design properly, it can handle more power than what is rated in free air and there is more to it than power compression.

Come back when you're able to properly transcribe your ideas into words.

Your scatter quotes have a problem if you do not put the question in the context of your own answers.

Sure, I'll account for the worst case, but we aren't given a worst case in this situation. jlee did not ask for a certain amount of SPL. Did he?

JBL, Polk, and Infinity just to name a few count for the worst case scenario. They count for the peaks which could be between 4 to 5 times greater than what the amplifier can handle, but the speaker driver has no choice to handle it. They do not state it on the box, but they do. I did not include SPL. I provided a design that the woofer can handle with out causing to distort or have other problems like extreme delays. Your design goes for complete SPL. Saying your design does not go for SPL clearly states your insecurities in order to stay correct.

Nope. No matter how strong the walls may be, they will still flex, unless you have found some wonder material that has an infinite flexural modulus.

Fine build your subwoofer with thin walls and add bracing. I will go with thick walls or go completely to composites.

 

[electroju]

electroju if your a designer can we see some pics of your designs? Seriously a picture of a sub isnt giving anyway anything so dont pull you want to be anonymous as an answer

I have subwoofer that I made with a Rubbermaid box. Yes, a Rubbermaid box and I know it is crazy. If I keep the volume low, noise does not become a big factor. Power tools and me do not mix well, so I went this route. I used two 8 inch Optimus woofers in a push-pull configuration. The volume of the box is about 38 liters. Sorry, I do not have any pictures and you may laugh if you want. At this time I am planning on making a subwoofer that looks better and functions better, so doing a few simulations with some woofers. I may have to go through a local cabinet maker or someone else to cut only the panels, so I can just assemble it.

 

[velillen]

Sounds interesting. But come on its 2010, everything has a freakin camera on it. Trust me i cant have cameras for work and its a PITA to find a half decent phone with a camera. I could get a camera phone and disable the camera but man i just feel dirty doing that.

 

[Howard]

Fine I wrong again, so the following equation is the right one.

(x - y) / y = percent difference

Another percentage difference comparing two different results of the same experiment.

(x - y) / ( (x + y) / 2) = percent difference

Other statements that you said is wrong.

Um, but the first equation is the one that I've been using all along.

The problem with shortening rings is they are costly.

A little bit.

Subwoofers still do not need shortening rings to achieve low distortions. It is better to design a subwoofer so it uses less Xmax of the desire woofer.

No, "low distortion" can be achieved without the use of shorting rings, but for any given excursion and power level, shorting rings can only help, and there is a reason why many drivers exist that do have them. If the lowest possible distortion was not necessary, we would all be using medium-grade drivers.

You do not think this way in the design, so I can see why you stating this way. You prefer to use the full Xmax, but only thing you are hurting is the distortion. Shortening rings are not the only way to reduce magnetic distortions. There are different types of motor structures like XBL^2 that helps reduce magnetic distortions at Xmax. Woofers using underhung will not be able to handle its full rated Xmax. It gets worst with overhung.

Don't get me wrong here. I prefer to use the full Xmax if my other constraints have already been decided. Artificially limiting the output of the sub, as you would, is not my thing. It's a self-limiting factor anyway - if they don't like high output because of distortion, they'll turn it down themselves. If they don't mind it, more power to them.

XBL^2 helps with BL(x) related distortions but not BL(i) related distortions, which is what shorting rings help.

Again low inductances are not required for low distortion for subwoofers until the woofer is used in a 2-way or 3-way design. In a multi-way loudspeaker low inductance woofers are better. Low inductance helps with the speed of the woofer, but I am not going to get in that one.

Low inductance, in itself, does not help distortion much as far as I know. It's the variation in inductance vs excursion, I believe, that is a factor.

I do not ignore, but most of your statements are bull.

Eh, look who's talking.

THD+N is THD that includes noise. IMHO, noise is still distortions.

So are you telling me that "anything that is not the signal is THD", because that's what you said earlier. Or is this latest revised statement of yours the correct one?

Changing the Qes can help the a desire woofer for an infinite baffle setup. If the Qts is 0.7, you can take advantage of the Fs.

And if it is not, then you can't?

Did you know that Linkwitz prefers low-Qts drivers for open baffle subs? Around 0.5 or less?

c also represents speed of light. The rho can also be Spearman Rho.

Are you crazy? What does the speed of light and some statistical measure have to do with audio?

You are the engineer. Why should I look it up.

The following is what it is.
http://hyperphysics.phy-astr.gsu.edu/hbase/electric/powerac.html

I don't want to haggle this point any more. It's basically the same shit and what I said earlier has not been invalidated.

There is no "strongly relationship" that dampening factor has a huge impact on Re.

You're breaking my balls here. I said there is a strong relationship between damping factor and power compression.

The dampening factor can change or increase the Q of the speaker, but it does not change the loudness of the speaker. You have to prove it because you are stating bull shit.

Can the damping factor really change the Q of the speaker? I would say that the Qes can change slightly because of the Re (not sure about this), but generally Qts stays the same for a driver that comes in 4ohm/8ohm, for example.

The reason why it is bull shit is because the resistance that make the dampening factor makes it too insignificant to change Re in any way. If you want to get dirty, there are cables and other passive components. We can also relate the speaker driver as a circuit to make the circuit ring with out causing Re to rise in any way.

Um, the impedance that you use to calculate damping ratio basically IS Re (plus some extra for inductance), and Re changes significantly as the coil heats up. Why can't you accept that?

Power compression is part of Sd deforming.

No it isn't. Stop regurgitating statements on websites that sell woofers, like this page here:

http://www.gr-research.com/index.asp?PageAction=VIEWPROD&ProdID=131

I'm sure this page is also where you "found out" that you shouldn't use more than an arbitrary fraction of a driver's Xmax, eh?

Yes, it is also cause by the voice coil heating up.

Good, this means that your mind is not totally closed to learning.

The following proves you are wrong on your statements on power compression.

http://www.harman.com/EN-US/OurComp...uments/Scientific%20Publications/7059.pdf

That is an old article because these days woofers includes vents to help cool it off.

I don't see what this disproves. I clearly see him provide an example where power compression robs max SPL by up to 7dB for a particular system. I've been saying all along that power compression can sometimes decrease max SPL by a significant amount.

One thing you miss is the difference between music and movie content. Movie content is not as intensive as music. Music is constant and peak outputs. Movies have a idle for audio. Subwoofer can easily dissipate the heat during movies, but not during music. Power compression does not matter at this time, so what does it matter to you. I think it matters to you because you feel you are more educated in the power compression area than others and you want to have a big head about it.

Yes, there are often instances in movies when the speakers have time to cool down. Power compression kicks in very quickly, though.

You know, I don't even know why we're talking about this. All I said in the beginning was that power compression, as a distortion element, goes up as more power is applied, and you go off on this wild tangent about how power compression is when the cone bends and all sorts of nonsense.

You said that any design can handle the power that the manufacture stated for its speaker driver. This is not true. It depends on the design. Then you said "of course it is the design". You are reversing your statements you are feel that you know better. The design that you suggested does not quite handle the wattage handling of what the manufacture stated. Also my does not too, but handle more power.

You're breaking my balls again. I guess I was wrong when I said your mind was not totally closed to learning.

If you will revisit my paragraph about power handling, you will understand when a power handling figure cannot be used. My design (and probably your design) is NOT one of those scenarios.

High dampening factor does not lower or reduce Qes. High dampening factor keeps the specs of the speaker driver consistent to what it has been measured. Also it helps keep the design consistent to simulations. Low dampening factor increase the Q of the speaker. If the speaker driver is related as a circuit, the dampening factor will make it ring or oscillate causing its Q to be higher than what has been measure before. Qes changes or increases because it does not have enough feedback to fight against the ringing caused by the other circuit the amplifier. The Re will not change because that is the speaker driver spec.

Are you blind, man? I said higher electrical damping reduces Qes. Where did I say higher damping factor reduces Qes?

By the way, Re changes all the time. Get over it. Look it up.

http://en.wikipedia.org/wiki/Temperature_coefficient

What you are smoking that power compression have to be brought up when the woofer is not even close to thermal damage. Also you do not know the difference between movie and music content.

A woofer doesn't have to get close to thermal damage to suffer from power compression. Power compression always exists, but in certain situations it gets to be bad.

I know the difference.

You first because again it about the design and your design is out of the passband. My design is in the passband.

Man, do you know how hard it is to argue with somebody for whom English is a second language? What about my design is "out of the passband"?

There is thing called mechanical Xmax after "70% said neutral Bl". The "70% of neutral Bl" is a manufactures opinions that the speaker driver can not go any further. It can go to mechanical Xmax, but this is then relying on the suspension for control instead of the voice coil.

Yes, mechanical Xmax is Xmech. Many manufacturers provide it, especially for car audio drivers.

It proves my point that any equation you stated in this thread does not matter to you in your design. For example distortion VS Xmax.

What? There's an equation for that? And what does that have to do with the other equations I (re)produced?

Not to you, but others. I specify a design using GR Research and someone acknowledge it as being good. All you said that design is too complex to build. Vented or bass reflex are complex. Have to use a stethoscope or the woofer tester to help tuning the box to the desire frequency. If you plug up one or two ports of my design, it can provide different tuning for different applications.

I didn't say it wasn't a good design. I just listed the pros and cons of your design compared to my design. A design can be good, even if the designer is clueless.

My designs uses less Xmax and has more control during its operation. Your design uses near Xmax. Xmax is the linear limits and it is also where it loses control.

Apples to apples, my friend. For a similar SPL, the Xmax is going to be pretty much the same. If you REALLY want me to check, I will, but I'm sure it's going to be very close.

Xmech or mechanical excursions is available for any driver. Xmech is usually not stated. You have to contact the manufacture or use the abuse technique to find out the Xmech. Xmech is not available to the so called "better drivers". In some cases Xmax is not stated, so you have to rely on something else like the voice coil height and magnetic gap.

Do you want me to find you a bunch of drivers that show Xmech or what?

Engineers always give and take in their designs. When I build or design a computer, I decide on a fast processors or a fast hard drive that suits my budget and what I need.

Good. Finding out what you want to do with it is the first step. This process, however, is not what you follow when designing subs, apparently.

Your point is...

My point is, if you set the gain properly you can make sure that you don't damage anything even when the volume knob goes to 11.

Your point is...that your Mr Know it all prefers to change issues to power compression because you are more familiar with that than designing a subwoofer.

This coming from the guy who thinks that power compression comes from the bending of the cone...

My point is about the design. If it is design properly, it can handle more power than what is rated in free air and there is more to it than power compression.

Anybody can put 10,000W into a driver, it just depends on how long it's applied. What I'm trying to explain to you here is that a driver will never have a higher thermal power handling inside a box than in free air (mechanical limits notwithstanding).

Your scatter quotes have a problem if you do not put the question in the context of your own answers.

So are you blaming me for your inability to go back and read, or are your blaming me for the fact that I can't understand your English?

JBL, Polk, and Infinity just to name a few count for the worst case scenario. They count for the peaks which could be between 4 to 5 times greater than what the amplifier can handle, but the speaker driver has no choice to handle it.

You mean a 1000W amplifier will be asked to output up to 5000W?

They do not state it on the box, but they do. I did not include SPL. I provided a design that the woofer can handle with out causing to distort or have other problems like extreme delays. Your design goes for complete SPL. Saying your design does not go for SPL clearly states your insecurities in order to stay correct.

Again I have no idea what you're trying to say here.

Fine build your subwoofer with thin walls and add bracing. I will go with thick walls or go completely to composites.

Who said I like building boxes with thin walls?

 

[Number1]

Sounds interesting. But come on its 2010, everything has a freakin camera on it. Trust me i cant have cameras for work and its a PITA to find a half decent phone with a camera. I could get a camera phone and disable the camera but man i just feel dirty doing that.

Very poor attempt at a joke.
The A/V/HT forum is a place to discuss just that, A/V/HT.
Take this nonsense to OT where it belong.

 

[velillen]

Very poor attempt at a joke.
The A/V/HT forum is a place to discuss just that, A/V/HT.
Take this nonsense to OT where it belong.

Nah not a joke. More just an overall statement cause I went phone shopping the day before and read this when I got home. He doesnt want to post pictures (even though the idea of a rubbermaid subwoofer sounds neato) so o well. Wasnt a joke so sorry if it came off at one. Then again its not like this thread isnt full of sarcasm/jokes anyways to be fair.....