Trying To Prove A Point

[Mr Peabody]

Seems we all may have had some bits of truth and inaccuracies. This question is also far more complex than it sounds. I've read this and need to read it again and maybe again to fully grasp it all. There's a table that shows at around 350 Hz power is even and then falls off in either direction. However, he does state that in lower frequencies peak power demands will be greater. What's cool is the example uses Geoffcin's analogy of all drivers being even sensitivity. Although this answers a lot it also arises new questions for me. I've got some feelers out will post additional info if received.

http://sound.westhost.com/bi-amp.htm

Although I found the entire two part article very interesting here's the link to part 2 where he goes onto more information about applying amplifier power. http://sound.westhost.com/bi-amp2.htm

[Mr Peabody]

This writer clearly states it takes more power for lower frequencies over mids or highs. He's not as big a fan of Bi-amping as the prior article. http://www.gcaudio.com/resources/howtos/biamping.html

So the prior link was not the definitive answer.

[poppachubby]

This writer clearly states it takes more power for lower frequencies over mids or highs. He's not as big a fan of Bi-amping as the prior article. http://www.gcaudio.com/resources/howtos/biamping.html

So the prior link was not the definitive answer.

This was a nice link. It didn't feel like I was arriving in the middle of a topic or discussion. He lays out everything simply and concisely.

Clearly, what I have been saying is at great odds with what this guy is saying. I have been busy this week but I will certainly be talking to Mike, my tech mate, soon. I can't imagine why he would explain everything azz backwards, there must have been some point I missed or misunderstood.

I'll report back what I find out...

[Geoffcin]

There's a table that shows at around 350 Hz power is even and then falls off in either direction. However, he does state that in lower frequencies peak power demands will be greater. What's cool is the example uses Geoffcin's analogy of all drivers being even sensitivity. Although this answers a lot it also arises new questions for me. I've got some feelers out will post additional info if received.

http://sound.westhost.com/bi-amp.htm

Although I found the entire two part article very interesting here's the link to part 2 where he goes onto more information about applying amplifier power. http://sound.westhost.com/bi-amp2.htm

Again, you've taken an article that you don't understand, and inferred things from it and are not what is stated. First off, he's stating that in the average program material, the center weighting of the signal is 350Hz. Half the tones will be higher, have the tones lower. That's IT!

The weighting scale givin in the article is in exact AGREEMENT with my statements. Taken that the program material center weighting is from 350Hz, as you move away from that hinge point you must provide more power to the respective drivers. Few speakers have their tweeter kick in at 350Hz, hence more of the TOTAL power would go to the woofer. This DOESN'T mean it takes more power to creat a bass note, only that the woofer's power requirements would be more based on it's GREATER coverage of the total bandwith of the program material.

I'm actually quite glad you've quoted this artical as the writer seems to have an in-debth understanding of sound reproduction, even if it's too much for some lay people to grasp.

As a matter of fact, if you would have grasped the meaning it would have bothered you, (but you miss-read it) so the author wrote a disclaimer;

"Since the last statement will possibly cause some discomfort or indeed confusion (after all, everyone knows that a subwoofer needs more power than the main speakers), I should explain myself. Most of the time in this article, I refer to power as average power, and indeed the average power falls with frequency below about 100Hz or so. The peak power is a different matter, and depends to a very great degree on the type of music"

And then goes on to telling you WHY a sub needs more power;

"
The table assumes equal efficiencies for the bass and mid+high drivers. Should they be different, then a correction factor must be added in. For example, if the bass driver were to be 3dB less efficient than the mid+high drivers, then the bass power must be doubled (and of course vice versa). If the difference is less than 3dB, you may safely double the power anyway, or calculate the actual power needed - this I shall leave as an exercise for the reader.
"
In addition he writes;

"In general, I suggest that the bass amplifier should have at least the same power as that used for the mid+high frequencies, but if any equalisation is used (such as the Project 71 Linkwitz Transform circuit), this may need to be increased dramatically. A boost of only 6dB may require that amp power be increased by 4 times."

Every bit of which I am in full agreement with.

[Mr Peabody]

Well Geoffcin, you should have gone ahead and also cherry picked from part two where even a lay person can understand he explains the tweeter will use way less power. Unfortunately, he only compares the tweeter to the mid which doesn't address the low end that is in question. Too bad your intelligence is so far above us lay people and you still don't have the answer and still misinformed in regard to the original question.

[audio amateur]

I can't really be bothered to read all that for the moment so I'll just go with whatever Geoff says, as I seem to have been in agreement with him since the start

[Geoffcin]

This writer clearly states it takes more power for lower frequencies over mids or highs. He's not as big a fan of Bi-amping as the prior article. http://www.gcaudio.com/resources/howtos/biamping.html

So the prior link was not the definitive answer.

OK, this writer is clearly not an engineer, but he does grasp the effect of driver sensitivity on audio reproduction.

"Passive crossover networks found in consumer speakers waste an enormous amount of power. The often complex network is made up of large coils, chokes, capacitors and resistors. The circuit splits the full range signal into different frequencies (low, mid and high) appropriate for the different drivers in the speaker. Further, a crossover network compensates for efficiency differences in the drivers; woofers demand mode power than midrange drivers which in turn demand more power than tweeters, etc. Further, each of the drivers has different sensitivities, with some requiring far more (of far less) power than other drivers in the same speaker system. In a passive crossover, the excess power not required is dumped into resistors and burned off as heat. This makes for an incredible waste of power."

From the other article we've learned that a woofer that is crossed over at 350Hz would be required to cover HALF the (average) total musical signal, so it would need at least HALF the power of the entire speaker design.

Furthermore, most modest designs with two drivers have the woofer crossed over from 2kHz-3kHz. This would put even MORE of the total energy into the woofer. Also with a woofer that has a LOWER sensitivity (most, but not all woofers are less efficient than tweeters) than a tweeter crossed over at the higher frequency you could have designs where fully 90%(or more!) of the power is going to the woofer. NONE of this means it takes more power to make a lower frequency than a high note, nor is that what this writer is saying by stating that more power is going to the woofer.

I hate to say it but this is clearly another case of you NOT understanding the writer, and inferring something that was simply not stated by him.

[Mr Peabody]

I hate to say it too but you are mixing the two articles together in a futile attempt to show something that isn't there.

Besides that if 350 Hz is mid point, and half your power was used to reproduce 350 down, that would certainly show it takes more power to reproduce low frequencies as there are more octaves below 350 than above to 20kHz.

[BallinWithNash]

a true sub is crossed below 100Hz mine is crossed at 90Hz and I have a bigger amp driving it ... so the mid's and high's together cover much more total frequencies, they cover everything 91Hz and up! and im pretty sure you can't play a negative Hz so the sub is only playing 1-90Hz so what your saying Geoff is that high's and low's require the same amount of power but how? the sub cover's less and is driven by more power? that must lead you to one conclusion ... it takes more energy (power) to produce a lower frequency!

[Geoffcin]

Hey, guys what takes more power to reproduce .... a high note or a low note?
So in other words the lower the note the more power or the higher the note the more power?

This was the question correct?

The answer is NO it doesn't take more power to produce a lower note. The reason your sub takes more power is that it's efficiency is less than your main speakers. If it were the same it would take EXACTLY the same power for every note, high or low.

[Geoffcin]

I hate to say it too but you are mixing the two articles together in a futile attempt to show something that isn't there.

Besides that if 350 Hz is mid point, and half your power was used to reproduce 350 down, that would certainly show it takes more power to reproduce low frequencies as there are more octaves below 350 than above to 20kHz.

No, I am mixing the two articles to show you how you've been mistaken. Obviously you can't stand the fact that your logic is in error, so you will refuse to believe the truth no matter how plainly it's set out for you. I'm not going to further waste my time trying.

[Mr Peabody]

Earlier I posted a link that explained lower frequencies take more power to reproduce due to the lower the frequency the further the woofer excursion so more power is meeded to push the woofer to those further reaches. I posed our question to a manufacturer's engineer and here's his response. I only asked about "low frequencies", after reading his response and Geoffcin's I reworded the question to specifically ask about woofers vs tweeters. I'll post that when and if received.

" For a given sound pressure level in a sealed subwoofer (the easiest example), cone excursion quadruples with each successively deeper octave. So going from 80 Hz to 40 Hz requires 4X more excursion, and going from 40 Hz to 20 Hz requires 4X more excursion again – so excursion at 20 Hz is 16X that of 80 Hz. More cone excursion generally requires more voltage of course.

There are compounding variables like the impedance load the subwoofer presents to the amplifier and the natural resonance frequency of the subwoofer system, which will exhibit a peak in the frequency response. Basically you look at the native (un-equalized) frequency response of the sealed subwoofer as an indicator of its relative sensitivity to a given drive voltage. Below the natural resonance peak of the subwoofer system (where it’s the most sensitive/efficient), the frequency response of the sealed subwoofer starts to roll-off at 12 dB/octave, which means that its becoming progressively less sensitive to a given drive voltage at progressively deeper frequencies. So if you wanted the subwoofer to maintain the same sound pressure at all frequencies below its natural resonance peak, you would need to feed it progressively more voltage (at a rate of 12 dB/octave) in order to accomplish that.

So it would be more accurate to state that if you wanted a subwoofer to reproduce frequencies below its natural resonance peak at sound pressures higher than it naturally wants to deliver with a given drive voltage, then yes – it certainly does require more power, and this would be reflected in the voltage transfer function (VTF) of the amplifier, which is essentially an inverse of the subwoofer frequency response and shows any cuts to reduce peaks at resonance, and also shows any boost (which by definition requires more power) to flatten/extend the frequency response deeper than the subwoofer wants to go naturally.

The below chart (courtesy of Linkwitz Labs) shows the natural FR of a sealed subwoofer, with a resonance peak of 55 Hz (system Q 1.21), and with a low-end response being -19 dB at 19 Hz. The equalized FR of the subwoofer shows a flat response with no resonance peak (i.e. system Q of around 0.5), and is now only -6 dB at 19 Hz, indicating considerably deeper extension than the native FR. The EQ curve (VTF) of the amplifier shows a cut at 55 Hz (where less power is required), and then a steadily rising power requirements showing 13 dB of boost at 19 Hz, in order to achieving the deeper extension. The boost curve levels off at about 10 Hz, but remains quite elevated with respect to the 0 dB reference level, indicating the higher drive voltage required to reproduce the deepest frequencies at SPLs louder than the subwoofer wants to play naturally."

I wish he hadn't gone on to talk about equalizing, at least Geoffcin agrees on that, but it doesn't void the statement in the first part of the response which reinforced the other link that the lower the frequency the further the excursion which requires more power. If no EQ it would still take more power to do 55Hz than 110Hz.

[Geoffcin]

So it would be more accurate to state that if you wanted a subwoofer to reproduce frequencies below its natural resonance peak at sound pressures higher than it naturally wants to deliver with a given drive voltage, then yes – it certainly does require more power, and this would be reflected in the voltage transfer function (VTF) of the amplifier, which is essentially an inverse of the subwoofer frequency response and shows any cuts to reduce peaks at resonance, and also shows any boost (which by definition requires more power) to flatten/extend the frequency response deeper than the subwoofer wants to go naturally.

Where does this say it take more power? When you go below the system's natural resonance. Duh!

[Geoffcin]

.

" For a given sound pressure level in a sealed subwoofer (the easiest example), cone excursion quadruples with each successively deeper octave. So going from 80 Hz to 40 Hz requires 4X more excursion, and going from 40 Hz to 20 Hz requires 4X more excursion again – so excursion at 20 Hz is 16X that of 80 Hz. More cone excursion generally requires more voltage of course. "

Yes, but remember, a woofer playing at 80Hz is required to cycle twice as fast as 40Hz wich requires 4X more power for the same travel. However it's also moving 1/4 the distance to make the same SPL so the equation balances.

[Mr Peabody]

Oh, I see, so in your world the power goes down when voltage goes up? Interesting statement from a higher being. How many times are you quitting and returning any way?

[Geoffcin]

Oh, I see, so in your world the power goes down when voltage goes up? Interesting statement from a higher being. How many times are you quitting and returning any way?

No, in my world we point out when someone takes something out of context and mis-understands what it means.

I can keep this up all day with you, but it's certain that you'll never get it. Mostly because you've set your mind to a point of view and refuse to budge from it no matter what proof is put before you. Actually you've managed to gather a great deal of good research, it's quite ironic that you refuse to take the time to understand it.

[Mr Peabody]

I am trying to understand it but you are the one taking things out of context and can't even get power formula correct. And, you have not provided any proof of anything. You can claim all you want that I don't understand but I do understand enough to know the links posted certainly show you have not been correct.

[BallinWithNash]

Peabody i love how he skips over my last post because there is no way he can prove it wrong haha!

[Mr Peabody]

Remember though earlier I said this question wasn't cut & dry. I'm trying to get to the root of things. It seemes some things are true that contradict each other. I will be posting some more info later but only if I can connect all the dots. Posting parts of the puzzle isn't getting us any where. I have found a helpful source, I only hope I can get everything answered before he gets tired of me.

[BallinWithNash]

Well ... have we come to a conclusion? any one still interested in this post?? hello??

[Feanor]

Well ... have we come to a conclusion? any one still interested in this post?? hello??

No really, but I this might be the fact of the matter.

It takes the same measured power for a bass note as a treble, however human hearing is not as sensitive to bass as treble, accordingly bass notes must be produced with more power to be perceived as loud as treble notes. Consequently, since bass requires more power to be heard, bass amplifiers need to be more powerful.

[Geoffcin]

No really, but I this might be the fact of the matter.

It takes the same measured power for a bass note as a treble, however human hearing is not as sensitive to bass as treble, accordingly bass notes must be produced with more power to be perceived as loud as treble notes. Consequently, since bass requires more power to be heard, bass amplifiers need to be more powerful.

When people listen at anything under 80dB or so the hearing response rolls off in both the bass and treble. To get around this many receivers have a "Loudness" control that boosts both the bass and treble. In addition when listening at low levels dynamic range is supressed too. You would have a tendency to loose low level detail. Many receivers also have a "midnight" or low level dynamic control that compresses the dynamics for better low level detail.

Audio engineers have know this for years, and when they mix a recording they listen at an exact median volume, (I believe 85dB) so that they can hear with the best ability that they can.

[Feanor]

When people listen at anything under 80dB or so the hearing response rolls off in both the bass and treble. To get around this many receivers have a "Loudness" control that boosts both the bass and treble. In addition when listening at low levels dynamic range is supressed too. You would have a tendency to loose low level detail. Many receivers also have a "midnight" or low level dynamic control that compresses the dynamics for better low level detail.

Audio engineers have know this for years, and when they mix a recording they listen at an exact median volume, (I believe 85dB) so that they can hear with the best ability that they can.

Further this subject, if you have a Rat Shack SPL meter, you'll recall that they have an 'A' and a 'C' setting. The former approximates the typical perceived human hearing and reports bass a lower lever than the true energy. The latter, 'C', is a true, flat reading and, of course, is what to use for setting equalization

[harley .guy07]

Well ... have we come to a conclusion? any one still interested in this post?? hello??

I gave up on my point a long time ago. I feel like I am back in damn high school in debate class reading this post. I know what I am talking about but some people are so set on what they want to say that they just can't realize that there is a lot of factors that can be variable in all of this and with human hearing being inaccurate at best then it also changes the simple part of the equation. So with taking that in to consideration I am done trying to make my points. Mr. Peabody has been the one that has shown the most data to back up his claims in my opinion and the rest of everyone is just hopping and skipping over things and arguing the points you want to argue and I guess that's fine but I have studied audio for many years and I know what I know to be true and that's all I could give two ^&*('s about and I will keep on keepin on.

[Mr Peabody]

I'm done. I've consulted three different person sources, two engineers and one in the technical area for a manufacturer. In addition to the web articles. I don't know if it's in the way the question is posed or maybe there are many aspects to the question but I have yet to get anything that completely answers the question. There's a part of the puzzle missing. I still have the link to this thread and will come back if I ever find what I'm looking for.

In practice, whether pro sound or car audio we've always used larger amps on bass. If we could have gotten away with smaller power I'm sure it would be done in a cost saving measure.

It would seem if power was irrelevant when using tweeters and woofers of same sensitivity that more manufacturers would do that in order to conserve amp power and save on parts used to even out the power distribution within a speaker.

Nash, for now you'll just have to weigh what we have and draw your own conclusion. If you are into car audio, you let me know when you see some one using larger or even same size amps on their mid/tweeter section as they do on the subs. Whenever you see Alpine, Rockford, PPI, or any reputable car amp manufacturer make a 5 channel amp, the sub channel is several times larger than the satelite channels and they don't even know what speakers you will use. If it's four channel, at least two can be bridge. No one bridges for tweeters. It's just not done. Bass needs more power. There's more to the story than just sensitivity. What the complete answer is I am not able to deliver at this time or maybe never. It depends on if I ever find any one who actually knows and capable of conveying it in a logical form.