Can someone please explain the excessive tweeter wattage in car audio?

[ozzuneoj]

This is something I can't wrap my head around so I figured I should finally just ask.

Why do some people talk about making hundreds of watts available to tweeters to improve sound quality? Often times they're talking about speakers with 92db or 94db of sensitivity. I'm not that educated on the technical\engineering aspects of this stuff, so I'm genuinely curious as to how this could possibly work. Are there any actual recorded measurements online showing a tweeter pulling hundreds of watts? I don't mean that it's simply hooked up to a 400 watt per channel amplifier... I mean that amount of power measured with a meter, or oscilloscope or something, being passed through a tweeter. Preferably done at actual listening levels.

An example discussion:
https://www.diymobileaudio.com/threa....391026/page-2

Specifically this post:
https://www.diymobileaudio.com/threa...2#post-5386930

I originally bought the sb 29s because of the advertised sensitivity and on axis frequency response chart thinking it would be a decent upgrade over the CT-2s but in a clear direct swap comparison, they definitely are not. They are far less sensitive than the massives. I even bridged one of my spare 4 channels to give them 450 watts each just to see if they wake up from that. No avail but they did sound A LOT better than being on 150 watts however the CT-2s even on 150 watts sounded overall better. It can also go way louder if i needed it to.

This is in reference to these tweeters:
https://www.madisoundspeakerstore.co...-dome-tweeter/

Even the manufacturer rated 100W power handling of that tweeter seems completely crazy to me.

I built a set of large Dayton 8" MTM speakers for my living room several years ago and they're connected to an old late 90's JVC RX-817VTN receiver rated for 105 watts per channel at 8 ohms in stereo mode (20hz-20khz, 0.007% THD). I can make stuff in the house rattle, and I get fantastic low frequency effects in movies with no subwoofer. The tweeters sound very clear to me, and if you crank it up you can fill the entire downstairs of my house with sound (while making the area in front of them uninhabitable due to the volume). This is with 105 watts x 2 channels into 41" tall floor standing full range speakers. I know that a car is a far less ideal listening environment, but how can a tweeter possibly pull (and dissipate) that same amount of power, while only playing frequencies above 3khz and being far more efficient? Where does all of that energy "go"? And how on earth could going from 150 to 450 watts improve sound quality noticeably?

Part of me feels like I wont be satisfied until I see the lights dimming on some dinky car with 2x500W RMS running exclusively to tweeters... but I have a feeling there's some complicated audiophile-technobabble answer to explain why tweeter watts are different or something.

'splain it to me guys. 'splain it like I'm five.

[nadams5755]

the sb-acoustic is 94db 2.83v/1m, which works out to be 91db 1w/1m. massive doesn't say how the ct-2's 92db sensitivity is measured. we could assume the ct-2 is maybe 1db more sensitive? also need to look at the response curve. is the ct-2 flatter? does it have more output at some frequencies than the sb-acoustics tweeter?

sb-acoustic is a ring-radiator, ct-2 is a dome. both have different output patterns for off-axis response.

in a car, installation is another variable.

lots of variables. you'll probably be more than fine with 100w to your tweeters.

[ozzuneoj]

the sb-acoustic is 94db 2.83v/1m, which works out to be 91db 1w/1m. massive doesn't say how the ct-2's 92db sensitivity is measured. we could assume the ct-2 is maybe 1db more sensitive? also need to look at the response curve. is the ct-2 flatter? does it have more output at some frequencies than the sb-acoustics tweeter?

sb-acoustic is a ring-radiator, ct-2 is a dome. both have different output patterns for off-axis response.

in a car, installation is another variable.

lots of variables. you'll probably be more than fine with 100w to your tweeters.

Sorry, I was just asking about tweeter power handling. I don't have any of the tweeters mentioned in this thread.

[Justin Zazzi]

OP is right, nobody is sending 100w to the tweeters in their cars.
If you substitute the word "volts" for "watts" then it makes more sense.

I routinely measure power going to automotive (and pro audio) tweeters during power testing. Most of the time I only need a 5w-10w amplifier to stress test a tweeter at max output. However, sometimes I need the voltage potential that a 200w amplifier has in order to make it loud enough.

Put another way, the current that a tweeter draws while playing music is incredibly small because the treble content of music is much lower in power relative to the bass content in music. However you still need the voltage in order to get loud enough and to reproduce the transients without clipping, which we are much more sensitive to in the higher frequencies.

For the technically inclined, you can see this as an increased crest factor in the treble. The peak voltage needed for treble is similar to the peak voltage for the bass, however the average voltage needed for treble is substantially lower than the average voltage needed for bass. Meyer Sound created a test stimulus called M-Noise that is supposed to be a music analog and it is similar to pink noise however the crest factor of the treble is much higher.
https://m-noise.org/

Once in a while you'll see a tweeter amplifier channel that has high enough voltage but low current capability (meaning low power or less watts). JL Audio had a multi-channel amplifier for this back in the 90's or early 2000's I think. It had a pair of channels with a large power rating and a pair of channels with with less power. I might have misunderstood that product since I just found it again and it's marketed as a "front" and "rear" channel sets. So nevermind about that. It was the JL 500/5 or the 450/4, slash series v1.

In order to the get the voltage high enough, you need an amplifier that is typically rated with lots of watts because that is how the industry rates things.

@Jdunk has some good thoughts on this too.

[ozzuneoj]

Ohhhhhhhhhhhhhhhhhhh....

Hey that actually makes sense: When talking about tweeters it has nothing to do with power (wattage) and everything to do with voltage.

So if this is the case, then the idea that you should match the rated wattage of an amplifier to the power handling of a set speakers is a bit misleading when trying to get the best sound quality? Ideally, you'd get the best amplifier with the highest voltage potential, which would likely end up being one advertised with a higher wattage? On that same note, is a more sensitive tweeter more likely to sound good with a low wattage amplifier than a less sensitive tweeter on the same amp?

Is there a way to know, realistically, how an amplifier performs in this area and what you need to make a specific tweeter "happy" at bearable listening volume?

As mentioned, I don't understand a lot of the technical terms here. Been working on computers and audio stuff for over 20 years but never dug deep enough into the fundamentals of electronics and how all this stuff interacts. A good explanation of transients would probably help me to figure this stuff out better.

...then again, it might also lead me to spending a boat load of money on an amplifier and a new set of tweets\mids to get more out of my system.

[Justin Zazzi]

So if this is the case, then the idea that you should match the rated wattage of an amplifier to the power handling of a set speakers is a bit misleading when trying to get the best sound quality? Ideally, you'd get the best amplifier with the highest voltage potential, which would likely end up being one advertised with a higher wattage? On that same note, is a more sensitive tweeter more likely to sound good with a low wattage amplifier than a less sensitive tweeter on the same amp?

Yes, matching wattage numbers can be misleading and frustrating and confusing. Some manufacturers try to make it clear what size amplifier you should use and I like JL Audio's graphics like the one below. However, it is not clear what standard was used to create these numbers so does "continuous power handling (rms)" mean this woofer can actually handle 500w? Or since it looks like your warranty is void if you use more than a 500w amplifier, is the manufacturer recommending the consumer should use an amplifier that is 500w capable?




Other manufacturers like Rockford Fosgate might list a testing standard like in the picture below, but then you have to know what that standard means and how to interpret the numbers. In this case, the power handling might not always mean the speaker can actually handle that much power and it might actually be a recommended amplifier size instead.



If you can figure out how many watts you actually want to use, then you need to look at the impedance of the speaker too. Most amplifiers will generate more power at lower impedances so if you really want to deliver 500w worth of voltage to a 4ohm speaker, then you might really need a 1000w amplifier because that 1kw amp only delivers half of it's power at 4ohms.

Is there a way to know, realistically, how an amplifier performs in this area and what you need to make a specific tweeter "happy" at bearable listening volume?

Assuming you can figure out how many watts and what impedance and so on ... then you need to consider that music is dynamic with loud parts and quiet parts (dynamic range or crest factor). Music, in general, has a crest factor of 6dB to 12dB. More agressive and more modern music has a smaller crest factor close to 6dB which is a power multiplier of 4x. But we're not really interested in having excess power, we want excess voltage potential.

This is where common rules of thumb appear. Some people will say use an amp that has twice as much power as your speaker can handle, and then keep the gains low ... because "headroom". This is one way to have extra voltage available for the transients and louder parts of the music so you don't clip the peaks and sound bad, while also keeping the average power lower so your speaker doesn't overheat

But what number should you double, if any? If the manufacturer has already doubled the power handling of their woofer and recommends you get an amp with twice as much power already, should you double that number again? How would you know if the manufacturer already added this 2x factor into their speaker power ratings? What if the manufacturer also lists a "max rms" power number like Rockford does? Do you double that number? If after all these considerations you arrive at a number like 500w ... how do you know it's a reasonable number and it hasn't already been doubled one too many times or if it was a bit too optimistic to begin with?

The frustrating part is, you'll almost never know. Unless you're working with pro-audio equipment where the specs are super clear and not dominated with marketing rainbows, it's hard to know for sure what any of the power numbers really mean. Even as a professional in the acoustic industry, I struggle with this all the time with consumer products.

I'm not really answering your question though. There are many ways to decide how big an amplifier to use on a given set of speakers. I like to think about it this way: if the manufacturer lists both an RMS and a PEAK or MAX power handling, or if they list an RMS and a MUSIC or PROGRAM power rating (which is more common in pro-audio), then I use the peak/max/music/program number. If there is not a peak/max/music/program number, then I'll double whatever number is listed.

However!!! and this is a huge caveat, I will make sure to not overdrive the speaker since I should have plenty of power available to overheat and smoke it if I'm not careful. Some people don't know how to do that, and if you or your friends like to turn every knob to 11 and blast it all day then maybe don't double any of the numbers and be a little more conservative.

As mentioned, I don't understand a lot of the technical terms here. Been working on computers and audio stuff for over 20 years but never dug deep enough into the fundamentals of electronics and how all this stuff interacts. A good explanation of transients would probably help me to figure this stuff out better.

...then again, it might also lead me to spending a boat load of money on an amplifier and a new set of tweets\mids to get more out of my system.

Transients, crest factor, and headroom are basically the same thing. It's the difference between the peak demand vs the average demand. Imagine you have a 300 horsepower motor in your car. If you take the freeway to visit a friend, how much of the 300hp are you actually using? You might be able to use all of it for a moment on the freeway onramp but you'll only use a tiny fraction of that power while you're cruising or on the surface streets.

In the car analogy, a transient would be a sudden surge in power like if you floor it on the freeway onramp. Headroom and crest factor mean the difference between the maximum available (300hp) vs the average amount you use (10hp?).

In music, a transient would be a sharp sudden event like a kick drum or the snap of a snare drum or crash of cymbals etc. Headroom and crest factor mean the same thing, it's the difference between the maximum power available (500w) vs the average power you're using over a long time (~100w). To learn more, look up the music loudness wars.



image borrowed from https://www.soundonsound.com/sound-a...age-be-stepped

[Holmz]

In the picture above it mentions crestfactor of 12-20 dB.
13dB is an easier number (20x over RMS),

then some SPL number is your friend here, as SPL, efficiency, and distance then allow one to compute RMS power.

and having dynamic range gets one from RMS to peek power.

Erin has a good video explaining it,..

[Jdunk54nl]

Here are a couple of good things to look at.

This is dynamic range in songs, you can look and see what type of music you most often listen to and what average dynamic range it has. This can lead to what Justin was talking about with the need for extra voltage to hit these dynamic ranges:
https://dr.loudness-war.info/


This is also a really good article by Andy Wehmeyer (AudioFrog)
https://www.audiofrog.com/pink-noise...get-150-watts/

Cliff notes of above article:
Basically, music has a natural slope of roughly -3db for each octave (An octave is a doubling of frequency). Every -3db is half the power needed.

So let's say you need 1200w at the lowest frequency of 20hz. (Common sub amps power for sq people)

0db 20hz = 1200w
-3db 40hz. = 600w
-6db 80hz. = 300w
-12db 160hz = 150w
-15db 320hz = 75w
-18db 640hz = 36w (I know it isn't this but math is easier with this number)
-21db 1280hz = 18w
-24db 2560hz = 9w
-27db 5120hz = 4w
-30db 10240hz = 2w
-33db 20,000hz = 1w

This assumes that the song actually has a 0db frequency, everyone that I have looked at. Most start off at like -12db and then follow the above pattern.

This first picture is Pink Periodic noise generated by REW, it is supposed to be set at -12db. We use Pink Noise because of it's -3db slope which resembles music as shown below.

Starts at -19db and ends at -50db, or about -30db difference or a -3db slope for each octave.




This is Michael Jackson's Billie Jean song. Goes from -14db at 70hz to about -50db for tweeter frequencies, a difference of -36db or 1/2000th the power, or you would need 2000w at 70hz to need 1w at 8000hz.
https://dr.loudness-war.info/album/view/185136




This is Billie Eilish Bad Guy. Peak at 50hz close to -10db. By 8000hz we are down to -60db, so -50db difference. I don't even know how little power you would need here....Fractions of watts.

https://dr.loudness-war.info/album/view/162536

[Jdunk54nl]

So, what does all of the above mean?

Take a look at the Michael Jackson song Billie Jean.
Here is billie jean from the loundess war website:
https://dr.loudness-war.info/album/view/185136

...........PEAK........RMS
DR13 -0.29 dB -14.13 dB 4:54 06-Billie Jean


Let's assume you set your gains using a -14db 70hz test tone on your subwoofers. That way you use all 1200w RMS for your sub amp in the example.

This also means that whatever volume level you set your gains, you need to be listening at that max volume for this to be true, otherwise you have already drastically cut needed power. So let's assume you are.

The power needed at 8000hz is 1/2000th the power at 70hz, so you need about 0.6w on your tweeter for RMS. Peak is close to 0db so you need about 14db for dynamic range (assuming the dynamic range is equal throughout the frequencies, which it is probably not) or 5 doubling's of the power (multiply by 2, 5 times) or 20w on your tweeter to meet this goal. Since music isn't a single frequency and your tweeter will probably have to play multiple at once, you will probably need some extra power to do this, but not a lot.

As Justin said up here:

I routinely measure power going to automotive (and pro audio) tweeters during power testing. Most of the time I only need a 5w-10w amplifier to stress test a tweeter at max output. However, sometimes I need the voltage potential that a 200w amplifier has in order to make it loud enough.

*Edited for reasons stated below*

[Holmz]

So, what does all of the above mean?
...
The power needed at 8000hz is 1/2000th the power at 70hz, so you need about 0.6w on your tweeter for RMS. Peak is close to 0db so you need about 14db for dynamic range (assuming the dynamic range is equal throughout the frequencies, which it is probably not) or 5x the power or 3w on your tweeter to meet this goal. Since music isn't a single frequency and your tweeter will probably have to play multiple at once, you will probably need some extra power to do this, but not a lot.

As Justin said up here:

10 dB is a factor of 10x.
13 dB is a factor of 20x
14 dB is a bit more...

0.6W RMS, with a crest factor of 14 dB is about 20x 0.6W... or 12W.

unless one were to band limit the measurements and run the music through it to noodle out the power, then I would stick to at least 10W on the tweeters, and 20/25W would be a bit safer for headroom (IMO). One could move from opinion to fact with some math and bandpassing whatever representative music they want in a 4K-22k range.