Making sense of power requirements, decibels and sensitivity

[DanielT]

You only need a moderate amount of power to get adequate loudness. You need a ton of power to get really loud! Get as much power as you can. If the speaker can't handle it, it will tell you (severe distortion).

Plus if you are simmering with low levels of volume, SPL together with very undynamic music you don't need that much power.
It probably won't be a particularly enjoyable music experience, but only a small Hifi wallet is needed then. That's probably the only advantage.

AND , I almost forgot, it promotes good neighborly relations if you live in an apartment complex.
(solution if thats the case and when you want to go all out loud - headphones)

 

[kyuu]

I posted this in another thread but it might fit better in this one. An example with a piece of music. Mostly to show that it may also be necessary to have a lot of amp power in the higher frequencies. If it is dynamic music that requires it, that is:

I will use the track Michael Ruff - Speaking in Melodies - Wishing Well (well-recorded fusion) as an example, as there are several high-frequency peaks that require a lot of power from the amplifier. First, the voltage, current and power for the entire signal are shown:
View attachment 444305

As you can see, we get an active power of 710 Wp at most.

Then we only look at frequencies above 7 kHz (HP filtering with 3rd order Butterworth):
View attachment 444306
For frequencies above 7 kHz, we get an active power of about 425 Wp. In addition, the strongest high-frequency peak does not occur at the same time as the strongest peak for the total signal.


We can also look only at frequencies above 10 kHz (HP filtering with 3rd order Butterworth):

For frequencies above 10 kHz we get an active power of about 225 Wp. Furthermore, the strongest high-frequency peak does not occur at the same time as the strongest peak for the total signal.

faktiskt.io • Visa tråd - Stereophile recension av Buckeye Purifi Eigentakt

Can you actually send ~500W into a tweeter, even for brief transients, without it going up in smoke? Not a rhetorical question, btw, I'm actually curious.

 

[bmc0]

Can you actually send ~500W into a tweeter, even for brief transients, without it going up in smoke? Not a rhetorical question, btw, I'm actually curious.

The short answer is yes. The slightly longer answer is that it depends on the tweeter's construction and the duration of the transient. Ferrofluid should, I believe, increase the maximum transient power. Morel happens to provide a transient power rating for their tweeters—the CAT298 is rated for 80W nominal (DIN) and 1kW for a 10ms transient.

 

[Old_School_Brad]

Is it not true that, once the capability of an amp's power supply rails is surpassed and the waveforms are clipped, a lot of high frequency harmonics are produced, and this is what generally blows a driver?

The key difference lies in the waveform shape and the resulting amount of power delivered to the driver. -Remember, it's power that burns voice coils.

Take the earlier example of a 25 W amplifier: to deliver 25 W into an 8-ohm load, it needs a supply voltage of around 20 V (√(25×8)=20 V).

If that same 20 V were used to output a square wave instead of a sine wave, the power could effectively double to 50 W, calculated as (20×20)/8.

On top of that, square waves contain strong high-frequency harmonics, which add significant energy in the upper spectrum precisely where tweeters operate. This sudden influx of power and high-frequency content is often what pushes tweeters beyond their limits… and makes them go tweet for the last time.

Can you actually send ~500W into a tweeter, even for brief transients, without it going up in smoke? Not a rhetorical question, btw, I'm actually curious.

Sure, it may sound like a lot, but since it's only for brief moments, it doesn’t really pose a problem.

Here’s a snippet from the datasheet of a driver I use -one I knew had excellent transient handling. It’s rated for up to 1000 V.

What’s interesting is that this isn’t just for a half-cycle spike, but for a full 10 ms -which is quite significant, especially in the high-frequency range where such durations are actually useful.

 

[DanielT]

Can you actually send ~500W into a tweeter, even for brief transients, without it going up in smoke? Not a rhetorical question, btw, I'm actually curious.

I saw that you got answer to your question. If a tweeter hadn't handled these short transients, there would have been blown tweeters here and there.

That Michael Ruff - Speaking in Melodies may be a dynamic recording but there is a lot of other music with similar dynamics that is listened to and then with amplifiers that have much more power than the RMS power that a tweeter is said to be able to handle.
I dare to listen to that Speaking in Melodies song with my 200 watt amp and these tweeters for example and they have:

Power Handling (RMS) 20 watts
Power Handling (max) 40 watts
(I have a fairly high crossover point plus steep crossover filter with them so it protects them from being blown)

Dayton Audio - ND25FW-4 1" Soft Dome Neodymium Tweeter with Waveguide 4 Ohm

ND25FW-4 1" Soft Dome Neodymium Tweeter with Waveguide 4 Ohm

www.daytonaudio.com

BUT in that case I would NOT turn the volume up too high. Transients are one thing, continuous power is another

 

[Old_School_Brad]

I saw that you got answer to your question. If a tweeter hadn't handled these short transients, there would have been blown tweeters here and there.

That Michael Ruff - Speaking in Melodies may be a dynamic recording but there is a lot of other music with similar dynamics that is listened to and then with amplifiers that have much more power than the RMS power that a tweeter is said to be able to handle.
I dare to listen to that Speaking in Melodies song with my 200 watt amp and these tweeters for example and they have:

Power Handling (RMS) 20 watts
Power Handling (max) 40 watts
(I have a fairly high crossover point plus steep crossover filter with them so it protects them from being blown)

Dayton Audio - ND25FW-4 1" Soft Dome Neodymium Tweeter with Waveguide 4 Ohm

ND25FW-4 1" Soft Dome Neodymium Tweeter with Waveguide 4 Ohm

www.daytonaudio.com

BUT in that case I would NOT turn the volume up too high. Transients are one thing, continuous power is another

Your ears would give out and your mids and woofers would be in flames long before your tweeter ever gets 20 continous watts.

 

[DanielT]

Your ears would give out and your mids and woofers would be in flames long before your tweeter ever gets 20 continous watts.

I was just going to add this as an edit to my #25 , post:

Or with that Dayton tweeter and my amp, the t.amp E-800, that has 490 W at 4 Ohm (which I normally use for my bass boxes) turned up really high is not something I would do. Besides, I could have much bigger problems then than a pair of blown cheap tweeters (or the rest of the blown speaker):
A pair of blown ears. Perhaps that is an exaggerated claim that they should be blown, but you should be careful of your ears because they can get damaged.

A general tip, reminder to those reading the thread. Don't forget when visiting a concert these:

 

[Anton S]

The key difference lies in the waveform shape and the resulting amount of power delivered to the driver. -Remember, it's power that burns voice coils.

Take the earlier example of a 25 W amplifier: to deliver 25 W into an 8-ohm load, it needs a supply voltage of around 20 V (√(25×8)=20 V).

If that same 20 V were used to output a square wave instead of a sine wave, the power could effectively double to 50 W, calculated as (20×20)/8.

Anecdotally, I used to experience power-related tweeter failures periodically at home. The T33 tweeters in my KEF 104/2's were notably weak, and I would occasionally lose one due to throwing too much clean power at it. The tweeter would be working fine and sounding great one moment, with no audible sign of distress, and then suddenly go completely silent. This became prohibitively expensive after a while, because KEF would sell drivers only in matched pairs, and they weren't cheap. Eventually, I replaced the T33's with a pair of Morel MDT29-4 units, and the problem never resurfaced, even with the same amp and same listening levels. Those 104's are still running fine, although now relegated to my bedroom system.

By the same token and speaking from practical experience, I have seen many, many more drivers trashed due to clipping from 20 - 30 Wpc amps and receivers than from being subjected to too much clean power. Back when I worked for a full-service audio shop, this was the condition we nearly always noted when a damaged speaker was brought in for service. A client would drag in a hurt speaker wearing an expression of bewilderment, saying something like, "I don't understand what happened. These speakers are rated at 100W, and my receiver is only 30." (Note that even when run into hard clipping, the power amp section of the client's receiver could pump no more than 60W into it.) It was generally easy to confirm clipping as the root cause by just connecting up the speaker, because tweeters subjected to too much unclipped power would always be completely silent due to their open voice coils (fuse effect - see above), while tweeters destroyed by clipping would still produce sound, but the sound was scratchy and distorted (blistered voice coils).

 

[DanielT]

From what I heard, the T27 tweeter burned out quite easily. Ok, it was used in many speakers/models. Here for example, a Kef Cadenza speaker with the T27 tweeter I had a couple of years ago:

... so if that happened every now and then maybe that's the explanation? It was so common that it was inevitable that some (how many?) could burn out. Note, what I heard. Not sure if that's true. Maybe you know more about that @DSJR ?

With RMS of 8 watts and maybe together with the first possible second-order crossover filter, which I guess many speakers/models had (wrong?), it must have been sweaty for them as well.

Edit:
Plus, if they burn out now in half-century-old speakers, I think they should actually be seen as having held up well since they lasted so long.

 

[DSJR]

From what I heard, the T27 tweeter burned out quite easily. Ok, it was used in many speakers/models. Here for example, a Kef Cadenza speaker with the T27 tweeter I had a couple of years ago:
View attachment 444441
... so if that happened every now and then maybe that's the explanation? It was so common that it was inevitable that some (how many?) could burn out. Note, what I heard. Not sure if that's true. Maybe you know more about that @DSJR ?

With RMS of 8 watts and maybe together with the first possible second-order crossover filter, which I guess many speakers/models had (wrong?), it must have been sweaty for them as well.

View attachment 444440

Edit:
Plus, if they burn out now in half-century-old speakers, I think they should actually be seen as having held up well since they lasted so long.

I'd suggest the burn-outs were more to do with clipping amps too low in power, as in the T27's production heyday, typical amps used were way under 50WPC I'd suggest and didn't clip that graciously, especially with the usual vinyl sources of the times, which also taxed bass drivers a goo dbiut back then.

 

[Anton S]

@DSJR

What about the T33? At the time, I was running my 104/2's with a Marantz Model 500, which was rated at 500 Wpc into 4 ohms. Nothing but clean power ever went into those speakers. I never put anywhere near 500 into them, because the KEF's were very efficient at 92 dB at 2 M with 2.83 V, as well as being rated for use with amps up to only 200 Wpc.

 

[Old_School_Brad]

...Back when I worked for a full-service audio shop, this was the condition we nearly always noted when a damaged speaker was brought in for service. A client would drag in a hurt speaker wearing an expression of bewilderment, saying something like, "I don't understand what happened. ...

Hehe, I know it -been there myself decades ago. Every time I cut open a blown tweeter, the voice coil was a charred, blackened mess. And the classic line? "I wasn’t playing it that loud.” Sure thing, buddy.

 

[DanielT]

Burning speakers/drivers is easy to do. Just feed them enough power and the right type of music for the purpose and boom.
Speaking of tweeters from 0:15 in the video below, a tweeter with a capacitor burns. OR actually more accurately in this case it seems to be the capacitor that explodes. You see it for yourself.

Later in the video, the man pours gasoline into a speaker and sets it on fire.
That man seems to want to take out his frustration or whatever it is and does so on some crappy drivers, speakers:

 

[Anton S]

Hehe, I know it -been there myself decades ago. Every time I cut open a blown tweeter, the voice coil was a charred, blackened mess. And the classic line? "I wasn’t playing it that loud.” Sure thing, buddy.

Yep, a common lament. At first, I thought the puzzlement was feigned and just some half-hearted attempt to score an undeserved warranty repair. After all, who could have such a tin ear as to miss the distorted sound of clipping and drivers in distress? (As you know, warranty nevers cover abuse.) But it became apparent that some people's' ears really are that undiscerning after I'd seen several older speakers that were well past their warranty periods come into the shop for repair. Their owners had every expectation of being expected to pay to have them put right, but accompanied them with the same sad tale.

 

[Greenman]

Hi there!

I'm creating this little thread to ask a question for what is probably a very dumb question, but I can't for the life of me figure it out.

Let's say you have speakers rated for 88 dB sensitivity (2.83 V @ 1m).
The speakers' maximum SPL is 112 dB.
Their power rating is 40 to 120 W.

Now, adding 3 dB means doubling the volume. Going (112 - 88) / 3 = 8, so going from 88 dB to 112 dB is multiplying the volume by 2^8, which is 256. Am I correct on that one?

If that's true, then I don't get the numbers in the spec sheet. 1W gives you 88 dB, so to get 112 dB you should need 256W, right? If so, why is the maximum power rating 120W?

I'm sure there's some dumb mistake somewhere in my understanding of how this works, but I'm not sure why.

Could someone please explain me? Thanks

Not sure if anyone else suggested this, but…. I think they are probably rating max SPL assuming you have 2 speakers. Then 120W into each channel will give you around 112db.

 

[Anton S]

Not sure if anyone else suggested this, but…. I think they are probably rating max SPL assuming you have 2 speakers. Then 120W into each channel will give you around 112db.

It should be specified per speaker. You can calculate each speaker's theoretical max SPL, based on on its rated sensitivity and power handling capability. The formula for this is

Max SPL = Sensitivity + 10 * log10(Peak Power / Reference Power)