Speaker efficiency vs sensitivity

[gene]

I read it, but about half way though everything got fuzzy and I forgot what I was reading. I understood a little of what he was saying but not enough to be able to,.... well, understand the overall picture.

Re-read this excerpt:

That compression driver will have different axial sensitivities on the different horns, but the exact same efficiency. (The difference is how the sound power is spread out). IF we have a speaker which is practically a point source, radiating its sound in every direction equally and simultaneously, then the efficiency can predict the axial sensitivity.

What he is saying is two identical drivers can have exact same efficiency but different sensitivity based on how the sound radiates out of the driver. The only time sensitivity and efficiency can be used interchangeably is when the driver acts as a point source which you can never have in a system that uses physically separated drivers to produce different frequency ranges.

 

[gmichael]

Re-read this excerpt:


What he is saying is two identical drivers can have exact same efficiency but different sensitivity based on how the sound radiates out of the driver. The only time sensitivity and efficiency can be used interchangeably is when the driver acts as a point source which you can never have in a system that uses physically separated drivers to produce different frequency ranges.

Thanks. So in techno-speak, they are not the same (unless more than one driver can radiate from the same point). But maybe they are close relatives? Outside of the techno world, maybe it's not so bad to use one for the other and most people will at least understand what you are saying.
In any case, it doesn't bother me if someone uses one for the other. It also doesn't bother me if someone with a techno background points out their differences. (Even if what they say goes right over my head)

 

[j_garcia]

Thanks. So in techno-speak, they are not the same (unless more than one driver can radiate from the same point). But maybe they are close relatives? Outside of the techno world, maybe it's not so bad to use one for the other and most people will at least understand what you are saying.
In any case, it doesn't bother me if someone uses one for the other. It also doesn't bother me if someone with a techno background points out their differences. (Even if what they say goes right over my head)

They are directly related and as mentioned earlier, a more efficient driver could yeild higher sensitivity. Related doesn't mean interchangeable though, since one sort of relies on the other.

 

[jostenmeat]

I read it, but about half way though everything got fuzzy and I forgot what I was reading. I understood a little of what he was saying but not enough to be able to,.... well, understand the overall picture.

I believe the nutshell is as follows:

Efficiency: power output, total, including ALL directions, for a specified power input.

Sensitivity: power output, for ONLY on-axis, for a specified power input.

Oh, people have already answered. Hm, I see you have already posted again. Ok. How about this analogy:

I have a specified power input of WATER, through my two identical water hoses, that are both putting out the same amount of water. However, one is using a wider nozzle, and its power is spread about more. The other hose is using a narrower nozzle, and the power is focused more directly out of it. IOW, If you sprayed me with both hoses, the narrower nozzle would make feel more pressure on any certain point of my body (higher sensitivity, as in ON-AXIS), than the other, even when the total output is identical when considering all angles. Well, I tried.

I think the whole Q thing he was talking about could describe the size of the nozzle. Sorta. You can instead think of two identical spray bottles, with different nozzle settings. Squirt vs spray. Same output overall, but when measured perfectly onaxis (perfectly in front of nozzle), the squirt is stronger.

 

[Vaughan Odendaa]

It seems like everyone has an opinion about this. The accepted usage is that efficiency is a measure of power output/power input. (we assume this is constant, but like all speaker parameters, it is not)

So then in the world of mathematics both terms scale with one another but in real world terms they don't ? Because impedance changes and that changes efficiency which changes sensitivity ?

Its like I mentioned on the first page about the 8 ohm speaker and 4 ohm speaker having identical sensitivities but different efficiencies. Is it inaccurate to state that the more efficient a speaker, the more sensitive it will be ? In my mind, I think it's a blanket statement because it depends on several different variables.

The EE that I've discussed this with has said that both terms are dependant on the varying impedance, that both are average values integrated over a frequency band and that both vary in the same manner, and when talking about relative differences between loudspeakers, they indicate the same tendencies.

He then mentions that doubling the efficiency gives a 3dB rise in sensitivity and vice versa. But that isn't telling the full story is it ?

 

[Vaughan Odendaa]

Just want to extend a big thank you for everyone who contributed to the thread. I appreciate it very much. Still have no idea but I think as the thread progresses I'll get a better picture.

 

[TLS Guy]

Sensitivity and efficiency are related but not the same.

Unfortunately there are two ways of rating sensitivity.

The db rating at 1 meter with a watt of power. There is is also the 1 meter db level produced by a voltage at the speaker of 2.83 volts.

2.83 volts into an 8 ohm resistive load will consume 1 watt of power.

So a four ohm speaker will consume twice the power to produce the same db output at 1 meter than an 8 ohm speaker on that rating.

Now the efficiency of a speaker is the amount of total power as sound energy radiated from the speaker as a percentage of the power input to the speaker.

This is usually of the order of around 1% the rest going as heat. Horns can do much better in this regard. So most speakers are inherently very inefficient and the vast majority of your amp power goes as heat.

The big point is that for the same sensitivity using the 2.83 volt at 1 meter rating a four ohm speaker will be half as efficient as an 8 ohm one for the same db reading at 1 meter.

 

[Vaughan Odendaa]

TLS Guy, that's what I thought, but then if that's the case then one can't compare sensitivity and efficiency (they don't scale correctly) in the real world. Then you have the extra variable of on-axis, off-axis dispersions.

What the EE was telling me was that both terms vary in the same manner. But that isn't true, at least that is what I'm getting from Paul. If you take this equation : Sensitivity(dB) = 112 + 10*log(efficiency) and then assume that that is a linear relationship between efficiency and sensitivity, well then in my mind that is false, because that excludes several variables that come into play in the real world.

Doesn't that math equation assume a constant, perfectly linear speaker with perfect dispersion in every direction ? I mean, one can't just say that a more sensitive speaker is more efficient because that isn't necessarily true. Just as it's not necessarily true that a more efficient speaker has higher sensitivity.

I await Pauls response for further clarification.

 

[highfigh]

Think of efficiency as the ratio of output:input and sensitivity as the SPL that results from a specific voltage at the input WRT a nominal impedance. 2.83V into 8 Ohms works out to 1.001 (plus a little) Watts but it will be double that with a 4 Ohm load. As long as the measuring mic is at the same distance and the acoustical environment is the same for all tests, it's valid. If one location is highly reverberant and the other is acoustically dead, the results will be somewhat different, depending on the directionality of the mic.

 

[Vaughan Odendaa]

I just need to know if I'm understanding this correctly (my last response).

 

[TLS Guy]

I just need to know if I'm understanding this correctly (my last response).

You have it right.

The total sound energy radiated from a speaker is hard to get and reproduce.

The sensitivity measurement is a at a point source 1 meter on axis.

So all you can say is that the more sensitive speaker will radiate more power at that defined point source then the less sensitive one for the same amplifier power.

 

[Vaughan Odendaa]

Frequency response will also play a factor in sensitivity, correct ?

 

[TLS Guy]

Frequency response will also play a factor in sensitivity, correct ?

If the speaker has a huge peak it will be more sensitive. So yes a speaker with a good smooth response will seem less sensitive.

Speaker manufacturers like Klipsch and Cerwin Vega know this, and give them a nasty mid range shout, which makes them seem louder as well as appear more sensitive.

Good speakers always appear to sound softer than poor ones as a rule.

 

[kavkazi]

I would prefer sensitivity

 

[highfigh]

I would prefer sensitivity

I feel your pain.

 

[Vaughan Odendaa]

So horn loaded speakers radiate sound more in all directions, at all angles than conventional speakers, which explains the efficiency increase and the sensitivity increase ?

This is confusing.

 

[Vaughan Odendaa]

So just to recap, the sensitivity and efficiency measurements are different, which will lead to different results. Speaker manufacturers don't usually measure efficiency do they ?

 

[Vaughan Odendaa]

Yet another question. If you have two speakers, one is a conventional woofer speaker and the sensitivity is 100 dB's and you compare a horn loaded speaker with identical sensitivity, the efficiency will be completely different between both because of radiation impedance ?

 

[TLS Guy]

Yet another question. If you have two speakers, one is a conventional woofer speaker and the sensitivity is 100 dB's and you compare a horn loaded speaker with identical sensitivity, the efficiency will be completely different between both because of radiation impedance ?

No horns are generally much more directional. Also their directionality can be tightly controlled. The other issue is that horns are the most efficient acoustic transformers, so efficiency is the highest of all designs. However to maximize potential, then all drivers need to be horn loaded. If you just have an HF horn it has to be padded down to match the other drivers. A speaker will all drivers horn loaded correctly can produce enormous spl with a watt or two.

 

[Vaughan Odendaa]

I'm just trying to understand what Paul was saying. So horns have more a focused beam of intensity directly in front and the sensitivity is extremely high but the efficiency, as I understand it now, is based on intensity from all angles. So if horns are the most efficient designs possible then they must focus well at all angles...

If a speaker does not radiate sound at other angles well then it's efficiency won't be as high. I could be completely off-base with my interpretation of all of this. I just figured that if horn-loaded speakers were the most sensitive speakers with the best efficiency that would imply that their on-axis sound intensity is very high as well as off-axis results...

Does any of that make sense ?