Speaker Sensitivity and Impedance

zombie boy 2000

I've noticed something a bit quirky in my audio travels. That being speakers with a high sensivity (say 92db) and a 4ohm impedance. And conversely, speakers with a very low sensitivity (84db) and an 8ohm impedance.

I suppose I was operating under the misguided illusion that the two went hand in hand. But upon discovering these anomalies, I think I had it all wrong.

Seinfeld sez.... "what's the deal?"

steveinaz

Sensitivity (aka efficiency) is a measure of output at a given input level and distance---OR how loud a speaker will play at a given wattage. Impedance is the total resistance (load) the speaker/crossover present to the amplifier.

zombie boy 2000

So basically the two can cancel each other out in the case of the two examples I gave. In other words, both would "perform" equally?

jakelm

I believe sensitivity has to do with the intire Theil parameters. Qms, Qts, Qes.

Factor in the ohms law and all parameters, there is a way to figure the sensitivity.

This was a cool little read.

http://www.integratedaudio.com/help/sensitivity.pdf

zombie boy 2000

I thought I had a firm grasp of the two concepts. I guess the only thing I'm trying to wrap my mind around is the concept of a speaker that is simultaneously "sensitive" and "resistive". For example: a speaker rated at 93db but with a nominal impedance of 4ohms.

AsSiMiLaTeD

I'm not sure if this answers your question, but there's no direct correlation between impedence and sensitivity. One can probably be calculated given the other and enough addition information, but that kinda goes for anything.

There is no direct corrrelation that allows us to say something like 'speakers with low impedence have higher sensitivity' or 'speakers with higher sensitivity must have a lower impedence'. You can't correlate one to the other without factoring in other values...

zombie boy 2000

That's a fair and easy answer. So if a direct correlation does not exist, but both attributes can serve as factors in determining the nature of the other -- this is why we traditionally see them going hand in hand (i.e. high sensitivity/high impedence and its converse).

Is that what you're saying?

AsSiMiLaTeD

I'm not sure really...

It's almost like those two attributes are each a result of other factors (components used in the crossover network, drivers, etc). However, maybe they're not a result of the same set of factors. So factors A, B, C, and D determine impedence, and factors A, B, C, and E determine sensitivity.

This could leave you with a scenario where there appears to be a direct correlation most of the time, but in certain circumstances where there is enough variance between the factors, you could see the anomaly where a speaker with higher impedence may also have lower sensitiivity.

zombie boy 2000

And a lightbulb just went off over my head...

thanks Pmaniac.

jrlouie

I've had one set of speakers that were 4 ohms with a 102 dB sensitivity and another set of speakers that were also 4 ohms but only 88 dB sensitivity.

steveinaz

zombie boy 2000 wrote: »

So basically the two can cancel each other out in the case of the two examples I gave. In other words, both would "perform" equally?

No, the 92dB speaker will play at 92 decibles with 1 watt of input at 1 meter*. Impedance has nothing to do with efficiency/sensitivity.

The 4ohm'er will require a robust amplifier if loud volume is wanted, due to the lower load. It also doesn't help that the 4ohm has a sensitivity of 84dB, which is on the low end, requiring more power for louder output.

In essence, the 8ohm speaker is playing 8 decibles louder than the 4ohm speaker with the same 1 watt of input. Therefore, more efficient.

*Typically the measurement is 1 watt/1 meter, some are based on voltage input.

Spacedeckman

Guys, you are doing too much thinking here, and not enough studying.

There are a ton of speaker parameters out there, most of them never get talked about. Some are important, some less so, but most exist outside of the sonic attributes of the speaker itself. So, let me give you a very, very short and rudimentary overview of driver and speaker engineering. You will learn something, even if you don't fully understand it today.

4 ohms vs sensitivity: Yes, most of the speakers with very high sensitivities are 4 ohms. The exact sensitivity spec is 1 watt, 1 meter, 1 kHz. Somewhere in the small print, they will define "watt" as 2.83 volts, and by maintaing 2.83 volts into 4 ohms, they have essentially doubled the power and gained 3 dB of "sensitivity". The temptation is there to alter the frequency response of the speaker at 1 kHz to get a couple of extra dB as well. Take such numbers with a grain of salt if the manufacturer is trying to sell a speaker on sensitivity alone. All manufacturers "fudge" a bit on this spec, some make products to look good there....sound usually suffers if it is a "goal".

High sensitivity speakers don't make much bass. It's a function of their driver design, and a real world engineering tradeoff. A very sensitive driver will have a thin top plate and a short voice coil. When you are really moving the driver (ie, playing loud or low), the voice coil spends much of it's time outside of the strong part of the magnetic field, meaning that it can't be well controlled. You can go the other way, and make a really thick top plate and add some windings on the voice coil and create a driver that will have tons of control and accuracy, but be a total pig to drive. Most manufacturers today split the difference, and usually use a long voice coil and a moderately heavy top plate. You get good sensitivity, good frequency response, and can adjust accordingly to the needs of the driver (tweeter, mid, or woofer). Funny thing, you tend to get a speaker with a sensitivity in the high 80s to low 90s. Look at the numbers for most quality speaker manufacturers. Go to the spec pages of this very site...yep, real world engineering.

Speaker impedence ratings often don't tell much of the story. Here again, there is a tendency to "fudge" a bit. Many receivers won't deal with a 4 ohm load very well. In order to keep the customer interested in buying your speaker, you may want to try to keep the impedence rating higher. Although there are official guidelines of sorts, many speaker manufacturers will use an "eyeball average". One company's 8 ohms is more conservative company's 6 or even 4. Impedence is less of an issue today, since many manufacturers use computer modeling and measurements to limit wild impedence fluxuations (which opens up another can of worms we won't even mention) and allow their products to work well with almost any available receiver. Less so with higher end speakers, since they figure you will have the amplifier to drive the speaker you chose.

Numbers don't really tell you beans about a speaker, especially the way it sounds....well, sort of. I did mention that high sensitivity speakers tend to make no real bass. Funny story. I was a "factory guy" for a speaker company a decade ago, and would use a high quality 6" two way bookshelf speaker volume matched against big Cerwins or Klipsch. The best one, which I got to use MANY times over, was against a Cerwin with 2 15" woofers. I asked the salesmen in the store which speaker made the most bass. They all pointed to that Cerwin Vega, without question or hesitation. I explained the volume matching capability of their speaker switching system, made the adjustments, and did an A/B. Without the advantage of sensitivity, the Cerwin had no place to hide. It was bested on all other fronts by a bookshelf speaker with 1/12 the volume, and IIRC, about $100 less money. Sonically, the bookshelf was a couple orders of magnitude better. Never failed to give a bunch of guys red faces.

When buying a speaker, you need to look at the following. Sound quality (to your ears, which btw, will change over time, not only ears, but preferences), can your recevier drive it, how loud do you want to play it, how big is the room you want to fill, and finally, does it fit your budget/system. Little has to do with specifications.

Another annoying thing is "watt ratings". I hesitate to call them "power ratings" because we typically don't deal with a lot of power in the real world. Most speaker damage is caused by trying to make a little amplifier do the job of a big one. However, in that situation, a "500 watt" speaker will blow a tweeter just as fast as a "100 watt" speaker, hooked up to your 50 watt receiver "turned to 11." (If you don't understand the reference, go rent the movie Spinal Tap) I had a pair of Bryston 7Bs (600wpc...a "meaningful" number, not "Japanese receiver watts") hooked up to my bookshelf speakers which are rated at 150 watts or some such BS. I ran those amps HARD, but stayed within the capabilities of the speakers. At times, that was all the AMP was capable of putting out. BTW, I've never blown a speaker in my life.

School's out.

Mark

heiney9

Spacedeckman- Define a high sensitivity speaker measurement. 90+ or 95+ what to you in terms of the above definition is high sensitivity.

H9

tonyb

Hmmm....while I agree with alot in your post,I will take issue with your statement of high sensitive speakers not putting out alot of bass.Maybe true years ago,but not today.My Legacy focus 20/20 puts out more bass than some subs and has a 95 db sensitivity rateing,and are 4 ohm.And in no way does sound suffer.Most speaker makers today will add a bass module to a horn speaker or planer that typically suffer in the lower bass dept.
btw-Welcome to our humble forum,
And as always,let your ears be the judge.

NeilGabriel

This idea of impedance and sensitivity confused me as well, even the idea that lower impedance speakers require more power is a bit counter intuitive. However, what helps me grasp the idea of impedance and sensitivity being different is that my 91 dB sensitive speaker is 91 if I run it at 8 ohms or in parallel with another speaker, making it 4 ohms. If you tell me I am wrong, my world comes crashing down. I recently read a reviw of someone complaining that too many speaker mfrs were making low impedance low efficiency speakers, thus demanding huge amps.

heiney9

NeilGabriel wrote: »

This idea of impedance and sensitivity confused me as well, even the idea that lower impedance speakers require more power is a bit counter intuitive.

They don't need more power they use more power....that is the power flows more freely and an underpowered amp runs out of juice quicker and shuts down because it can't continure to provide the power (current and to a much lesser degree wattage) to run them at higher levels. A simpleton synopsis but you get the idea.

Impedence and sensitivity is very much like horsepower and torque in a combustion engine. Once you understand how these affect perfromance you begin to understand the differences in engines. Same goes for imp. & sens. in speakers. The numbers alone on a spec sheet don't tell you much other than the general picture.

H9

WilliamM2

Somewhere in the small print, they will define "watt" as 2.83 volts, and by maintaing 2.83 volts into 4 ohms, they have essentially doubled the power and gained 3 dB of "sensitivity".

They haven't doubled the power, or gained 3db of sensitivity. It is still 1 watt, whether it is 4 ohm or 8 ohm.

NeilGabriel

heiney9 wrote: »

They don't need more power they use more power....that is the power flows more freely and an underpowered amp runs out of juice quicker and shuts down because it can't continure to provide the power (current and to a much lesser degree wattage) to run them at higher levels. A simpleton synopsis but you get the idea.

Impedence and sensitivity is very much like horsepower and torque in a combustion engine. Once you understand how these affect perfromance you begin to understand the differences in engines. Same goes for imp. & sens. in speakers. The numbers alone on a spec sheet don't tell you much other than the general picture.

H9

I read a nice article a few weeks ago that used the water pipe and water analogy...with water being the power. Made sense to me, but my daughter is the one studying to be an engineer...the article was actually by a Polk employee now that I think about it.

dkg999

I'm a little iffy on the statement that high sensitivity speakers don't put out a lot of bass. I think back to the big old Klipsh speakers that were very high sensitivity and had tons of bass. I'm currently considering Paradigm Monitor speakers for my HT, and they are around 96db efficient and seem to have good bass output. I'm interested in going to smaller more efficient HT speakers so I can go back to a high quality receiver to drive them, with maybe just a separate power amp of 150wpc or so for the front mains. I'm working on downsizing my HT :eek:

ben62670

WilliamM2 wrote: »

They haven't doubled the power, or gained 3db of sensitivity. It is still 1 watt, whether it is 4 ohm or 8 ohm.

Thank you for keeping it simple! 2.83 volts at one meter at 1k= sensitivity.

ben62670

dkg999 wrote: »

I'm a little iffy on the statement that high sensitivity speakers don't put out a lot of bass. I think back to the big old Klipsh speakers that were very high sensitivity and had tons of bass. I'm currently considering Paradigm Monitor speakers for my HT, and they are around 96db efficient and seem to have good bass output. I'm interested in going to smaller more efficient HT speakers so I can go back to a high quality receiver to drive them, with maybe just a separate power amp of 150wpc or so for the front mains. I'm working on downsizing my HT :eek:

Can you say Polk. They dig pretty deep. I am running 2 channel, and turned up thebass a little. It sounds like I have a tight mid powered sub!

ben62670

Here is a very interesting article on line arrays if you really want to rack your brain with sensativity.
http://www.audiodiycentral.com/resource/pdf/nflawp.pdf
BTW most Polk MW's have a fz of 29-31 Hz

lanion

Just to add, that if you have an ideal amplifier (doubles from 100W at 8 ohms to 200W at 4 ohms), then a 88db 4 ohm speaker would be the same as a 91db 8ohm speaker if you don't change your volume dial.

ben62670

lanion wrote: »

Just to add, that if you have an ideal amplifier (doubles from 100W at 8 ohms to 200W at 4 ohms), then a 88db 4 ohm speaker would be the same as a 91db 8ohm speaker if you don't change your volume dial.

Yes this is true. Now we could open a debate on energy consumed from the power co via more currant used, and AC to cools such amps, unless of course you are into those switching type of amps!
Ben

Spacedeckman

tonyb wrote: »

Hmmm....while I agree with alot in your post,I will take issue with your statement of high sensitive speakers not putting out alot of bass.Maybe true years ago,but not today.My Legacy focus 20/20 puts out more bass than some subs and has a 95 db sensitivity rateing,and are 4 ohm.And in no way does sound suffer.Most speaker makers today will add a bass module to a horn speaker or planer that typically suffer in the lower bass dept.
btw-Welcome to our humble forum,
And as always,let your ears be the judge.

Considering it's a big 4 ohm speaker, 95 isn't out of line with what can be expected. You are agreeing with me.

Mark

tonyb

4 ohms vs sensitivity: Yes, most of the speakers with very high sensitivities are 4 ohms. The exact sensitivity spec is 1 watt, 1 meter, 1 kHz. Somewhere in the small print, they will define "watt" as 2.83 volts, and by maintaing 2.83 volts into 4 ohms, they have essentially doubled the power and gained 3 dB of "sensitivity". The temptation is there to alter the frequency response of the speaker at 1 kHz to get a couple of extra dB as well. Take such numbers with a grain of salt if the manufacturer is trying to sell a speaker on sensitivity alone. All manufacturers "fudge" a bit on this spec, some make products to look good there....sound usually suffers if it is a "goal".

High sensitivity speakers don't make much bass. It's a function of their driver design, and a real world engineering tradeoff. A very sensitive driver will have a thin top plate and a short voice coil. When you are really moving the driver (ie, playing loud or low), the voice coil spends much of it's time outside of the strong part of the magnetic field, meaning that it can't be well controlled. You can go the other way, and make a really thick top plate and add some windings on the voice coil and create a driver that will have tons of control and accuracy, but be a total pig to drive. Most manufacturers today split the difference, and usually use a long voice coil and a moderately heavy top plate. You get good sensitivity, good frequency response, and can adjust accordingly to the needs of the driver (tweeter, mid, or woofer). Funny thing, you tend to get a speaker with a sensitivity in the high 80s to low 90s. Look at the numbers for most quality speaker manufacturers. Go to the spec pages of this very site...yep, real world engineering.

Maybe I am reading this wrong,but how do you figure we agree?
You said high sensitive speakers at 4 0hm don't make much bass.
I said some do...like mine.
Thats about as far from agreeing as you can get.
While I do get the rest of your well thought out explanation,and can agree on most,this particular thought mentioned above,I do not.No biggie though,we can agree to disagree.Cheers.

steveinaz

Spacedeckman wrote: »

Guys, you are doing too much thinking here, and not enough studying.

There are a ton of speaker parameters out there, most of them never get talked about. Some are important, some less so, but most exist outside of the sonic attributes of the speaker itself. So, let me give you a very, very short and rudimentary overview of driver and speaker engineering. You will learn something, even if you don't fully understand it today.

A bit of a pompus statement.

4 ohms vs sensitivity: Yes, most of the speakers with very high sensitivities are 4 ohms.

That's a HUGE generalization with which I disagree.

High sensitivity speakers don't make much bass.

While sometimes true, not always the case; define "high sensitivity."

Numbers don't really tell you beans about a speaker, especially the way it sounds....well, sort of.

Agree!

I did mention that high sensitivity speakers tend to make no real bass.

Disagree, again. At least you said "tend" this time.

When buying a speaker, you need to look at the following. Sound quality (to your ears, which btw, will change over time, not only ears, but preferences), can your recevier drive it, how loud do you want to play it, how big is the room you want to fill, and finally, does it fit your budget/system. Little has to do with specifications.

Agree!

Now, school's out.

tonyb

Oh come on Steve....I just got out my coloring book.:p

don2dusk

Hey guys. I am somewhat of a newbie on the topic of speaker selection. However, I am trying to do my homework. Currently, I have a Kenwood KR-V9030 receiver with JBL HP420's (all over 14 years old). Needless to say, I am looking to upgrading my entire home theater system to take advantage of HDMI, ProLogicII, THX Ultra2, etc. I am strongly looking at the Yamaha RX-V2700 for my receiver, and considering the Polk LSi series for the 5.1 surround. However, this bit with the specifications of the LSi series being rated at 4 ohm impedence is concerning me. Other manufacturers rate theirs at 8. If I go ahead with purchasing the Lsi's, will my Yamaha be able to keep up continuously with the more power required of the LSi's as a result of the 4 ohm impedence? Do you know if the Yamaha's heat sink will dissipate the additional heat, or am I missing the boat entirely? As I say, I am in the shopping mode right now, but I want to get a great system with a lot of bang for the buck.

Early B.

don2dusk wrote: »

If I go ahead with purchasing the Lsi's, will my Yamaha be able to keep up continuously with the more power required of the LSi's as a result of the 4 ohm impedence?

No, not really. You'll need separate amplification for the Lsi's.