Driver Sensitivity

[RedJacket]

What are some things to keep in mind while designing for max sensitivity? In other words, what do certain manufacturers do to achieve high sensitivities in their drivers, given that they're limited by Copper (i.e. power and magnetic compression)

 

[TLS Guy]

What are some things to keep in mind while designing for max sensitivity? In other words, what do certain manufacturers do to achieve high sensitivities in their drivers, given that they're limited by Copper (i.e. power and magnetic compression)

These issues are well known, so I don't know why you keep asking questions you could easily research on your own. If you are interested in transducer design, then I suggest you enroll at a university that offers a degree within your sphere of interest.

You are I think confusing sensitivity and efficiency. They are not the same thing. Sensitivity is a mess as there are two ways of looking at it. I db 1 watt 1 meter, and 2.83 volts one meter. Obviously on the later a 4 ohm speaker will be 3 db more sensitive than an identical 8 ohm speaker, but it will take twice the power. It does not stop there as if one speaker is highly directional and the other has good dispersion, the sensitivities may be the same, but obviously the speaker with wider dispersion is more efficient.

Efficiency is the percent electrical energy fed to the driver converted to sound energy.

Now the way to build up efficiency is to increase the flux density in the gap, though very powerful magnets and VC design, especially having a very small air space between pole and coil. You also make the moving parts, especially the diaphragm as light as possible. This light diaphragm raises the free air resonance and also limits bass response.

Now as you build up the flux density electrical Q is lowered and total Q is lowered. So a high efficiency driver is a low Qt driver. The take home is that bass response is inversely proportional to flux density. So if you make a high efficiency low Q driver, then you have to horn load it to get any bass, as it requires a highly efficient acoustic transformer to bring up the bass. If you make a three or two way speaker with high efficiency mids and tweeters, you only get the gain from the mid and high frequency units with an active design. In a passive speaker, the mids and highs have to be padded down and the efficiency gain wasted as heat in the L-pad resistors, unless the system employs a huge horn for a high efficiency woofer.

The best way for you to look at this is to look at the Lowther drivers. These were first produced by Paul Voight in the 1930s. He really has to have the credit of designing the first high fidelity loudspeakers. Lowther is easily the worlds oldest loudspeaker company. They produce highly efficient drivers for horn loading. The flux densities are enormous. As you go though the specs of the units you will see that as the Tesla in the gap goes up, efficiency goes up and Qt drops.

Take a look at this driver.

This is the enclosure it takes to load it.

You have an interest in Ribbon speakers. The first was Stan Kelly's Decca/Kelly horn. These units date back to the fifties. I knew Stan. He was one of the great pioneers. To up the efficiency and get the crossover point to 2.5 K he horn loaded his ribbon speaker. Horn loading will greatly increase the efficiency of any speaker drive unit.

For more information on the subject you need a period of didactic study.

In essence the answers to you question have actually been known for over 80 years.

 

[herbu]

These issues are well known, so I don't know why you keep asking questions you could easily research on your own. If you are interested in transducer design, then I suggest you enroll at a university that offers a degree within your sphere of interest.

You are I think confusing sensitivity and efficiency. They are not the same thing. Sensitivity is a mess as there are two ways of looking at it. I db 1 watt 1 meter, and 2.83 volts one meter. Obviously on the later a 4 ohm speaker will be 3 db more sensitive than an identical 8 ohm speaker, but it will take twice the power. It does not stop there as if one speaker is highly directional and the other has good dispersion, the sensitivities may be the same, but obviously the speaker with wider dispersion is more efficient.

Efficiency is the percent electrical energy fed to the driver converted to sound energy.

Now the way to build up efficiency is to increase the flux density in the gap, though very powerful magnets and VC design, especially having a very small air space between pole and coil. You also make the moving parts, especially the diaphragm as light as possible. This light diaphragm raises the free air resonance and also limits bass response.

Now as you build up the flux density electrical Q is lowered and total Q is lowered. So a high efficiency driver is a low Qt driver. The take home is that bass response is inversely proportional to flux density. So if you make a high efficiency low Q driver, then you have to horn load it to get any bass, as it requires a highly efficient acoustic transformer to bring up the bass. If you make a three or two way speaker with high efficiency mids and tweeters, you only get the gain from the mid and high frequency units with an active design. In a passive speaker, the mids and highs have to be padded down and the efficiency gain wasted as heat in the L-pad resistors, unless the system employs a huge horn for a high efficiency woofer.

The best way for you to look at this is to look at the Lowther drivers. These were first produced by Paul Voight in the 1930s. He really has to have the credit of designing the first high fidelity loudspeakers. Lowther is easily the worlds oldest loudspeaker company. They produce highly efficient drivers for horn loading. The flux densities are enormous. As you go though the specs of the units you will see that as the Tesla in the gap goes up, efficiency goes up and Qt drops.

Take a look at this driver.

This is the enclosure it takes to load it.

You have an interest in Ribbon speakers. The first was Stan Kelly's Decca/Kelly horn. These units date back to the fifties. I knew Stan. He was one of the great pioneers. To up the efficiency and get the crossover point to 2.5 K he horn loaded his ribbon speaker. Horn loading will greatly increase the efficiency of any speaker drive unit.

For more information on the subject you need a period of didactic study.

In essence the answers to you question have actually been known for over 80 years.

I couldn't have said it better myself.

 

[PENG]

Obviously on the later a 4 ohm speaker will be 3 db more sensitive than an identical 8 ohm speaker, but it will take twice the power.

I hate to correct you but if a 4 ohm speaker takes twice the power to produce the same SPL at 2.83V vs an identical 8 ohm speaker then it is not 3 dB more sensitive.

 

[TLS Guy]

I hate to correct you but if a 4 ohm speaker takes twice the power to produce the same SPL at 2.83V vs an identical 8 ohm speaker then it is not 3 dB more sensitive.

No, I'm correct.

If you take two drivers identical except for the voice coil winding and one is 8 ohm and the other 4 ohm, then if you specify sensitivity as 2.83 volt drive and spl at one meter on axis, then the four ohm speaker will show a sensitivity 3 db higher than the 8 ohm one, but it will take twice the power from the amp. However the acoustic efficiency will be the same.

 

[PENG]

No, I'm correct.

If you take two drivers identical except for the voice coil winding and one is 8 ohm and the other 4 ohm, then if you specify sensitivity as 2.83 volt drive and spl at one meter on axis, then the four ohm speaker will show a sensitivity 3 db higher than the 8 ohm one, but it will take twice the power from the amp. However the acoustic efficiency will be the same.

Haha, I thought you might say that.. Fair enough, in that case you are correct, though confusing without elaborating a little. However, you were talking about sensitivity, not acoustic efficiency in that post so may be we shouldn't confuse the issue more with another term. Now, even if the 4 ohm speaker produce 3 dB more at the same input voltage, the sensitivity base 1W/1M will be the same as that for the 8 ohm version, no more no less. I just thought Comparing the so called sensitivities of speakers of different impedance using the 2.83V/1M way is really not that great either, plus the fact that impedance varies with frequencies it can come close to being meaningless. So I have to agree with you that it is a mess.

 

[TLS Guy]

Haha, I thought you might say that.. Fair enough, in that case you are correct, though confusing without elaborating a little. However, you were talking about sensitivity, not acoustic efficiency in that post so may be we shouldn't confuse the issue more with another term. Now, even if the 4 ohm speaker produce 3 dB more at the same input voltage, the sensitivity base 1W/1M will be the same as that for the 8 ohm version, no more no less. I just thought Comparing the so called sensitivities of speakers of different impedance using the 2.83V/1M way is really not that great either, plus the fact that impedance varies with frequencies it can come close to being meaningless. So I have to agree with you that it is a mess.

You are correct on the 1 watt 1 meter sensitivity spec. However manufacturers almost always give the 2.83 volt spec, as most speakers are less then 8 ohms and it gives the highest number!

 

[RedJacket]

Thanks for the lesson/overview.

I understand that sensitivity is a measure of loudness (how much SPL at 1 m for either 1 W or 2.83 V), whereas efficiency is the percentage of how much energy is actually converted. My question is more related to power implications; achieving a higher sensitivity basically means less power to achieve the same SPL correct (every 3 DB increase leading to a power reduction by half). Would using a different voice coil material (one whose conductivity is on par with copper at room temperature, but keeps it more constant as temp increases - as well as a lower density) keep sensitivity higher, thus bringing down the power requirements (which would be important for portable applications)?

These issues are well known, so I don't know why you keep asking questions you could easily research on your own. If you are interested in transducer design, then I suggest you enroll at a university that offers a degree within your sphere of interest.

You are I think confusing sensitivity and efficiency. They are not the same thing. Sensitivity is a mess as there are two ways of looking at it. I db 1 watt 1 meter, and 2.83 volts one meter. Obviously on the later a 4 ohm speaker will be 3 db more sensitive than an identical 8 ohm speaker, but it will take twice the power. It does not stop there as if one speaker is highly directional and the other has good dispersion, the sensitivities may be the same, but obviously the speaker with wider dispersion is more efficient.

Efficiency is the percent electrical energy fed to the driver converted to sound energy.

Now the way to build up efficiency is to increase the flux density in the gap, though very powerful magnets and VC design, especially having a very small air space between pole and coil. You also make the moving parts, especially the diaphragm as light as possible. This light diaphragm raises the free air resonance and also limits bass response.

Now as you build up the flux density electrical Q is lowered and total Q is lowered. So a high efficiency driver is a low Qt driver. The take home is that bass response is inversely proportional to flux density. So if you make a high efficiency low Q driver, then you have to horn load it to get any bass, as it requires a highly efficient acoustic transformer to bring up the bass. If you make a three or two way speaker with high efficiency mids and tweeters, you only get the gain from the mid and high frequency units with an active design. In a passive speaker, the mids and highs have to be padded down and the efficiency gain wasted as heat in the L-pad resistors, unless the system employs a huge horn for a high efficiency woofer.

The best way for you to look at this is to look at the Lowther drivers. These were first produced by Paul Voight in the 1930s. He really has to have the credit of designing the first high fidelity loudspeakers. Lowther is easily the worlds oldest loudspeaker company. They produce highly efficient drivers for horn loading. The flux densities are enormous. As you go though the specs of the units you will see that as the Tesla in the gap goes up, efficiency goes up and Qt drops.

Take a look at this driver.

This is the enclosure it takes to load it.

You have an interest in Ribbon speakers. The first was Stan Kelly's Decca/Kelly horn. These units date back to the fifties. I knew Stan. He was one of the great pioneers. To up the efficiency and get the crossover point to 2.5 K he horn loaded his ribbon speaker. Horn loading will greatly increase the efficiency of any speaker drive unit.

For more information on the subject you need a period of didactic study.

In essence the answers to you question have actually been known for over 80 years.

 

[TLS Guy]

Thanks for the lesson/overview.

I understand that sensitivity is a measure of loudness (how much SPL at 1 m for either 1 W or 2.83 V), whereas efficiency is the percentage of how much energy is actually converted. My question is more related to power implications; achieving a higher sensitivity basically means less power to achieve the same SPL correct (every 3 DB increase leading to a power reduction by half). Would using a different voice coil material (one whose conductivity is on par with copper at room temperature, but keeps it more constant as temp increases - as well as a lower density) keep sensitivity higher, thus bringing down the power requirements (which would be important for portable applications)?

Again you are confused. Higher efficiency only relates to dynamic thermal compression in so far as there being less power required for a given spl the heating effect will be less. The heating effect is resistance X the square of the current.

As far as wire, you need good electrical and thermal conductivity. Copper fits the bill the best on both counts.

Thermal compression is minimized by flattening the wide and having the coil as close to the pole piece as possible. For woofers venting the gap also helps.

 

[RedJacket]

I am getting my information from this website - goo.gl/YiIi7z

Again you are confused. Higher efficiency only relates to dynamic thermal compression in so far as there being less power required for a given spl the heating effect will be less. The heating effect is resistance X the square of the current.

As far as wire, you need good electrical and thermal conductivity. Copper fits the bill the best on both counts.

Thermal compression is minimized by flattening the wide and having the coil as close to the pole piece as possible. For woofers venting the gap also helps.

 

[TLS Guy]

I am getting my information from this website - goo.gl/YiIi7z

Well you got your information from a good source.

Basically that is the long version of what I told you. I don't think the author really makes the point about the disadvantage of high efficiency woofers. The problem is low Q and no bass without very efficient bass loading.

The curse of good audio is that everyone wants small speakers. If you use small bass enclosures then you will never achieve really good reproduction. It is well know that I do not care for big heavy drivers in small boxes with a lot of power. They do not sound good. You can force out bass, but it is of awful quality. In addition to driver compression in those boxes there is dynamic compression due to physical air compression in the box. I don't care how you Eq the box, servo or not, the result will be equally bad.

This trend to small speakers has really made PA and cinema systems sound much worse than they did fifty to sixty years ago. There is nothing like a large efficient low Q driver in a large horn in a big auditorium. Those wonderful Altec Voice of the Theater systems are long gone.

The take home is that for good driver design, you need above all else precision. This makes good drivers expensive. You need a well wound precision VC with flattened wire. You must have minimal gap and get the VC former as close as technically possible to the pole piece. This relates to mitigating the magnetic flux problems and thermal problems.

You probably know from the pictures of my system, that I do not use sub drivers, but high quality units with light metal cones. The drivers are loaded in very large transmission lines. The total system performs like nothing else I have heard, with very life like dynamics throughout the range. The bass is order of magnitudes better then any sub system I have ever heard.

Driver Qt is 0.35, so I don't have to use massive horns which I don't have room for, so the sensitivity ends up at 90 db 1 watt 1 meter and the also 90 db 2.83 volts 1 meter. For the room the size it is, then that leaves plenty of reserve before running into the issues we are talking about.

If it is accurate bass you want with realistic dynamics, you have to think big. Small will always fail you.