Impedance compensation needed for active speakers?

[tcpip]

I would ask this question in the solid state forum (assuming you have a SS amp). Any sound quality improvements would be from less distortion from your amp rather than your speakers

I would also phrase it in terms of output stage stability/oscillation.

You are already way ahead of the curve with active XO

To better answer your question, you should state the amp you are using or provide schematics, and list drivers and XO points

A well designed amp should not have output stage instability problems.

Actually, this is not what I wanted to know.

My question relates to stable amps, irrespective of their topology. What I wanted to know is how is the power delivered to a driver kept constant across the freq spectrum, when we can see that the driver's impedance is varying with frequency? Since power is proportional to Z, won't the increase of impedance lead to a change of power delivery, given the fact that solid-state amps have very low output impedance?

 

[richie00boy]

The acoustic power output is determined by the voltage drive applied to the drive unit. Think about it, all cheap speakers with no compensation would have no mid or upper treble if what you are thinking is true

 

[phase_accurate]

What I wanted to know is how is the power delivered to a driver kept constant across the freq spectrum, when we can see that the driver's impedance is varying with frequency?

The simple answer is: Power vs frequency neither isn't constant at all nor does it have to be constant.
At fs for instance the driver takes much less power but has increased efficiency that makes up for it.

Driver FR measurements are usually made at a constant VOLTAGE driving them.

So if your amp doesn't mind impedance changes (in terms of sonic performance and stable operation) you don't need any compensation at all with active topologies.

Regards

Charles

 

[soongsc]

Actually, this is not what I wanted to know.

My question relates to stable amps, irrespective of their topology. What I wanted to know is how is the power delivered to a driver kept constant across the freq spectrum, when we can see that the driver's impedance is varying with frequency? Since power is proportional to Z, won't the increase of impedance lead to a change of power delivery, given the fact that solid-state amps have very low output impedance?

See Wireless World 1970 "The design and Use of Moving-coil Loudspeaker Units" by E. J. Jordan

 

[rcw]

re sound presssure

A loudspeaker is over a part of lts range a linear transducer.
What this means is that it produces a sound pressure on axis that is an analogue of the input signal voltage, this happens over its linear range.
The acoustic power it radiates is not linearly related to the electrical input power.

 

[forr]

Hi Soongsc

"That is why I wonder why the response of the driver differs so significantly depending on which amp it's connected to."

A bit like in photography where you cannot consider the sensitivity of a black and white film without its revelator, a loudspeaker must not be considered on its own but always associated with the amp by which it is driven. It is the whole electro-mecano- acoustic parameters which define the response
I am in the camp that thinks that all audible differences are measurable.

Do you have a scan of the Jordan's article ? I would be interested to read it. It happens that I am familiar with the name of Ted Jordan as I bought eight of his 50 mm alu cones at the begining of the eighties.

~~~~~~~~~ Forr

§§§

 

[DSP_Geek]

What I wanted to know is how is the power delivered to a driver kept constant across the freq spectrum, when we can see that the driver's impedance is varying with frequency? Since power is proportional to Z, won't the increase of impedance lead to a change of power delivery, given the fact that solid-state amps have very low output impedance?

Ayup, the real power going to a driver changes with frequency. That's why drivers with high voice coil inductances tend to roll off quite early -- see TC Sounds woofers for examples.

On the other hand, driver dispersion tends to narrow with increasing frequency, so smart driver designers try to have the effects of directivity and voice coil impedance offset each other.


Francois.

 

[soongsc]

Hi Soongsc

"That is why I wonder why the response of the driver differs so significantly depending on which amp it's connected to."

A bit like in photography where you cannot consider the sensitivity of a black and white film without its revelator, a loudspeaker must not be considered on its own but always associated with the amp by which it is driven. It is the whole electro-mecano- acoustic parameters which define the response
I am in the camp that thinks that all audible differences are measurable.

Do you have a scan of the Jordan's article ? I would be interested to read it. It happens that I am familiar with the name of Ted Jordan as I bought eight of his 50 mm alu cones at the begining of the eighties.

~~~~~~~~~ Forr

§§?

Forr,

I know that there are lots of interaction between speaker and drivers, Since I have not really studied power amps that much, I am curious as to what ampe characteristics really cause uncompensated speakers to react so differently on different amps. Is it the damping factor? The speakers I am talking about has impedance between 7 and 10 ohms, the only thing far from that is the Resonant which goes to about 14 ohms.

I don't have a scan of the article, it was recommended by collin in another thread. Did you get the unshielded or shielded version of the 50mm ones? He certainly lives closer to you than I.

 

[mbutzkies]

tcpip:

Impedance compensation does not change the voltage or the current that the speaker sees. In common amplifiers, Voltage not electrical power is held constant with changing Z, excluding damping factor of course.

Most SS amps have large damping factors so you usually do not have to worry about voltage drop with large impedances.

 

[tcpip]

Re: re sound presssure

The acoustic power output is determined by the voltage drive applied to the drive unit. Think about it, all cheap speakers with no compensation would have no mid or upper treble if what you are thinking is true

I'm not sure I understand why, but I'm trying to figure this one out.

The simple answer is: Power vs frequency neither isn't constant at all nor does it have to be constant.
At fs for instance the driver takes much less power but has increased efficiency that makes up for it.

I think I understand the drift. Basically, you're saying that the speaker doesn't try to be linear w.r.t. power dissipation.... it tries to be linear with respect to the voltage across its terminals (assuming a very low-Z drive source). In that case, of course, the varying Z of the speaker becomes irrelevant.

A loudspeaker is over a part of lts range a linear transducer.
What this means is that it produces a sound pressure on axis that is an analogue of the input signal voltage, this happens over its linear range.
The acoustic power it radiates is not linearly related to the electrical input power.

Yes, this is what I'm beginning to understand. Thanks for the clear answers, everybody. In my mind, I think I'll close this question and treat it settled now.

 

[forr]

Hi Soongsc

My 1981 Jordan were not shielded. They were in an array of four for each channel. I kept them for about six years, they were good but at least at a distance of four meters. To me arrays are not good for smaller distances.

~~~~~~~ Forr

§§§

 

[soongsc]

Hi Soongsc

My 1981 Jordan were not shielded. They were in an array of four for each channel. I kept them for about six years, they were good but at least at a distance of four meters. To me arrays are not good for smaller distances.

~~~~~~~ Forr

§§?

I still have two of the unshielded 50mm that has a thin cloth and a small hole in the middle. The current JX53 is much better. I would like to try and get a hold of this later 50mm version that was shielded. It would be interesting to see how a driver evolves through the years.