CW1526C (K-33) efficiency?

[Arash]

vast majority of Klipschophiles use CW1526 as a replacement of K-33 in their speakers. we all know CW 1526 is a woofer of about 95.2db of efficiency (k-33 is probably the same) but La Scala and Klipschorn are speakers of 103-104db of efficiency. will Folded Horn loaded structure really increase the efficiency that much that a 95db woofer sing 104db at the same amplitude?!

someone light me up

[djk]

http://www.parts-express.com/pe/showdetl.cfm?Partnumber=293-656

Sensitivity

85.2 dB 1W/1m

Subtract 26.38dB for the 1W/1M, so it's 106.92dB/1W/1M at 50hz, ie: 20dB more efficient.

[Arash]

I didn't get it. write a line or two instead of posting that much of graphs

will a La Scala Horn structure increase 95db K-33 up to 104db?

[djk]

What part of

http://www.parts-express.com/pe/showdetl.cfm?Partnumber=293-656

Sensitivity

85.2 dB 1W/1m

Subtract 26.38dB for the 1W/1M, so it's 106.92dB/1W/1M at 50hz, ie: 20dB more efficient.

didn't you understand?

[WMcD]

Yes, the horn really does increase the acoustic output.

PWK analgized it to putting a piston in the middle of a lake and moving it back and forth. It doesn't move very much water. But if the put the piston in a tube, it is an effective pump. (A bigger piston would do better though.)

Going further (my explanation), we can put the piston-like speaker diaphragm with a diameter of about 1 foot at one end of the pipe (throat of a horn), and then increase the area of the pipe by flaring it out along its lenght (the horn) of three or five feet until its diameter (mouth) is over two feet or more (call that a virtual diaphragm). From this, you have the better pump and a bigger diaphragm.

And if you put the 2-foot or greater diameter virtual diaphragm in a corner, the trihedral corner forms a longer horn with a bigger diaphragm.

There is an electrical analogy to the acoustic loading of the driving system by the throat (say the opening of the tube) and the voice coil resistance. We're going to look at the effect of improving the acoustic load on the front of the diaphragm. We'll pick some numbers for acoustic resistance presented to the diaphragm in two conditions. The first is the bass driver in free air or with one side in a box (0.08 ohms) and attached to the horn (8.0 ohms). The fundamental results requires just two fundamental concepts. (1) Ohms law says that voltage drop across a resistance is Voltage = current Intensity x Resistance. And (2) Power = current Intensity times Voltage drop. V=I x R and P=I x V

Bass driver in free air or in a box with one side open to the air:

(+)- - - (8 ohms from the voice coil) - - - - (0.08 ohms from the air load on a free diaphragm) - - - (-)

We apply an a.c. voltage between (+) and (-) in the diagram above. The overall series resistance is just above 8 ohms. Equal current flows through the 8 ohm resistance and the 0.08 resistance. But there is very little voltage drop across the 0.08 ohm resistance. It is 1/100th of that across the voice coil, and the voice coil just makes heat, not sound).

Remember that power absorbed is the current times the voltage drop across the resistance. So the 1/100th voltage drop shows very little power is going to the air load, or acounstic resistance, or sound being created.

Bass driver :mounted to the optimum sized tube (at throat of horn)

(+)- - - (8 ohms from the voice coil) - - (8.0 ohms from the air load on a diaphragm mounted to tube) - - - (-)

We've increased the acoustic load resistance by a factor of 100 by virtue of horn loading.

The first thing you say is: "Ya can't fool me; the total load resistance is now 16 ohms, and there is only half the current in this new circuit. Overall, it looks like the total circuit is absorbing less power from the source. How the heck can this increase acoustic output if electrical input is reduced?) True. But there is more.

Now, the voltage applied is being dropped equally across the voice coil and the acoustic load. The voltage drop across the acoustic load is very much higher. (I think it is 50 times higher than the first instance and the current is only halved. Very rough numbers. But we get 25 times the power into the acoustic load. And that would be a bit over 13 dB gain.).

Again, these are rough numbers, but you see that this is not magic, just good engineering.

Sorry if I got pedantic; some people like it.

WMcD

[Arkytype]

[Arkytype]

There's not much I can add to Gil's cogent argument. I had forgotten PWK's piston in a lake example.

One of the key benefits to horn loading a driver is higher acoustic output compared to simply mounting the driver on a baffle. The other benefit is lower distortion. PWK wrote extensively about modulation distortion and re-reading those papers are well worth the time. The less the diaphragm has to move, the lower distortion it will generate.

In one of the Dope From Hope (ca. 1971?), Paul wrote that the Klipschorn could generate a 32 Hz note at one acoustic watt output with a woofer diaphragm motion of 1/16" peak-to-peak! One acoustic watt is the equivalent of a sound pressure level of 120 dB. For the Klipschorn, Belle and La Scala, that would require only 40 watts input power!

djk's response was both uncivil and cornfusing. How does posting the data for a 10" woofer help with arash's valid question?

While I've not used the Horn Response software, I noted a glaring error in the data window labeled "Sample". A sound pressure level of 133.3065 dB does not translate to an acoustic output power of 129.8809 watts; it corresponds to an acoustic power level of 21.42 watts. To be more precise, acoustic power is expressed in watts per square meter.

The best web site I've found for understanding the relationship between acoustic and electrical measurement and properties is http://www.sengpielaudio.com

Lee

[djk]

"uncivil and cornfusing (sic). How does posting the data for a 10" woofer help with arash's valid question?"

So giving an example of how horn-loading a 10" woofer can increase the sensitivity 20dB over the same driver (as a direct radiator) is confusing?

And asking what he didn't understand is being uncivil?

Either you are being obtuse or uncivil, which is it?

As regards McBean's Hornresp, I suggest contacting David through the DIYaudio subwoofer forum, he is very responsive. I think in 2Pi 1W=112dB, so 133.3dB would be +21.3dB, or roughly 125W acoustic (Hornresp says 129.88dB).

[WMcD]

Hey guys, it is a bit disheartening when people get wound up too much.

To answer the question, yes the LS does the same type of thing.

In the circuit, the voltage drop across the acoustic load is pressure, IIRC. The diapragm can do more work because it generates force against the pressure.

In the analogy the acoustic resistance is just an electrical resistance which typically has a fix value in elecronics. More accurately, its value varies. We see that in the many illustrations of throat impedance of a horn where it decreases very much as we go down to the fc of the horn. PWK wrote a paper showing that in finite lenght horn, the throat impedance does not completely go to zero ohms. But there are wiggles around fc.

The diagrams posted by djk have value in this discussion.

As far as output, there is final bump up at about 40 Hz which I believe is the resonance of acoustic and mechanical system. You can get this lower is you have a larger back box and/or a floppier suspension. Therefore PWK looked to increase both.

djk's excursion plot is interesting too. We see that above 100 Hz or so it is small and this is because throat resistance is reaching it's optimum.

Below 100 Hz we see excursion going up and reaches a peak at about 40 Hz, were we see the aforementioned system resonance. Part of the reason for the increased excursion is that the driver is experiencing the effect of decreasing throat resistence. It is more free to move.

This is a complicate juggling, dance. The Khorn is starting to work like a direct radiator in a box below Fc. We also have room effects to goose up the effective output.

WMcD

[Axz Hout]

g

[djk]

The biggest excursion peaks in the impedance of the TC 10 horn design are at 38hz, then 56hz, then 93hz, then 125hz.

If you look at the Acoustical Reactance plot (not shown) you will note the impedance is just coming up from being negative, and crossing through zero at those frequencies, and the Acoustical Resistance plot (not shown) is essentially zero.The zeros being why the excursion is so high.

The peak at 38hz is the reactance annulling frequency, the other peaks and zeros are a consequence of reflections from the small mouth of the horn.This design has a tiny mouth for a 40hz horn in 2Pi, only about 30" square, which is where the source of the reflections back down to the throat come from. Because the mass corner of the TC Sounds driver is so low you would need a 150hz midrange horn (like the Oris), and the cabinet would have to be about a 3' cube externally.

The Hornresp for the LaScala and vented LaScala for comparison. An 8' ceiling will give a peak around 71hz and room gain and a long room dimension of 16'~20' will fill in the bottom end and make the vented LaScala flat to 30hz or so, the venting alone adds about 7dB at 40hz (over a stock LaScala).

[geoff.]

another sticky

[pzannucci]

Why?

[Deang]

Did he mean to pin it?

[geoff.]

Lol! 

 

LOTS of good information in these archives I stumble across that I can’t always rediscover later.

 

Was on a device (at work, gasp!) that would not allow me to “follow” it so I did a workaround until I could get back here.

 

 

 

 

[Woofers and Tweeters]

RIP Dennis. It's great that your information is still being used. 

[WMcD]

I think I have done a poor job of explaining things.  Let me return to the original question about when we see efficiency ratings of a driver of 4 % or so, how does a horn increase that ?

 

The answer is that the efficency of the driver is when it is in free air or a box with nothing but air in front of the diaphragm.  Importantly, we actually want to look at the efficiency of the entire system.  We want to put as much energy as possible into the acoustic load.  For the PWK analogy that is the the piston in a tub of water versus a piston in a tube (throat of the horn).

 

The best way of doing this analysis is an electrical model of the acoustic situation.  This all tracks with Bell Labs paper and Don Keele's paper.  Here I've stolen a schematic from the Internet.

 

It is important to realize that the driver is modeled at a voltage source and an impedange which is usually Rg for R generator.  There is simple electrical loop which include the load or Rload.  Other letters used are used but you'll see it.

 

It is a classic issue of "impedance matching."  The maximum power into the load is achieved when the Rgenerator is equal to the Rload.  The important issue is that we want to maximize the voltage drop across Rload and also the current thought Rload.  In the loop circuit they are sometimes in conflict.

 

You have to consider the driver as having an internal impedance, which is also called the output impedane of the generator.   The answer to efficiency of the system is when Rgen is equal to Rload.  

 

Stepping back a minute.  PWK's major thesis is the doppler effect, like a train with a whistle.  Bass notes on the diaphragm are like the train going by.  Treble notes are like the train's whistle.  When the train speeds by we hear the pitch of the whistle chage to higher and lower notes which the whistle is not actually producing.  Therefore we want to stop the train (diaphragm) from moving.  That occurs when the diaphragm does not move very much because the system is acoustically efficient.

 

Now we have to step back to an electrial analogy.  I know it is complicated but it is the true explanation.  We have to study the electrical analogy of the acoustic situation to thereby understand why the efficient horn sounds best.

 

 

 

 

 

 

 

 

[JohnA]

No denigration to anyone, especially Gil, but I struggle with the electrical analogy.  Maybe it's a EE thing, they think everything is an RLC circuit. 

 

I like to say that the horn traps a mass of air in a column.  The confined column of air BEGINS to behave as an incompressible fluid closely following the motion of the driver's diaphragm.  The expanding area of the horn makes the sort of incompressible fluid look like a much larger diaphragm at its mouth, increasing spl because there is a much larger area moving the same distance in and out. 

 

Perhaps an amplifier is an adequate analog; same voltage (pressure wave amplitude) but more current (area vibrating).  Though the effect is really about more effective/efficient coupling rather than adding energy like an amplifier does. 

 

So, yes, a horn can increase sensitivity from 95 to 104 dB/w/M.  

 

Wonder where Arash is these days?  And how long ago we lost him?  😀

[glens]

It's merely an acoustic transformer...