Constrained throat riddle

[John Warren]

The cost of the woofer was *the* reason for the throat change.

This message has been edited by John Warren on 04-22-2002 at 07:01 PM

[Al Klappenberger]

John,

Thanks for taking the other plot down. The "glitch" casued by the out of phse squawker didn't look very pretty! Anyhow, be carefull about the anechoic response plots on the Khorn. I understand that the Khorn goes lower than the reliable low frequency limits of the chamber that was used in making the plots on my network. Better low frequency plots were done outside using the corner of two walls. I would use that plot, if possible, for Khorn woofer analyisis.

AL K.

[John Warren]

I thought it appropriate to post a few Google quotes regarding this topic-

Posted by Bruce Edgar on September 25, 2001 at 18:36:06 In Reply to: Re: Klipschorn Drivers posted by Tom Brennan on September 25, 2001 at 18:01:32:

If the throat is 3"x13", the k-33 is ok. If the

throat is 6"x13", then you have to use a regular 15"

horn driver like Altec 515B, JBL 2220, JBL D-130, D-

140. Bruce

Posted by Bruce Edgar (M) on January 16, 2002 at 09:45:12 In Reply to: K-Horn posted by vincolise on January 16, 2002 at 01:00:40:

The K-33 driver needs to see a 3"x13" throat to get

any bass. The old original plans specified a 6"x13"

throat. Klipsch changed to the constricted throat in

order to make their cheaper K-33 driver give better

bass.

the full threads can be found here:

http://www.audioasylum.com/forums/HUG/messages/16877.html

http://www.audioasylum.com/forums/HUG/messages/21750.html

This message has been edited by John Warren on 04-24-2002 at 05:39 AM

[WMcD]

I'd like to finish up this thread with a technical disussion of why a front chamber, in theory, should increase the high end response of the system.

It is fairly obvious that without the 3 x 13 inch slot, there is not a chamber formed in front of the diaphragm. With it, there is a chamber. But why should that make a difference?

The diagrams are from Leo Barank's "Acoustics". It tells the story about midrange drivers. I can only think that PWK was applying this to the bass driver.

As you can see, if there is a front chamber, the high end cut off is increased, per the graph. However, the eventual roll off is 12 dB per octave rather than 6 dB.

The electrical diagram is a model of the system overall. It is a bit beyond the subject to explain it all. Suffice to say, the electrical components (inductors, capacitors, and resitors) represent parts of the works. Rg can assumed to be zero for modern amps. L is the voice coil inductance and we should ignore that also.

At the far right we see rmt, which is the resistance created by the horn. This complicated analogy seems like a cross over circuit. And the graph looks like one too.

Let's take a side trip, having nothing to do with the chamber issue. But it will describe some things we've read about.

Re is the resistance of the voice coil. Let's put that at 4 ohms. If we have no horn, the diaphragm does not encounter much mechanical resistance to its movement and we can assume rmt is 0.1 ohm. The amp sees a total resistance of 4.1 ohms. Most power is being used to heat up the voice coil resistance. (See O.Gadfly Hertz, space heater, Dope from Hope.)

What does that say about power delivered into rmt? Actually, there is little voltage drop across it (at 0.1 ohms), thus little power delivered into it.

With a horn attached, the diaphragm meets great resistance to movement, pumps air better, and the value of rmt rises to 4 ohms if the throat size is optimzed. Now the total resistance "seen" by the amp is 8 ohms. You may have read that use of the horn increased the electrical impedance, this is why.

The first thought is that the amp is delivering less power into an 8 ohm load (total) than a 4 ohm load. True. However, if rmt is 4 ohms, there is a much greater voltage drop across it, and it absorbs more power than the 0.1 ohm resistor. Therefore, raising the value of rmt by means of a horn allows more acoustic power to be delivered, even though the system resistance, as seen by the amp, is increased, and it is delivering less electrical power!

Those familiar with ohms law will see what is going on. The current in the loop (Re and rmt) is halved, but the voltage drop across rmt is very, very much higher. More power.

Now we have to look at the electrical schematic as a whole. It looks very similar to the schematic of a speaker crossover. Mmd is the mass of the diaphragm. Cm1 is the front chamber. We are going to look at the effect of Cm1 inductor. (BTW, it uses a C because it is a compliance, or spring, a compression chamber). Let's assume it is not there and there is a short circuit, wire, in its place.

Without Cm1, we have capacitor Mmd dominating the crossover circuit. It is wired to ground. As the frequency increases, the impedance of the cap decreases and more and more power is being shorted to ground and not delivered into rmt (the throat).

This makes sense in the mechanical analogy we're working from. At high frequencies, we're trying to move the mass of the diaphragm back and forth with greater acceleration. F=MA. But we have a limited amount of force available from the motor.

How much force do we have available? It depends on motor strength (B and l) and, whatever might be absorbing force. Yikes! Remember that we used the horn to impose more of a load on the diaphragm, creating force so it can pump air?

It seemed so wonderful to get more efficency and power into rmt. But rmt robs power to move the diaphragm back and forth. Horn loading has increased efficency but it decreased the high end output from the diaphragm.

Therefore, the same driver would have good high frequency response without the horn, as we see in a Cornwall.

But we're stuck with values of rmt, and Mmd. And that is the big problem.

We have to go back to our appreciation of the electrical analogy circuit which looks like a cross over. So we have this oddball crossover circuit which is just a capacitor to ground, forming a 6 dB low pass circuit.

What to do? Oddly, it works out that we can create a 12 dB, second order cross over by inserting the inductor of Cm1. Remember, in this analogy, it is the front chamber.

It is not obvious at all at the start. If you were presented with a 6 dB crossover using capacator Mmd to ground, you'd think there is nothing possible to improve the high end roll off point. Yet by making it a second order circuit, where you can play with the inductor Cm1, you can increase the high end roll off point.

I have to explain it to myself, again. If you had a first order crossover and couldn't alter cap Mmd, you still could tweek it by adding inductor Cm1 and increase the high end roll off point. So adding a front chamber can increase the high end. In theory.

But how much of this works in fact? This is all so theorectical.

It appears from the Bruce Edgar graphs that the restricted throat area created its own loses, for good or bad. In fact it looks as if we can accept some decrease over most of the band in question, we get a little gain up toward 400 Hz.

Now you know what I know.

Regards,

Gil

This message has been edited by William F. Gil McDermott on 04-30-2002 at 10:58 PM

[Tom Mobley]

So, the inevitable question arises:

does anybody have have hard information indicating that a KH-LS-Belle would actually, noticeably sound better if upgraded to Altec 515B or similar and the throat opened up to six inches? Or would it require new/different crossovers or something else out of scope of a woofer upgrade?

I've acquired a set of LS cabs, and most of the components for the Al K. crossovers have rolled in. I'm about ready to buy woofers, go with the K-33-E's or move to something supposedly better?

expiring minds want to know....

Tom

[John Warren]

Tom-

WIth the LaScala, the bass output is so minimal the issue is boderline moot. I wouldn't "waste" an Altec 515B in a LaScala cabinet.

[djk]

Altec 515B in a LaScala? Not unless you re-designed it and built it from scratch. The woofer is too wide.

[JohnA]

Excellent post Gil! I understood it! I'd always had trouble with the electrical to mechanical analogies. civils don't get much circuits training.

John

[John Warren]

515B came in two sizes,

standard 15" (15 3/16" OD) from 1955-1977

and 16" 1977-1978.

the K33E is 15 3/16" OD.

Altec 515 revision history here:

http://www.soundpractices.com/images/515history.pdf

Altec YOM dates here:

http://www.soundpractices.com/images/largedriv.pdf

This message has been edited by John Warren on 05-01-2002 at 06:06 PM

[Tom Mobley]

Thanks for clarifying that size issue. I don't see that anybody is volunteering hard info that there would be an improvement by making this change, guess I'll do the 33 thing.

Anybody want to sell me a set of 400/401's cheap? Don't everybody volunteer at once now....

Tom

[BeFuddledinMn]

Just stumbled upon this golden nugget of a topic from way back and thought I’d add my recent findings on the topic - about 16 years late, lol.

 

Here are some relevant woofer/motorboard test results from my efforts last year: