results of ALK mods on K-Horns

[sunnysal]

Finally I have finished (or better said settled) on the setting up of my k-horns with the ALK upgrade x-over and I can now make the following comments about the results:

1. the new x-overs are much clearer and sound more "direct" than the original AK-3 x-over...especially in the mids and highs...improved slam and pace (sorry for the mag speak)

2. the bass is somewhat more extended and less bloated in mid-bass (this is one reason I felt a percieved loss in bass) I also feel the bass is less "one note" than before my upgrade...I like it!

3. After much playing around I found my optimum setting for the squawker was 3-0 (the 5-2 was too forward) though I have a note on this issue: since the tweeter is not adjustable the right setting for the squawker seems 5-2 in relation to the tweeter BUT NOT in relation to the bass...when using my two other speaker sets for low-mid-high level referencing (a pair of RB-5´s and a pair of Magneplanar SMGc´s) the proper bass-mid relation means a 3-0 but this leaves the tweeter too high in level...so I have to see if I can pad down the tweeter a couple of DB (AL?)...

3. MUCH BETTER in terms of the glare, smear, and straining I sometimes heard in the mids-highs before the ALK mod...I can listen much longer now without strain than before and I feel a real liquidity on the top end I did not have before...

Summary: Great upgrade! much bigger difference in the sound of my K-horns than any other tweak I have made to date...highly recommended...

Now, AL K. tell me how to pad down the tweeter a little and we are home free!

Thanks all, Tony

[Al Klappenberger]

Hi Tony,

I'm glad you like the new crossovers Tony. Thanks for the

compliments!

I have just learned something that might have a bearing on why they

seem to smooth out some of the "glare" as Tony described it. It seems that

some of the Klipsch squawkers, which are supposed to cut off at 6000 Hz, come

back to life to emit a narrow band around 9000 Hz. My network has about 7 dB

of rejection at 9000 Hz which helps this. The Klipsch networks have no

rejection at all at that frequency!

There is no way I know of to reduce the levels to the tweeter without

the use of some additional hardware. The choices are some form of reisitive

attenuator or another transformer. Some have suggested the use of a specially

chosen cable from the network to the tweeter. I think I would try getting the

smallest speaker wire you can find to replace the tweeter cable. Maybe

even as small as AWG #22! If that doesn't do it, I could easily calculate the

values for a "T" or "Pi" attenuator made from resistors. You can also order

adjustable pads from companies like Solen.

Maybe a big Persian rug hanging on the back wall will do it too!

Al K.

[brian_donaldson]

I had some time ago asked two questions and only got a response from one, I think, so I will ask a few others:

Will the ALK crossovers work in the belles, how much are they, and are they obtained from Klipsch directly or thru a third party--like Al himself...?

[Al Klappenberger]

Brian,

The crossover networks Tony has installed and referred to these days as the "ALK" is available only from me. Klipsch company has nothing to do with it! Don't bother them!

I have Belle Klipsch myself and use a set in my own speakers.

For the price information, please email me and I will give you all the info.

name@domain.com

[Tom Brennan]

Parallel a .1mh coil and a 3uf cap, wire this circuit in series with the + terminal of the Atlas midrange. This circuit is called a P Trap and was designed by a friend who's an old time horny years ago to suppress the 9khz flare the Atlas often exhibits and has been sucessfully used on many Khorns and LSs, ought to work on CWs and Hereseys too.

[Randy Bey]

I built a P trap to attempt to reduce high end on my Khorns but didn't keep it in.

It might be that my Rat Shack SPL meter isn't sensitive enough to notice the 9K spike, but subjectively I could not hear an improvement.

If Al would post a simple resistive L pad circuit to drop the tweeter output 2 or 3 DB I think a lot of folks would benefit.

[Klewless]

Randy, ALK,

Couldn't you just take the tweeter input from the xformer at the -3 db tap rather than at the full input? Al, would that mess things up?

My thoughts.

------------------

John P

St Paul, MN

[Al Klappenberger]

Hi guys,

You can't tap down on the transform for the tweeter because it is

located only in the squawker channel on my network. The Klipsch networks use

it for both. This is the main reason I have not offered a network for the

Cornwall. The tweeter runs wide open as does the woofer. Both are supposed to

have equal efficiency with the squawker being higher efficiency. The

transformer cuts down only the squawker to match the other two.

Well, I sort-of hate to encourage the use of resistive attenuators

since they are counter to the design philosophy of my network. I will do it

if everybody twists me arm real hard though! I think you would do best to

just get an adjustable one. Try Solen. The part number for an adjustable 8

Ohm, 15W "L" pad is LP-15W8. My price list says it costs $2.06.

Al K.

[sunnysal]

al, hate to twist your arm but...no really, I thnk with a few dbs of tweeter padding I will be in audio nirvana, but your reservations make me pause, would the passive resistor cause some kind of sound change? if not then I would love to give it a try...where can I buy that item on line? thanks, tony

[Klewless]

sunnysal,

Well here I am just squirming in my chair and can't keep my mouth shut.

A resistor would reduce the current into the tweeter, hence making the sound a little bit duller, which is effectively what you are attempting. As such it is changing the sound. Perhaps you could try using your tone control just a tad bit to see if you like the reduced sparkle. Actually IMHO speakers should not have sparkle. Do people's voices sparkle? One should hear the source NOT the speaker. That means one of the better tools for testing speaker systems is vocals not orchestra!

Good luck in your quest. One other point is that if you installed a T-pad, which would keep the impedance from the tweeter to the xover constant and allow you to adjust the volume of the tweeter, then you'd be in the infamous "Which setting do I like, and why or why not?". Similar to constantly twiddling the volume or tone controls. And "Do I have both systems set the same?", etc.

------------------

John P

St Paul, MN

[Dick]

John,

Well, yes people's voices do sparkle. Think Kathleen Battle, or Jussi Boerling. or even Leonard Warren. That's sparkle! And they really shine on La Scalas with ALKs in the box. But then when you listen to a great talent like Diana Krall the warmth that comes out of the system is amazing.

Remember that the human voice doesn't generally go over about 1200Hz-1400Hz (without overtones. By the time you get into the tweeter you're looking at nothing but overtones and they are what give a singer their distinctive difference from others.

For my money the de-emphasized sqwauker is the greatest reason for using the ALKs.

------------------

Richard Hemmings

[Steve Phillips]

sunnysal,

I got the"Prototypes"and I love them.What a diiference.Smooth and very clean.Made a nice improvement to my 77'K-Horns.I too noticed the cleaner and more"direct"sound.I haven't found any music that it hasn't improved yet.I do prefer the 2 5 setting over the "Normal"5 2 setting.Tryed the 5 1 just for fun and changed it back.That realy laid the mids back too much.Think I'll try it your way just to see(or hear)what that does.I'm going to rewire the tweeter and squawker to network this weekend.

[Al Klappenberger]

Hi guys,

It sounds like everybody is tyring to talk Tony out of damping down the tweeter!

As to "T" pad Versus "L" pad. The "L" pad match impedance between unlike impedances while providing attenuation. The "T" is constant impedance from both ends. I don't think the tweeter cares. It can be, and usually is driven from a very low impedance. It's the tweeter filter in the network that needs to see 8 Ohms. So, an "L" pad should do it. Keep arm twisting and will post some options of you decide you really want to try it.

Al K.

[sunnysal]

klewless, I know that backing off the tweeter would, in effect, be changing the sound...but when I compared what was coming out of the tweeter after the mod with what came out before AND I compared the balance between low-mid-high with my other speakers, I came away convinced that the tweeter is running hot, hotter than with the original x-over and hotter than with my other speakers...also, I rather like the ability to adjust level for the squawker, it may not be puristic (a word?) but I get the sound I like...however the tweeter level on the ALK cannot be adjusted like the squawker in kind so...Al, another twist, an "L" pad or "T" pad, something that lets me goof around to my hearts content to get the balance the way I like it...I hope it is something I can do with my tremendously limited soldering ability! thanks all for the comments, tony

[Randy Bey]

Tony,

I think the L or T pad would be fairly easy to implement.

I would not say it would "dull" the tweeter, just reduce the voltage level in relation to the other drivers.

If the tweeter is running too hot to begin with, I think this is a reasonable change.

[Al Klappenberger]

Tony,

Ok, Here's an "L" pad that will cut the tweeter down by about 2.3 dB. It's

made up of 27 Ohms across the tweeter and three 5.6 Ohms in parallel (1.86

Ohms) connected between the tweeter and the network. You can use the barrier

block mounting screw as a binding post for the tweeter + wire and the two

resistors. See the drawing:

Al K.

[Klewless]

Yes guys,

I have heard Al's xover, including the change in tweeter output. And I liked what I heard to the point that I am going to build myself a set of them. My main reason for building my own is that I want to move the bass section next to the woofer (my convoluted form of bi-wiring I suppose).

My comment regarding "sparkle", while probably poorly worded, was in the context of real life, not "reproduction". I still stand by my idea that real live people don't have "sparkle" (not the same as high pitched voice). So all that being said, I agree with most of the ideas on this topic. For sake of clarification my idea of a good speaker system is that one should never be able to "hear" individual components. One only "hears" a single coherent sound.

I don't see the difference between cutting the high frequencies at the amp or reducing the level of the tweeter. The highs have been cut in both cases. Of course, there is the possibility of the amp cutting in the wrong places, in which case the only recourse is to do it within the xover, which surely must affect the damping. There are no free lunches.

Al's voltage divider is a good solution but is not adjustable. It also add a minor advantage in added protection to the vulnerable tweeter.

You guys may now consider me "dull - no sparkle", until I actually encounter somebody in real life who does sound that way to me. Then I'll just have to change my opinion!

------------------

John P

St Paul, MN

[Al Klappenberger]

John,

If you are building my network design and want to Bi-Wire it, you can do it this way:

Separate the connection between the big 2.5 mHy inductor and the 39+1 uFd cap. Connect the woffer cable between the inductor and the amp. Connect the high frequency cable between the 39+1 uFd cap and the amp.

It's that easy!

Figuring out where and if you want the fuse is you problem though!

Al K.

[sunnysal]

THANKS AL I WILL TRY THE RESISTORS AND LET YOU KNOW WHAT HAPPENS...TONY

[sunnysal]

weekend tweaking (with a battery to check + and - on woofers) results in correct bass polarity and BOOM!, bass is back! now I will listen with everything finally in correct polarity!