JBL 2404 and ALK in KHORN

[markus111]

Hi All,

I have a pair of the JBL 2404 baby cheeks on the way, and was giving some thought as to how to integrate them into my khorns with ALK crossovers. I figure I'll first try it the straightforward way - replace the t35 with the jbl. The second option would be to build John Warrens tweeter crossover with the LCR, and tap that into the ALK after the 40mF cap. Has anyone tried this? The LCR crossover apparently changes the average impedance of the tweeter to 6 ohms. Is this a concern? Would this franken-crossover idea mess up the rest of the ALK? Any experiences or insight would be appreciated. I'm really looking forward to trying these out!

Mark

[3dzapper]

Mark, You are going to want to pad the 2404s by a couple of Dbs in the long run. You can use an L-pad calculator and build a fixed pad or as I did put an adjustable L-pad in the tweeter line.

Right out of the box you will note an improvement. When they are dialed in........

[jorjen]

Rick,

You know, I was prepared to have to pad my 2404's when I was awaiting their arrival. But once I connected them to the ALK's in my Khorns, I found after extended listening that I did not need to at all. K55's are on 5-2 setting and things seem to blend rather nicely. Go figure! Hey, maybe it's the rolloff in these 48 year old ears!.

[Al Klappenberger]

Mark,

You will need to draw out that RLC stuff you are thinking of adding. My guess is that you may need an L pad, but that should be all. My network runs the tweeter hot.

Al K.

[Deang]

Al, what's the acoustic output (spl) of the K-400/K-55 using the ALK?

I think the notch circuit is a must to damp the resonance. I wouldn't run the JBL without it.

[Al Klappenberger]

Dean,

The squawker level is adjustable in all my networks. That means you set it where you want it. Hopefully, that will be to match the output of the woofer.

Al K.

[Deang]

I asked a question I already knew the answer too! Yeesh.

[Al Klappenberger]

Dean,

Old age maybe?

Al K.

[Deang]

Hell no, it comes from talking too much to really old people. LOL!!

I had a really good time today, did you hear about it?

[Daddy Dee]

Well, I think it may come from having a sense for the hereafter...

For example, this morning I went in the other room....

... and I said, what'd I come in hereafter.

[3dzapper]

You too huh? A mind is a terrible thing to loose.

Rick

[BEC]

"What a waste it is to lose one's mind. Or not to have a mind is being very wasteful. How true that is."

-Former U.S. Vice-President Dan Quayle

Bob

[markus111]

Hi Everyone,

Thanks for all of your comments. The bass and midrange are the same as the ALK. The tweeter portion has been replaced by John Warren's JBl mod. It is basically an 18db high pass followed by a 1.5db l-pad followed by an LCR. The LCR reduces an impedance rise at 4.7 kHz. The LCR reduces the average impedance of the tweeter to 6 ohms, and the filter and l-pad are calculated for 6 ohms. Will this scheme mess with the ALK bass and midrange? Please forgive my drawing abilities - I can't draw for beans

Thanks,

Mark

[Al Klappenberger]

Mark,

The tweeter filter you have grafed into my netwrok is WRONG! The correct values should be 2.2 uF, .2 mHy and 7 uF. What you have there will generate a screwed up tweeter / squawker transition! What you connect to the filter does NOT justify the value changes!

Al K.

[Al Klappenberger]

Here's a computer analysis of the tweeter mods you posted. As you can see it's a mess! There is no good reason to put an R-L-C network broadly resonant at the 7 dB down point on the tweeter filter's skirt! It is the kind of thing that is allover the "AL" network found in LaScalas and what gives it it's bad reputation. It is not functioning as an equalizer becasue it is out of the tweeters frequency range. It also makes no sense as an impedance corrector becasue it is doing more harm than good. If a big JBL tweeter needs compensation like that it isn't as good a tweeter as it's price tag would suggest! The K77 or Beyma CP25 needs no such corrections!

Al K.

[John Warren]

Gee Al, we would of NEVER thought about modelling the response(?)

First off, you modelled it wrong. You assumed the LCR is working into a resistor, it is not!

The graph below is the 2404H filter. It is a SPICE model that assume the filter is working into a 6ohm load. The LCR + the tweeter is the 6Ohm load.

How do I know that? because we worked the LCR until the impedance was "essentially" a flat 6 Ohms between 3-8kHz.

[John Warren]

Here the SPICE model (red) and the actual filter measurements (blue).

The filter response (blue) is the entire circuit, filter + LCR + 2404H tweeter.

The measurements are calculated dB responses based on impedance.

Reasonable agreement.

[Al Klappenberger]

Marilyn,

As usual, you don't know what you are talking about!

Put the tweeter filter back the way it is supposed to be!

Al K.

[John Warren]

Well...

Here's your plot (-5dB at 7kHz)

And here's a SPICE model with the LCR feeding an 8 Ohm resistor, which is NOT the way to model the 2404H filter.

Gee Al, they look the IDENTICAL.

[Al Klappenberger]

Mark,

Warren can't see the forrest for the trees. The element values in the tweeter filter are as they are to be compatible with the squawker filter associated with it. To change them for any reason is to mess up the smooth transition between the tweeter filter and the squawker filter. They are an inseperable pair! The only variable is the termination (driver) you apply to that filter. It should, in theory, represent a purley resistive load to the filter. Most quality tweeters are close enough to 8 Ohms to provide a good enough load. In short, you make the load right, you do NOT screw up the filter! Put the 2.2 uf, .2 mHy and 7 uF value back in the tweeter filter! The R-L-C to ground is doing nothing but screwing things up. A quality tweeter like the JBL doesn't need that crap. If it does, go back to a K77 / T-35!

Al K.