How far back?

[Islander]

The length of the La Scala woofer horn is a bit over 3 feet.  The green arrows in Marvel's drawing above shows the signal path correctly.  This suggests that for proper time alignment, the squawker driver should be located past the rear of the cabinet.  However, the difference should be the difference in length of the horns, because that horn you're using is fairly long.  Moving the horn way back on the cabinet also introduces reflections from the top of the cabinet, which is why the fronts of all three horns are in the same vertical plane in the stock speaker.  That way, each horn opens into clear space, with nothing to reflect off.

 

What caused you to reject the stock HF cabinet, with its carefully chosen crossover values and generally harmonious sound?  Does that vintage horn sound better than the K401/K-55 combo?  A lot of development went into the La Scala II, and the improvements relative to the first-gen La Scala are obvious to most listeners, so it's surprising to see your setup.  What are you using for a crossover?  The stock one, or something totally different?

 

My 2-way bi-amped La Scala IIs are modded with the big Jubilee tweeters (as you can see in the picture to your left), and the delay is corrected in the Electro/Voice active processor/crossover, using delay times determined by Roy Delgado in the Klipsch lab, accurate to the millisecond.  There was no experimenting on my part.  Even so, the difference is not obvious to my ears, except with certain types of music, usually tunes that have a bass beat that keeps time with a midrange beat.  To some other listeners, the effect of correct time alignment is obvious with all kinds of music.  Every ear/brain combo is different, right?

 

As well, the EQ in the processor is pretty detailed, correcting various dips and peaks in the frequency response of the stock speaker.  The result is better and smoother sound than is available from any passive crossover, even one as refined as the one in the LS2.  Those settings were provided by Roy, who used the same E/V Dx38 processor when developing the Jubilee, but had also tested La Scalas, original and LS2, with the Jubilee tweeter, and supplied those settings to us guys who followed the "recipe" to adapt the wide-range Jubilee tweeter to the Scalas, making it possible to cover the range from about 480 Hz to 20 KHz with one driver.  This eliminates one driver-to-driver transition, with its accompanying phase shift issues.

[Chief bonehead]

9 minutes ago, captainbeefheart said:

 

I left room for the exception by saying it's "not typically detectable by ear".

 

There are many variables and of course a large enough time error will be audible.

What’s “large”?  Just curious. 

[Marvel]

4 minutes ago, Islander said:

The length of the La Scala woofer horn is a bit over 3 feet.  The green arrows in Marvel's drawing above shows the signal path correctly.  This suggests that for proper time alignment, the squawker driver should be located past the rear of the cabinet.  However, the difference should be the difference in length of the horns, because that horn you're using is fairly long.  Moving the horn way back on the cabinet also introduces reflections from the top of the cabinet, which is why the fronts of all three horns are in the same vertical plane in the stock speaker.  That way, each horn opens into clear space, with nothing to reflect off.

 

For sure, having the three horn mouths in the same plane and using a digital crossover to adjust for the time errors would be the best way. With my tweeter on the back edge of my LS, I could never detect the reflections off the top of the cabinet. But I sat with my head slightly below the top, so there is that.

 

Krock has the tweeter and mid aligned, but that setup would need to hang way off the back to included aligning with the woofer. At least it's not an MWM woofer bin.

[captainbeefheart]

13 minutes ago, Chief bonehead said:

What’s “large”?  Just curious. 

 

Well that's a tricky question because some people have incredible hearing ability, but I would say the average person isn't going to hear a delay of less than 5mS. When getting around 1-5mS it can now cause more of a Chorus effect than an audible delay effect. So I feel safe saying <1mS isn't going to be audible to the vast majority of population.

[Islander]

1 minute ago, Marvel said:

 

For sure, having the three horn mouths in the same plane and using a digital crossover to adjust for the time errors would be the best way. With my tweeter on the back edge of my LS, I could never detect the reflections off the top of the cabinet. But I sat with my head slightly below the top, so there is that.

 

Krock has the tweeter and mid aligned, but that setup would need to hang way off the back to included aligning with the woofer. At least it's not an MWM woofer bin.

 

Marvel, having your tweeter well above the cabinet could be greatly reducing any reflections, especially if you put some soft sound-absorbing material on the cabinet from the tweeter to the front.  The short wavelengths of the tweeter's sounds are easily absorbed by slightly thick cloth.  It might be something to try when you have time.  When the house is quiet, check the sound with and without a big soft cloth, and maybe even a folded towel, and see if you can hear any difference.  And let us know what you find.

[Invidiosulus]

I don’t think it’s a matter of most people hearing it and saying “wow, I can tell your mid/tweets are time aligned!”

 

I think it’s a matter of when it’s right it sounds right and the stereo imaging(when using two speakers) through the crossover frequencies is that much clearer and everything sounds more cohesive.

[KROCK]

Islander. With my setup, Prima Luna HP & Don Sachs preamp the 401/55 combo & the original crossover the 401 wasn't producing clear vocals. Now, being I changed the crossover first to a ALK ES5400, the M55I think was the problem. I have no said distortion from how it's setup now.

[Invidiosulus]

3 minutes ago, Islander said:

 

Marvel, having your tweeter well above the cabinet could be greatly reducing any reflections, especially if you put some soft sound-absorbing material on the cabinet from the tweeter to the front.  The short wavelengths of the tweeter's sounds are easily absorbed by slightly thick cloth.  It might be something to try when you have time.  When the house is quiet, check the sound with and without a big soft cloth, and maybe even a folded towel, and see if you can hear any difference.  And let us know what you find.

The LS with the outboard tweeters are now at my place and @Marvel’s grandkids have been kind enough to place all manner of dispersing materials on top of the cabs for me.

[Islander]

19 hours ago, KROCK said:

This is how I have it now;

10" off the rear wall from what I've tried to gather off drawings on the net.

 

I like your driver support rack.  It looks similar to the factory unit, but was probably much less expensive.

[KROCK]

Invidiosulus

I have pyramid  foam on top of the 511. Just removed it to take a pic.

Also I have the tweeter on a slight tilt to my ear level.

[KROCK]

[Islander]

20 minutes ago, KROCK said:

Invidiosulus

I have pyramid  foam on top of the 511. Just removed it to take a pic.

Also I have the tweeter on a slight tilt to my ear level.

 

I've got my tweeters angled forward 7 degrees, which aims them directly at the seated listening position.  Since they're sitting on top of the disconnected HF cabinets (I've got nowhere to put those things, especially when they're in their factory boxes), the tweeters are closer to the ceiling than the floor, so the tilt should reduce the chance of any ceiling bounce.  All these barely audible details, but they add up, and the result is great sound.

[Edgar]

26 minutes ago, Chief bonehead said:

What’s “large”?  Just curious. 

From Audibility of Group-Delay Equalization, Juho Liski, Aki Makivirta, and Vesa Valimaki, IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, VOL. XX, NO. X, JANUARY 2021:

"The listening test contained four different signals: the unit impulse, the pink impulse, a castanet recording, and a synthetic hi-hat cymbal sound. The two former signals are known to be the most critical for group-delay audibility, whereas the latter two resemble real-life musical signals and thus help to generalize the results better. Group-delay audibility was tested at the frequencies 500 Hz, 1 kHz, 2 kHz, 3 kHz, and 4 kHz. The results indicate that the audibility thresholds for local group-delay variation are less than ±1 ms [1 ms is approximately 13½ inches - Edgar] for the most critical signals, and approximately 1.5 ms to 4.5 ms for a local positive group-delay peak and between -1.0 ms and -2.3 ms for a local negative group-delay peak for real-life signals."

[Islander]

KROCK, have you tried locating your LS2s closer to the front wall (the wall behind the front speakers, of course)?  Mine are just 5" from the wall, measured from the nearest corner of the speakers.  I found that closer to the wall gives smoother bass response, and once you pull them out past 6', there's no improvement, just more floorspace lost.

 

Also, that foam looks like it should be totally effective.

[KROCK]

Not an option;

[Chief bonehead]

53 minutes ago, Invidiosulus said:

I don’t think it’s a matter of most people hearing it and saying “wow, I can tell your mid/tweets are time aligned!”

 

I think it’s a matter of when it’s right it sounds right and the stereo imaging(when using two speakers) through the crossover frequencies is that much clearer and everything sounds more cohesive.

Exactly. 

[Chief bonehead]

36 minutes ago, Edgar said:

From Audibility of Group-Delay Equalization, Juho Liski, Aki Makivirta, and Vesa Valimaki, IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, VOL. XX, NO. X, JANUARY 2021:

"The listening test contained four different signals: the unit impulse, the pink impulse, a castanet recording, and a synthetic hi-hat cymbal sound. The two former signals are known to be the most critical for group-delay audibility, whereas the latter two resemble real-life musical signals and thus help to generalize the results better. Group-delay audibility was tested at the frequencies 500 Hz, 1 kHz, 2 kHz, 3 kHz, and 4 kHz. The results indicate that the audibility thresholds for local group-delay variation are less than ±1 ms [1 ms is approximately 13½ inches - Edgar] for the most critical signals, and approximately 1.5 ms to 4.5 ms for a local positive group-delay peak and between -1.0 ms and -2.3 ms for a local negative group-delay peak for real-life signals."

If you measure the freq response, does that jive with the test?

[Edgar]

1 minute ago, Chief bonehead said:

If you measure the freq response, does that jive with the test?

 

For the paper, they used allpass filters to create the delay. So the magnitude response was unaffected, but of course the phase response reflected the delay.

[Islander]

22 minutes ago, KROCK said:

Not an option;

 

That's a nice setup you have.  Those corners are ideal placement for most Heritage Series speakers, as they give the best bass response when in corners, La Scalas in particular.  In my room, that's not possible, due to one side that opens to the dining room and kitchen, and the other side has a baseboard heater.  Lucky you!

[Chief bonehead]

1 hour ago, Edgar said:

 

For the paper, they used allpass filters to create the delay. So the magnitude response was unaffected, but of course the phase response reflected the delay.

Oh I see. Different animals. I only deal with getting two or more sections to get them not only time aligned but “period” aligned to that the amplitudes at crossover add up to the proverbial 6 db while eliminating as much interference at crossover.