My 1977 LaScalas finally got their original Autoformers T2A, what a surprising reveal!

[KT88]

22 minutes ago, mikebse2a3 said:

 

Heinz I checked the driver wires which appear to be stranded tin coated and they appear to be slightly loose in the 18ga “stranded stripper slot” and tight in the 20ga “stranded stripper slot” so I would go with 18ga IMHO.

 

miketn

 

 

 

 

Thanks a lot, Mike!

[Dave MacKay]

5 hours ago, KT88 said:

I'm into day 2 and my first positive impression from yesterday was not a short-changed showstopper just because it's "different". I got a lot more timbre, a great 3D stage and not a single 60's trumpet dissecting my ear.

Have you taken any measurements, e.g., with REW? If so, please share a plot that shows the performance measured with each autoformer.

[KT88]

11 minutes ago, Dave MacKay said:

Have you taken any measurements, e.g., with REW? If so, please share a plot that shows the performance measured with each autoformer.

 

I did no measurements, just used my ears. There was a thread where Autoformers were measured. TBH I believe my ears. On the thread below I lost somehow the plot after a while which does not mean that sensible measurements are not worth it. I also are interested to see measured results with significance as a proof regarding what I here.

 

 

[Deang]

4 hours ago, Dave MacKay said:

Have you taken any measurements, e.g., with REW? If so, please share a plot that shows the performance measured with each autoformer.

Kind of confused. You were following Mike’s thread on Audiokarma, and even “liked” the post with the plots. 

 

[Dave MacKay]

3 hours ago, Deang said:

Kind of confused. You were following Mike’s thread on Audiokarma, and even “liked” the post with the plots. 

 

Was wondering if @KT88 had before/after measurements to show the differences he hears.

 

Thanks for the reminder about the plots and for posting them. The differences looked to be pretty minor (to my eye) which is why I was  curious. I thought that a single chart showing before & after would be clearer.

[henry4841]

As I stated previously I played with a number of different X-over designs for my '86 LaScala and the one I thought best sounding to my ears was the one with no autotransformer just as Klipsch has abandoned it in it's present line of speakers. But that being said I am using an AA being old and wanting the speakers stock as much as possible for my kid after I no longer need them. Sound plenty good enough but I did think without the autotransformer did sound somewhat better but I am sensitive to phase shift which any transformer is going to have. 

 

Not trying to make a point, just info for anyone such as KT-88 that might want to play with the network for their older Heritage line. Designing and building different networks can be fun and entertaining if one has the time. No scientific analysis was done, just a layman's ears for the final judge. I do not put much emphasis on others doing measurements of speakers no matter what kind of test instruments used being most do not have the proper room to do such test that will certainly be different for someone else's room no matter what that find. Very few have an anechoic chamber to do proper speaker analysis. Testing speakers is not like testing electronic gear unless one has the proper facility. Just my thoughts. 

[KT88]

13 hours ago, Dave MacKay said:

Have you taken any measurements, e.g., with REW? If so, please share a plot that shows the performance measured with each autoformer.

Very interesting, I have read the Audiokarma thread.
I think when it comes to measuring my positive listening impression of the T2A vs 36xx, other measurement criteria could play a role, more involving the overall context in the way the LaScala performed as a whole.

 

So we're talking about things other than just SPL. An SPL change can make a big psychoacoustic effect. But what I hear with the T2A is a qualitative difference that encompasses more than if I, for example, cleanly change only the SPL of the K402 at the DSP on my UJ.

 

Phase shift

I'm only a technical layman, but when it comes to whether you can hear phase rotation or not, I mean you can hear it very easily in comparison. That the T2R makes a different phase angle than the 3654 could be shown by my simple mono test (left T2R, right 5436). The sound is no longer coming exactly from the center but a bit out of space.

Now the question is, what effect a certain phase rotation in the range of the crossover frequency of the K400 can have compared to the phase rotation of the bass (and the tweeter). Because the bass also has a phase rotation over its frequency response, I assume. So can it be sound-decisive that at e.g. 400Hz the bass horn and the K400 show a "happy" well-matched phase angle? At least I believe that in interaction the phase is not irrelevant...even if it may not be so important when considered in isolation (you only hear the K400 for example).

Ringing

Honestly, the 3654 makes very unpleasant resonances in the upper frequency range of the K400, making everything sound more washed out and indistinct. And that with the higher SPL of the 36xx by only 3dB reduction. In the other forum someone said that more iron could make more ringing, I don't know if that's true but that's how I heard it with the T3654 before! I read the thread. So it is a confirmation and not self-deception in the sense because I wanted to hear something I read.

Balance of bass, mid and high.
It seems to me that the bass is fuller, almost like an EQ.

 

My suggestion for measurement criteria:

 

All of the following criteria refer to T2A vs. 3654 - installed in the network and measured at the loudspeaker.

SPL/ frequency response over all
Frequency response of each of the three drivers bass, mid, high, looking at whether the bass SPL changes when the autoformers are swapped.

Phase response of all three drivers individually, especially in the crossover area to the bass.

If possible, measure something like this Sawtooth impulse with measuring points from 400 Hz to 6000 Hz

Waterfall plot, maybe then you can see the larger ringing of the 3654.

Perhaps these measurements let understand more the positive sound impression than if only the autoformers are measured in isolation.
 

 

 

 

 

 

 

[Deang]

12 hours ago, Dave MacKay said:

Thanks for the reminder about the plots and for posting them. The differences looked to be pretty minor (to my eye) which is why I was  curious. I thought that a single chart showing before & after would be clearer.

 

Like Mike said, since the 3636 has exactly twice the inductance, the plots look symmetrical - but that does not mean they are the same. If you look closely, you'll see the rate of attenuation below the cut-off frequency is different. Compare the capacitor only line, to tap 4 at 400Hz and 200Hz. 10dB/octave is not 12dB/octave. Since this relates to rate/amount of attenuation being applied just below frequency cut-off, maybe think about this in terms of what is happening at the diaphragm. It sounds different because it measures different. Now, there is a way to band aid this, and if you were paying attention, Mike discussed this in the AK thread.

 

Same with the capacitors. Modest change, but very audible - and of course all of these changes are accumulative in their effect.

[mboxler]

1 hour ago, Deang said:

 

Like Mike said, since the 3636 has exactly twice the inductance, the plots look symmetrical - but that does not mean they are the same. If you look closely, you'll see the rate of attenuation below frequency cutoff is different. Compare the capacitor only line, to tap 4 at 400Hz and 200Hz. 10dB/octave is not 12dB/octave. Since this relates to rate/amount of attenuation being applied just below frequency cut-off, maybe think about this in terms of what is happening at the diaphragm. It sounds different because it measures different. Now, there is a way to band aid this, and if you were paying attention, Mike discussed this in the AK thread.

 

Same with the capacitors. Modest change, but very audible - and of course all of these changes are accumulative in their effect.

 

Dean's right.  The larger the inductance between the output taps to more the filter behaves like a first order vs a second order.  Unfortunately, I can't remember if that test was with a new T2A or the way-out-of-spec one I removed from my Heresy.  I'll rerun that voltage test with the Heresy one just to be sure.

[Deang]

Be sure to use the Boxler Alphabet System, lol. 

[mboxler]

 

Deleted 

 

 

[Trey Cannon]

On 5/17/2023 at 5:58 PM, Dave MacKay said:

Have you taken any measurements, e.g., with REW? If so, please share a plot that shows the performance measured with each autoformer.

REW can only test SPL at a 2.7sec time laps...You had better be outside or in a VERY big room no to get the ROOM in the curve... This is because it is a ROOM EQ WIZARD...Not a speaker measuring tool.

 

[Dave MacKay]

9 hours ago, Trey Cannon said:

REW can only test SPL at a 2.7sec time laps...You had better be outside or in a VERY big room no to get the ROOM in the curve... This is because it is a ROOM EQ WIZARD...Not a speaker measuring tool.

 

What tool would you recommend to see any audible difference between autoformers? 

[HiFi Heaven]

AudioTool is a cheap 1/12-octave RTA that runs on Android phones and tablets.

Includes pink-noise to inject into your amp + sine & square-wave generators.

The electret mic in your device is excellent to 10kHz, fine for these tests.