AA Network sounds better with…

[mustang_flht]

Hi @mboxler

 

I do not know 

[Klipschguy]

3 hours ago, mboxler said:

 

Although this simulation shows the voltage across an 8 ohm resistor, it gives you an idea of the change.  Red is .245mh, blue .34mh.

 

Mike

 

Hi Mike. Thank you for running the curves. From the graph, it appears the .34 mH inductor shifts the crossover frequency from about 5800 Hz to about 4800 Hz relative to the .245 mH inductor. I might add the iron core inductor would have a lower resistance versus the air core which would shift the .34 mH curve up a bit. 

 

Again, I appreciate you taking the time to run the simulation. 
 

Andy

[billybob]

Non metallic screw can be the baseline for your test...

[pcbiz]

On 7/31/2022 at 6:03 PM, Klipschguy said:

Recently, I was a bit surprised to discover the AA Network in the Klipschorn sounds perceptibly better to my ear with a steel screw (not stainless or brass) securing the .245mh inductor.  It seems to give a little more energy to the lower treble  that to my ear is necessary to that “magic Khorn sound”; with the brass screw it sounds a just a tad “too polite” (again, to my ear).  My inductance meter shows .24mh (brass screw) versus .34mh (steel screw). 
 

I realize this post is going against the grain, so I was a little hesitant to post. If anyone is interested, the screw is a #8 1.5” steel Phillips pan-head. 

 

My AA experiments revealed the same. Materials do make a sonic difference. I'm relatively new to the tube amp/horn speaker scenario, so I take every tweak I can get since  I have no treble, mid, bass, presence, gain... or all those nice eq tricks I had in my solid state days.

[Klipschguy]

Yes. Pushing the crossover point down by raising the inductor value, or by just keeping where the factory originally had it in many cases, is easily audible (and for the better, in my opinion). As for others seeing the value….well…

 

[Deang]

You using the K-77? If you are, I can think of two reasons not to do this.

[Klipschguy]

K77-M.  What reasons?  It seemed OK from the factory with the ferrous screw. 

 

 

 

[tigerwoodKhorns]

1 hour ago, Klipschguy said:

K77-M.  What reasons?  It seemed OK from the factory with the ferrous screw. 

 

 

 

K-77 is very limited as to how low it can go.  Not so sure about the K77M, I have a pair and I am curious about the lower x over point.

 

Crites sells a modified Type A (or AA, I forget which) that crosses at 4500 Hz, but you have to switch the tweeter out to handle the lower x over. 

 

Seems like the screw does the same thing from the factory, intentional or not.  To the old hats here, was there a Dope from Hope on this instructing to change out the screw to protect the K77M? 

[tigerwoodKhorns]

Here is the 4500 x over, a modified Type A.

 

If the 2.4 MH is only to the woofer as a low pass, how does it change the squaker / tweeter x over point? 

 

 

[Deang]

1 hour ago, Klipschguy said:

K77-M.  What reasons?  It seemed OK from the factory with the ferrous screw. 

 

Like I said, they were being built by some employees at home for extra money and it was simply missed during QA. It wasn't "OK" because they caught it and fixed it.

 

As already pointed out, you can see how it shifts the response, which probably increases distortion. It also increases summing errors, which could include phase (I can't remember). There is also a peak in the K-77 that the filter is designed to bring down. Your "tweak" moves the target frequency.  

 

 

K77 Peak.webp

[Klipschguy]

Hi TigerWoodKhorns,

 

I do know that Paul Klipsch liked shallower slope crossovers for sound quality. I think he went to steeper slopes mostly to protect tweeters. At 18dB/octave coupled with the zener diodes, I think the K77 would be quite safe at any reasonable sound pressure levels in home use. 
 

My brother’s vintage Khorns use a K77 with a 500–5000 crossover, which has the crossover set at 5000Hz 6dB/octave for the tweeter. No zener diodes.  He cranks them up all the time and there has NEVER been a problem (200wpc available, but not used).   He does keep it reasonable though (sub-ridiculous). The Type A is 6dB/octave at 6000 was used for years and is a very reasonable crossover for home use; it provides WAAAY less protection than the AA with the ferrous screw. 
 

At a 4800Hz crossover frequency at 18dB/octave, the K77 would be 18dB down at 2400Hz, 36db down at 1200Hz, & 54dB down at 600Hz.
 

At a 6000Hz crossover frequency at 6dB/octave (Type A) would only be 6dB down at 3000Hz, 12dB down at 1500Hz, &  18dB down at 750Hz. 

 

The K77 may have a little trouble reaching below 5000Hz, but then again, the K55 has some issues reaching 6000Hz, so a little lower crossover point for the K77 is probably a good thing. 

[Klipschguy]

7 minutes ago, Crankysoldermeister said:

 

Like I said, they were being built by some employees at home for extra money and it was simply missed during QA. It wasn't "OK" because they caught it and fixed it.

 

As already pointed out, you can see how it shifts the response, which probably increases distortion. It also increases summing errors, which could include phase (I can't remember). There is also a peak in the K-77 that the filter is designed to bring down. Your "tweak" moves the target frequency.  

 

 

K77 Peak.webp 74.35 kB · 1 download

Dean, I thought you liked the Type A? It would have a lot more IM distortion vs a 18dB/octave at 4800Hz. I think the more favorable frequency response would far outweigh the slight increase in IM distortion. BTW, thank you for your response. 

[Deang]

3 minutes ago, Klipschguy said:

Dean, I thought you liked the Type A? It would have a lot more IM distortion vs a 18dB/octave at 4800Hz. 

 

I like them at very low volumes. I pretty much used the AA for over 15 years. 

[Klipschguy]

I understand. I would think a Type A should do just fine at 110dB peaks, if the power is clean and the treble control is set to flat. As you know, 110dB is VERY loud. I know my brother pushes his Khorns pretty hard at times (6dB/oct at 5000Hz), but always clean. 
 

As always, thank you for your thoughts and taking the time to reply. 

[OO1]

On 7/31/2022 at 6:03 PM, Klipschguy said:

 sounds perceptibly better to my ear with a steel screw 

gotta try it 

[Klipschguy]

3 hours ago, tigerwoodKhorns said:

Here is the 4500 x over, a modified Type A.

 

If the 2.4 MH is only to the woofer as a low pass, how does it change the squaker / tweeter x over point? 

 

 

I think Bob intended his crossover to be used with a different tweeter. A 6dB slope at 4500Hz with a K77 is probably pushing the limit. Although, as stated earlier, my brother’s Khorn do just fine at 6dB/oct at 5000hz. 
 

The 2.4mH inductor won’t change the tweeter crossover point but it does cause a 90 degree phase shift in the woofer leg of the circuit. 

[tigerwoodKhorns]

45 minutes ago, Klipschguy said:

I think Bob intended his crossover to be used with a different tweeter. A 6dB slope at 4500Hz with a K77 is probably pushing the limit. Although, as stated earlier, my brother’s Khorn do just fine at 6dB/oct at 5000hz. 
 

The 2.4mH inductor won’t change the tweeter crossover point but it does cause a 90 degree phase shift in the woofer leg of the circuit. 

 

 

Correct, he has a different tweeter. 

 

So how does the x-over go down from 6K to 4800HZ with the metal screw in the 2.4 inductor?  Or did I read this wrong. 

 

[Klipschguy]

48 minutes ago, tigerwoodKhorns said:

 

 

Correct, he has a different tweeter. 

 

So how does the x-over go down from 6K to 4800HZ with the metal screw in the 2.4 inductor?  Or did I read this wrong. 

 

Not the 2.4mH inductor in the woofer leg, but rather the .245mH in the tweeter leg of the AA network (notice the decimal). The Type A network does not have an inductor in the tweeter leg because it is a simpler design with less roll off per octave.

 

The .245 inductor is the round thing in the top right corner of the picture BESIDE the 2.4mH inductor. Note the shiny ferrous screw in the middle; that little guy is what changes the inductance depending on whether is a material that is attracted by a magnet (like good ole steel), or not (like brass or stainless steel).

[tigerwoodKhorns]

51 minutes ago, Klipschguy said:

Not the 2.4mH inductor in the woofer leg, but rather the .245mH in the tweeter leg of the AA network (notice the decimal). The Type A network does not have an inductor in the tweeter leg because it is a simpler design with less roll off per octave.

 

The .245 inductor is the round thing in the top right corner of the picture BESIDE the 2.4mH inductor. Note the shiny ferrous screw in the middle; that little guy is what changes the inductance depending on whether is a material that is attracted by a magnet (like good ole steel), or not (like brass or stainless steel).

 

I am so used to looking at the Type A schematic and not the AA.  Thanks. 

 

 

[mustang_flht]

Hi 

 

Type AA, man ☺️😉