Klipsch AA net attenuation mod

Discuss modeling, designs and measurement schemes.
User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Klipsch AA net attenuation mod

Postby jwarrenNRE » Mon Jan 02, 2017 10:40 pm

If you're using the AA network (Klipsch, diy, etc.) and find that the -3dB setting is not providing sufficient attenuation for the particular midrange assembly your using, you can "tweek" the attenuation a wee bit lower by adding a series resistor as shown in the figure below.

The power dissipation across the resistor is small.
Attachments
AA_mod.jpg
AA_mod.jpg (312.95 KiB) Viewed 4791 times

User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Re: AA net mod

Postby jwarrenNRE » Mon Jan 02, 2017 10:42 pm

So what values to use.

Going to the simulator (these are all electrical domain simulations!, not acoustic, repeat x3).

Below is the predicted voltage response of the AA at the positive terminal of the compression driver with 5 different resistors (1, 2, 3, 4 and 5 Ohms). As the resistance increases so too the attenuation. The hot pink plot is the AA without the resistor. The blue plot is the voltage at the positive terminal of the Klipschorn with the series inductor shown in the AA schematic.

The attenuation is about 0.5dB/Ohm of series resistance. It's really no different than the attenuation (i.e. loss) associated with the resistance of speaker cabling.

The crossover point changes a bit but the change is something that needs to be listened to. So f you don't try the experiment you won't know if it's significant. Obviously, the larger the resistor the greater the change.
Attachments
AA_mod_1.jpg
AA_mod_1.jpg (388.04 KiB) Viewed 4790 times

User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Re: AA net mod

Postby jwarrenNRE » Mon Jan 02, 2017 10:44 pm

For those of you that like this sort of thing, here's the impedance as the value of the resistor changes. No big deal.
Attachments
AA_mod_2.jpg
AA_mod_2.jpg (379.23 KiB) Viewed 4789 times

User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Re: AA net mod

Postby jwarrenNRE » Mon Jan 02, 2017 10:59 pm

Here's the impedance as a function of series resistor value compared to the stock AA. As is evident in the simulation, the impedance magnitude at 5kHz is predicted to be about 24 Ohms and with the 5 Ohm series resistor (a -6dB attenuation setting) it's about 31 Ohms.
:o
:D
Attachments
AA_mod_3.jpg
AA_mod_3.jpg (415.84 KiB) Viewed 4780 times

User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Re: AA net mod

Postby jwarrenNRE » Mon Jan 02, 2017 11:08 pm

Try it with a couple of alligator clip patch cords and a handful of resistors. The simulations suggest it'll work in a predictable way. If you decide to simply hang a resistor off the terminal of the compression driver and clip it to the speaker lead that would work too.

If you'd rather just switch to the -6dB tap then you'll have to change the capacitor from the stock 13uF (see schematic above) to about 9uF. The electrical simulation of that change is shown below. The hot pink plots are the series resistor plots above.
Attachments
AA_mod_4.jpg
AA_mod_4.jpg (340.93 KiB) Viewed 4771 times

User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Re: AA net mod

Postby jwarrenNRE » Mon Jan 02, 2017 11:34 pm

You really shouldn't change the midrange tap without changing the cap otherwise this is what happens.

That will likely sound odd and could cause your midrange to "bark".
Attachments
AA_mod_5.jpg
AA_mod_5.jpg (296.8 KiB) Viewed 4766 times

User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Re: Klipsch AA net attenuation mod

Postby jwarrenNRE » Fri Jan 13, 2017 10:21 pm

Here's how the autoformer with and without the resistor is modelled. The {R} shown in the resistor R6 allows the magnitude of the resistor to step from 1 to 5 Ohms.
Attachments
T2A_autoformer.jpg
T2A_autoformer.jpg (172.74 KiB) Viewed 4614 times

User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Re: Klipsch AA net attenuation mod

Postby jwarrenNRE » Fri Jan 13, 2017 10:23 pm

Output (phase and magnitude) from the sim model shown above.
Attachments
T2A_autoformer_1.jpg
T2A_autoformer_1.jpg (314.78 KiB) Viewed 4613 times

User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Re: Klipsch AA net attenuation mod

Postby jwarrenNRE » Sat Jan 14, 2017 1:16 pm

So let's see if it's bullshit.

Here's the setup...
The blue resistors are a 13 Ohm resistive load (a non reactive load that's simulating the K55X, close enough for our purposes).
The spectrum analyzer is clamped across the 13 Ohm load.
The white resistors are 1 Ohm, 2.7 Ohm and 4.5 Ohms and each is placed at the location shown in the schematic at the top of this thread and the response across the 13 Ohm load measured.
The two black caps are the 13uF load on the "primary" side of the autotransformer.
The signal is a sweep from 10-80kHz, the amplifier is a hot-rodded Crown D45.
Attachments
resistor_load_AA.jpg
resistor_load_AA.jpg (469.93 KiB) Viewed 4589 times

User avatar
jwarrenNRE
Site Admin
Posts: 185
Joined: Thu Nov 03, 2016 5:32 pm

Re: Klipsch AA net attenuation mod

Postby jwarrenNRE » Sat Jan 14, 2017 1:19 pm

Here's the analyzer results from the setup above.

Red is no series resistor
Green is the 1 Ohm
Blue is 2.7 Ohm
Gold is 4.5 Ohm.

So its not perfection but for small changes like a dB or two in the AA midrange output it's works fine.
Attachments
resistor_load_AA.jpg
resistor_load_AA.jpg (99.84 KiB) Viewed 4589 times


Return to “Crossover Networks”

Who is online

Users browsing this forum: No registered users and 1 guest