JBL 2445 HF boost with autoformer in reverse mode

[Elac310]

In case someone is interested in the idea of using an autoformer in reverse mode (i.e. to increase the signal and not to attenuate it) to boost the very high frequencies on a 2445J, I've created a thread on the matter at DIYaudio: http://www.diyaudio.com/forums/multi-way/242408-jbl-2445j-hf-boost-autoformer-reverse-mode.html#post3656995

[ivica]

In case someone is interested in the idea of using an autoformer in reverse mode (i.e. to increase the signal and not to attenuate it) to boost the very high frequencies on a 2445J, I've created a thread on the matter at DIYaudio: http://www.diyaudio.com/forums/multi-way/242408-jbl-2445j-hf-boost-autoformer-reverse-mode.html#post3656995

Hi Elac310

I think that 2-inch (4-inch voice coil) drivers such as 2441-2445-2446-2450 drivers are very good for use up to 10kHz~12kHz, but over that frequency, it seems that in the FR response a lot of unexpected peeks and deeps can be get. May be they are so narrow that would be difficult to resolve
but a kind of "audible ringing" ("metallic sound") can be get.
May be our less sensitivity to the frequencies over 12kHz would "help" to neglect that

http://www.audioheritage.org/vbulletin/showthread.php?34180-2445J-with-2311-Horn&p=345210&viewfull=1#post345210
http://www.audioheritage.org/vbulletin/showthread.php?34180-2445J-with-2311-Horn&p=350117&viewfull=1#post350117

Depending to the sensitivity of the rest of the used drivers ( say less then 98dB/2.83V/1m), headroom of around of 12dB is available for HF correction, so may be the use of the auto-transformer would not be necessary (if up to 15Khz UHF is acceptable )

Regards
Ivica

[Elac310]

The rest of the system includes 2x15" per side, i.e. more than 100dB at 4 ohms impedance.
Working in passive filtering mode as I do, the 2445 in J version (16 ohm) has insufficient sensitivity in the HF region. In normal mode, it needs no attenuation at all to match the sensitivity of the bass section so I have no headroom that would allow to compensate the FR by lowering the driver's overall level. JBL themselves used on such cinema systems as the 4670 (also a 4ohm bass section with 2x15" in parallel) a HF boost circuit comprising a capacitor followed by a small inductor.

I also have heard of the reputation of harshness in some cases but I cannot really confirm this, even despite the exta boost I apply. I also found out that an old fashioned capacitor (Russian cap in metal can said to be paper in oil or something like that) has considerable positive influence on the overall smoothness (I never had very noticeable changes in the last 20 years with different types of capacitors).

[ivica]

The rest of the system includes 2x15" per side, i.e. more than 100dB at 4 ohms impedance.
Working in passive filtering mode as I do, the 2445 in J version (16 ohm) has insufficient sensitivity in the HF region. In normal mode, it needs no attenuation at all to match the sensitivity of the bass section so I have no headroom that would allow to compensate the FR by lowering the driver's overall level. JBL themselves used on such cinema systems as the 4670 (also a 4ohm bass section with 2x15" in parallel) a HF boost circuit comprising a capacitor followed by a small inductor.

I also have heard of the reputation of harshness in some cases but I cannot really confirm this, even despite the extra boost I apply. I also found out that an old fashioned capacitor (Russian cap in metal can said to be paper in oil or something like that) has considerable positive influence on the overall smoothness (I never had very noticeable changes in the last 20 years with different types of capacitors).

Hi Elac310,

May be the easy solution would be to 'introduce' 2402H driver(s) for over 10kHz region.

Using auto-transformer (non ideal) with expectable small inductance ( say less then 2mH) can produce not-expect-able behavior of the total network, as there is influence of the transformer inductance too.

Reagrds
Ivica

[Elac310]

Hi Elac310,

May be the easy solution would be to 'introduce' 2402H driver(s) for over 10kHz region.

Using auto-transformer (non ideal) with expectable small inductance ( say less then 2mH) can produce not-expect-able behavior of the total network, as there is influence of the transformer inductance too.

Reagrds
Ivica

Na...For the time being, it works fine without a tweeter. If I add one, in comparison it sounds like adding HF noise but nothing genuine. Besides, with the HF Boost, the high frequencies are really as good as with a tweeter.

Good point: I'll have a look at what the actual inductance of the auto-former is.At least by curisosity.