crossover improvement help needed

[tom1356]

I have a pair of coaxial speakers that use this circuit as a crossover.
The parts quality is top shelf so there is no where to go there.
I was wondering if any experts could see any areas that could use improvement or tweaking?
Thanks
Tom

[darkfenriz]

I am no expert, but this series xover connection with a resistor directly at amp's output looks bad for me.
I would build a typical parallel second order instead.

[tom1356]

darkfenriz
The 1 ohm R1 resitor is there to match the low output impedance of the leak TL-12 amp to the circuit. It is bypassed for amps with higher output impedance.
Thanks
Tom

[Brian Knight]

If it is being used with the intended drivers I would not fiddle with this crossover. The parallel inductor/capacitor/resistor in series with the high-frequency unit appear to form a resonant trap which is probably designed to control some peak or dip in the HF unit's output. The non-standard component values suggest to me that the designer has already taken some trouble to get the crossover right and it is probably best to leave well-enough alone. If the speaker system does not perform well it is probably because of limitations with the drivers: crossover design can only do so much to fix this.

[tom1356]

Originally posted by Brian Knight
If it is being used with the intended drivers I would not fiddle with this crossover. The parallel inductor/capacitor/resistor in series with the high-frequency unit appear to form a resonant trap which is probably designed to control some peak or dip in the HF unit's output. The non-standard component values suggest to me that the designer has already taken some trouble to get the crossover right and it is probably best to leave well-enough alone. If the speaker system does not perform well it is probably because of limitations with the drivers: crossover design can only do so much to fix this.

Hello Brian,
The crossover is working perfectly. It was designed by a master about 57 years ago. The complete system is the best I have ever heard.
The tweaker in me has a real need to try to "improve" things I should leave alone.
I will fight the urge to mess it up.
Thanks
Tom

[darkfenriz]

The 1 ohm R1 resitor is there to match the low output impedance of the leak TL-12 amp to the circuit.

That's exactly what I don't unerstand. Most speakers are designed to be driven from voltage source i.e. zero impedance, just for higher Q and better mechanical resonance damping.

[tom1356]

Originally posted by darkfenriz

That's exactly what I don't unerstand. Most speakers are designed to be driven from voltage source i.e. zero impedance, just for higher Q and better mechanical resonance damping.

darkfenriz
According to the original monograph...the 1 ohm resistor is included "to ensure the filter (crossover) is always effectivly fed from the same source impedance". Apparently they switched between two different amps but wanted nothing to change.
The BBC were real sticklers.
Thanks
Tom

PS: I have been searching for a pair of the special Leak TL-12 amps for quite a while. I would like to hear the speakers driven with the amps designed for them.

[kevinkr]

These speakers were designed at a time when the "perfect" voltage source (amplifier) did not exist. Better then to design the network to perform consistently with a known and reasonable source impedance that could be padded as needed to compensate for amplifier source impedance variations.

I'll bet that horn has a pretty big peak right around 12kHz and that trap was intended to remove it.

I would not modify the existing x-over at all, however you might try replicating it with modern components and see if the overall result is different.

The only way you could do anything meanfully better/different beyond modern components would be if you could measure both the responses and impedances of the drivers and develop an x-over that performed measurably better, even then you might be limited by the performance of the drivers themselves.

[tom1356]

Originally posted by kevinkr

I'll bet that horn has a pretty big peak right around 12kHz and that trap was intended to remove it.

Hi Kevin,
You are correct the trap is there to tame a 12kHz peak. I'm not going to touch them.
One last question is the positive input no.1 or no.2?
As for beter drivers...if someone should build some I will gladly buy them.
Thanks
Tom

[sreten]

Originally posted by tom1356

One last question is the positive input no.1 or no.2?
Tom

By convention input 1 is the postive connection, +dc = bass unit forward.

Originally posted by tom1356

As for beter drivers...if someone should build some I will gladly buy them.
Tom

Hmmm.......

Better drivers ? = new crossover = new speaker.

/sreten.