How to modify this x-over ??

[Tob Toon]

Hi there !!



This crossover is from a Kef 103.2
I made the drawing, but left out the S-Stop protection circuit,
because I am planning to bypass it.

The used units are:
Tweeter T33A SP1074
Mid/woofer B200G SP1075

* Is it possible to remove those 2x 300mfd and the 680ohm resistor without negative side effects ?

* Why is the tweeter out of phase ?

*Are there any other modifications possible, in order to get this "old" x-over (1985) in a up to date version.


Thanks for helping me !!

Toni.

[Tob Toon]

Update: to make it more clearly,
I have drawn the position of the relais (S-Stop) into it.

Toni

[Ron E]

If you are not using the same drivers (model # and manufacturer) You probably shouldn't use this crossover.

There's zobels and notch filters and contour circuits in there. A crossover is not some "magic recipe" that works with any driver combination, it is a design driven by the drivers used and their unique response characteristics.

Whether you have the identical drivers or not, modeling the response in a simulator with measured data is the only way you will know whether this circuit will work or not. It will also show you why the tweeter is inverted.

Try Speaker Workshop. I think it may even have a Dutch language add-in.

[Tob Toon]

Hi Ron,

I am sorry, I think I was not clear enough. My mistake.

What I mend is, I AM using this speaker, the Kef 103.2,
which consists from Tweeter T33A (SP1074) and
Mid/woofer B200G (SP1075)

And I don't want to use any other drivers, I just want you guy's to have a critical look at this x-over filter, and come up with idea's in order to get this "old" x-over (1985) in a up to date version.
(it is off course quite possible that this design has no flaws)

From my (very) limited knowledge about x-over's, I have to following question: Is it possible to remove those 2x 300uF and the 680 Ohm resistor without negative side effects ?
What will happen, when I remove those 2x 300uF ?

Ron, You say: There's zobels and notch filters and contour circuits in there.
Could You Please tell me with parts are doing what ?
In order for me, to understand and learn.

Thanks !

[planet10]

Originally posted by Tob Toon
From my (very) limited knowledge about x-over's, I have to following question: Is it possible to remove those 2x 300uF and the 680 Ohm resistor without negative side effects ?
What will happen, when I remove those 2x 300uF ?

The 600 uF and the relay bypassed resistor are there for safety. I would have no hesitation removing them. IIRC the 103 was a sealed system... are yours as well?

dave

[Tob Toon]

Yep, it's a sealed 2way speaker.

I am planning on removing the S-Stop protection circuit (not completely drawn on the attached picture), and the
680 Ohm resistor anyway.

But that leaves me with those 2x 300uF .

Some people say it's a form of DC protection for the woofer,
and others say, it's a trick to let the woofer have a lower -3db point, when using a smaller cabinet.

My idea is, to remove those 2x 300uF, and replace the original foam damping with BAF (polyfill).

So in that case, the closed box is too small when those 2x 300uF are removed, but by using BAF (which will let the woofer "see" + 15% more air space) or wool it's compensated.

Good thinking ?

[planet10]

Originally posted by Tob Toon
My idea is, to remove those 2x 300uF, and replace the original foam damping with BAF (polyfill).

The 600 uF cap is a 30-35 Hz high pass filter... it could be bringing down a resonant peak from too small a box. But it has to be a bipolar electrolytic. I'd take it out, stuff it with wool and if it is too peaky on the bottom, make the enclosure aperiodic.

dave

[SY]

I'd model it before pulling out the caps (or making any other modifications). The capacitors will certainly interact with the impedance of the woofer, which goes through some complex gyrations in the lowest several octaves. It is unlikely to be a simple high-pass filter, but is more likely used to shape the LF response, not unlike the Clark-Geddes alignments.

Measure the drivers' complex impedances and frequency responses, then put these and the crossover circuits into a good loudspeaker CAD program. That way, you can see what any proposed changes will do to frequency response, impedance, driver excursion, and power handling. You'll also see why the tweeter and woofer are connected in opposite polarity.

[Tob Toon]

Hi Dave,

It's indeed a bipolar electrolytic.

You know what....I removed the protection circuit this afternoon,
and bypassed the 600 uF cap.

What happens is, the -3db point is a little higher, but the definition and information in the lower region has increased.

Originally Kef used very thick foam as damping. (all sides)
When used, it dampes to much (to my taste that is), but has a really good punch.

Using BAF with the 600 uF cap in place sounds fine, less damping,
more definition.

But when using BAF without the 600 uF cap, it lacks punch.

So what I did, was I used Pritex (thinner special foam) on all sides, and now I have it all, definition (because of the removal of the 600 uF cap) and punch, because of using some kind of foam.

Any more idea's ?


Toni

[planet10]

I figured the removal of the big elco would be well worth the price, but you seem to have found a good fix for the downside... you also might want to try thick, dense wool-felt in place of the thin foam.

A lot of people have used foam just because it streamlines the production process.

dave