Idea for "Balanced" Passive crossover- opinions?

[clm811]

A while back I asked a knowledgeable friend about placing crossover components in the negative("ground") leg of a passive crossover, and he said it "didn't matter", as the positive and negative legs of the filter are the same in a AC circuit (Duh!).

On the drive home I had (too much) coffee, and got to thinking:

In active electronics and cables, much favor seems to be given to balanced circuits (positive and negative circuit legs being treated in a "symmetrical" way)...

[clm811]

WHAT IF...

One were to make a passive crossover (e.g. third-order) and put the second series cap (coil) in the negative(return) leg of the crossover? I drew a sketch on a napkin, and became intrigued. I found nothing similar searching both here and on other fora.

I figure that this places the resistance (reactance) of the loudspeaker voice coil in-between the caps(coils) of the crossover circuit, rather than after them; Consequently(I'm assuming), this may affect the "Q" or "Zeta" of the crossover circuit...

Any opinions on how workable this circuit would be, and what (if any) effect the driver voice coil resistance(reactance) would have on the circuit “Q” (“Zeta”)?

Thanks in advance for indulging (humoring) me!

-Chas

p.s. Note how this differs from the "classic" balanced circuits(fig.3,4), as shown on wikipedia:

File:Balance and symmetry.svg - Wikipedia, the free encyclopedia

[clm811]

OK.

Maybe I'm a little off-base here...

I just took a look at flipping the circuit diagram vertically, and it doesn't look as good as I first thought.

Maybe the "balanced/symmetrical" circuit (fig.4 in above wikipedia link) is
a better illustration of what I'm interested in doing (?).

I know that this takes almost twice as many components to realize; However, I'm sincerely interested in how such a crossover would perform...

Anyone care to model it and critique the result here? (Or think it's a waste of effort?)

Thanks again for indulging my whim!

-Chas

[bjorno]

Hi clm811,

Whim is allowed as long as the XO functionallity remains the same..

b

[bjorno]

Originally Posted by bjorno

Hi clm811,

Whim is allowed as long as the XO functionallity remains the same..

b

..and the correct headlines.

b

[Pano]

Thanks for the examples Bjorno.

This would work fine, but would it be any better? I can't see that it would. It would cost more, tho....

[clm811]

...It would cost more, tho....

My original idea (3rd order) still uses the same number of components; Consequently (provided placing the V.C. resistance/reactance within the filter doesn't hurt the response), the parts count/cost would remain the same(Why I was excited enough to risk being bashed here on this forum)!

Thanks, Bjorn, for the ecouraging response!

Now if we could just model it to check for any flaws...

I'm serious about building (actually, modifying my existing 3rd order crossover) this, provided there's no (un)forseen problems. I'm hoping that it wouldn't be a waste of time. That's where this forum comes in handy!

-Chas

[clm811]

Hoping someone here with CAD software is intrigued enough to model it,
here is the crossover with actual component values:

[bjorno]

Originally Posted by clm811

Now if we could just model it to check for any flaws...

-Chas

Is this good enough?:

b

[clm811]

Wow!

Thanks, Bjorn.

I guess I'll get to work on an extra set of crossovers!

Then comes the listening comparison, to determine if there is any audible difference...

-Chas