Overlapping of crossover frequency

[Aucosticraft]

I am planing to overlap cross over frequencies. ie mid cutoff at 5Khz and tweeter at 4 Khz. I would be using first order BPF for midrange and second order HPF for Tweeter. My tweeter is fabric dome with ferrofluid. also I am adding L-Pad attenuator for tweeter. and all the drivers are in phase with each other.

I decided the above method after actually summing the response of drivers ( from graph), network and Lpad attenuator. The overall response is close to flat.

Now I request you all to suggest me if there is any hazel in doing this or any spl care is to be taken.

I imagine that when two crossover frequencies are overlapped there could possibly be addition of some distortion ( adding or cancelation due to phase differences ) because of different transient response of drivers, as well of filter used for each driver.
but this is just what I imagine pls correct me.

[Inductor]

There's no golden rule, also for generics (no-name) driver-parts (if that's the case or not). Each one (driver/speaker/xover) is a special case. The same thing applies for simulations. Unless you give specs.

[gedlee]

If you can make it work then thats fine, but with two drivers operating in the same frequency band, there will be lots of polar effects, which is not usually a good thing.

[sreten]

Hi,

What you are saying does not make much sense.

After applying the c/o components each driver has an acoustic response
and it is this, not the crossover components, that determines the filters
(order, frequency and type) that are actually being used.

AFAIK no overlapping topology sums to a flat response, if it does
sum to flat, then it is not acoustically an overlapping topology.

/sreten.

[gedlee]

Sreten - it is possible to do, it would just not be adviseable so its never done. The sum could be flat along one axis, and peak up either side of that. If the two drivers were in quadrature (90 dgrees from each other), for example. NO crossover configuration is flat everywhere!

[tinitus]

Originally Posted by Aucosticraft

The overall response is close to flat.

Then you may actually have a filter with perfectly matching slopes, on the acoustic side
What you think it does on the component side is a different matter
And what you think is 1.order could easily result in 12db, but probably something different, like 8 or 10db, or maybe even 15db
But if mid driver rolls off naturally around there, it could also be 18 or even 24db, or something in between
And probably at a different frequency than you expect

hey, just a teaser

[Aucosticraft]

well guys. i ll share datasheet pls suggest me the design.cause I used 3Khz as crossover. the response was not linear. as well while listening i could feel mid vocal was taking back seat. also hi frequencies were not enjoyable. also the higher-mid that appeared from tweeter were not smooth.

3FULLRANGE8E10WSH.pdf - 4shared.com - document sharing - download
D21FNRDFP.pdf - 4shared.com - document sharing - download

datasheet for tweeter and midrange .

[LineArray]

Some options, which may be tested even in connection:

- Try padding tweeter down using an L-Pad
- Try even lower crossover frequency (2.3 Khz)
(might be worse due to distorsion of tweeter, but worth a try)
- Try higher crossover >6 Khz
- Try compensating fullrangers voice coil inductance before applying filters

[dannyjmf]

I advise you better not to do so. The summed output you will get will not be flat. In other words, you would not get flat response from your midrange + tweeter. (Serious, I tried it before). This also happen (in biamping) when your midrange gain is not equal to your tweeter amp gain.

Never mind, you still can give it a try. If you like the sound then stick to this config or just revert to your old config. Good luck!!

[Aucosticraft]

ok well then any words on using second order filter for tweeter and first order for midrange. I used second order for tweeter to avoid it from damaging.