Crossover Slope Question

[69stingray]

Hello everyone. I have what I hope is a simple question. I have not been able to find the answer in an article on the web.

When picking a crossover point and slope, how far “down” should the signal be at the speaker’s frequency limit? What is the minimum?

For example, it you have a midrange driver that has a lower limit of 100 Hz and a slope of -6dB/Octave, how many Octaves should you back off from 100 Hz?

[Cal Weldon]

This is difficult to answer because of the nature of the question but:

A simple rule of thumb is to cross it over no lower than twice the resonant frequency. Same with tweeters. The higher, the slope, the better the power handling and less overlap so a good place to start is 12dB.

If you are just starting out, this is as simple as the explanation can get. It also leaves out a mitt full of factors so don't take these to heart. There are so many things that determine what frequency/slope you cross it over at.

Cal

[qwad]

spot on T.C.

[69stingray]

Thanks for the quick responds! I understand that this is a very difficult question that I am trying to get a simple answer to.

Looking at it in very similar terms, is there a rule of thumb such for how many "-dB" the signal should be at the drivers frequency limit (in dependant of slope)?

For example, let say I have speaker X, whose frequency response is 100 Hz to 100,000 Hz and for the purpose of this question, I have a crossover with -6db/slope (the -6db has a better demonstrative purpose).

At 100 Hz, should the speaker signal be down:

-6 db (200 Hz crossover)
-12 db (400 Hz crossover)
-18 db (800 Hz crossover)
-24 db (1600 Hz crossover)
-30 db (3200 Hz crossover)
-36 dB (6400 Hz crossover)

[Cal Weldon]

Zach,

Answer #1: You know there really isn't an answer to your question. You're designing a crossover by going through the back door. What you have to do is pick two drivers that you think will go well together. Start with the woof and mid. Have a look at their graphs and make your decision about the XO point by analyzing the strong and weak points and at what point you believe one should take over from the other. That's your XO point.

Answer #2: 500 Hz with 12dB/oct XO

Cal

[mike.e]

Electrical filter + acoustic filter = final response

Fc ::
lower Fc depends on displacement requirements while upper F3 of driver depends on dispersion* and presence of cone modes.

see
http://www.woodartistry.com/linkwitzlab/
and in particular
http://www.woodartistry.com/linkwitzlab/x-sb80-3wy.htm

*when wavelength produced approaches cone diameter the cone will have a narrow dispersion approaching 90°

Its easy to produce a flat on axis response but flat off axis is another matter,crossing over to a tweeter which is then largely omnidirectional mucks up the reverberant energy received at listening position. the off axis response must be smooth also unless you listen outdoors with no boundarys or anechoicly.

[69stingray]

Mike-

Thanks for the "easy" answer That is some heavy reading, it will give me something to do during lunch tomorrow.

The "2 octaves" over the the F3 point with a -12db/octave seems to very common (ie. -24db) standard.

[mike.e]

Doing an average passive crossover is easy-Calculate values for non reality impedence values that you havent measured,giving cutoffs not where you specify,then the circuit can act as a resonant circuit with the driver causing funny response shapes...

or learn heaps and learn how to do moderalteywell!

4th order Linkwitz riley has a good place thesedays
odd order shifts the acoustic axis at Fc,while even order have their own niggles.