Crossover math check

[jmikes]

I'm still getting a handle on reading schematics and doing the math required for crossovers and I have a question. I was looking at the Angelfire crossover on the PartsExpress project showcase page, and the builder states the tweeter is crossover at 5150. When my rusty brain clanked into gear and worked out the numbers, I got a answer more like 3200. Can someone take a look at it for me?

Thanks in advance.

[Cal Weldon]

I'm no expert on Xovers but I was wondering why the take off for the tweeter is after the mid cap. Is this a common practice?

Cal

edit: The mid and tweet are only a 6 dB while the woof is 12? Is this also on purpose?

[Bill Fitzpatrick]

It is common practice but put the 28mfd (most certainly electrolytic) in the tweeter line which some would consider very bad.

Even if you consider the series combination of the 2 caps, the crossover freq. disparity is probably due to the fact that the tweeters unfiltered response is also considered.

[jmikes]

Originally posted by Bill Fitzpatrick
It is common practice but put the 28mfd (most certainly electrolytic) in the tweeter line which some would consider very bad.

Not sure I understrand. Can you translate?

Even if you consider the series combination of the 2 caps, the crossover freq. disparity is probably due to the fact that the tweeters unfiltered response is also considered.

I'm lost again. The tweeter (Morel 37) shows good response well below both of the stated crossover points. Why should it perform so badly with that circuit?

[Cal Weldon]

You will note the signal has to go through the 28 cap before going to the 7.7 for the tweeter. Bill mentioned it was a common practice. I'm not sure why, it doesn't seem to make sence on the surface.

Nor do I understand using a bypass resistor on the woofer.

[planet10]

Originally posted by Cal Weldon
You will note the signal has to go through the 28 cap before going to the 7.7 for the tweeter. Bill mentioned it was a common practice

I've seen this quite often... nets you a 6 uF cap on the tweeters (and no doubt means that the 7 uF cap has to handle a lot less power.

If you are calculating a text-book XO using an 8R loas you have to realize that the real impedance of the tweeter actually has little resemblance to an 8R resistor so the text book formulas are at most an optimistic guideline.

dave

[Bill Fitzpatrick]

It's not a bypass resistor on the woofer, it's a zobel.

And now that I look at PEs project page I see that the 28mfd is a poly cap so that negates my comment about it being a bad idea.

So, I'd say that they did all that they did based on parts on hand and the need to do some response tailoring. The exact crossover freq. is often a nebulous number.