Crossover inductor question

[activexp]

Hope someone can help me with this.

If a woofer has an inductance (Le) of say 1mh, do I reduce the calculated value of the crossover inductor accordingly?

e.g. Woofer Le = 1mh
Calculated value of low pass inductor = 3mh
Actual value of inductor = 2mh?

Or am I barking up the wrong street!?

Cheers,
Steve

[jrockhead]

Hi Steve, I've just completed a tweeter replacement in my Large Advents and have looked at all sorts of online crossover calculators.

Not once have I seen the Woofer crossover inductor value compensated by the Woofers actual inductance. Same with the Tweeter. So the short answer is no.

As an aside most all recommendations for inductors say to use air core inductors for the lowest harmonic distortion.

Marcus

[activexp]

Hi Marcus,

Thanks for the helpful reply. As you say, there's no mention of compensating for a driver's inductance on any of the online calculators or articles regarding crossover design so one has to assume that it can be ignored.

That's put my mind at rest

Regards,
Steve

[DcibeL]

FYI, speaker crossover "calculators" are completely useless for the most part.

[PLB]

Hi Steve,

In a circuit model of a loudspeaker, Le appears in series with the speakers resistance Re. It therefore has the effect of increasing the impedance of the speaker as frequency increases, which effectively unloads the crossover and changes its response. To overcome this, a Zobel network, comprising of a series resistor and capacitor, can be connected across the speakers terminals, thus flattening the impedance curve at high frequencies. So if you add a Zobel network, Le can be ignored and you can assume that the speakers impedance is purely resistive. I should also point out that the speaker will also have a low frequency resonance that may affect the theoretical response of the crossover, but this too can be corrected by an series LCR network, but you may find this is unnecessary in your application.
Hope this helps.

Best regards

Peter

[activexp]

Hi Peter,

That's very useful information. I had a feeling that Zobel networks fitted in somewhere. From now on I'll be using them in my projects.

What prompted me to ask the question in the first place is that I've got some cheap Skytronic 8" kevlar woofers with no quoted Le (all other T/S parameters are available). According to my LCR meter they have an inductance of 2.8mH. That strikes me as rather high compared to other drivers of similar size but they do have 4-layer voice coils, no doubt for high power handling.

Cheers,
Steve

[PLB]

Hi Steve,

After I finished work today I did a quick Internet search on Zobel networks and came up with this link.

http://www.trueaudio.com/st_zobel.htm

Have fun.

Best regards

Peter

[activexp]

Hi Peter,

Thanks for the link. I also found this one:

http://diyaudioprojects.com/Technical/Speaker-Zobel/

All useful stuff.

Regards,
Steve

[panoramamonk]

May I have some advice and help to understand:

what inductor and cap values should be used in association with

4.5" mid cone
6.0" mid cone
8.0" woofer
tweeter (just any tweeter)

There's no lack of xover calculators and articles floating around, but more I read more dumpish I feel I'm. Here I need answers to a simple example. Say, I have in a cabinet a 8" woofer and a 4.5" mid and a tweeter. From a table I found on website, 8" woofer works within 80-1250 Hz and 4.5" is 160-2500 Hz. For an inductor allowing the low frequency to pass thru the 8", which frequency should I take into the calculator (80 or 1250, the ceiling)? and same question applies to the 4.5", and tweeter which is just the opposite, a cap to allow the lowest frequency of its range?

I may be terribly wrong. I was thinking to use 1250 Hz for the inductor calclator so frequency from 80 - 1250 can pass thru and a cap to exit anything over 1250 point.

Thanking you in advance.

[Lynn Olson]

One of the evil things that happens when you simply double the woofer's inductance is you end up with a weird shelved-down response, instead of the desired early rolloff, like you'd expect.

What's going is the woofer has a peaked mechanical response that almost exactly compensates for the inbuilt voice-coil inductance. This is not an accident; driver engineers play around with cone materials and profiles so this happy result happens, yielding a nice flat frequency response that in a well-designed speaker goes right through the breakup region. (The main purpose of cone damping is smoothing - but not removing - the breakup region.)

The problem starts when you want to modify this response - to cross it over, in other words. It's not obvious you're really dealing with two systems, one electrical, one mechanical, and they've already been carefully balanced against each other by the driver engineer.

What the Zobel inductance corrector does is decouple these systems, so you can go ahead with your crossover, and pretend it's a resistive load in the crossover region. The Zobel is simple enough; the value of R is similar to the DC resistance of the driver, and the capacitor, to a first approximation, cancels most of the inductance. In practice, thanks to the fact that the voice-coil inductance doesn't quite behave like a real inductor, there are ripples in the impedance curve after correction. But you can get close enough.