zobel circuits and series notch filters for active networks?

[pwnt by pat 10+ year member]

I've been doing some reading and zobel filters has caught my attention.

A little intro:

In normal crossovers, caps, resistors, and inductors are used to filter out unwanted frequencies. However, a woofer by itself is an inductor. Due to the nature of inductors, the higher the application frequency, the higher the resistance.

This is where zobel circuit come in to play. The zobel network uses a resistor and capacitor based on the woofers re and le values.

On the other end of the spectrum, the large impedeance peak apparent at the speakers fs also causes crossovers to be inaccurate and less effective. This is supplimented with a series notch filter, also using a cap, resistor, and adding an inductor.

Now, in active circuits, the crossover happens prior to amplification. In doing so, any post-amplification crossover is eliminated, thus the removal of the zobel circuit and series notch filters.

The benefits of both circuits is obvious, as the natural resistance of the woofer can be almost completely elminated, allowing for more efficient use of the amplifiers power.

Now to my questions:

1) In crossover construction, the notch filters are applied after the crossover circuit (further from the source). Would the zobel circuit be closer or further away from the source than the series notch filter? My assumption would be as close to the source as possible?

2) Is it possible to apply a zobel circuit and series notch filter to a driver post-amplification using an active application while having no or minimal adverse effect to the post-amplification crossover?

edit: quick links:

http://www.carstereo.com/help/Articles.cfm?id=36

http://www.carstereo.com/help/Articles.cfm?id=23

Click to read more...

 

[Beat_Dominator 5,000+ posts]

Technically the Zoebel needs to become "the driver" as far as the amp is concerned. That means the zoebel needs to be attached at the midrange drivers.

 

[pwnt by pat 10+ year member]

As far as making it become the driver, how's within several inches? Surely the added resistance and capaciatance from a couple inches of wire would fall in to the "good enough" category of physics.

But as far as you're concerned, the zobel can be included unconditionally since the amplifier will see a consistand load?

 

[Beat_Dominator 5,000+ posts]

It can be as far from the driver as you want but it's the first thing after the terminals, then you factor it into the specs for a passive x-over. But if you're using it active that wouldn't be an issue.

The amplifier would still see an impedence spike at the resonant freq. of the driver, but there would be little impedence rise as you increased in freq.

 

[helotaxi 5,000+ posts]

A zobel filter is used with a passive crossover to keep the impedance spike at resonance from messing up the crossover slope and point of the tweeter since the slope and corner freq of the crossover is determined by the impedance of the driver. Since the resonant freq of a tweeter is going to be well below the crossover point and the point and slope of the active filter is not affected by the electrical properties of the driver, a zobel filter is not needed when running active.

 

[Beat_Dominator 5,000+ posts]

Since the resonant freq of a tweeter is going to be well below the crossover point and the point and slope of the active filter is not affected by the electrical properties of the driver, a zobel filter is not needed when running active.

Good point, but wouldn't the driver still have imp. rise across the frequency range? Eliminating that can save you some Eq'ing I would think.

 

[helotaxi 5,000+ posts]

Depending on the driver it might have a gentle rise. This is basically the effect of the Le of the driver and equates to the high end rolloff that all speakers exhibit. It won't be noticeable on most tweeters and EQ won't fix it anyway.

 

[pwnt by pat 10+ year member]

Helotaxi, I think you just confirmed what I was thinking after reading some more papers on the subject.

That is:

since the crossover happens before the impedeance rise of the speaker is takin into consideration, the position and slope of the crossover is uneffected. The zobel filters doesn't change the actual impedeance rise of the driver, but rather, masks the rise. The woofer will still exhibit the characteristics that it would have in an open circuit, just the filters would "trick" the passive "crossover". The notch filter at fs does the same thing.

I could be wrong and that "tricking" the amp in an active application would actually change the amount of power applied... or would that extra power just be dissipated in the zobel network.

 

[helotaxi 5,000+ posts]

The power would be dissapated. IIRC the impedance is raised across the frequency spectrum so the impedance at the lowest point is basically the same as the impedance after the rise. It doesn't lower the rise is raises the imp over the rest of the range to the level of the rise and robs the whole setup of power. Really a waste of power in an active system where the taming of the resonance isn't needed anyway.