QuestionI've built many active systems and want to try my hand at a passive one. I've decided on a 3 way 2nd order system with linkwitz-riley alignment. This seems to be pretty close to a compromise of all pros and cons of passive crossovers. not too complicated, not to many parts, good compromise of slopes vs musicality vs phase problems etc.
one thing I'm seeing is that I really should use a zobel circuit on my woofer and mid to help the lowpass sections perform properly (correct me if I'm wrong). my problem is this: the midrange that I want to use is not rated for its inductance so I don't know what capacitor to use in the zobel. my woofer does come with a rated inductance. SO... I purchased a meter designed for reading inductance. however, when I hooked the meter to the woofer it read significantly higher inductance that what the spec sheet showed (spec was .45 mh, meter read 1.2 mh). so my questions are this. is there a reason the rated mh and the metered mh are so different? does inductance vary with frequency? if so, what number do I go by? the rated or the metered? if I cant go by the meter, what do I do for my midrange?
and if there's anything else about what I said above about passive crossovers in general you can give input for PLEASE DO!!! my next step was to buy the inductors for the crossover and measure them with the meter just to see if the meter is off. however the meter does read capacitance as well and I had a bunch of caps that I tested and the meter was spot on with them....
THANK YOU!!!!
AnswerInductance on voice coils is quite variable. It will vary with drive level, frequency and the magnetic field in the coil area.
Inductance meters are not all that accurate either, where there is iron and magnetic fields involved. They are fine for air core inductors. In fact, air core inductors will usually measure as accurately as do capacitors because much of the variable effects are out of the picture.
The difference in H readings from the manufacturers specs are likely mostly due to the manufacturers rating being off. They might have changed something about the voice coil and not updated the specs. I would go with the measured result - it is likely to be a little closer.
More Comments:
For voicing out the system the SPL level of the midrange is super critical; more critical than the crossover frequency. In listen testing the tonal balance the human ear can easily perceive a difference of only .5db between the mid and bass and tweeter.
In the factory, acoustic and loudspeaker engineers spend hours and hours changing the components in the crossover and slight adjustments in the voice coil until they get the right voicing. It is all done impirically - after the math is cranked through.
So, expect the basic design of the crossover to be only the objective part of the work; plan on spending a few hours tweaking the parts for the right sound.
Back to the mid-range. Get the sensitivity as closely matched as you can to the high end. If the tweeters are more efficient than both the woofers and mids you can always tone them down, but you can't boost them up. So, mid-ranges should be higher in sensitivity than the woofers and the tweeters should be highest of them all. Does that make sense?
Personal comment: I am not much for exotic crossover designs. They are mostly a math exercise and not terribly influential in the overall sound performance of speaker systems. Why? Because the performance of the drivers are so far removed from ideal electronic performance the filter designs are almost totally mitigated by their erratic behaviors. The solution is to use drivers that have smooth performance around the crossover points you are going to be using. OR, use crossover frequencies that are well outside the range where the driver starts getting ratty. The real result for home grown speakers is in the voicing procedures. If you run a factory where mass production is going on the needs are all together different; there consistency is important and repeatability an essential. But for custom design the most important element is fundamental performance.
Best wishes,
Cleggsan
