Originally Posted by chasw98
I agree with you about an accurate model not having been done (or not being publicly available yet). From what I have seen (not heard or measured) these woofers can go to 2K which would mean a target point between 500 and 1K is feasible. I learned on my previous 3 way project that a passive xover on the bottom is not feasible in todays economy. The copper and the losses in the huge inductors needed is not worth the money. You can get an adequate amp and electronic crossover for what you might put into parts for a passive. But a 400 to 800 Hz or so might have possibilities.
If and when I get some TD15's I will definitely post my measurements of the raw driver so that we can at least have a comparison to go by.
As most know, I prefer to throw "convention" out the window and go with what has scientific basis and agrees with how/what we hear. My point here was more that we are talking about 3 way designs, and while a 2 way passive crossover is a very managable task, many more things interact if you add a high pass on the midrange, and target a lower frequency crossover on a woofer.
The hurdles I'm hinting at are entirely separate from the component values. This is related to the low pass proximity and interaction with the LF corner of the woofer being used. I would say the impedance minimum just below the inductive rise is about the range where things get complicated. Complications are two fold. First you have the matter of getting a smooth response that is close to the target without peaking or other oddities in the response. Secondly, you have to keep the minimum impedance above an acceptable minimum. There are a few brute force means, which will involve lots of big and expensive components, but some departure from common targets in woofer alignment intended to beneficially combine with the passive crossover can make a 12 component solution turn into a 2-4 component solution. In effect this is a bandpass system involving electrical and acoustic filters, not just acoustic like some of the subwoofers I've been designing. Just as with bandpass designs, you can't take either the high or low pass in isolation and expect accurate results.
If you are after easy intergration and smooth on/off axis response, crossovers below 500Hz are required, desireable, and easily accomidated by midranges currently available. Jeff Bagby had noted on HT Guide that his free software would allow you to import the impedance of a measured or modeled woofer to design the crossover around. I suggest that some invest the time to attempt a 2nd, 3rd or 4th order crossover on the woofers they have in mind with the modeled impedance of the alignment they intend.
I have a handful of models for the 10, 12, & 15" TD drivers that will work nicely for very high sensitivty, vented bass alignments which will nicely cross in the 200-350Hz range. Low frequency extension is limited, but that's what subwoofers and kWs of power are for.
96-103dB @ 2.8V @1m is not all that hard, and can be had in anywhere from 60-260L depending on how low you really want to go. Since there is a lot of interest outside of the DIY world, I'm looking at such a configuration to allow a passive variation on my Catalyst, which John's drivers do a nice job of enabling.