Please review the site Rules, Terms of Service, and Privacy Policy at your convenience. Rules, TOS, Privacy
Get familiar with the reaction system: Introducing the Reaction System

The minimum phase panacea of 1st order crossovers...

So when we get past electrical characteristics and deal also with the mechanical output of the actual drivers, we find that the ‘text-book’ electrical filter designed coil and/or cap often need to be adjusted to get the net amplitude output of both drivers to sum flat.

OK, So to what extent does having a less than ‘text-book’ electrical filter mess with phase unity of a 1st order crossover network at the speakers’ output?  

As, my underlying question (assuming drivers in this example can handle additional loads put on by a 1st order network) is; is this adjusted less than ‘ideal’ 1st order still better on a relative basis as compared to the phase issues (and associated off-axis peaks and troughs…) associated with higher order filters?



  • edited March 15
    Unless your frequency response and impedance looks like this, :

    And you have no acoustic offsets between drivers, you can be assured that you will never have a perfect 1st order response from a simple cap and coil. More often than not, the natural roll off of the woofer and tweeter get you closer to a 2nd order acoustic response with a simple 1st order electrical filter, with some lumps where there is impedance peaks in the tweeter impedance.

    Minimum phase is directly correlated to the frequency response, so the key is to focus solely on the acoustic frequency response, and use whatever necessary electrical filter necessary to get there.
    I'm not deaf, I'm just not listening.
  • So advantages or no, with Thiel and Vandersteen designs? 

  • I heard a set of 'steens and thought they weren't all that. The Thiel stuff sounded good.

    In terms of 1st order, you have other criteria to do first. You have to compensate for a lot in conjugate/shunt networks in order for 1st order to work properly most of the time.

    As an example:
  • thx for the link, I'll take a look
  • tajanes said:

    So advantages or no, with Thiel and Vandersteen designs? 

    I would focus more on the overall power response or directivity index of a speaker. The filter order ultimately does affect the power response, but there are other factors there as well like relative driver size, distance between drivers, horn/waveguide loading, and a lot of drivers simply don't have the wide bandwidth smooth low distortion performance to allow for a 1st order filter to be the best choice for the job.
    I'm not deaf, I'm just not listening.
  • edited March 15

    All true.

    So my quest to keep the affects of a crossover to the minimum, and what can’t be avoided out of the highest hearing sensitivity range (1kHz to 5Khz) have me looking at broad ranges for extended mid coverage (to cross with a tweeter), and/or coax’s for their minimization of driver offset, and for this project, to take a look at what a 1st order may (or not) bring to the table.

    While there has been some discussion here on affects of age on our hearing, high frequency perception or not, I’d suggest our ability to hear phase inconsistencies between drivers especially at upper frequency ranges (causing both constructive and deconstructive interference, exasperated in part by driver separation) is something worth ‘hobbying’.    

    And as such I really appreciate the opportunity to post at this forum, as I’ve found great advice given based upon real experience.  

  • The primary advantages of first order filters to me is power response which closely tracks the on-axis response, low Q factor of the filter, and ease of voicing.  The ease of voicing is very important.  I find that I frequently end up with around 45 degrees of simulated phase offset between drivers, it doesn't seem to be a problem though.  The group delay of a first order system is usually pretty low, I think that helps also.
  • With my beyma 5in coaxials I'm moving away from 1st order (although it sounded nice) biz a) the tweeter has a bump up in its lower-end not sufficiently attenuated by the 1st order HP filter, and b) not wanting to find out how robust the tweeter actually is on these puppies ($$$).  

    I've been working on a 2nd order LP in combo with a 3rd order HP.  Using the LP to clean up what is the natural roll off of the mid driver, and then bringing in the tweeter where I find a good fit.  Not my intended 1st order, but I can get the crossover point out of the 1-5Khz range.
Sign In or Register to comment.