JohnK has posted spreadsheet for Thiele's CE filters

[capslock]

... and seems to have grasped the difference between these target functions and Jon's somewhat more efficient approach. He said he'd include calculators for Jon's filters over the weekend.

http://www.madisound.com/cgi-bin/discuss.cgi?read=365892

[john k...]

[QUOTE=capslock]... and seems to have grasped the difference between these target functions and Jon's somewhat more efficient approach. He said he'd include calculators for Jon's filters over the weekend.

(JPK) Not exactly what I said, :-) I will be posting spreadsheets that compute targets based on Thiele's 4th, 5th and 6th order alignments. These are all imlpimented, electrically, with topologies somewhat different than Jon's. However, the spreadsheets can be used to generate targets with initial roll offs close to LR6, B7, and LR8, respectively. It is surprising how close Jon's simple 4th order LP topology can come to the 6th order alignment and when the natural driver roll off is considered it would appear that it would be possibe to acousticly match the Thiele 6th order roll off with Jon's topology very closely. The primary difference, on the electrical side is that, for the LP filter, above the notch the true 6th order alignment rolls off at 24dB/oct where as the asymptotic roll off of Jon's 4th order alignment is 18dB/oct, with accordingly less phase rotation (Thiele's 4th order asymptotic roll off is 12). As you can see, adding is the typical natural roll off af a driver can bring things nicely into alignment. So I belive that acoustically a 4th order electricaly topology such as Jon's (or Thiele's, depending on the driver characteristics) is well suited for achieving the 6th order alignment.

The spreadsheets compute complimentary HP/LP pairs whcih can be exported as FRD files and then used/imported as desired. The HP/LP pairs sum flat. Input paraments are x-o frequency and the ratio of the LP notch frequency to the x-o frequency. Graphs of amplitude, phase and group delay (of the summed response) are presented.

They can be downloaded at https://www.musicanddesign.com/codes...otched1.00.zip

[TacoD]

Very nice work JohnK, I am also experimenting with Jon's filters. In my experience 6th order acoustically is easy to obtain. The phase behavior in the x-over region depends on the driver layout. I had both cases, one case where I had to invert the tweeter polarity and also cases where everything was in the same polarity. Maybe this is due to the fact I did not always got "true" 6th order acoustical behavior (combination of driver type and filter).

[john k...]

If you are having a problem...

(JPK) A couple of guys over at MaD were having problems running the spreadsheets. If you encounter a problem it is likely that the Analysis ToolPak - VBA is not loaded. Go to the Excel Tools menu and click the Add Ins item and when the window open make sure Analysis Tool Pac -VBA is checked. You may have to exit excel and reopen it to make the tool pac load.