Lecleach paper on crossovers

[swak]

I have read his very good presentation and I have been asking myself if there are not other variables afecting phase response (I am almost sure I am missing something...). When I model an enclosure I can see phase shift, drivers alone as BP devices induce shifts, op-amps, output transformers, etc. Now I think the spreadsheet asumes all these as linear phase devices, but to my surprise the measured responses he shows very good phase tesponse troughout the freq range.

I would appreciate if someone could help me with my difficulty.

[planet10]

Link to paper please?

dave

[sreten]

Document and Speadsheet (doc, xls): http://freerider.dyndns.org/anlage/LeCleach1.zip
Theory of operation (ppt): http://freerider.dyndns.org/anlage/LeCleach2.zip


/sreten.

[sreten]

Originally posted by swak

I would appreciate if someone could help me with my difficulty.

Hi,

As far as I can tell it represents YATFF (yet another theoretical filter
function) and as such it would be your final acoustic target, or put
another way the drivers / enclosures are idealised to be perfect.

/sreten.

[swak]

Serten, that is what I thought. What has been puzzling me are the very nice MEASURED results he gets with those horn systems (as shown in his presentation).

[sreten]

Hi,

The spectograms seem to be derived from the above impulse response.

The second is certainly well aligned, where it comes from I'm not sure.

/sreten.

[Jmmlc]

Hello Swak,

on the document downloadable at:
http://freerider.dyndns.org/anlage/LeCleach.htm

On page 52: this is a pulse response I measured on a two ways system using a Le Cléac'h crossover (the bass enclosure is a bass reflex quasi 2nd order using a low tuning frequency, its response is nearly the one of a closed enclosure, low group delay below 70Hz).

Page 68 of the same document: this is the pulse response measured on the system I brought at ETF2004. THe measurement was done the day before the conference I gave on "phase distortion".

You can see others pulse measurements and curves derived from pulse measurements on the document:

http://www.melaudia.net/zdoc/distorsion_de_phase.pdf

page 78: this is the result of the sommation performed with a simple resistor sommator at the outputs of a DCX2496 digital crossovers. Different classical crossovers are compared to the Le Cléac'h crossover, only the first order Butterworth (LP and HP in phase) is a bit better than the Le Cléac'h crossover .

page 82 and 83 using the spectrogram to align a 2 ways system using a Le Cleac'h crossover

Notice: page 46 group delay curve (delay T is expressed in equivalent length d calculated by d = T . c with c = celerity of sound) for Marco Henry's Jerzual horn (reference J321). If you use the horn above 750Hz the delay variation is less than 6 centimeters (expressed in equivalent length). If you use a high pass frequency lower than 750hZ then there will be some difference with the simulated response obtained using my excel spreadsheet. Same thing happens if you use a too low frequency for the low pass of a bass driver (when the group delay rises).

Best regards from Paris, France

Jean-Michel Le Cléac'h

Originally posted by swak
I have read his very good presentation and I have been asking myself if there are not other variables afecting phase response (I am almost sure I am missing something...). When I model an enclosure I can see phase shift, drivers alone as BP devices induce shifts, op-amps, output transformers, etc. Now I think the spreadsheet asumes all these as linear phase devices, but to my surprise the measured responses he shows very good phase tesponse troughout the freq range.

I would appreciate if someone could help me with my difficulty.