Export minimum phase as .txt-file for passive crossover configuration

[buzzdee]

I'd like to start my first thread thanking all developers and supporters of REW for their great job. I started using it for DIY speaker projects (and only them as of yet) and am fascinated by the functional range of this freeware, but got to a point where I couldn't help myself and didn't find any answers online neither. I must admit that I am not too familiar with using REW (yet), so please pardon if my question is just to obvious to be asked in any other thread (or I didn't find the answer in another one).

I am trying to configure a passive 3-way frequency crossover for a new DIY speaker with LSPCad. For simulating the frequency response, I need a .txt-file with 3 columns frq - SPL - (min) phase and frq - mag - imp. The impedance measurement I conducted with a Klippel-system, the SPL using REW and the UMIK.

I ended up with 3 measurements each for woofer, midrange and tweeter. When trying to export the frq - SPL - (min) phase as .txt I realised tat I couldn't export the min phase data. I tried the workaround (export min phase as .wav - import impedance - export .txt) but the phase graph does not match with the original min phase. When creating a min phase version of the imported impedance, I do get the original graph though. If its of any help, I could add some screenshots later.

Q: Is their any way to export the required .txt file with the original min phase data in 'frq - SPl - (min) phase' format?
I am using REW v.5.19

Thank you in advance.

 

[John Mulcahy]

I wonder whether minimum phase data is really what is required, rather than data with time delays removed. The latter can be generated by using the Estimate IR delay feature and removing the estimated delay. That leaves the actual phase response of the drivers, which would seem more relevant than the phase response of a minimum phase system which isn't the drivers but just has the same magnitude response.

 

[buzzdee]

Hi John,

Obviously, I don’t have a thorough understanding of the difference between minimum phase and delayed phase, despite the fact that there are lots of websites explaining it (e.g. https://www.roomeqwizard.com/help/help_en-GB/html/minimumphase.html). I just tried to match the required input from LSPCad with the REW function that seemed most fitting. Reconfirming with someone else who knows the difference, your requested solution is exactly what I need for the crossover configuration.

Thank you very much for your immediate help

 

[Lex_audio]

I wonder whether minimum phase data is really what is required, rather than data with time delays removed. The latter can be generated by using the Estimate IR delay feature and removing the estimated delay. That leaves the actual phase response of the drivers, which would seem more relevant than the phase response of a minimum phase system which isn't the drivers but just has the same magnitude response.

***Sorry John! Isn't that the same thing? It always seemed to me that if we remove the delay and move the beginning of the impulse response to time zero, then we get the minimum phase! Is not it? What am I wrong about?

 

[John Mulcahy]

***Sorry John! Isn't that the same thing? It always seemed to me that if we remove the delay and move the beginning of the impulse response to time zero, then we get the minimum phase! Is not it? What am I wrong about?

Room responses are typically not minimum phase even with time delays removed.

 

[Lex_audio]

***Thanks for the quick response! I read! Some of it is not clear to me yet, but one of the indicators of the minimum phase is group delay. Right? On the graph, we can observe the frequency regions of NOT the minimum phase. Right? However, group delay is always present, Not big but present! How to determine that, for example, a given level is large or small, and if it is large, then this area is NOT the minimum phase !? Is there a border?

 

[John Mulcahy]

You need to look at excess group delay. Anywhere the excess group delay plot is (fairly) flat is a minimum phase region of the response.

 

[Lex_audio]

***I saw it! Cool! I took another small step up to understanding the measurement results! Thank you very much!
: