REW Beta Release export of measurements for simulation in the crossover

[karoles75]

Hello
Sorry for my english but i need to use google translator.
I need help exporting measurement files of gated loudspeakers in the final enclosure for use in a crossover simulator, but I don't know how to do it correctly. Could anyone explain me how to do it step by step. I would like to add that I use a loopback as a reference in measurements and my sample measurement file looks like below

 

[John Mulcahy]

Looks like you need a timing offset to allow for the distance between mic and speaker, you can use the Offset t=0 option in the graph controls to apply that and get a more meaningful phase response.

What you export depends on what your crossover simulator needs, most common would be File -> Export -> Export measurement as text.

 

[karoles75]

John, thank you for your answer.
I was wondering a bit and I would do this:
In the picture, the phase waveform after removing the delay by the value resulting from the feedback loop, and the minimum phase diagram. As you can see, they differ from each other. Okay I think now I have the actual phase waveform?

 

[John Mulcahy]

Looks like the delay figure is a tiny bit too large, would expect some phase lag at the top end more in accordance with the roll-off (and so closer to the min phase plot there).

 

[karoles75]

Or maybe you have an idea what this could be the result of? Maybe I set something wrong in the program?

 

[sm52]

How did you create a loopback? Like in post #7 or like I described in post #8? https://www.avnirvana.com/threads/rew-loopback-connection-setup.9967/post-75966

 

[karoles75]

sorry but i need to remember ...

 

[sm52]

Do the wires on the right connect together like a Y-cable and go to the amplifier?
Input 2 can be instrument or line input. In the picture, the switch makes this input instrumental. It's not good for measuring. It is necessary to switch to linear or use input 1 to enter.

 

[karoles75]

I think so

 

[karoles75]

Or maybe for the loop I need the output that goes to the amplifier?

 

[sm52]

As you showed in red - yes. Do so. Or use input 1, which is on the left.
As shown in blue, I do not advise. Although it should also work well. But it's better to make a Y-cable that goes from the input of the sound card to the output (for example, R) and from the R output goes to the amplifier.

 

[karoles75]

The easiest way, I think, will be to use the reference: t = 0 at IR peak. I am only wondering about this option in Xsim:
Does anyone know what delay it removes and should I use it when simulating a crossover?

 

[sm52]

karoles75
Has the latency that John wrote about changed?

 

[karoles75]

I think I figured out where it came from. Before generating the minimum phase, I set a very short gate after pulse as I had poor measuring conditions which resulted in a false phase picture. My conclusion is this: too short a gate causes the phase to be displayed incorrectly. It is smoother but not appropriate.
I haven't figured out yet why the delay given in the chart description is not appropriate and using it causes the phase chart to rise too much on the right-hand side. Unfortunately, now I do not have much time and the possibility of making accurate measurements and I rely on the previous trial, on which I learned a very nice program, which is REW. I used to play Speakerworkshop before, but compared to rew it was torture.
I only have a theoretical question: Will the minimum phase of this speaker be different for the measurements of the same loudspeaker and from different distances?

 

[sm52]

Will the minimum phase of this speaker be different for the measurements of the same loudspeaker and from different distances?

In accuracy will not coincide. The overall slope will be about the same.

 

[karoles75]

I sit and read and read and I still cannot understand one thing. Is the measurement with the minimum phase of the tweeter and midrange drivers, performed under the same conditions in the final enclosure, one that I can use to simulate the crossover?

 

[sm52]

This will not be correct. It is better to use measurements with a normal phase, and after loading into the simulator, you can do this and that. Someone denies the use of the minimum phase, someone uses it.

 

[karoles75]

This will not be correct. It is better to use measurements with a normal phase, and after loading into the simulator, you can do this and that. Someone denies the use of the minimum phase, someone uses it.

Thank you very much, I think I get what this minimal phase is about. To simulate a crossover, I need measurements that take into account the different time of arrival of the signal from the speakers to the microphone. I just need to rethink how to do it in REW and have some ideas to test.

 

[sm52]

You argue correctly. REW makes this easy. But first, something needs to be done to trust the measurements. That delay in measurement suggests that he should not be trusted. Need to be redone.