IR Windows

[bowl_actually]

Can someone explain the nuances of the "left window," "window ref time," and "right window?"

From my understanding the left window is the time before the sweep starts, which if that's correct, I'm not sure why anything before the sweep starts matters. So clearly I'm missing something.

The "window ref time" I don't really understand, so I won't attempt that one.

The right window, as I understand it, is the time after the sweep, so if you limit that you limit the number of reflections you're seeing. So if I wanted to measure a speaker with a mic that is maybe only 10 inches away, I might change that to prevent the room from adding to the response. But I'll limit how low of a frequency I can measure.

I'm not real sure why that limits the low frequency I can measure. I also don't actually understand what the time window that's defined relates to. It can't be the amount of time from the beginning of the sweep as the sweeps take longer than that window and are also variable. It doesn't make sense that it's from the end of the sweep?

Sorry for the confusion, but the settings on the window affect all the graphs so I'm trying to wrap my head around all the nuances of how that works.

 

[John Mulcahy]

The help has a relevant topic.

 

[bowl_actually]

The help has a relevant topic.

Thank you John, I apologize for missing that, for some reason I looked at the help but didn't find that. Blind as a bat as my mother would say!

So if I read that correctly, the reference time defines where the divider is between left and right. So that's generally set at 0 or at the main spike of the impulse. From that main spike, everything to the left (generally noted by the setting of the left window as long as the ref is at the main spike) represents "distortion?" So by setting the ref and the left to "0" you effectively remove all the distortion from the other graphs?

Then the right window (again assuming the ref is at 0 or the main spike) sets the resolution and if you find where the reflections are you can effectively filter out some of those if you so desire?

I'll have to reread and think about that Gaussian and I'll have to find more info about the other choices there as well (blackman, hann, flat-top, etc. etc.)

 

[John Mulcahy]

Distortion images are further back in time. The left window needs to be wide enough to capture the onset of the impulse response, which may be a significant period for something with limited bandwidth such as a subwoofer. Correct on the right window.

 

[bowl_actually]

Distortion images are further back in time. The left window needs to be wide enough to capture the onset of the impulse response, which may be a significant period for something with limited bandwidth such as a subwoofer. Correct on the right window.

Ah, thanks for that. Any guidance on typical settings for the left? Is it just milliseconds, 10s of ms, 100s? Maybe a rule of thumb for woofer sizes? Like 125 for a 15" inch sub, 80 for a 12, or maybe some clue from looking at the graph as to where to set it?

 

[John Mulcahy]

REW's 125 ms default is generally fine for the left window.

 

[bowl_actually]

Thanks.

In case anyone else see this thread and is curious about more details, this thread gives more detail about a number of things regarding the window settings.

How to make quasi-anechoic speaker measurements/spinoramas with REW and VituixCAD

Quasi-anechoic measurements are a way to capture the 'real' frequency response of a speaker without the influence of a room. They are called 'quasi-anechoic' because they have the potential to get you really close to the results one might achieve with a non-reflective environment like an...

www.audiosciencereview.com