Window choice for outdoor measurements

[Matthew J Poes]

any thoughts on how the different window options make sense for outdoor measurements and if different left window lengths make sense when in a free field environment. The outdoor measurements I took were reflection free out to 8.5ms (7 feet off the ground or so). I’ve been using Tukey .25, but I’m trying the other options the response shape does change some. Especially the absolute level of the bass (such as with Gaussian).

What about frequency dependent windowing? Any benefit to this with an outside measurement.

 

[John Mulcahy]

The aim of the left window setting is to ensure the IR onset is captured. It is more important for low frequency measurements where the peak usually occurs well after the response onset, for full range measurements the left window can be fairly short, it doesn't usually have much influence on the resulting response. To get a more or less anechoic response the right window can be set just shorter than the first reflection, so 8.5 ms or just under for that setup. Window shape shouldn't have much effect above the frequency corresponding to the right window width, 117 Hz for 8.5 ms. The response isn't really valid below that.

 

[Matthew J Poes]

The aim of the left window setting is to ensure the IR onset is captured. It is more important for low frequency measurements where the peak usually occurs well after the response onset, for full range measurements the left window can be fairly short, it doesn't usually have much influence on the resulting response. To get a more or less anechoic response the right window can be set just shorter than the first reflection, so 8.5 ms or just under for that setup. Window shape shouldn't have much effect above the frequency corresponding to the right window width, 117 Hz for 8.5 ms. The response isn't really valid below that.

Thanks John, helpful as always.

Some of the window shapes did change the LF shape and absolute level within the range where it should be valid. I had it in my head I was valid down to 80hz, so I think I miscalculated something (and may have seen the change down in that range).

Any chance you also know what effect wind would have on the response if using a wind screen. Earthworks insists that anything under 100mph is just fine as far as wind across the mic (though suggested we might induce resonances in the test rig or other such issues). There was some signs of problems from wind at the low frequency range well below the valid frequency range, but nothing noticed above say 50hz. We took double measurements frequently, especially when there was a lot of wind in one. The only difference was below 50hz or so, everything else laid right on top.

It appears none of the distortion data was all that valid but we can take that in my soundproof room.

 

[John Mulcahy]

Afraid not, haven't done outdoor measurements.