REW Beta Release How to simulate anecoic chamber using Rew 5.20 beta 61

Cippino

New Member
Thread Starter
Joined
Sep 14, 2020
Messages
9
More  
Front Speakers
2
hello, I need help to understand which is the best configuration to use to simulate the results of an anechoic chamber in REW v5.20 beta 61. I have to test and calibrate a 3-way system with active filter minidsp 4x10hd. microphone positioned 1 meter away 1.12 meters high (ear height) on a carpet in the center of the room measuring 5.1 meters x 4.5 meters x 2.8 meters. umm6 Dayton microphone. I am attaching what I have mesured and I thank's you for your assistance. Daniele.
 

Attachments

  • 3 vie.mdat
    2.3 MB · Views: 19

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
To generate a pseudo-anechoic result from an in-room measurement the impulse response right hand window needs to be adjusted to exclude the room's reflections. You can change the settings using the IR windows dialog or, more interactively, by dragging the "R" marker at the top edge of the impulse graph, which shows a preview of the effect the change will have on the response.
 

Cippino

New Member
Thread Starter
Joined
Sep 14, 2020
Messages
9
More  
Front Speakers
2
Thank you so much, you obviously mean this window I think. In your opinion which do you think is the best window for my case? Or how do you think it is best to proceed? Thank you very much
 

Attachments

  • Window.JPG
    Window.JPG
    194.5 KB · Views: 51

Martin.NWUK

New Member
Joined
Apr 27, 2020
Messages
3
Firstly, according to the RT60 and RT60 decay plots from your measurement, you're in a significantly reflective environment, you want to minimise that as much as possible. The carpet is absorbing some of the higher frequency reflections but only beyond 8K ish by the looks of it Try putting as much absorption material on the floor (thick and large area), or at least a decent area and thickness over the reflection point between loudspeaker system and mic. Keep the system as far away from any boundaries as possible, which you seem to be doing. Measure again then chop the impulse response to before the first reflections if you can identify them. The low frequency limit will always be restricted unless you measure outside, way up. Near field measurements can be made in rooms below 300Hz or so, then corrected for level and diffraction at your measuring distances for anechoic responses at low frequencies.
 

Attachments

  • 1.RT.png
    1.RT.png
    24.6 KB · Views: 35
  • 2.RTD.png
    2.RTD.png
    87.2 KB · Views: 33
  • 3.IR.png
    3.IR.png
    26.3 KB · Views: 29
  • 4.WF.png
    4.WF.png
    168.7 KB · Views: 29
  • 5.AR.png
    5.AR.png
    7.4 KB · Views: 28

Cippino

New Member
Thread Starter
Joined
Sep 14, 2020
Messages
9
More  
Front Speakers
2
Thanks, tomorrow I will proceed as indicated and I will send you the material to verify that I am doing well. Thanks.
 

Cippino

New Member
Thread Starter
Joined
Sep 14, 2020
Messages
9
More  
Front Speakers
2
Thanks, it's a great read. As soon as the measurement methodology is completely clear to me, I will measure this. For the moment I have redone the measures by improving the soundproofing of the floor, I am attaching the file. What I notice to be more evident now is that as the time of the analysis window changes, deep signal drops appear in correspondence with the crossing zone, which does not appear in the complete measurement. I used 48db BW filters. Time alignment was done on the tweeter and midrange. is that correct? can it be improved?
thanks.
 

Attachments

  • 3 vie con imbottitura.mdat
    9.4 MB · Views: 10

Cippino

New Member
Thread Starter
Joined
Sep 14, 2020
Messages
9
More  
Front Speakers
2
good evening, I performed all the measurements in the near field, very useful to arrive at the result that I am attaching (complete 3-way system) as I now understand that this measurement always made at a distance of one meter is true, that is, that it looks like or otherwise do you approach one made in an anechoic chamber? In addition, in your opinion, how can I improve it in terms of measurement? What other setups or tests can I do to test the speaker properly with rew?

thanks.
 

Attachments

  • misura 14122020.mdat
    1.8 MB · Views: 14

Martin.NWUK

New Member
Joined
Apr 27, 2020
Messages
3
Looking at the waterfalls in (3 vie con imbottitura...) , it still looks to me like your absorption is only effective down to about 1.6kHz, which is an improvement but isn't low enough to say for sure whether the dip centred on ~750Hz is due to the loudspeaker system or out of phase reflections, assuming nothing else is interfering. Do you have very bare, highly reflective walls and ceiling? Try sweeps at a closer distance, say 50cm, experiment with 30cm also, keeping and possibly thickening your floor absorption. There's also the possibility of aiming the loudspeaker upwards to avoid a bit more floor reflection maybe. The aim is always to avoid and reduce as many reflections as possible and down to as low as possible. Where there are reflections and resonances, you're in guessing and approximation land. For a quick look at what might be happening, try using pink noise while shortening the measuring distance and see if the dip remains. Go in very close to the driver centres and flush with port output, if there is one, to get an idea of the nearfield responses.
 

Attachments

  • fr.icfrs.jpg
    fr.icfrs.jpg
    75.1 KB · Views: 19
  • wf.abs.1k6.jpg
    wf.abs.1k6.jpg
    93.8 KB · Views: 18

Cippino

New Member
Thread Starter
Joined
Sep 14, 2020
Messages
9
More  
Front Speakers
2
Thanks for your analysis, I tried to do some new measurements, how do you find them? Are there any points where I can improve?
Thanks.
 

Attachments

  • Misura varie.mdat
    14.7 MB · Views: 7

Martin.NWUK

New Member
Joined
Apr 27, 2020
Messages
3
I think your full range measurement looks excellent and maybe all you need for the systems on axis response. The rest seem pretty good too. I've had a look at your file and uploading a version of it. The useful measurements with comments are the top four.
 

Attachments

  • Misura varie.ms.mdat
    16.4 MB · Views: 8
Top Bottom