How to measure soundproofing with REW?

[michelpayou69]

I had this simple idea :
Put a speaker in room A. Use REW to measure a "reference level" in room A. So far so good.
Now I wish to measure the level in room B to check how much soundproofing I need.

Speaker stays in room A but since my mic is connected to my laptop which is in room A...I can't record the signal in Room B

I wish there were a way to connect 2 laptops in Wifi with 2 REW : 1 is the source (sweep) in ROOM A, and 2nd is the receiver in ROOM B.

Any other way to do this?
Hope I'm clear

 

[Wayne A. Pflughaupt]

Assuming the amplification system is in Room A with the speaker - Whatever is getting the REW signal from the computer to the sound system (HDMI? RCA cable?), just get an extension cable. Move the computer and mic to Room B, add the extension cable to get sound from the computer to the amplifier, take second measurement.

Regards,
Wayne

 

[ddude003]

You could also use the Generator function to create a .wav file of the sweep... Copy the sweep.wav to a jump drive and play it from your kit in room A while taking the computer and mic into room B for a measurement...

 

[John Mulcahy]

play it from your kit in room A while taking the computer and mic into room B

Probably won't be able to pick up the timing reference though, wired is the safe option.

 

[Wayne A. Pflughaupt]

Probably won't be able to pick up the timing reference though, wired is the safe option.

I assumed an A/B for soundproofing purposes would only require a SPL measurement. Is that incorrect?

Regards,
Wayne

 

[John Mulcahy]

No, but the discussion has been about sweep measurements and sweep measurement with file playback uses the timing reference to synchronise capture.

 

[michelpayou69]

No timing reference needed. Just wanted to do a basic operation : ROOM A - ROOM B = actual soundproofing.

 

[michelpayou69]

John, you recommend simply to get a longer cable?

Thanks guys!

 

[BenToronto]

Sound loss for what reason?

SPL dBA is the traditional assessment of human annoyance. Corrected for airports by how often planes fly... and for other places by all kinds of assumptions about noxiousness, but you get the idea.

With a sweep, you get to see what band is troublesome. And then.....

you can tune the music room (like fiddling with your music or theatre room sound?)
you can dampen the target room (almost any absorbent almost anywhere in the room will help)
you can address the transmission between rooms

Many years ago, I worked on highway noise annoyance. The only barrier that people said worked was the fence or planting you build yourself.

Ben

 

[linearphase]

Probably won't be able to pick up the timing reference though, wired is the safe option.

John, lacking a timing reference and just looking at the freq domain magnitude component how would you use waveform math to subtract the in room response and the outside the room response to determine transmission loss?

 

[John Mulcahy]

Magnitudes get divided. If there is no timing reference with file playback there will not be any sweep measurement, wired is the only viable option for a sweep measurement. An alternative would be to use the RTA with periodic noise, but unless the transmission losses are low it would typically be difficult to generate a PN signal at a high enough level to measure the actual loss rather than just ending up measuring the receive chain noise floor.

 

[BenToronto]

Having trouble figuring out why a time reference matters. Isn't a simple sweep all you need for the basic physics (assuming identical mic calibration in sending and receiving rooms)?

Much, much bigger issues to think about like location of receiving-room mic (I am assuming everything in the sending-room will be glued in place and aren't being studied as variables here (although they should be)), background noise level, and the many and important site/person specific issues I alluded to earlier.

BTW, it is possible the source sound (say rock music versus speech) matters in a way beyond just the physical.

 

[John Mulcahy]

There was a suggestion early in the thread to get around the problem of running a cable by using file playback for the sweep. The receive side of a file playback measurement uses detection of the timing reference to start capture.

 

[linearphase]

Magnitudes get divided. If there is no timing reference with file playback there will not be any sweep measurement, wired is the only viable option for a sweep measurement. An alternative would be to use the RTA with periodic noise, but unless the transmission losses are low it would typically be difficult to generate a PN signal at a high enough level to measure the actual loss rather than just ending up measuring the receive chain noise floor.

Perfect I have done this before with the RTA Function. I measured the noise floor outside the room first then set the SPL so that with noise on it was 10dB above that floor. It was loud but not loud enough to damage anything. Then I measured just inside the door and compared the curves. Next time I will divide.