Early Impulse Response - Can't Time Align

jimndeb

Registered
Thread Starter
Joined
May 2, 2021
Posts
4
I am attempting to Time Align my measurements using the Impulse Overlay window. My understanding is that I can use this window to measure the delay and difference in time the impulse is arriving at the listening position for each channel. Once known I can adjust each channel measurement in the Impulse tab of the main window using the Offset t=0 button.

I'm having a problem similar, if not the same as in an earlier thread: Impulse response anomaly? (I am a new user so I cannot post links)
My impulse measurements appear before (to the left of) t=0.
The suggestion from John Mulcahy was to:
...You can ask REW to estimate and then remove any delay (positive or negative) in the impulse responses using the Estimate IR delay button in the graph controls.
I'm wondering which graph controls are meant or where to find them. The other suggestion in this thread isn't helpful because I am using the same speaker (R) as a Reference Output. I think this means Chirp will always be on the (R) default speaker.

Below you can see the overlay window, my Analysis Preferences, and that I'm using the same (R) speaker for a Reference Output.

This is making me crazy. Any help is appreaciated.

2Time alignment issue.png

Screen Shot 2023-04-29 at 9.07.02 AM.png
Screen Shot 2023-04-29 at 10.06.23 PM.png
 

Attachments

  • Time_Alignmenet_early.mdat
    1.6 MB · Views: 25

John Mulcahy

REW Author
Joined
Apr 3, 2017
Posts
8,033
That's a little odd, you could see whether it is the same for the V5.20.14 early access build.

Estimate IR delay is in the Impulse graph controls and the SPL & Phase graph controls.

If you are aiming to align left and right just use the difference between their peaks since the responses are so similar, but it is only 100 us or so.

There's no need to use such long sweeps by the way, 256k is fine.
 

sm52

Member
Joined
Mar 14, 2019
Posts
1,008

jimndeb,​

if the mdat file contains measurements of the left and right sides, then no timing is required from the point of view of improving the sound picture. If you move the microphone a little, it turns out that the impulses will be at the same moment. That is, the graphs of the impulses of the left and right sides will coincide. And if you want to train with REW, then it's for a long time, it takes a lot of effort.
 

jimndeb

Registered
Thread Starter
Joined
May 2, 2021
Posts
4
Problem solved thanks to Serkan et al over on the Roon forums. The problem was that the USB input to my amplifier was being looped out to my DAC and back into the amp via XLR. This created a distortion in the timing. Having the USB input play through my integrated amp's native DAC solved the timing issue.
 

ManCaveAudio

Moderator
Staff member
Joined
Mar 12, 2023
Posts
75
That's a little odd, you could see whether it is the same for the V5.20.14 early access build.

Estimate IR delay is in the Impulse graph controls and the SPL & Phase graph controls.

If you are aiming to align left and right just use the difference between their peaks since the responses are so similar, but it is only 100 us or so.

There's no need to use such long sweeps by the way, 256k is fine.
Hi John,

Can you or anyone guide how to correct phase response in the home theatre AVR setup like 5.1.4. I am using denon x4700h.
 
Top Bottom