REW Alignment tool - Guides or Manual?

[Duper]

I hate to say what you don't want to hear, but if I had to guess, the issue is that you are not using a calibrated microphone.
Are you using a reference speaker for the time alignment?
There is an old thread (2010) on AVSforums from a few members that were trying to use REW using the microphone supplied with their AVRs. They all seemed convinced that the Frequency response graph they generated were pretty accurate, but also agreed that the SPL measurements were not accurate.

Before going any further, I'll ask if you've already read AustinJerry's simplified REW manual? That manual can atleast help you get all of the other step correct if you don't have the proper mic, so that the only error could come from the mic and not necessarily the process that you used. I've attached it to this post in case you need it.

All that aside, if your setup is not very complex, the auto-calibration in your Onkyo AVR will get you very close with both the SPL levels and the speaker distances. The subwoofer distance is usually a little tricky because if it is a self-powered subwoofer, the amplifier that it has can add delay to the signal, which is why the subwoofer distance when measured via a microphone is usually not the same as the actual physical distance of the subwoofer itself.

Having access to both a UMIK-1 calibrated microphone as well as the AVR supplied microphone, I can tell you that the result are different when using them with REW.

Yes I am using reference speaker for time alignment. All the readings were also adjusted using Estimate IR delay. Of course by not using calibrated mic results will not be perfect especially SPL levels. But why timing seems to be jumping forth and back by almost 20ms when all the graphs looks almost identical?

I definitely read AustinJerry's simplified REW manual like 3-4 times and referring to it if I am trying something new.

I am looking in potentially acquiring calibrated mic but for now that is all that I have and wanted to experiment with the software and my setup.

Thanks

 

[jtalden]

I was trying to time align my L and R speakers (they do not look too aligned in the graphs). And then time align my only SW to the mains. I have few questions:
1 - Does the Left and Right look aligned to you guys? Crossover on the receiver is set to 80hz. Right speaker is a bit further from the Left one.

The left and right measurements are at about the same timing (look at your impulse overlay chart). You apparently have adjusted for any significant difference in distance in the AVR. Your measurement #3 L+R is time shifted 0.83 ms delayed from the L and R mains measurements for some reason. This is more than expected but not a significant impact for SW timing work. The SWs channel measurement timings look consistent, so it appears acoustic timing is worked as expected except for that one measurement. If you want to confirm the repeatability, just take a larger number of measurements on the same speaker. The impulses should all fall on top of each other when zoomed in on the time scale.

2 - When I do "trace arithmetic /2 to get average" it is way lower than the actual signal from L + R played together. Is that normal?

It's normal. It will fall between the L and R in most places. In some areas the room will create a large phase difference and a sag or null will occur. The rolloff above 2 kHz is due to the slight difference in timing of the impulses. This is not significant for SW timing setup.

3 - when I was trying to align SW to the Left speaker. I was changing distance for the SW in the receiver. I think I got it close to 1.xx ms at 10.6ft. The question is, if I take multiple measurements without changing mic position or anything, I am getting different result. Sometimes it shows delay in 18ms range and sometimes 1.xx ms why is that? You can see that in 5.6 and 10.6 measurments.

I am not sure what you are seeing. The overlay of the 2 set of 3 SW impulses fall on top of each other and spaced 5 ms apart as labeled. This indicates good repeatability. Don't read the delays in the REW measurement notes if that is want you are doing. They are not used for this type of timing analysis.

 

[Duper]

The left and right measurements are at about the same timing (look at your impulse overlay chart). You apparently have adjusted for any significant difference in distance in the AVR. Your measurement #3 L+R is time shifted 0.83 ms delayed from the L and R mains measurements for some reason. This is more than expected but not a significant impact for SW timing work. The SWs channel measurement timings look consistent, so it appears acoustic timing is worked as expected except for that one measurement. If you want to confirm the repeatability, just take a larger number of measurements on the same speaker. The impulses should all fall on top of each other when zoomed in on the time scale.

I see what you mean. Impulse overlay chart was very helpful, I did not know it existed. I can see that "L+R together" is time shifted for some reason by 0.82ms no idea why it happening. Nothing was moved or changed in the settings. I was taking 3 measurements when adjusting the the SW distance in AVR. Maybe need more than 3 but not sure which one to trust. See more below.

I am not sure what you are seeing. The overlay of the 2 set of 3 SW impulses fall on top of each other and spaced 5 ms apart as labeled. This indicates good repeatability. Don't read the delays in the REW measurement notes if that is want you are doing. They are not used for this type of timing analysis.

Please see the screenshots, I was talking about the Delay in the alignment tool. It shows different delays. Looking at timing impulse overlays I can see that they do fall right on each other (just like in your screen shot).


Another question: should I be time aligning my SW to Left Speaker (as I am doing now), Actual combined output from L+R together, or as it was suggested at the beginning of this thread using trace arithmetic /2 to get average A+B?

 

[jtalden]

Please see the screenshots, I was talking about the Delay in the alignment tool. It shows different delays. Looking at timing impulse overlays I can see that they do fall right on each other (just like in your screen shot).

There are several settings that provide favorable timing for SPL support. Below is the timing that I found favorable. It requires inverting the SW and reducing the delay 7.5 ms.

Another question: should I be time aligning my SW to Left Speaker (as I am doing now), Actual combined output from L+R together, or as it was suggested at the beginning of this thread using trace arithmetic /2 to get average A+B?

I used measurement 10 (L,R Average) and picked SW measurement 12 for the initial timing. The XO range is about 60-130 Hz. The L,R average results in the correct gain level in the case of 1 SW or a group of SWs that has already been summed as a group.

 

[Duper]

There are several settings that provide favorable timing for SPL support. Below is the timing that I found favorable. It requires inverting the SW and reducing the delay 7.5 ms.

I used measurement 10 (L,R Average) and picked SW measurement 12 for the initial timing. The XO range is about 60-130 Hz. The L,R average results in the correct gain level in the case of 1 SW or a group of SWs that has already been summed as a group.

Thanks for your input jtalden. I am also able to get almost identical SPL graph by using these settings (my crossover is set to 80hz in receiver). By selecting 80hz on the graph and clicking "Align phase at cursor". Does it mean my sub is almost aligned to the mains?

My previous question remains the same. Why do I get different results selecting measurement # 8: Sub..... It is not different from measurement #7. It was taken right after taking measurement #7. Following the same procedure as above I get positive 13ms delay, and SLP is lower after 80hz. So I do not know which measurements to trust #7, #8 or #9. #7 and #9 looks like I am golden, but #8 suggest I need delay. I hope I am explaining this correctly.

 

[jtalden]

That setting has the SW delayed an extra wavelength at the XO frequency so there is more unneeded phase rotation and group delay. The SPL and sound quality is usually pretty similar though so it can be used also. Most of us would prefer to avoid that unneeded additional SW delay. If you overlay those 2 SPL responses, you can also see that the second option is not quite as good as the first.
Your first chart SPL will match very closely with the timing I suggested using the inverted SW polarity. I think either of those 2 alignments are the best choice.

 

[Duper]

That setting has the SW delayed an extra wavelength at the XO frequency so there is more unneeded phase rotation and group delay. The SPL and sound quality is usually pretty similar though so it can be used also. Most of us would prefer to avoid that unneeded additional SW delay. If you overlay those 2 SPL responses, you can also see that the second option is not quite as good as the first.
Your first chart SPL will match very closely with the timing I suggested using the inverted SW polarity. I think either of those 2 alignments are the best choice.

Understood, thanks. So basically the way I have everything set is as good as it is going to be. Regarding difference in timings(for the last graph) , I guess REW for some reason is suggesting extra wavelength, instead of the negative delay to reach the same goal.

 

[jtalden]

Yes.
For low frequency XO from the LP position the REW automated alignment solution can give different timing depending on the position of the cursor position and the room responses at that frequency. Different cursor positions in the XO range can provide different results. With some experience we can chose among the options. With more experience we can just choose the timing manually using the slider.

 

[Duper]

Yes.
For low frequency XO from the LP position the REW automated alignment solution can give different timing depending on the position of the cursor position and the room responses at that frequency. Different cursor positions in the XO range can provide different results. With some experience we can chose among the options. With more experience we can just choose the timing manually using the slider.

That makes so much more sense now. I was going crazy scratching my head on why I am getting different results when nothing got changed. Thanks for taking time to explain it.

if only it could be that easy to correct the 10db bump around 50_60hz that I have from the room mode. I guess only MiniDsp could fix it.

 

[UKSPAWN]

yes use MiniDSP to EQ that lump out (try to use no boost but rather cut)

 

[clubfoot]

Jtalden in your opinion is it important/necessary to also align the surrounds with the sub?

 

[jtalden]

I would guess this is a trivial issue. This is only speculation however as I have no direct information to support it one way or the other.

 

[clubfoot]

OK, thanks.

 

[ece2k2]

Thank you all for the wonderful discussion that helped me to navigate the mains and sub alignment process to some extent. Using the tool, I am getting a subwoofer delay of -25ms for a good summation with the mains at 80Hz with LR4 applied for speakers and sub. But this delay sounds a bit too steep and I would like to know if I am supposed to try with a different crossover frequency/slope or if I need to do any other tweaks to improve the outcome. I have attached the mdat file.

I am using Trinnov Altitude processor and I have a good amount of flexibility with the type of crossover filters and slopes that I could use. Thanks again for all your help!

 

[jtalden]

Well done. Your posted sub delay needs to be adjusted about -25 ms to provide the most favorable alignment. I found -24.6 ms (a trivial difference).
Below are 3 charts to illustrate the impact:

SPL of Current vs aligned SPL

Your XO range is from about 35-150 Hz so shown below is the phase tracking differences.

Current

Aligned

FYI. All speakers currently show to be inverted polarity. This could be in the measurement setup or the audio system. Some say they can hear a difference with test signals. I am personally very skeptical of this at least regarding the practical impact on the listening experience.

 

[jtalden]

You could try a sub delay change of -35.6 ms to reduce the overall group delay if that is your concern. This also appears a little improved in the spectrogram as well. The phase tracking and SPL would not be quite as favorable, but they are still very acceptable. I think the room acoustics is the primary influence here so it is questionable that using lower slope XO filters would improve the situation, but you could experiment with that.

 

[ece2k2]

You could try a sub delay change of -35.6 ms to reduce the overall group delay if that is your concern. This also appears a little improved in the spectrogram as well. The phase tracking and SPL would not be quite as favorable, but they are still very acceptable. I think the room acoustics is the primary influence here so it is questionable that using lower slope XO filters would improve the situation, but you could experiment with that.

Much appreciated jtalden. With the mains delayed by 25ms to implement the negative delay on the sub channel, the bass quality has improved drastically. Really happy with the outcome so far

I will try with the -35.6ms sub delay as well to see how that sounds. The room is a dedicated HT space designed by Nyal Mellor and the mains are JBL M2. I will also experiment with the lower slope XO filters. I already tried higher slope XO filters but they didn't improve things.

Thanks again for your kind support!

 

[sm52]

ece2k2,​

write what you heard with a delay of 24 ms and with 36 ms. Which option seems best to you. From my point of view, the correct option is 36 ms.

 

[ece2k2]

ece2k2,​

write what you heard with a delay of 24 ms and with 36 ms. Which option seems best to you. From my point of view, the correct option is 36 ms.

I listened to some music with -25ms and -36ms sub delays. For some reason, I was unable to find any significant audible difference between them. Also, it looks like -25ms sub delay yields better centre channel response when compared to -36ms.

To keep the timing consistent among all the speakers for the multichannel/movies content, I have added a 25ms delay to every speaker channel in my system. Hope this is the right approach.

 

[sm52]

For some reason, I was unable to find any significant audible difference between them.

Thank you. If you have time, I'm wondering why both major channels measured with negative impuls. Could you, when selecting input and output devices, in the Preferences window, select the same devices, but without the Excl prefix, and make measurements of the right and left channels. And post here.

 

[ece2k2]

Thank you. If you have time, I'm wondering why both major channels measured with negative impuls. Could you, when selecting input and output devices, in the Preferences window, select the same devices, but without the Excl prefix, and make measurements of the right and left channels. And post here.

Hi sm52,

I have attached the measurements taken by selecting the devices without the Excl prefix. Please let me know if you see any issue.

Thank you!

 

[sm52]

devices without the Excl prefix

Thank you. But nothing has changed. All impulses are negative. So use devices with the Excl prefix as before. The reason for the negative impulse can be in different places. Perhaps in the measuring equipment. To adjust the delays, this is not a hindrance. That is, you don't have to worry. If the speakers receive the correct music signal (positive polarity).