REW Alignment tool - Guides or Manual?

[jtalden]

I don't see how a new external soundcard helps your situation. How does that help disconnect a speaker?
The external soundcard also requires an XLR measurement mic. It is also a little more difficult to calibrate and thus most prefer to use a USB mic like the UMIK-1.

 

[AustinJerry]

@jtalden: sorry for the late reply. I thought I would be receiving email notifications when a post was made to this thread, but I am not receiving anything. I appreciate your insights and need a bit more time to comprehend everything.

 

[jtalden]

@AustinJerry
I just reviewed our previouse posts and noticed a directional error in Post-138. I mistakenly said that a reduction of the mains distance in an AVR will increase the SW delay. That has now been corrected. A decrease in mains distance in an AVR causes a decrease in delay of the SWs.

I looked at your setup equipment list and see you are using a 8802A pre/pro. You should thus be able to either:

• Decrease all the mains distance by 1 ms or,
• Increase the SWs distance by 1 ms

Either adjustment would be the same a decrease to the SWs delay in the MiniDSP.

 

[AustinJerry]

@AustinJerry
I just reviewed our previouse posts and noticed a directional error in Post-138. I mistakenly said that a reduction of the mains distance in an AVR will increase the SW delay. That has now been corrected. A decrease in mains distance in an AVR causes a decrease in delay of the SWs.

I looked at your setup equipment list and see you are using a 8802A pre/pro. You should thus be able to either:

• Decrease all the mains distance by 1 ms or,
• Increase the SWs distance by 1 ms

Either adjustment would be the same a decrease to the SWs delay in the MiniDSP.

Thank you for the follow up. While it is correct that I have a Marantz 8802a, I am also using a MiniDSP 88a and Dirac Live. The mains distances in the 8802a are all zeroed out, which is the usual setting when using Dirac. The delays established by Dirac in the 88a cannot be altered by the user, and the delay settings in the 88a output stage cannot have negative values. Unfortunately, it doesn't seem that I can implement your recommendation.

 

[jtalden]

No need for a change. Your results are outstanding and will not be impacted by such a trivial change.
I only wanted to clarify how it could be done in your setup since you had asked.

That is:

• Increase the SW distance by 1 ft in the 8802a

That would create approximately 1 ms delay in the LFE channel

 

[Harrycr]

I don't see how a new external soundcard helps your situation. How does that help disconnect a speaker?
The external soundcard also requires an XLR measurement mic. It is also a little more difficult to calibrate and thus most prefer to use a USB mic like the UMIK-1.

Might be missing something here.
Thought that the loopback was done internally in the outboard soundcard?
I can turn the speakers of via the Z9 menu but only both left and right.
I could use the left and right in REW for measuring the mains?

Thanks

 

[AustinJerry]

No need for a change. Your results are outstanding and will not be impacted by such a trivial change.
I only wanted to clarify how it could be done in your setup since you had asked.

That is:

• Increase the SW distance by 1 ft in the 8802a

That would create approximately 1 ms delay in the LFE channel

Thanks, I'll give it a try.

 

[jtalden]

Thought that the loopback was done internally in the outboard soundcard?

Most simple USB soundcards use a cable for loopback - output to input on the reference channel.

I can turn the speakers of via the Z9 menu but only both left and right.

Using an internal soundcard stereo output allows connection to 2 channels - a measurement channel and a reference channel. If you connect them to the Z9 multi-channel inputs that allows switching the measurement channel by moving the cable connection.

I could use the left and right in REW for measuring the mains?

Yes - as noted above or:
Most people are using an HDMI connection for the REW output. That makes it easy to pick which channel they want to measure within REW.

 

[Harrycr]

Most simple USB soundcards use a cable for loopback - output to input on the reference channel.

Using an internal soundcard stereo output allows connection to 2 channels - a measurement channel and a reference channel. If you connect them to the Z9 multi-channel inputs that allows switching the measurement channel by moving the cable connection.

Yes - as noted above or:
Most people are using an HDMI connection for the REW output. That makes it easy to pick which channel they want to measure within REW.

Thanks
Can I use the 9 channel stereo setting in the menu which can be done via 2 channel input?
I can turn off via the menu:
Both mains
Centre
Both surrounds

 

[jtalden]

Which of the 2 options are you thing of?
For mulli-channel inputs we would need to need to us the AVR multi-channel mode. Then physically move the REW measurement channel output cable to the desired AVR measurement channel input on the AVR.
For HDMI output using a USB mic. Refer to the videos on YouTube.

 

[Harrycr]

Which of the 2 options are you thing of?
For mulli-channel inputs we would need to need to us the AVR multi-channel mode. Then physically move the REW measurement channel output cable to the desired AVR measurement channel input on the AVR.
For HDMI output using a USB mic. Refer to the videos on YouTube.

I was going to try both.
There is no HDMi connection on the Z9 don't think I can get adaptor for this?

 

[jtalden]

I doubt it.

 

[mattel]

I doubt it.

Hi, Noobie here.

I am also new to forums and the correct way to reply, etc. So, forgive me for my ignorance.

I have been reading this thread quite a bit and have some questions regarding alignment of dual subs and subsequently alignment with mains using the alignment tool.

When you are using the time alignment tool for aligning subs, you are to move the sliders to determine the flattest FR and add the delay to the sub accordingly?

Also, while moving the sliders you should be looking at the phase graph to determine what exactly?

In some of your previous replies "FDW" is mentioned. What is it and how do I use it.

 

[JStewart]

When you are using the time alignment tool for aligning subs, you are to move the sliders to determine the flattest FR and add the delay to the sub accordingly?

You could but if you have more than 2 subs it would be a lengthy trial and error process. With more than 2 subs MSO (multi sub optimizer) is freeware that will find a solution for you. There is a bit of a learning curve for most. It’s primary objective is a smooth and very similar response at all measured listening positions.

A different method is to align the subs so as to produce the maximum output. This will mean they are in phase to the extent possible and can be eq’d to a desired response shape (i.e. flat or with a house curve) after.

This video demonstrates the process, but where the author makes delay changes and remeasures to zero in on the best solution, the alignment tool is much more easily used.

After the subs are aligned to each other then the alignment tool can also be used to align the subs group to the mains.

Plenty of help around here if you have questions as you get into it!

FDW is Frequency Dependent Window. In the context of using it for subs/mains alignment, It’s a setting in REW that can limit the number of cycles in the measurement for the purpose of more closely examining the phase and response of the direct sound vs the phase and response of the direct sound combined with room reflections.

 

[mattel]

I have two subs, 5.2 setup
So, the alignment tool is only for aligning subs with mains, but not alignment of two subs?

Thanks for your help. I appreciate it.

 

[JStewart]

I have two subs, 5.2 setup
So, the alignment tool is only for aligning subs with mains, but not alignment of two subs?

Thanks for your help. I appreciate it.

Yes the alignment tool can be used to align the subs to each other. The procedure I’d recommend is covered in the video. Just substitute using the alignment tool for the “trial and error” method the video’s author used to align subs to one another.

 

[mattel]

Ok, I will look at the video and make measurements accordingly, hopefully no later than this weekend or whenever my wife is out of the house.

 

[mattel]

Okay, I got a window and took some measurements. I would like your input. I measured both subs individually and together per video. Looking at the measurement results, it looks to me that the better responses are the "inverted" response and the 11s delay response. I have attached the mdat file as well.
I did a cut and paste for the image below. Is there a better presentation?

 

[JStewart]

Okay, I got a window and took some measurements. I would like your input. I measured both subs individually and together per video. Looking at the measurement results, it looks to me that the better responses are the "inverted" response and the 11s delay response. I have attached the mdat file as well.

That was quick. Nice!

Using measurements 3 & 4 from the .mdat along with the alignment tool a 9.5ms delay to measurement 4 looked best, but there was not much difference at all between it and 11ms.
Measurement 4 (LT Sub) has that dip you've likely noticed at 46hz. Often space placement options are limited, but you may get a better result with different placement if possible and not already tried.

And now for the not so great news... One thing I noticed is that the alignment tool's combined measurement with a 9.5ms delay to measurement 4 (LT Sub) does not precisely match measurement 10 which is labeled as 9.5ms delay. I would expect them to match exactly so long as neither the mic position nor subwoofer positions changed and assuming I'm understanding your measurement labels correctly. I would recommend:
1. Measure each sub separately again.
2. Use alignment tool to calculate combined response without any delay.
3. Measure combined response.
4. Verify alignment tool and measured response match.
If they do not there is something up with the measurement system that needs found and corrected.
If they do then use alignment tool to find best combined response or verify you've already got it.
Set DSP with best delay and measure combined response to alignment tool calculations. They should be the same.

May I ask what device you're using to set the delays?

 

[Harrycr]

That was quick. Nice!

Using measurements 3 & 4 from the .mdat along with the alignment tool a 9.5ms delay to measurement 4 looked best, but there was not much difference at all between it and 11ms.
Measurement 4 (LT Sub) has that dip you've likely noticed at 46hz. Often space placement options are limited, but you may get a better result with different placement if possible and not already tried.

View attachment 34788

And now for the not so great news... One thing I noticed is that the alignment tool's combined measurement with a 9.5ms delay to measurement 4 (LT Sub) does not precisely match measurement 10 which is labeled as 9.5ms delay. I would expect them to match exactly so long as neither the mic position nor subwoofer positions changed and assuming I'm understanding your measurement labels correctly. I would recommend:
1. Measure each sub separately again.
2. Use alignment tool to calculate combined response without any delay.
3. Measure combined response.
4. Verify alignment tool and measured response match.
If they do not there is something up with the measurement system that needs found and corrected.
If they do then use alignment tool to find best combined response or verify you've already got it.
Set DSP with best delay and measure combined response to alignment tool calculations. They should be the same.

May I ask what device you're using to set the delays?



View attachment 34790

With the measurements for phase alignment is this the best way to test if the measurements is usable?
Is there any other aspects of the measurement that can be looked at?

Thanks

 

[JStewart]

With the measurements for phase alignment is this the best way to test if the measurements is usable?
Is there any other aspects of the measurement that can be looked at?

Thanks

Not sure if I understand what you’re asking ...
If you’re asking if this is “the best” method to set up subs my answer is who knows?
Are there other methods? Absolutely.
Geddes method
Time align using impulse response, manual or with dsp software
MultiSub Optimizer
Double bass array
Sub alignment built in to room correction such as Dirac Live Bass Management
And others I’ll wager.
And let’s not forget the group that insists subwoofers are detrimental to sound.

In the absence of consensus (scientific or otherwise) as to best, I suggest just pick a method that you have the resources needed on hand, or fits your budget or objectives and skills. If you become dissatisfied or curios there’s always something else to try!

 

[mattel]

That was quick. Nice!

Using measurements 3 & 4 from the .mdat along with the alignment tool a 9.5ms delay to measurement 4 looked best, but there was not much difference at all between it and 11ms.
Measurement 4 (LT Sub) has that dip you've likely noticed at 46hz. Often space placement options are limited, but you may get a better result with different placement if possible and not already tried.


Yes, I know but the combined sub response eliminates some of the drop. I thought it would not be such and issue after eq'ing.


And now for the not so great news... One thing I noticed is that the alignment tool's combined measurement with a 9.5ms delay to measurement 4 (LT Sub) does not precisely match measurement 10 which is labeled as 9.5ms delay. I would expect them to match exactly so long as neither the mic position nor subwoofer positions changed and assuming I'm understanding your measurement labels correctly. I would recommend:
1. Measure each sub separately again.
2. Use alignment tool to calculate combined response without any delay.
3. Measure combined response.
4. Verify alignment tool and measured response match.

I probably bumped the mic

They match...

If they do not there is something up with the measurement system that needs found and corrected.
If they do then use alignment tool to find best combined response or verify you've already got it.
Set DSP with best delay and measure combined response to alignment tool calculations. They should be the same.

May I ask what device you're using to set the delays?

Mini DSP HD

I think I followed your instructions. I really don't understand the alignment tool. When moving the sliders, it appears that there are several different delay values where the graph looks good. So, how do you know which value is the best?
Also, what information is the phase plot telling you?

I hope my post is proper.

 

[JStewart]

Yes, I know but the combined sub response eliminates some of the drop. I thought it would not be such and issue after eq'ing.

Agree and you've got a good starting point for EQ.

 

[mattel]

Agree and you've got a good starting point for EQ.

Thanks for your help. I buried my other response inside the quote portion of my previous post. Did you see them? Still learning post presentation.
Did I do everything correctly? Am I ready to move on to main/sub alignment Is the procedure the same?

 

[Harrycr]

Not sure if I understand what you’re asking ...
If you’re asking if this is “the best” method to set up subs my answer is who knows?
Are there other methods? Absolutely.
Geddes method
Time align using impulse response, manual or with dsp software
MultiSub Optimizer
Double bass array
Sub alignment built in to room correction such as Dirac Live Bass Management
And others I’ll wager.
And let’s not forget the group that insists subwoofers are detrimental to sound.

In the absence of consensus (scientific or otherwise) as to best, I suggest just pick a method that you have the resources needed on hand, or fits your budget or objectives and skills. If you become dissatisfied or curios there’s always something else to try!

Sorry my question not clear.
Post 169
Were you were saying that when combining the two subs with the alignment tool and align sum it should match the measured two subs graph this is a sign that measurement is ok.

Thanks