Help with 2 sub alignment using REW and AVR distances for delay

[nadrealista]

Question for the experts.

I have 2 different subs in asymmetrical locations in my HT, one in the left corner ( SVS PB2/plus - dual 12") and second one (velodyne cht10) about 7' from the front wall on right side wall. those two locations give me best complimentary response without any tweaks. Recently I have gotten UMIK and wanted to see if I can align the subs for better response, then align the virtual sub with mains.

I am a bit suck on aligning the sub with each other as the aligned response I get in REW does not match what I get when I run the actual measurements with REW calculated delays between the subs. REW file attached

sub 1 and sub 2

REW aligned vs measured:

combined subs no delay (better response) vs REW delay

What am I missing/doing wrong here?

Given the data I provided what delay would you chose using the REW alignment tool?

@OCA or @Keith_W maybe you can take a look?

 

[90sNintendoKid]

What AVR do you have? If it's Denon or Marantz, I recommend OCA's A1 Evo Acoustix. I have used it with 2 SVS 3000 Micros and 2 SVS SB1000s. The 3000 Micros fill in a 50hz null of the SB1000s.

 

[nadrealista]

Pioneer Elite SC-95 with MCACC Pro

 

[FrantzM]

Hi

I do not know the Pioneer Elite SC-95 with MCACC Pro. I would suggest you to try MSO (Multi Sub Optimizer) . A free software that uses REW for its measurements. The most tedious part would be to measure the subwoofers, it seems you know your way around REW...
You , need a DSP, one of the best and with a decent price, is the miniDSP 2x4 HD around $300 on Amazon, MSO is free and it will require some time to master, but it will provide you with a linearized "virtual" subwoofer, that you would present to the AVR and let it align that virtual subwoofer with the mains..
I did that with my Denon AVR for a long time, until I discovered @OCA Scripts. These are the real deal and also free, however they seem to work only with Denon or Marantz AVR

 

[nadrealista]

Thanks for the advice but I would like to just use my AVR distances to set delays and not purchase any additional hardware if possible.

 

[Keith_W]

For sub alignment, you need a timing reference with your measurement. None of your measurements have a timing reference, so it can not be used for time alignment. Since you are using a UMIK-1, an acoustic timing measurement is your only option. Read the REW Manual for instructions on how to use an acoustic timing reference.

Not mentioned in the manual: REW's timing measurement is usually inaccurate with subs. For example, a reflection greater in amplitude than the main impulse can trick REW into thinking that the reflection is the impulse peak. For this reason, I prefer to MANUALLY read the timing of the sub impulse, but there is no easy way to do this with REW. It is possible to do this with a complicated signal chain and third party software, but for now the best advice I can give is to repeat your measurement a few times and make sure REW's timing is the same between all the measurements.

I don't have my interface with me at the moment, but at some point in the future I will run some experiments and discuss this with John.

 

[Chrispy]

Personally having used different subs with a Pioneer avr with MCACC vs Denons with Audyssey, particularly with multi sub capability (XT32), I do prefer the Denons. As to weird rooms or limited positions for subs, hard to know.

 

[nadrealista]

@Keith_W ok I will retake the measurements with timing reference. how do I set REW driver in preferences to have ability to set up output to individual channels including subs so I can just send timing reference signal to say L speaker and test signal to the sub I am measuring. Right now I only have L,R and L+R output options with Java driver?

for now I can set AVR XO to 200hz for sub sweeps in order to separate sub responses from reference channel since we are interested in 0-120hz region?

 

[Keith_W]

How is your sub connected to the AVR? Is there no dedicated output channel for the sub? If the sub can not be controlled independently, how are you going to align it?

These are the measurements you need:

- individual measurements of each sub with a timing reference
- individual measurements of each speaker with a timing reference

 

[nadrealista]

AVR has 2 independent sub pre-outs/outputs so I can set individual sub level/4 band eq(32,63,125,250)/distances and 3 standing waves corrections (which are for both subs).

AVR calibration set distances vs actual distances:

SPEAKER	ACTL DIST	AVR DIST
L
C
R
SW1
SW2

 

[nadrealista]

@Keith_W I am having trouble making measurements using acoustic reference signals. It keeps giving me this message and does not want to complete the measurement? I check the levels before the measurement and they are fine. I hear the chirps before and after measurement but REW keeps waiting for the second one?

 

[Keith_W]

Yeah, I don't know why that happens. I used to get that issue occasionally. I tried increasing the ref level trim so that it's unbelievably loud and it seemed to work a bit. But otherwise, I don't know ... that's a question for John Mulcahy. You can ask him here.

 

[nadrealista]

@Keith_W after some trial and error I was able to make measurements with time ref. new files attached - had to split them due to the upload size limit.
for these measurements all dsp is off and all distances in AVR are equal, set to 10ft

let me know if you have what you need now?
thanks for taking the time!

 

[neRok]

Question for the experts.

I have 2 different subs in asymmetrical locations in my HT, one in the left corner ( SVS PB2/plus - dual 12") and second one (velodyne cht10) about 7' from the front wall on right side wall. those two locations give me best complimentary response without any tweaks. Recently I have gotten UMIK and wanted to see if I can align the subs for better response, then align the virtual sub with mains.

View attachment 499366

I am a bit suck on aligning the sub with each other as the aligned response I get in REW does not match what I get when I run the actual measurements with REW calculated delays between the subs. REW file attached

sub 1 and sub 2
View attachment 499347

REW aligned vs measured:

View attachment 499349

View attachment 499348

combined subs no delay (better response) vs REW delay

View attachment 499351


What am I missing/doing wrong here?

Given the data I provided what delay would you chose using the REW alignment tool?

@OCA or @Keith_W maybe you can take a look?

The Room Sim doesn't account for the subwoofers having different filters, especially the subsonic filter required to protect a ported sub from over-extension. These different filters mean the subs have different timing characteristics (group delay). So Room Sim is kind of showing you how 2 identical subs interact with the room, whereas you don't have 2 identical subs. So you need will need to measure the subs with a timing reference, which you can use then use to align their timing at the necessary point.

So if the timing offset required to the suit room is 6ms, then that's 6ms for equal subs — but if one sub is naturally 10ms slower than the other, then your actual DSP offset might be 4ms or 16ms (depending on what's-what).

One sub may also be wired up with "opposite polarity", which means you might have to account for that too.

Lastly, when you look at various phase plots, you can see the "slope" of the phase line, which corresponds to the shape of the group delay graph. If one sub has a 2nd order BW subsonic filer, and the other has a 4th order LR filter, then these slopes will likely diverge. So even if you align the subs at 60Hz say, their timing might have drifted apart by 30Hz. If this is the case, you may be able to add an additional subsonic to the "faster" sub (which may slow it down to match the slower one), or correct it by generating an FIR with tools like RePhase.

View attachment 499847

If you're trying to use the subwoofer as the timing ref, then it likely means the sweep cant distinguish the tweet from the background noise. More volume might help like Keith suggested.

There's probably a better way to do this, but I usually start with speakers+sub on, so that the tweet comes out the speakers. then I quickly mute the speakers, so only the sub sweeps, and then a few moments after the sub finishes I un-mute the speakers in time for the 2nd tweet. But If you mute too late or un-mute too early, you can pollute the sub sweep with bass from the speaker(s).

 

[levimax]

There's probably a better way to do this, but I usually start with speakers+sub on, so that the tweet comes out the speakers. then I quickly mute the speakers, so only the sub sweeps, and then a few moments after the sub finishes I un-mute the speakers in time for the 2nd tweet. But If you mute too late or un-mute too early, you can pollute the sub sweep with bass from the speaker(s).

While the acoustic timing reference has been improved in REW and in many cases works well, there are cases it is hard to get it to work reliably especially with USB microphones and subwoofers. For FR measurements it is fine, but if you are trying to get accurate timing measurements the UMIK 1 has issues. https://www.avnirvana.com/threads/acoustic-phase-measurements.13300/#post-99941. For accurate timing measurements the best solution is a regular MIC with a 2 channel interface where you can use loopback for timing, loopback is much easier and more accurate and repeatable. For USB MIC timing measurements the UMIK 2 is more suitable as it has its own clock so is not affected by USB issues.

 

[Keith_W]

OK those measurements are much better. I am feeling generous tonight so i'll talk you through the process.

Step 1: make sure the microphone is adequately centred. Both these measurements were taken with a left speaker acoustic reference. See, this is what I don't like about REW. How can a measurement of a left speaker, using the left speaker itself as the timing reference, have a greater time discrepancy than the right speaker? The delta T should be zero. Anyway, this is something i'll ask John at some point in the future. For now, the mic is +/- 10mm so it's well centred.

Step 2: sum the left and right speakers, or use the combined measurement of both speakers. Here, you can see the comparison of my vector sum of L+R mains (pink) compared to the actual measurement of your L+R mains. You can see that the two measurements are identical, except for slight discrepancy at the top end. Probably a measurement error.

Step 2: determine subwoofer delay. In the "Overlays" section, select "Impulse response". Overlay the subwoofer with the L+R vector sum. You can see that the start of the impulses do not line up. Read the start of the impulse, I usually use the zero crossing point after the initial deflection as impulse start. So in this case, it's 21.3ms. Write this down.

Step 3: In REW's "All SPL" tab, right click and choose the Alignment tool. We will align the SW1 sub with the L+R Vector Sum. Note that I have restricted the delay range to +/- 20ms. Since we determined that the sub is 21.3ms too late, we need to delay the main speakers by 21.3ms. Play around with the gain, polarity, and fine delay whilst keeping an eye on the black curve which updates in real time. The goal is to find the combination that produces the fewest dips (not necessarily the flattest response!). I ignore peaks because you can chop them off with PEQ's. But dips ... this is where you fix them.

The best combo in my opinion was to invert the polarity of the sub and delay the mains by 28ms. Once you have found the correct combination of delays/gain/polarity, write them down. Click on "Aligned Copy" for both, then close the Alignment Tool.

Step 4: Sum the aligned sub and aligned LM speaker with "Trace Arithmetic A+B". Name this something like "L+R+SW2 Aligned".

Step 5: Repeat the exercise with the remaining subwoofer, but this time align it to "L+R+SW2 Aligned". Continue until you have aligned all subwoofers.

Step 6: (Optional) Load all these values into your DSP and take a measurement of ALL subs and speakers to confirm. If you have correctly loaded the values into your DSP unit, I assure you it will look exactly the same as REW's simulation.

Step 7: Use REW's EQ to smooth out any remaining dips.

 

[Paffi]

Interesting. As I'm always curious about different alignment techniques, may I ask:

- why you've choosen the zero crossing point for the alignment? I understand, that you've choosen it as a start for the alignment tool - I've often seen, that people align with impulse peak / valley (for that I would decrease the frequencybandwith of the L+R sweep to get a "better" suited impulse response), using the crossing is new for me.

- Why did you choose to align sub after sub with l+r instead of all subs together first and then the aligned subs in total to l+r?

 

[Keith_W]

Interesting. As I'm always curious about different alignment techniques, may I ask:

- why you've choosen the zero crossing point for the alignment? I understand, that you've choosen it as a start for the alignment tool - I've often seen, that people align with impulse peak / valley (for that I would decrease the frequencybandwith of the L+R sweep to get a "better" suited impulse response), using the crossing is new for me.

When aligning subs to mains in the impulse response, you have two goals:

1. No part of the subwoofer response should occur before the main response. If it is unavoidable, it should be low in amplitude and within the psychoacoustic threshold of inaudibility (about 20ms of pre-masking).
2. The highest peak of the subwoofer response should not occur after the main response. Ideally, it should occur together. If this is unavoidable, then it should be within the psychoacoustic threshold of inaudibility (generally about half a period of the wavelength in question).

Both goals are mutually contradictory with minimum-phase DSP. You have to choose one or the other - either the small wiggle occurs before the main impulse, or you align the small wiggle to the main impulse, then the big wiggle occurs later than ideal. One would give you "pre ringing", the other would make the subwoofer sound "slow". I prefer prioritizing the second goal because the initial deflection is small and short.

Sometimes, subwoofers have additional peaks that may be even higher in amplitude than the initial deflections (e.g. the 3rd or 4th peak). This is due to reflections and not the direct sound of the subwoofer because your brain will perceptually sum them. Some caution needs to be taken when interpreting the response.

A good example is in his MDAT. Notice how the green subwoofer response is distorted compared to the red sub? Assuming it's the same brand of subwoofer with the same response, this is due to a reflection distorting the response of the sub. It's not so clear where the main impulse is.

- Why did you choose to align sub after sub with l+r instead of all subs together first and then the aligned subs in total to l+r?

Think of it this way: each speaker and sub is a "subwoofer" with different frequency responses thanks to different positions in the room. If we are talking about mono bass, then you can think of the sum of L+R mains as a single subwoofer. Every time you add another subwoofer, the sum of all the speakers and subs is a single subwoofer.

Of course you could do it the way you describe. This way is how I do it. That post will get very long if I were to go through every possible permutation of how to align subs to mains. I am a bit of a collector of different sub aligning methods, every time I see something new, I go and experiment with it. Some methods work better than others, some are easier than others.

 

[Keith_W]

Actually, wrt my previous post, I was a bit of a numbskull. I forgot to check if the sub response would look better if the polarity was inverted. Here it is inverted - you can see that the zero crossing point is 15.8ms (and not 23.1ms in the previous post). This gives you much shorter "pre-ring" AND brings the main impulse of the sub closer to the main impulse of the speaker. Sorry about that.

 

[neRok]

A good example is in his MDAT. Notice how the green subwoofer response is distorted compared to the red sub? Assuming it's the same brand of subwoofer with the same response, this is due to a reflection distorting the response of the sub. It's not so clear where the main impulse is.

Have you applied an "octave window" to those IR's though? You can't visually/directly compare IR's that have different "windows".