Understanding sub phase/group delay for placement

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
I'd be grateful if you have any observations or ideas.

SW to mains XO timing looks ideal.

Misc thoughts:
  • This SPL result is actually more in line with what I would expect based on the FDW measurements. SPL charts I had seen posted by others previously were perfectly flat SPL results. I had wondered how any EQ method can impact strong room effects like both you and I have. That shouldn't be possible. Those with flat SPL results apparently have very favorable room setups.
  • Your RT60 is near the low end of the recommended range and that suggests you have significant room treatments so they must not be a solve-all solution either.
  • Your room curve doesn't rolloff at higher frequencies. That is a bit unusual for most tunings at the LP.
 

FargateOne

Member
Thread Starter
Joined
Jun 5, 2017
Messages
223
More  
Preamp, Processor or Receiver
Rotel RSX-1562
Additional Amp
Bryston 3B3 for fronts mains
Universal / Blu-ray / CD Player
Cambridge CXU
Front Speakers
B&W 804 D2
Center Channel Speaker
B&W HTM4D2
Surround Speakers
B&W 705
Subwoofers
SVS PC-2000 and SVS PC-2000 Pro
Other Speakers or Equipment
10 PEQ filters/channel in receiver with REW
Video Display Device
Samsung UN55ES8000
Other Equipment
miniDSP 2x4 HD
May I ask a question about Group delay. I read the REW help about it and some others sources. The goal is to get the most flat line possible; this is possible , among other things, by choosing the right place for the sub. If I see a big spike or an abyssal dip those FR area can't be EQ because, very often they are at the same FR than a dip in FR response which can't be EQ. In the capture here, from the mdat at post #37, I see a flat GD graph between 100Hz and 200Hz for the L,R which tell me that I can set a filter Q and a gain of 2 or 3 db without messing with phase, ... isn't it?:hide: For the CC the dip at 164Hz says not to EQ higher? Or Is it the alpha male for all mains and it forbid to EQ those 3 channels higher than 164Hz? Or should have generated excess groupdelay response?
 

Attachments

  • GD.jpg
    GD.jpg
    63.7 KB · Views: 29

JStewart

Senior AV Addict
Supporter
Joined
Dec 5, 2017
Messages
2,040
Location
Central FL
SW to mains XO timing looks ideal.

Credit Dirac for that.

Those with flat SPL results apparently have very favorable room setups.

This was my assumption. What I did notice one time looking at some measurements Sonnie had posted (good room) was the range (peak to peak) of the HF comb filtering on his response was about 1/2 or less of what mine is using the same smoothing.

Your RT60 is near the low end of the recommended range and that suggests you have significant room treatments.

Yes it is. A little overdone for 2 channel but good for multichannel is my listening impression.

Your room curve doesn't rolloff at higher frequencies. That is a bit unusual for most tunings at the LP.

Also correct and the speakers are not toed in either which makes it more surprising. About 8ft from MLP to the speaker plane and the speakers about 8ft apart to be on either side of the sceen. I believe it to be the speakers. Other speakers I've had in the room have shown more roll off. With 62 year old ears it doesn't sound bright at all to me. So far the one benefit of aging I can point to. :)

Amazing what you are able to deduce from a set of measurements!
 

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
I don't use GD or excess GD to decide any EQ adjustments. Maybe there are others that do and will chime in. I am more comfortable using the other charts. [In my room and many other marginal rooms the unwindowed GD is basically meaningless for me. When windowed, the GD chart makes more sense, but still is difficult to interpret as the GD spike size and frequency location changes greatly as the FDW width is changed. The SPL, Wavelet Spectrogram, and waterfall charts are easier for me to interpret.]

PEQ between 100 and 200:
The sharp notch in the FL and CC SPL at 109 Hz cannot be removed using EQ. The sharp notch in the CC SPL at 164 Hz also cannot be removed. Any EQ those areas can effect the surrounding frequencies however. Changes to the speaker and LP positions can effect these notches, but often it just changes the frequency location of them. For a clearer understanding we can open the EQ panel and manually place a peak filter at a strong notch and view the SPL impact with various gain and Q changes to the filter.
 

FargateOne

Member
Thread Starter
Joined
Jun 5, 2017
Messages
223
More  
Preamp, Processor or Receiver
Rotel RSX-1562
Additional Amp
Bryston 3B3 for fronts mains
Universal / Blu-ray / CD Player
Cambridge CXU
Front Speakers
B&W 804 D2
Center Channel Speaker
B&W HTM4D2
Surround Speakers
B&W 705
Subwoofers
SVS PC-2000 and SVS PC-2000 Pro
Other Speakers or Equipment
10 PEQ filters/channel in receiver with REW
Video Display Device
Samsung UN55ES8000
Other Equipment
miniDSP 2x4 HD
With 62 year old ears it doesn't sound bright at all to me. So far the one benefit of aging I can point to. :)
You bet! I am 62 too and are my ears too!
 

thexder

New Member
Joined
Jun 19, 2017
Messages
76
There is not enough information in that chart to make any comment regarding delay timing between SW and mains. If you feel you may have a problem with SPL in the XO range and want some to confirm the delay timing, please start a new thread. Provide the basic info on your setup and attach an mdat containing 3 measurements.
Okay, I'm now ready to learn more about time aligning.
I made some measurements. Left, right and each sub separately using timing reference.
Subs are already using some HP/LP filters and phase/polarity settings. FR looks quite decent but I'm afraid the timing isn't right,
My first question is if those peaks I marked with red should all be at the same place when subs are properly time aligned?
 

Attachments

  • timing3.mdat
    12.8 MB · Views: 18
  • time-align.jpg
    time-align.jpg
    201 KB · Views: 33

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
Please confirm the situation:
  • This setup is for AV use including an AVR XO at 80 Hz
  • All 4 SWs are connected to the same LFE output
  • There is no device like a MiniDSP in the setup to change the relative delay timing between the SWs
  • The SW phase settings of measurements 3-6 are settings you chose to provide relatively smooth SPL
 

thexder

New Member
Joined
Jun 19, 2017
Messages
76
Hi,

No, no crossover is used in this case. Well, it's set to the max value - 250 Hz.
All 4 subs are connected to the same LFE output, however, my AVR is supporting two sub channels so delay for front and rear pair can be used (currently all distances are set to 0).

Yes, SUB1 phase 120, SUB2 phase 120 (later changed to 180), SUB 3 phase 0 and SUB4 phase 0 provide good SPL and good sound as well. But I fear something isn't right. I believe rear subs are misaligned.
When SUB2 phase was changed from 120 to 180 both front subs (1 and 2) got well aligned impulse plots. I'm attaching the picture.

But when I tried to align subs 3 and 4 with subs 1 and 2 (I assume the initial peaks have to be aligned) I'm getting nowhere.
Phase settings doesn't help so I tried using the delay. REW calculated the delay and when I inserted it to my AVR the plot shifted to the LEFT? In my opinion it should move to the right?
 

Attachments

  • subs4-delay.jpg
    subs4-delay.jpg
    174.4 KB · Views: 26
  • subs1+2matchingphase.jpg
    subs1+2matchingphase.jpg
    152.1 KB · Views: 27

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
I'm still not understanding completely.
In answer to your original question the alignment of the 'initial rise' (rather than the initial peak) of the impulse of each SW is a good first step, but often some adjustment from that setting will be needed to find a favorable SPL response. The final timing shift between the SW should best be kept within one wavelength of the XO frequency. The differences should be less than that if possible.
Some general comment on manually getting 4 SWs to work well together
  1. Initially align the start of the rise of the 4 impulses to the same time.
  2. If the SPL is not favorable for EQ when all 4 are measured together then:
    • Adjust the delay of 1 or more SWs and see if that helps the SPL.
    • Adjust the phase of 1 or more SWs and see if that helps the SPL.
    • Invert the polarity of 1 or more SWs and see if that helps the SPL.
When SPL looks favorable for EQ then confirm that the ETC, step response and spectrogram charts are reasonable.​
Manually doing this can be very time consuming.​
After the 4 SWs are working well together then it is time to align the SWgroup with the mains.​

Some observations on the Post-56 data
  • The SWs are all rolling off as if for a 80Hz so maybe the SWs LPF is set there.
  • It would be better to have closer timing between the 4 SWs You have no ability to time align all 4 independently, but you can manually change the delay of front vs rear SWs and adjust that delay timing as needed. Audyssey can do this automatically I think. SWs 1 and 2 are delayed roughly 8 ms more than SWs 3 and 4. I would initially try to keep the differences within 2 or 3 ms of each other if possible.
  • The current SPL of measurements 3-6 looks passible for EQ, but could probably be better. The other charts suggest the same. They will look better once the EQ is applied.
 

thexder

New Member
Joined
Jun 19, 2017
Messages
76
Well, I did this backwards which was probably a mistake. I first set up filters and gains and then checked alignment. All this with mains included. I checked the impulse response at the end.
But why the impulse response of the second (rear) pair shifts to the left when delay is set? I don't see any other way to get those initial rises together. Phase settings aren't enough and the delay is shifting them in the wrong direction.
And why there are many people saying that concept of time alignment at less than 200 Hz is not important?

What am I missing here?

XO in the AVR is not activated, however, subs do have HP/LP filters and one sub goes to 40 Hz and others to 80/85 Hz. This was necessary due to better SPL.
I'm leaving Audyssey out of this, this is my stereo setup and Audyssey does not sound good when listening to stereo.

Just to be sure: Initial irse is first big peak? And the little rise tells us if polarities match?
Can timing reference be seen on an impulse response?
 

Attachments

  • initial_peak.jpg
    initial_peak.jpg
    153.4 KB · Views: 22
Last edited:

FargateOne

Member
Thread Starter
Joined
Jun 5, 2017
Messages
223
More  
Preamp, Processor or Receiver
Rotel RSX-1562
Additional Amp
Bryston 3B3 for fronts mains
Universal / Blu-ray / CD Player
Cambridge CXU
Front Speakers
B&W 804 D2
Center Channel Speaker
B&W HTM4D2
Surround Speakers
B&W 705
Subwoofers
SVS PC-2000 and SVS PC-2000 Pro
Other Speakers or Equipment
10 PEQ filters/channel in receiver with REW
Video Display Device
Samsung UN55ES8000
Other Equipment
miniDSP 2x4 HD
Just to be sure: Initial irse is first big peak? And the little rise tells us if polarities match?

If I am right, the big peak is the initial signal (if I am wrong we will know quickly :redgrin:); all peaks must be at the same place. Polarity is easy to see: the little rise and the big one are upside down from the reference.
 

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
Well, I did this backwards which was probably a mistake. I first set up filters and gains and then checked alignment. All this with mains included. I checked the impulse response at the end.

No matter the order, a favorable result is the important thing. The order is just to suggested a more logical and efficient approach to finding a favorable setting.

But why the impulse response of the second (rear) pair shifts to the left when delay is set?

How was the delay changed, if it was by a LFE-2 (rear) distance increase in the AVR, that results in a reduction in delay (IRs moves left). So the correct method would be to instead increase the distance on LFE-1 (front). That will reduce the delay on the front SWs to match the timing of the rear SWs.

[/QUOTE]Phase settings aren't enough and the delay is shifting them in the wrong direction.[/QUOTE]

Phase settings on the SW probably do not change the actual timing, but instead delays the arrival of the lower frequencies longer than they would be with a 0° phase setting.

And why there are many people saying that concept of time alignment at less than 200 Hz is not important?

I don't know, but probably because the SPL is the more dominate characteristic for sound quality. That said, It is best to not ignore the other charts I mentioned above. We don't want excessive group delay or excessive ringing to occur due to excessive delay differences or excessive EQ.

XO in the AVR is not activated, however, subs do have HP/LP filters and one sub goes to 40 Hz and others to 80/85 Hz. This was necessary due to better SPL. I'm leaving Audyssey out of this, this is my stereo setup and Audyssey does not sound good when listening to stereo.

Understood.

[/QUOTE]Just to be sure: Initial irse is first big peak? And the little rise tells us if polarities match?
Can timing reference be seen on an impulse response?
[/QUOTE]

The relative timing can normally be seen by the initial rise in the IR as in this case). In some cases it may be difficult to identify the initial rise time due to various influencing factors. REW can make the alignment automatically for you using 'All SPL/Controls/Align IR Start' with the 4 SW selected. You are past that feature now unless you intend to start the process over again.

Below is the result of aligning the initial timing using SW measurements 3-6 in Post-26 the resulting SPL is not particularly favorable as there are large dips at 56 and 97 Hz. It still represents a good reference point. Don't stray any further than necessary from this timing to achieve a more favorable SPL.

29512


29513
 

Attachments

  • ja2post56.mdat
    12.9 MB · Views: 13

thexder

New Member
Joined
Jun 19, 2017
Messages
76
OK, thank you very much for your effort.
Looking at the plots you kindly provided I don't see a point bothering with the delay as it creates less favorable SPL.

I did some more testing: closed the ports on mains, left the delay untouched and changed some slopes.
I got this. Sounds pretty cool, I believe the best so far. My target is Harman curve and I came pretty close.

PS: I have learned that looking only at the SPL is not enough. This way you can't see ringing issues which may be solved by placing treatment or equipment on slightly different locations. That would not change the SPL but int might improve decay times.
 

Attachments

  • noBRtry.mdat
    2.8 MB · Views: 15
  • house.jpg
    house.jpg
    68.5 KB · Views: 24
  • harman.jpg
    harman.jpg
    13.9 KB · Views: 21

thexder

New Member
Joined
Jun 19, 2017
Messages
76
@jtalden ... just to confirm, when matching rises in the IR plots you synced the first peaks? Many other articles suggest this should be done for first big "wave"?

I have never looked the step response plot. Is there perhaps a quick explanation what should one pay attention at?
 

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
I suggested you use the initial rise of the impulse as a starting point for any timing adjustments. The first chart in post-62 shows that alignment. The cursor is at 0 ms. All 4 SWs are aligned such that the initial sound arrives at the that point. In some other cases I have suggested the first peak as good starting point as these often result in very similar timing.
I normally start my personal alignment efforts by instead finding the best phase tracking of the direct sound through the XO range. This is a more complicated starting point to find and may not really be significantly better than others in finding a favorable SPL at the LP. I still often chose to adjust timing from that point the minimal amount needed to find a more favorable phase tracking and SPL at the LP.
If all subs were identical and setup identically for polarity, phase setting, LPF, HPF and whatever other settings they may have, and were placed in anechoic space, then the IR shapes would all be identical. It would not matter which IR feature was chosen to time align them. Once different; SW types, SW settings, room locations or EQ is introduced then the IR shapes are greatly changed. I am not aware of any simple IR feature that will result in the best initial timing alignment in every situation.
I didn't choose aligning the largest peak only because I have see situations where the largest peak is hard to identify and others where it occurs very delayed from first sound arrival. However, I do not know that choosing that as a starting point is necessarily less attractive in most cases. I have no significant experience actually using that as a starting point.

Regarding step response:
For multi-SW setups we again are looking to keep the final step response reasonably compact as opposed to multiple significant peaks spread out over time. Since room effects, EQ and other factors influence this, this is more helpful when comparing alternative setups. There is no absolute target. It is another chart that helps us to understand the overall tradeoffs.
 

thexder

New Member
Joined
Jun 19, 2017
Messages
76
Sorry, I don't really understand this.
I get that initial rise or first peak should be a starting point for further time adjustments. But what else is there to align if they're already aligned?

What do you think of this alignment? It's obvious that two subs have reversed polarity but the sound is better. The bass sound more synced, I guess.
However, SPL is slightly better than yours but worse than the one in post-63.
Decay time at 43 Hz is better too and I'm starting to suspect that this is a vital part when it comes to liking the bass.

Another problem here is that I cannot set the distance more than 6 meters. AVR suggests to relocate the speaker.
And why such long distances? All subs are less than 4 meters away! Front subs are 35 cm away from mains and yet I need 6 meters of distance to align the impulses. What am I missing here?
 

Attachments

  • 4subs_neg.jpg
    4subs_neg.jpg
    154.4 KB · Views: 27
  • subs1234average_reversed_polarity.jpg
    subs1234average_reversed_polarity.jpg
    67.3 KB · Views: 23
  • delayd.jpg
    delayd.jpg
    98.6 KB · Views: 20
Last edited:

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
Sorry, I keep forgetting or not fully understanding the capabilities and limitations of your stereo setup. I have no experience with this type of setup. My understanding is that you have a double bass type of setup that will send the bass frequencies to the L, R mains and also to the 2 LFE outputs. The LFE output frequency range will be dependent on the setting of the XO in AVR even though the entire range is also sent to the L, R.

I get that initial rise or first peak should be a starting point for further time adjustments. But what else is there to align if they're already aligned?

If I'm thinking correctly you have the ability change the timing between the 2 front and 2 rear SWs. Presumably, the SPL response with the 2 sets best aligned was not favorable for SPL. It was probably similar to what I found when I time aligning all 4 SWs using REW trace arithmetic. You also have already experimented with; offsetting LFE timing, adjusting the phase controls, and changing the polarity of one or more SWs. There is nothing else to adjust in setting up the 4 SWgroup unless you want to change SW/LP locations or room treatments.

What do you think of this alignment? It's obvious that two subs have reversed polarity but the sound is better. The bass sound more synced, I guess.
However, SPL isn't better.

You may want to post this mdat for this setup along with the mdat for the previous best setup. We can't really tell anything much by just this impulse response chart. With 4 SWs setup differently and distributed differently in the room, an IR chart like that may represent one of the best setups or one of the worst setups. There is nothing inherently wrong with 1 or 2 of the SW being the opposite polarity if that is the concern. Sometimes that is helpful to the overall response.

It was suggested to initially start with all the same SW settings and close initial timing as a reasonable starting point. If it does not look good for SPL make adjustment using one or more of the available controls mentioned above to try and find a better SPL response. When one is found that setup can be used as the new target and the search continued to find even better settings. At some point it is no longer easy to improve on the target and we accept that last target as our favorable alignment. You have already been doing this a have moved from those initial settings to gain improvement. Ideally each new improved target was compared to the previous one using all the relevant charts to confirm no serious negative tradeoffs were made. Have you now ended this process and identified the SWgroup settings that will be used?

Another problem here is that I cannot set the distance more than 6 meters. AVR suggests to relocate the speaker.
And why such long distances? All subs are less than 4 meters away! Front subs are 35 cm away from mains and yet I need 6 meters of distance to align the impulses. What am I missing here?

??, I may not follow. There appears to be either 2 subjects intermingled here or we are now discussing the subject of timing the 4 SWgroup to mains. The setup of the 4 SWs to each other is not impacted by the timing to the mains. If you haven't yet settled on the settings for the SWgroup then complete that process first.

The next step it then to adjust the timing between the SW group and mains. If your AVR auto setup routine is suggesting more than a 6 m change to the timing of L and R or LFE-1 or LFE-2 you are correct it suggests a problem. What is the AVR XO setting? I think it should be 80 Hz or so to match the SWgroup setup that is rolling off at about that frequency. Why is the AVR suggesting distances? I thought you were going for a completely manual setup with no EQ. Maybe you ran it provide EQ settings? If you want that EQ help then possibly the SWgroup to mains delay can still be manually reset to the proper values after the AVR auto setup routine is run?

Remember I am new to this type of setup and may not fully understand the situation, so if we are not making progress that may be the reason.
 

thexder

New Member
Joined
Jun 19, 2017
Messages
76
Sorry, I keep forgetting or not fully understanding the capabilities and limitations of your stereo setup. I have no experience with this type of setup. My understanding is that you have a double bass type of setup that will send the bass frequencies to the L, R mains and also to the 2 LFE outputs. The LFE output frequency range will be dependent on the setting of the XO in AVR even though the entire range is also sent to the L, R.
No need to be sorry, I don't see you having any issues understanding. So far your help was spot on. My lack of understanding measurements and a language barrier are more likely the reasons for some extra words here. And probably some strange questions.
All these questions are referring to stereo listening.
For movies I will simply run Audyssey (hence the equdistant sub locations), changed it from reference to flat and set all crossovers to 80 Hz. And perhaps turn up sub volumes for a couple of dB.

Do you think this setup is all that different? Yes, bass is redirected to mains and LFE but that's also the case with a regular setup, isn't it? The only difference is the lack of XO. The method I'm trying was proposed by dr. Geddes. He suggests at least 3 subs should be placed randomly, each with different HP/LP filters and then globally EQd at the end. He thinks time alignment below 200 Hz is not important. Mains should be non ported and full range. More here: https://mehlau.net/audio/multisub_geddes/


If I'm thinking correctly you have the ability change the timing between the 2 front and 2 rear SWs. Presumably, the SPL response with the 2 sets best aligned was not favorable for SPL. It was probably similar to what I found when I time aligning all 4 SWs using REW trace arithmetic. You also have already experimented with; offsetting LFE timing, adjusting the phase controls, and changing the polarity of one or more SWs. There is nothing else to adjust in setting up the 4 SWgroup unless you want to change SW/LP locations or room treatments.
You are thinking correctly. I can change timing for 2 front and 2 rear subs which are equidistant. I didn't bother much with timing as I got pretty good results with phase/filter settings (so I was not using delay/distance). When we started to talk about IR plots and alignment, I tried to align impulses but my AVR settings are obviously too clumsy for that. Probably why many people are using miniDSP 2x4 for delay settings.


You may want to post this mdat for this setup along with the mdat for the previous best setup. We can't really tell anything much by just this impulse response chart. With 4 SWs setup differently and distributed differently in the room, an IR chart like that may represent one of the best setups or one of the worst setups. There is nothing inherently wrong with 1 or 2 of the SW being the opposite polarity if that is the concern. Sometimes that is helpful to the overall response.
Happy to. Please, take a look. Since this is a stereo setup, the lack of SPL below 45 Hz may not be an issue.
My ears can confirm that oposite polarity can sound good. Probably reversed polarity isn't a good thing when using auto correction systems. Hence, if I'm not mistaken, Audyssey suggests all subs should be set to 00+ prior to auto calibration.


It was suggested to initially start with all the same SW settings and close initial timing as a reasonable starting point. If it does not look good for SPL make adjustment using one or more of the available controls mentioned above to try and find a better SPL response. When one is found that setup can be used as the new target and the search continued to find even better settings. At some point it is no longer easy to improve on the target and we accept that last target as our favorable alignment. You have already been doing this a have moved from those initial settings to gain improvement. Ideally each new improved target was compared to the previous one using all the relevant charts to confirm no serious negative tradeoffs were made. Have you now ended this process and identified the SWgroup settings that will be used?
I agree. I will stop when my ears will be happy no matter what plots will say. I'm pretty close now. I'm eager to try my initial setup I had couple of years ago where front subs were next to the mains (now they're behind mains). I'm curious what will happen in the impulse plots.
If I may ask, what was (is?) your setup technique? Crossover at 80, all subs with same phase/polarity, no HP/LF/slope settings, just individual delay for each sub?


??, I may not follow. There appears to be either 2 subjects intermingled here or we are now discussing the subject of timing the 4 SWgroup to mains. The setup of the 4 SWs to each other is not impacted by the timing to the mains. If you haven't yet settled on the settings for the SWgroup then complete that process first.
I'm aware of that. Since you tought me more about IR plots I looked and compared impulse plots of mains and subs. More on that below.


The next step it then to adjust the timing between the SW group and mains. If your AVR auto setup routine is suggesting more than a 6 m change to the timing of L and R or LFE-1 or LFE-2 you are correct it suggests a problem. What is the AVR XO setting? I think it should be 80 Hz or so to match the SWgroup setup that is rolling off at about that frequency. Why is the AVR suggesting distances? I thought you were going for a completely manual setup with no EQ. Maybe you ran it provide EQ settings? If you want that EQ help then possibly the SWgroup to mains delay can still be manually reset to the proper values after the AVR auto setup routine is run?
That's what I was trying to do but was dissapointed when I saw limited options in the AVR. AVR didn't suggest anything (yes, I'm going for a full manual setup with XO set to max value - 250 Hz). I checked the impulse of mains and subs and measure it with REW. Only then I was able to see I need more than 6 meters of distance to align subs and mains.
Which is still a mystery to me since actual distance is 3-4 meters but on a plot we can clearly see REW measured 6,5 meters.

If everything fails I was going to leave everything the way you see in the mdat file above.


What is the AVR XO setting? I think it should be 80 Hz or so to match the SWgroup setup that is rolling off at about that frequency
Does a crossover affect the time alignment between subs and mains?


Remember I am new to this type of setup and may not fully understand the situation, so if we are not making progress that may be the reason.
Oh, we're making progress, I'm learning a lot here! I'm really thankful for your time and effort.
 

Attachments

  • testnobrB.mdat
    1.4 MB · Views: 18
Last edited:

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
Do you think this setup is all that different? Yes, bass is redirected to mains and LFE but that's also the case with a regular setup, isn't it? The only difference is the lack of XO.

I have a modest DIY AV 5.1 system setup for the highest quality stereo sound my system can muster. If I were to recommend or design another stereo system it also would be a 2.1 system as I think that decoupling the bass from the mains speaker locations is normally a big advantage due to SW location flexibility. I am open to the idea that in some cases a larger overlap of the bass range may be beneficial, but wouldn't expect it to have an inherent advantage in most cases. It is just another setup option in my book. The lack of EQ in the bass range is a significant factor unless a smooth response can be found without it.

The method I'm trying was proposed by dr. Geddes. He suggests at least 3 subs should be placed randomly, each with different HP/LP filters and then globally EQd at the end. He thinks time alignment below 200 Hz is not important. Mains should be non ported and full range. More here: https://mehlau.net/audio/multisub_geddes/
https://mehlau.net/audio/multisub_geddes/

There are lots of ways to find a favorable multi SW setup.

https://mehlau.net/audio/multisub_geddes/
My ears can confirm that oposite polarity can sound good. Probably reversed polarity isn't a good thing when using auto correction systems. Hence, if I'm not mistaken, Audyssey suggests all subs should be set to 00+ prior to auto calibration.https://mehlau.net/audio/multisub_geddes/
https://mehlau.net/audio/multisub_geddes/

There is nothing inherently bad or wrong with using opposite polarity if it results in a more favorable tradeoff of response factors. I don't know if it confuses the auto setup routines or not; I wouldn't think so. Their recommendations most likely are just reflecting the industry standard target of an acoustical LR4 XO.

https://mehlau.net/audio/multisub_geddes/
If I may ask, what was (is?) your setup technique? Crossover at 80, all subs with same phase/polarity, no HP/LF/slope settings, just individual delay for each sub?https://mehlau.net/audio/multisub_geddes/
https://mehlau.net/audio/multisub_geddes/

My equipment is listed. For the SWs XO my current approach is; time align 2 passive SWs, set a 100 Hz AVR XO, augment the resulting acoustic XO using a DCX filters and EQ to best approach the targeted LR4 shape acoustically. So, that entails using the same polarities for my SWs and midwoofers. I have had excellent result using several other XO options as well and also find the bass sound quality to be noticeably different between some of the favorable options.

https://mehlau.net/audio/multisub_geddes/
That's what I was trying to do but was dissapointed when I saw limited options in the AVR. AVR didn't suggest anything (yes, I'm going for a full manual setup with XO set to max value - 250 Hz). I checked the impulse of mains and subs and measure it with REW. Only then I was able to see I need more than 6 meters of distance to align subs and mains.
Which is still a mystery to me since actual distance is 3-4 meters but on a plot we can clearly see REW measured 6,5 meters.https://mehlau.net/audio/multisub_geddes/
https://mehlau.net/audio/multisub_geddes/

Per the note below, will provide a setup option for you to consider if you can accommodate it in your setup.

https://mehlau.net/audio/multisub_geddes/
Does a crossover affect the time alignment between subs and mains?https://mehlau.net/audio/multisub_geddes/
https://mehlau.net/audio/multisub_geddes/

I would not think it is an issue in this case as the < 250 Hz frequencies will be set to the SWs, but the SWs LPF will effectively filter out the ones above 130 Hz anyway. The 250 Hz XO filter setting would impact the phase of the signal directed to the LFE output so a 80 or 100 or other setting would have some effect on the relative timing. The mains and SWs distance settings also impact that so it is not possible to know what settings are best without a full analysis. The handoff with your current and previous XO settings appear to work favorably so there is nothing to worry about until there is an issue.

Regarding the new measurements
The previous setup vs this new setup both appear to me to be reasonable settings. Chose the one you prefer. I did use this data to find another optional setting that is a little better in reducing group delay and providing a smoother SPL response above 40 Hz. I will post the details later. It has been a long session this morning and I have other responsibilities.
 

thexder

New Member
Joined
Jun 19, 2017
Messages
76
Okay, no problem, absolutely no rush here. Been dealing with this for a couple of months so it's not really a priority. ;)
I'm now off to bed and will reply to post #69 tomorrow.
If you didn't start with the measurement yet - here is fresh one. Slightly improved.
Thank you very much!

EDIT: added modified mdat file with separate subs measurements.
 

Attachments

  • 25-2su.mdat
    528.4 KB · Views: 16
Last edited:

thexder

New Member
Joined
Jun 19, 2017
Messages
76
I have a modest DIY AV 5.1 system setup for the highest quality stereo sound my system can muster. If I were to recommend or design another stereo system it also would be a 2.1 system as I think that decoupling the bass from the mains speaker locations is normally a big advantage due to SW location flexibility. I am open to the idea that in some cases a larger overlap of the bass range may be beneficial, but wouldn't expect it to have an inherent advantage in most cases. It is just another setup option in my book. The lack of EQ in the bass range is a significant factor unless a smooth response can be found without it.
By 2,1 you of course mean more than 1 subwoofer?
What is your suggestion when using EQ? SInce I haven't got any useful information my theory is to treat only peaks at frequencies below 200 Hz.
I have very little experience with EQ as I always hear some anomalies in the sound. The best sound so far I had was with no EQ. Probably I was doing something wrong. I didn't use any external device. Just built-in 8-band PEQ on the subs.

My equipment is listed. For the SWs XO my current approach is; time align 2 passive SWs, set a 100 Hz AVR XO, augment the resulting acoustic XO using a DCX filters and EQ to best approach the targeted LR4 shape acoustically. So, that entails using the same polarities for my SWs and midwoofers. I have had excellent result using several other XO options as well and also find the bass sound quality to be noticeably different between some of the favorable options.
Looking at the DCX2496 rear panel. This device is able to set the individual settings (PEQ, high/low pass, delay, slope ...) for 4 individual subs?

Per the note below, will provide a setup option for you to consider if you can accommodate it in your setup.
Thank you, today I uploaded a new file. The one posted yesterday had no sub measurements.
I wonder why there is a lack of SPL between 40-80 Hz. I can't get this area up no matter what I do. Could taming the 45 Hz with EQ and then raising the subs volume help?
 
Last edited:

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
By 2,1 you of course mean more than 1 subwoofer?

Yes, 2.1 just indicates the processor outputs (L, R, SW). The number of SWs attached is optional. I know others often use other notations and don't know what notation is more correct. It's a big advantage to have the .1 output split in 2 (LFE-1, LFE-2) only if there is separate delay/distance settings provided for them in the AVR. It may help avoid the need for a separate box for SW delay settings.

What is your suggestion when using EQ? SInce I haven't got any useful information my theory is to treat only peaks at frequencies below 200 Hz.
I have very little experience with EQ as I always hear some anomalies in the sound. The best sound so far I had was with no EQ. Probably I was doing something wrong. I didn't use any external device. Just built-in 8-band PEQ on the subs.

Room EQ to address SPL <200 Hz is commonly accepted as a good idea.
There shouldn't be any disadvantage in using SW PEQs unless they don't provide enough flexibility of the parameters. It is important to get a good starting point with room treatment and/or SW/LP locations. Modest PEQ is then very effective. We all work within our various constraints and make compromises. I do prefer to weight cuts a little more than boosts as it seems to work better when considering the overall impact on the SPL and spectrogram. Each situation is different however.

You have been having trouble finding favorable settings with the current layout of the room. The 2 front SWs have the identical large SPL peak and it would be good to find a different placement for one of them. If your LP is too close to the back wall that can be a big part of the issue.

Looking at the DCX2496 rear panel. This device is able to set the individual settings (PEQ, high/low pass, delay, slope ...) for 4 individual subs?

Yes, the LT version has all the controls needed for 4 SW management. It is not a very reliable unit however and there are many detractors. The MiniDSP line is much more popular.

Thank you, today I uploaded a new file. The one posted yesterday had no sub measurements.
I wonder why there is a lack of SPL between 40-80 Hz. I can't get this area up no matter what I do. Could taming the 45 Hz with EQ and then raising the subs volume help?

I only took a quick look at that setup. It did not appear to be a particularly favorable one, but it was not a careful analysis.

The more favorable settings I found based on Post-58 may not be possible depending on those AVR speaker distance settings. If you provide those distances, I will see if it is worth any further consideration.
LFE-1 (Front SWs) = ?
LFE-2 (Rear SWs) = ?
L (Left main) = ?
R (Right main) = ?
 

thexder

New Member
Joined
Jun 19, 2017
Messages
76
You mean current distance settings in AVR?
They're all set to 0.
Impulses were aligned with sub settings.
 

thexder

New Member
Joined
Jun 19, 2017
Messages
76
Before I put these tests to rest I'd like to share some (probably very uncommon) facts I encountered yesterday.
1. EQ experiment
I tried to boost 63 and 80 Hz at the FR I posted in post #70. I did three tests with Q1 Q2,5 and Q5 on both front subs.
The result was even bigger peak at 45 Hz. I find this very weird.

2. Location experiment
Now, time for some heavy lifting. I moved front subs to a new place and re-measure them. Turned out their response is quite good - way better than before. I used delay to time align them with mains. To get even nicer align I changed phase on one sub as well. Both needed reversed polarity. IR graph is now perfect.
But when both subs were measured with mains - the FR is very bad.

@jtalden , when you'll have some spare time, would you please care too look and make a comment? This could look as a phase issue ...
 

Attachments

  • experiment2.mdat
    386.9 KB · Views: 12

jtalden

Senior Member
Joined
May 22, 2017
Messages
887
Location
Arizona, USA
More  
Preamp, Processor or Receiver
Marantz AV7705 Pre/Pro
Main Amp
VTV 6 chnl NC252MP P-amp x 2
Additional Amp
Behringer DCX2496 x 2
Universal / Blu-ray / CD Player
OPPO BDP-103 Universal Player
Front Speakers
DIY SEAS H1456/H1212 Spkr x 5
Subwoofers
DIY JBL 2235H 15" SW x 2
Video Display Device
JVC DLA-X790R
Screen
Da-Lite Da-Snap 39105V - 92"
Just finished my effort to suggest favorable settings based on the Post-68 mdat:
Below are 2 favorable setup adjustments.
29606


Option 1 vs Option 2
29615


Option 1
29616


Option 1
29611


Option 2
29617


Option 2
29613


Both these options look favorable to me. The last 3 columns detail the changes needed. Note that the rear SWs started as negative polarity in the reference settings so they remain that way in option 1. They changed to positive polarity in Option 2. The SPL is similar, but 2 setting options are provided as there may well be sound quality difference between these 2 setups.
 
Top Bottom