Setting up 4 subs and phase alignment...

[jtalden]

i am unfamiliar with the phase tab analysis , usually i align the subs using the impulse tab overlay with a separate measurements(with timing reference).

can you please further explain the principle of using the phase tab with FDW and its benefits ?

As you suggested below there are several possible delay settings that will result in favorable SPL support. Using the FDW on the phase in the alignment tool is just one way to determine which delay setting is closest to the conventional target timing where the direct sound phase response tracks as closely as possible throughout the XO range. When the SPL support is not favorable when the FDW is removed, then it is best to adjust the delay timing as need from that point to get favorable SPL.

The method you mention below is a good an alternate method for this preliminary alignment. That is, align the initial rise the 2 impulses so they start at the same time then adjust timing from there as needed. Some people prefer to align the impulse peaks and adjust as needed from there. That can also result a favorable timing for SPL and sound quality. I have seen many impulse shapes that make this method less reliable for finding the conventional delay timing, but it often still works fine to find a favorable delay timing.

i have download the Sonnie's REW measurements ,but when overlay them at the phase tab - i just see the FR and the FL - don't see the SW,can you please advice on that ?

You can see the relative phase either by using the REW overlay phase chart or using the alignment tool.
If you want to better understand the phase and alignment tool I suggest the exercises found here. That thread has a lot of discussion regarding the alignment tool.

finally, i have always use the "sub distance tweak" (adjusting timing between the mains and subs) till i have got the flattest FR at XO region. i have read many times that you can have a very flat FR but still be off by one cycle (hence effect badly the sound quality and not being aware of that from the "graph") - what do you think of that ?its seems to me that this method of checking phase relationship between the mains and the subs -could help in that region .

There is nothing wrong with that method. It finds a favorable delay setting for SPL. If we are off by one cycle from the conventional target timing that is not likely to impact the sound quality very much. There are arguments that decreasing the delay of the SW by 1/2 or 1 cycle can sound better because there is less overall group delay. There a tradeoff of characteristics and it is difficult to know from a technical perspective what may sound better to an individual listener. I don't like to delay the SW more than a 1/2 cycle from the conventional target alignment as it is not longer a tradeoff and thus it is harder to see why that setting is more favorable from a technical perspective. I doubt that the sound quality is significantly different however.

 

[dandi36]

thank you very much for your time and answers ....i will go over the link you have suggested

 

[Sonnie Parker]

Ring a ding a ling...

Had the timing on the subs set... looked good pre-Dirac Live, but with Dirac settings (no equalization), the timing is off.

 

[AustinJerry]

This doesn't surprise me at all. I did a fairly detailed analysis as to whether my sub response was better when all subs were precisely aligned, or whether it was better with the alignment delays zeroed out. The response was significantly better without alignment, which is not the result I was expecting. I can provide a link to the analysis, if you would be interested.

I would expect Dirac Live to produce an optimized sub response but, based on the analysis I just referenced, optimized may not mean precise time alignment.

 

[Sonnie Parker]

I was thinking that was going to be the case. I remember when Mark Seaton was here, he expressed that same sentiment... time alignment is not always necessary. Of course he went into more detail.

But yes... please share that link.. I need something to read while listening a bit later.

 

[AustinJerry]

I agree that $4K for the HTP-1 plus $500 for the DLBC is getting pricey. I am only interested in bass response at the MLP, so a solution that produces more balanced bass cross the entire listening area is not high on my list. My results with Dirac Live ver 1.0 (MiniDSP 88A) with a 2x4 to consolidate my four subs is excellent. I will need serious convincing before I pay that much money to replace my setup. However, I remain open-minded and am following the HTP-1 thread on AVS very closely.

Some of you may have seen that I finally made the upgrade and have replaced my MiniDSP 88A with the Monolith HTP-1. Audio quality has improved, likely due to fewer A-D-A conversions, and overall better electronics. I have not invested in Dirac Live Bass Control at this time--not sure it will buy me anything. My frequency response measurements are fairly similar when comparing the HTP-1 and the 88A, which is not unexpected--same room, sake treatments, same speakers, same version of Dirac Live.

 

[AustinJerry]

Here it is: https://www.dropbox.com/s/xyn7vbkll...time alignment and polarity matching.pdf?dl=1

I think how alignment affects sub response is highly dependent on a number of factors, including listening room dimensions, sub and seating placement, etc. The lesson learned is not to take anything for granted--experiment, and base final decisions on measurement results.

 

[FrankV]

@ Sonnie, have you tried Multi-Sub Optimiser from Andyc? I've got great results from it with four subs and optimising together with readings from the mains.

 

[Sonnie Parker]

Some of you may have seen that I finally made the upgrade and have replaced my MiniDSP 88A with the Monolith HTP-1. Audio quality has improved, likely due to fewer A-D-A conversions, and overall better electronics. I have not invested in Dirac Live Bass Control at this time--not sure it will buy me anything. My frequency response measurements are fairly similar when comparing the HTP-1 and the 88A, which is not unexpected--same room, sake treatments, same speakers, same version of Dirac Live.

Congratulations of course. I had a feeling you were going to pull the trigger based on your participation in the AVS thread. I'm liking what I'm hearing with mine thus far, although I still have some tweaking to do.

Here it is: https://www.dropbox.com/s/xyn7vbkllny82ki/Case study--Sub time alignment and polarity matching.pdf?dl=1

I think how alignment affects sub response is highly dependent on a number of factors, including listening room dimensions, sub and seating placement, etc. The lesson learned is not to take anything for granted--experiment, and base final decisions on measurement results.

Interesting read Jerry. What I wonder is if it is okay to have the timing off on the subs. For example, mine is off by 4.6ms now. Granted, that is not much and I probably could not identify it with testing... as there are 1,000 ms in just 1 second. 4.6ms is miniscule for sure. HOWEVER, the audiophile elitists will say the subs need to be time aligned. THANKFULLY, I'm not an elitist and have used timing to help with summed frequency response since Mark Seaton shared that with me. Yet... recently, I've tried to do a better job of aligning the subs, then allowing Dirac to do it's thing, which I've not learned is the fact that DLBC using delay to achieve a better response.

From Dirac...
One part of the DLBC algorithm is the calculate a delay for each subwoofer that maximizes the energy of the sum response.
From the plot it looks like this is what has happened.
Difficult to say for sure of course but it looks like the "after" plots will sum better overall (fewer cancellations).

Seems reasonable to me.

@ Sonnie, have you tried Multi-Sub Optimiser from Andyc? I've got great results from it with four subs and optimising together with readings from the mains.

I have not... but I have looked at it. I don't think it's difficult to do by any means, but I was curious about Dirac doing it. My approach is a bit different, and only involves one listening position, but I can accomplish a very good outcome.

 

[FrankV]

I have not... but I have looked at it. I don't think it's difficult to do by any means, but I was curious about Dirac doing it. My approach is a bit different, and only involves one listening position, but I can accomplish a very good outcome.

I saw the multiple seats in your photos at the start of this thread and assumed you were looking for multi-seat fixes.

 

[Sonnie Parker]

I saw the multiple seats in your photos at the start of this thread and assumed you were looking for multi-seat fixes.

Nah... I'm the only one in the room about 999% of the time. Occasionally kids or some friends might come over, but they think it sounds good regardless of the seat, so I optimize solely for my primary listening position.

 

[FrankV]

You are fortunate! Usually it's only my wife an I but even so, trying to get even bass has been a nightmare. My room is virtually square and height is around half the width/length, so not a good start. Layout is pretty much fixed as needs to also be a "liveable room". Room modes are very strong, if I play a tone around 80Hz for example at an uncomfortably loud level when roughly in the middle of the room, when I move back and sit down it has pretty much completely gone. Quite an astonishing demonstration of how the room takes over control from whatever kit is in there! Multiple subs with MSO has done an amazing job of levelling out the modes.

 

[Sonnie Parker]

Yep... well... my wife told me many years ago she was NOT going to put with a loud subwoofer in the great room... lol. And I'm not sure I'd ever get to watch movies or listen to music like I do now because she wouldn't like it, and I would never be able to get the right placement for speakers or subs so that I can enjoy excellent sound stage, imaging and depth acuity. A dedicated room is pretty much a must to accomplish it.

 

[AustinJerry]

HOWEVER, the audiophile elitists will say the subs need to be time aligned.

I know what the elitists say, and I used to think the same way. However, the point of my exercise was to suggest that one must experiment, measure the results, and come to an educated conclusion. In my room, where there is a 9ft difference between the front and rear subs, adding the time ligament delay made the response worse. In a different room, the results could be completely different.