or Connect
AVS › AVS Forum › Audio › Subwoofers, Bass, and Transducers › phase vs. distance?
New Posts  All Forums:Forum Nav:

phase vs. distance? - Page 3

post #61 of 95
Quote:
Originally Posted by Danne2 View Post

or adjust to get the best compromise in the whole bass region? that is, not max SPL at the XO point, but overall the "best" FR in the bass region?
That would be it. Keep in mind that every response dip and peak is caused by time/phase interactions, and that not only does every source introduce time/phase issues, but so does every boundary within the room, each acting differently with every source, and with every listening position. The more sources you have the more they tend to even each other out, but you still have to make compromises in trying to get the best overall result.
post #62 of 95
Quote:
Originally Posted by Bill Fitzmaurice View Post

That would be it. Keep in mind that every response dip and peak is caused by time/phase interactions, and that not only does every source introduce time/phase issues, but so does every boundary within the room, each acting differently with every source, and with every listening position. The more sources you have the more they tend to even each other out, but you still have to make compromises in trying to get the best overall result.

+1
post #63 of 95
Quote:
Originally Posted by braveheart123 View Post

Just phase align the sub with mains at the crossover point. Your surrounds are always misaligned even in a perfectly phase matched 5.1 or above system. Let the avr take care of that. You cannot phase align all the speakers with the sub. Although, one can achieve that also but then the only listening option is 'All Channel Stereo" where every speaker is playing the same set of frequencies. You cannot use the same setting for DD and DTS; as it will ruin your sound stage.

I do only phase align the subs with the mains. The surrounds are not into that mix.
But can someone give a good explanation why it's ok/good to create a rise or dip in the frequency response at other places than the XO point, just to get the XO point to play max SPL?
post #64 of 95
Quote:
But can someone give a good explanation why it's ok/good to create a rise or dip in the frequency response at other places than the XO point, just to get the XO point to play max SPL?

Coz you can handle only one XO point. What happens at other frequencies depends on your room acoustics. A PEQ device such as BFD, SMS, miniDSP can be used to flatten the response.
post #65 of 95
Quote:
Originally Posted by Danne2 View Post

I do only phase align the subs with the mains. The surrounds are not into that mix.
But can someone give a good explanation why it's ok/good to create a rise or dip in the frequency response at other places than the XO point, just to get the XO point to play max SPL?

All things being equal, it's better to have your subs and fronts in phase at the xover than not, but it's only part of a comprehensive solution. Like I mentioned earlier, real phase knobs on subs are vestiges of when people used a sub in a stereo system and it's purpose was to put the sub and fronts in proper phase. The introduction of 5 speaker and higher systems made the phase knob obsolete. You can adjust the sound by playing with the phase knob, but it's probably best to leave the phase at 0 and let the receiver adjust the delays to each channel to keep things in phase. Otherwise, you're more likely to screw things up than help if you change the phase knob. There are too many sound sources to try to manually adjust everything. Software is much better at it than people.

The only exception is if you have more than one sub on a single channel whereby you can't have different delay settings for each sub. In this case, it may be a good idea to adjust the phase to keep your 2 subs in phase with each other. Ideally you would want your phase knob to be a delay knob instead since there is no xover frequency between your subs. You're trying to get them in phase across their entire frequency domain.
post #66 of 95
Quote:
Originally Posted by KidHorn View Post

The only exception is if you have more than one sub on a single channel whereby you can't have different delay settings for each sub. In this case, it may be a good idea to adjust the phase to keep your 2 subs in phase with each other. Ideally you would want your phase knob to be a delay knob instead since there is no xover frequency between your subs. You're trying to get them in phase across their entire frequency domain.

I have more than one sub on a single channel, that's why this is an issue.

Brings up another interesting thing. With the above sentence in mind, some measure the distance between the subs to the listening position and average those. That average distance is put into the AVR as sub distance, for the room correction.
Why not instead put in the distance to the closest sub, and then just increase Delay on the the other subs, increasingly as they are further away from the listening position ?
Anyone has any ideas about that?
post #67 of 95
Quote:
Why not instead put in the distance to the closest sub, and then just increase Delay on the the other subs, increasingly as they are further away from the listening position ?
Anyone has any ideas about that?

You can only do that with receivers that have independent distance settings for 2 subs. For any other avr, treat both subs as one and use PEQ to tame the combined response with one set of FEWER filters.
post #68 of 95
Quote:
Originally Posted by braveheart123 View Post

You can only do that with receivers that have independent distance settings for 2 subs. For any other avr, treat both subs as one and use PEQ to tame the combined response with one set of FEWER filters.

I meant to set the distance to the closest sub into the AVR as the sub distance, and then increase Delay on the subs themselves, on the subs Delay dials. Never touch the AVR delay.
To me that sounds like a better idea. Are there any pitfalls with that, which I have missed? Guess there are some reason why people average the sub distances instead of my suggestion, but I have never seen an explanation to why they do it.
post #69 of 95
Quote:
Originally Posted by Danne2 View Post


I do only phase align the subs with the mains. The surrounds are not into that mix.
But can someone give a good explanation why it's ok/good to create a rise or dip in the frequency response at other places than the XO point, just to get the XO point to play max SPL?

For reference see my first post in this thread, http://www.avsforum.com/t/1465159/whats-more-important-flat-response-or-phase-timing-of-subwoofer#post_23124743

 

You can ignore the blue line for now, but the black line is with the audyssey set sub distance measuring the combined sub/left speaker interaction.  Orange line is my tweaked distance/phase setting.  Notice that the change is not focused on the 80hz  crossover setting used, rather I got a more broad boost in bass of about 3-5db across most of my subwoofer range.

post #70 of 95
Quote:
Originally Posted by Danne2 View Post

But can someone give a good explanation why it's ok/good to create a rise or dip in the frequency response at other places than the XO point, just to get the XO point to play max SPL?
It isn't. In the vast majority of cases users have only one sub, and the main concern is time/phase aligning that sub to the mains at the crossover frequency, as that's where issues are most likely to arise. If you have more than one sub you introduce more time/phase issues, being between not only each sub and the mains but between the subs themselves. You then set up the system to get the best overall result, not just the best mating of one sub to the mains. The good news is that multiple subs tend to lessen room modes, so what additional concerns they raise usually are more than made up for by their benefits.
post #71 of 95
Quote:
Originally Posted by Bill Fitzmaurice View Post

It isn't. In the vast majority of cases users have only one sub, and the main concern is time/phase aligning that sub to the mains at the crossover frequency, as that's where issues are most likely to arise. If you have more than one sub you introduce more time/phase issues, being between not only each sub and the mains but between the subs themselves. You then set up the system to get the best overall result, not just the best mating of one sub to the mains. The good news is that multiple subs tend to lessen room modes, so what additional concerns they raise usually are more than made up for by their benefits.

Thank you! What I thought, and several subs certainly helps to flatten FR.
Then I will continue like I have done, with the whole bass region in mind when adjusting phases on the subs, to not introduce bigger problems somewhere else.

Quote:
Originally Posted by primetimeguy View Post

For reference see my first post in this thread, http://www.avsforum.com/t/1465159/whats-more-important-flat-response-or-phase-timing-of-subwoofer#post_23124743

You can ignore the blue line for now, but the black line is with the audyssey set sub distance measuring the combined sub/left speaker interaction.  Orange line is my tweaked distance/phase setting.  Notice that the change is not focused on the 80hz  crossover setting used, rather I got a more broad boost in bass of about 3-5db across most of my subwoofer range.

Interesting. Similiar to my question, but this was with one sub.
I wanted to have confirmed if there is a thinking error on my part, since I use several subs, I thought just averaging the sub distance seemed so rough, and wondered if just taking the closest sub distance and put that into the AVR. Then adjust the other further away subs with their own Delay dial.
Would like to understand if this is theoretically a bad idea. I have tried some different approaches, but not as the one I suggest here.
post #72 of 95
Quote:
Originally Posted by Danne2 View Post


Thank you! What I thought, and several subs certainly helps to flatten FR.
Then I will continue like I have done, with the whole bass region in mind when adjusting phases on the subs, to not introduce bigger problems somewhere else.
Interesting. Similiar to my question, but this was with one sub.
I wanted to have confirmed if there is a thinking error on my part, since I use several subs, I thought just averaging the sub distance seemed so rough, and wondered if just taking the closest sub distance and put that into the AVR. Then adjust the other further away subs with their own Delay dial.
Would like to understand if this is theoretically a bad idea. I have tried some different approaches, but not as the one I suggest here.

Actually, I do have two subs, just using y splitter on sub output from AVR.  So those graphs are both subs plus left speaker.  Subs are located along front wall, 1/4width, and therefore both same distance from listening position.  My problem was because of the large phase change with my ported mains that go down to 40hz or less.

post #73 of 95
Quote:
Originally Posted by primetimeguy View Post

...Subs are located along front wall, 1/4width, and therefore both same distance from listening position...  

That still makes it as a one sub setup in this regard, since they have the same distance to the listening position. And you added on to sub distance, that might be the same as I ask for, but I'm not sure if taking the distance to the neareast sub into the AVR as sub distance, and then just add on the subs Delay dail for further away subs. Maybe someone with more theoretical knowledge than me can provide some input to that.
Also if someone knows the reason why averaging the sub distances for input to AVR is a good thing, given the subs have different distances to the listening position. In my case one at front wall, one at back wall.
post #74 of 95
Quote:
Originally Posted by Danne2 View Post


That still makes it as a one sub setup in this regard, since they have the same distance to the listening position. And you added on to sub distance, that might be the same as I ask for, but I'm not sure if taking the distance to the neareast sub into the AVR as sub distance, and then just add on the subs Delay dail for further away subs. Maybe someone with more theoretical knowledge than me can provide some input to that.
Also if someone knows the reason why averaging the sub distances for input to AVR is a good thing, given the subs have different distances to the listening position. In my case one at front wall, one at back wall.

Agree acts as one sub for me and much easier.  If it were me in your situation I would integrate one sub alone the best you can using distance setting in receiver.  Then add the other one into the mix and play with position and delay/phase knob on sub and see what you can do to make it even better.  Then run Audyssey.  Maybe tweak sub distance in receiver if needed after Audyssey otherwise call it good.

post #75 of 95
Quote:
Originally Posted by Danne2 View Post

.....Why not instead put in the distance to the closest sub, and then just increase Delay on the the other subs, increasingly as they are further away from the listening position ?...

My understanding is that you should do the opposite; set the distance in the AVR to the furthest sub then add delay to the closer sub(s) so the they behave as if they are all the same distance from the main LP (you want their sound to arrive at the main LP at the same time). If you set the distance to the sub closest to the main LP then add delay to the further subs(s), then they won't behave like that are the same distance from the main LP; the sound from the further subs will arrive at the main LP after the sound from the closer sub(s).
post #76 of 95
Quote:
Originally Posted by Saturn94 View Post

My understanding is that you should do the opposite; set the distance in the AVR to the furthest sub then add delay to the closer sub(s) so the they behave as if they are all the same distance from the main LP (you want their sound to arrive at the main LP at the same time). If you set the distance to the sub closest to the main LP then add delay to the further subs(s), then they won't behave like that are the same distance from the main LP; the sound from the further subs will arrive at the main LP after the sound from the closer sub(s).
The issue is all the reflected waves as well as the direct waves, which greatly complicates things. I agree that you should dial in the furthest sub first. At that point it becomes a seat of the pants affair. Bring the second sub up to the same level at the LP, then adjust distance, polarity, phase, EQ, whatever you have on it, while doing RTAs, until the best overall result is realized. Repeat the process with any additional subs.
post #77 of 95
+1, adding distance to receiver sub out means signal arrives earlier from sub. Less distance means that signal arrives later (add delay)

Often times the correct acoustic subwoofer distance is longer than physically measured distance to account for electronic delay of subwoofer amp and delay from boundary reflections.

When i setup my duals i measured subs without speakers and added delay to nearest sub for best response. Then ran autosetup for distance settings and tweaked subwoofer distance for best response ( i think it ended up being 1.5 ft more than receiver set) with subs and center and then verified with mains. I do the center because i am primarily movies. Here is a post by seaton. Basically i follow his advice.smile.gif

"For the vast majority of multi-channel playback, the center channel effectively anchors the system, with all other speakers providing the soundscape around it. I spend much more time than most with the overlap region and overall integration of the subs and main L/C/R. So many speakers perform and behave much better with a crossover above 60Hz (usually 80Hz). The resulting overlap region has a definite impact on the subjective balance, clarity, and impact of the system. Having had the opportunity to hear some very clear and balanced systems outdoors makes such problems more readily apparent by comparison. For a great home theater system, first and foremost I expect excellent dialog reproduction. Hence I start with the center channel.

As I mentioned above, this provides a less confusing point to start from, as you only have one speaker to balance. Once I have the center & sub playing nicely, I know I can generally get the L & R to match well enough, even if it requires some slight deviation from the initial settings. The problem of integrating a pair of speakers which often behave differently in the room and with the subwoofer is lessened by starting from a point that you know gives decent results for the center channel. For just about any movie soundtrack I can more easily hear problems with the center-sub integration than with the L/R & sub.

Unfortunately there are still too many cases where the center is less capable than the left and right speakers. I have helped fix many systems where the dialog sounded weak and there was little subjective impact and weight to soundtracks. In combining hefty main speakers with a smaller center manyl set the crossover too low and the upper bass from the subwoofer too weak. Starting with the center channel gets around this easy mistake, saving you from having to re-work the main L/R settings when you later find they won't work for the center channel.

Of course some general knowledge of the loudspeakers, amplifiers, subwoofer and pre-processor's capabilities is useful, and may even dictate a different starting point. In my own experience I have most often had to backtrack when I did not follow the above method."

http://www.avsforum.com/t/595341/how-to-eq-the-1-channel

Here he shows results from placing subs and adding delay to nearest sub for best results.

http://www.whatsbestforum.com/showthread.php?7501-Seaton-Submersives-amp-Rockport-Arrakis-amp-Mira-Grands/page3
post #78 of 95
Quote:
Then ran autosetup for distance settings and tweaked subwoofer distance for best response ( i think it ended up being 1.5 ft more than receiver set) with subs and center and then verified with mains. I do the center because i am primarily movies.

How do you phase align CC with the sub first and then the mains with the sub?
post #79 of 95
Measure and tweak sub distance, its a balancing act. Save well labeled measurements of the different distances, speakers, crossovers... and compare. Obviously someone 95% 2 channel with sub wouldnt favor the center channel. Someone 50/50 might want the best compromise. It may be that one sub distance is best for all mains. Basically measure and measure some more. Good thing is when done, dont need to do it again unless change something. biggrin.gif
post #80 of 95
And how do you phase align the sub with the mains now?
post #81 of 95
Quote:
Originally Posted by braveheart123 View Post

And how do you phase align the sub with the mains now?

i measured center with sub. Saved measurements at different distances. Now i can pull up any combination and compare. I rinse and repeat with right and sub, then left and sub. now i can compare different sub distance affects between each front and sub.
post #82 of 95
Quote:
i measured center with sub. Saved measurements at different distances. Now i can pull up any combination and compare. I rinse and repeat with right and sub, then left and sub. now i can compare different sub distance affects between each front and sub.

IMO there is not much to gain from this tedious exercise. I mean if your front three channels are equidistant from MLP (and they must be); phase aligning the mains with the sub is enough.
post #83 of 95
Id say yes if your speakers are all the same and identical room acoustics in all locations.

For me the audyssey set distance of sub was perfect for my center but I got much better response with mains using tweaked distance. This tweaked distance causes a nasty null at my crossover for the center. So need to compromise.
Quote:
Originally Posted by braveheart123 View Post

IMO there is not much to gain from this tedious exercise. I mean if your front three channels are equidistant from MLP (and they must be); phase aligning the mains with the sub is enough.
post #84 of 95
What are we discussing here?

Are you guys changing the sub distance wrt to mains for phase matching the sub with mains OR changing CC distance wrt to sub to bring CC in phase with sub?

We can only bring the mains in phase with sub by changing the distance of sub in AVR. As regards CC, we cannot change sub distance now coz it would invalidate the sub/mains phase matching. For it to work properly, we cannot change sub distance. We will have to tinker with the distance of CC keeping the distance of sub constant. This way all front 3 channels can be in phase with the sub.

The only way a sub can be in perfect phase with FR, FL, and CC at the same time by changing the distance of sub is when room acoustics have same effect on all front three channels wrt MLP. That is very difficult to achieve as our rooms are far from perfect.
Edited by braveheart123 - 7/17/13 at 10:07am
post #85 of 95
Tweaking with the distance that MCACC, or Auddysee set in essence is altering the entire system. Especially subwoofer difference. Using a spl meter or measuring tape is primitive compared to what autocalibration does with the system.
post #86 of 95
I am no expert but like to see what they (such as bill or mark) have to say. I have the measurement equipment and $ invested in my system. I want to test what they think on my own and try and get all i can from what i got. If its important enough that Mark does it and has taken the time to discuss on several occasions, its good enough for me. auto cal says 10 is best, center is best at 12, mains best at 14, maybe 13 is a good compromise. Its tedious, just run autocal and enjoy. rolleyes.gif

Primetime has it right. The sub cross is below schroeder, room boundaries, different speakers... autocals dont check interactions between speakers and sub, just arrival time. This discussion has become tedious, i'm out. Have a good one. smile.gif
post #87 of 95
Quote:
Originally Posted by derrickdj1 View Post

Tweaking with the distance that MCACC, or Auddysee set in essence is altering the entire system. Especially subwoofer difference. Using a spl meter or measuring tape is primitive compared to what autocalibration does with the system.

Quote:
Originally Posted by Luke Kamp View Post

I am no expert but like to see what they (such as bill or mark) have to say. I have the measurement equipment and $ invested in my system. I want to test what they think on my own and try and get all i can from what i got. If its important enough that Mark does it and has taken the time to discuss on several occasions, its good enough for me. auto cal says 10 is best, center is best at 12, mains best at 14, maybe 13 is a good compromise. Its tedious, just run autocal and enjoy. rolleyes.gif

Primetime has it right. The sub cross is below schroeder, room boundaries, different speakers... autocals dont check interactions between speakers and sub, just arrival time. This discussion has become tedious, i'm out. Have a good one. smile.gif

Good input in the thread Luke. There has been plenty of discussion as to why most auto-cal systems have limits. In short they are limited to the assumptions made and measurement data they are given. None take measurements to confirm the expected behavior. In many cases there are not clearly "correct" answers and certain acoustic issues or mic placement can confuse the measurements used to generate the settings. Very obvious issues can be different level and distance settings for left & right speakers. There may be differences due to exact mic placement and room interaction, but in a mostly symmetrical room you will want to match the two speakers, often some average between settings. If you make the change and it gets worse, it's easy to go back to the Audyssey settings. As you can see with some searching, that's rarely the case.

Subwoofer distance is a very fuzzy setting with such long wavelengths and the interaction with the main speakers of various types. The selected crossover frequency can often shift the optimal setting. Just remember that at 80Hz, ~3.5' is only 1/4 wavelength. Impulse measurements from a main speaker make sharp spikes with the first sound arrival. With a subwoofer an impulse is measured as a ramp/mound of energy as high frequencies are required to make any sharp shapes. The peak of that mound is not always the best setting, especially when the bass from the main speakers always lags the high frequencies they use to set the distances.
Edited by Mark Seaton - 7/17/13 at 1:18pm
post #88 of 95
Quote:
Originally Posted by Luke Kamp View Post

autocals dont check interactions between speakers and sub, just arrival time.
They can't even do that with a high degree of accuracy, as there isn't one arrival time, but separate arrival times from the speaker directly and from each reflective surface in the room, even when a gated pulse measurement is used. They can give a reasonable approximation, but they're not perfect. Required reading:
http://www.ethanwiner.com/believe.html
post #89 of 95
Since the Bass FR at a given listening position is influenced by the sub phase relationship with all speakers in the room, it seems the goal would be to flatten out the FR around the XO freq with all speakers fed a single source tone during a frequency sweep. What I have observed doing tis is that I was able to flatten out the respons of the entire system by just shifting the sub distance one foot. from where the peak output was obtained at the XO freq.

However, I do agree that the phasing with the CC is most important if I am mostly watching movies. So, now I am thinking I should remeasure and re-tweak with just the CC for best FR. I believe that gave me a difference of 6 feet from what I have now.
post #90 of 95
Thanks for chiming in gentlemen. Very interesting information on impulse measurements and sub being a ramp/mound of energy. No doubt that autocals can be a good starting point, but can be improved upon. I had not seen that article, Ethan's site has a plethora of information and learn (or remember:)) something every time.
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Audio › Subwoofers, Bass, and Transducers › phase vs. distance?