AVS › AVS Forum › Audio › Subwoofers, Bass, and Transducers › Audyssey with multiple, different size subs.
New Posts  All Forums:Forum Nav:

Audyssey with multiple, different size subs.

post #1 of 42
12/3/12 at 6:00pm
Thread Starter 
AVR-3310CI, 2 crown amps, Behringer dual 31 band EQ, 2 18s and 2 15s. Trying to figure out the best way to set my system up properly. I have the sub output split and going to each channel of the behringer EQ. Each EQ output is split and goes to its own Crown amp which powers a pair of subs. I figure there is no way for Audyssey to properly handle this so what I've done so far is set up my DB meter, pick a target DB of 95, and play test tones while adjusting the EQ to archive as close to 95 at each frequency I can get. I did it once with just the 18s powered and once again with just the 15s powered. My plan is to now run Audyssey and see what it does but I am very open to suggestions or even a blatant "hey dumbass, do it like this". The system sounds great but I have the equipment so I figure I may as well make it as good as possible.
Reply
Reply

AVS Top Picks

post #2 of 42
12/4/12 at 5:49am
Thread Starter 
If this is by chance posted in the wrong section, could someone please tell me the best place to put it, or could a mod/admin move it. Thanks.
Reply
Reply
post #3 of 42
12/4/12 at 5:56am
I think Audyssey would handle this just as it does with other amps. Do your equalization with the Behringer's first then run Audyssey. I do this with miniDSPs. I have two different sized subs too. Generally your results will be limited by the smaller subs.
Reply
Reply
post #4 of 42
12/4/12 at 6:12am
Thread Starter 
Does my method of equalization sound correct? Or is there a better way?
Reply
Reply
post #5 of 42
12/5/12 at 6:23pm
Quote:
Originally Posted by Locoelectrician View Post

Does my method of equalization sound correct? Or is there a better way?
Are the subs sealed or ported? Do you have the 2 18's on one amp and the 2 15's on the other? Do you know the frequency response and output capabilities of the subs? Most importantly, do you have any measurement gear other than the SPL meter and test tones. If not, the biggest improvement you can make to your system is to obtain the ability to measure it.
Reply
Reply
post #6 of 42
12/5/12 at 7:03pm
Thread Starter 
What would be recommended to measure it? The 18s (behringer) are ported as well as the 15s (pyramid). Each pair is on their own amp and own channel of the EQ. As of now I'm using a frequency generator and a Extech, calibrated SPL meter. Amps are crown. XLS202 at 4 ohms stereo for the 15s and XLS402 at 8 ohms stereo for the 18s.
Reply
Reply
post #7 of 42
12/5/12 at 7:37pm
Quote:
Originally Posted by Locoelectrician View Post

What would be recommended to measure it?
http://www.parts-express.com/pe/showdetl.cfm?partnumber=390-790
http://www.acousticfrontiers.com/Acoustic-Measurement/XTZ-Room-Analyzer-II-Pro.html
http://www.hometheatershack.com/roomeq/
Quote:
The 18s (behringer) are ported as well as the 15s (pyramid).
What are the ports tuned to? I'm going to guess that the 18"s are more powerful, and are tuned lower. That will make them more difficult to integrate with the smaller, less powerful 15's.
Quote:
Each pair is on their own amp and own channel of the EQ. As of now I'm using a frequency generator and a Extech, calibrated SPL meter. Amps are crown. XLS202 at 4 ohms stereo for the 15s and XLS402 at 8 ohms stereo for the 18s.
Measuring individual tones will get very tedious very fast. They also don't tell you anything about the time domain response. The measurement systems I mentioned above will give you a lot more information than an SPL meter and test tones. The last one, REW, will give you the most info, but it's also more compllex to install and optimize. The other 2 are "plug 'n play".

Craig
Reply
Reply
post #8 of 42
12/6/12 at 3:59am
Quote:
Originally Posted by Theresa View Post

I think Audyssey would handle this just as it does with other amps. Do your equalization with the Behringer's first then run Audyssey. I do this with miniDSPs. I have two different sized subs too. Generally your results will be limited by the smaller subs.

Sometimes this doesn't work. In my case, have a room mode that produces a nasty null at ~30hz at the LP. Audyssey didn't fix that. I had to run Audyssey first and then play with the distance setting (phase) and levels to remove it. I was also able to get a more extended response doing that. I'm waiting for a miniDSP so I can do a better job of this.

Cheers,
OldMovieNut
Reply
Reply
post #9 of 42
12/6/12 at 4:26am
Thread Starter 
The 18s unfortunately cut off sharply at 40hz, which is the low scale of their rating in the manual. I'll look at the links after work today, thanks.
Reply
Reply
post #10 of 42
12/6/12 at 5:39am
Quote:
Originally Posted by OldMovieNut View Post

Sometimes this doesn't work. In my case, have a room mode that produces a nasty null at ~30hz at the LP. Audyssey didn't fix that. I had to run Audyssey first and then play with the distance setting (phase) and levels to remove it. I was also able to get a more extended response doing that. I'm waiting for a miniDSP so I can do a better job of this.
Cheers,
OldMovieNut

be aware that a null caused by room interaction cannot be EQed out. when you increase the level ot the relevant frequency the corresponding standing wave that is subtratcing from the direct sound increases exactly as much. It's a lose lose situation that's best addressed, AFAIK, with sub placement etc. You can overheat your amps but you cannot elimiate the dip. Audyssey knows that and has a limit on how far it will EQ something that looks like a dip. Six dB if memory serves . . .
Reply
Reply
post #11 of 42
12/6/12 at 6:10am
Audyssey recommends you set the phase to 0 on all your subs and place them all equidistant from the primary listening position before running Audyssey. Any third party equilization should be performed prior to running Audyssey.
Reply
Reply
post #12 of 42
12/6/12 at 6:13am
I do it before and after running Audyssey, after all my miniDSPs are also crossovers.
Reply
Reply
post #13 of 42
12/6/12 at 6:41am
Thread Starter 
All 4 subs are equal distance from the listening position. They are beneath the screen and arranged from left to right as follows. Wall, space, 15, 18, audio rack, 18, 15, wall. No space on the right side.
Reply
Reply
post #14 of 42
12/6/12 at 7:04am
Quote:
Originally Posted by JHAz View Post

be aware that a null caused by room interaction cannot be EQed out.
Strictly speaking it can, but only at specific spots in the room. That's not a problem if your listening area is small enough so that you can get good response within that area, but otherwise using multiple identical subs at multiple locations to reduce or eliminate room mode issues is the only way to go.
Quote:
Audyssey knows that and has a limit on how far it will EQ something that looks like a dip. Six dB if memory serves . . .
That's to protect the amp from overload and the drivers from over-excursion, struggling to fix with sheer power what it can't fix.
Reply
Reply
post #15 of 42
12/6/12 at 7:59am
Quote:
Originally Posted by Locoelectrician View Post

AVR-3310CI, 2 crown amps, Behringer dual 31 band EQ, 2 18s and 2 15s. Trying to figure out the best way to set my system up properly. I have the sub output split and going to each channel of the behringer EQ. Each EQ output is split and goes to its own Crown amp which powers a pair of subs. I figure there is no way for Audyssey to properly handle this so what I've done so far is set up my DB meter, pick a target DB of 95, and play test tones while adjusting the EQ to archive as close to 95 at each frequency I can get. I did it once with just the 18s powered and once again with just the 15s powered.

.......................................eek.gif......................................eek.gif.................................


Start around 67dB as opposed to 95dB.

The following is my drill. Do with it what you will.

Setting dual subs, I use the sub's gain to separately (all other subs turned off) set each subwoofer's test tone output, at a measured 39" or 1m, to 67dB. If I were running four subs, I'd dial it down to 65dB @ 1m. Doing this offsets the combined subs room gain. Now that the subs have been balanced at 1m, balance their output a second time at the main listening position so the output of each sub is individually the same at the main listening position; 65dB. The combined total of all the subs running, at main listening position, to keep Audyssey happy, should be in the 70-73dB. In my opinion, the subs are now balanced and ready for Audyssey to be run.

The above is an imperfect science/art as each room has it's own room acoustic characteristics (gain) so the final result will be dictated by how the room responds to the combined subs output; the need for user intervention.

When Audyssey has been run, go into the AVR's main menu, to the Audio section, bring up the subwoofer slider and adjust the subwoofer channel, all subs running, at main listening position, to 73dB. Yes, that's an odd number which is purposefully intended to split the difference between 70dB and 75dB; the two reining standards. Now the subs are balanced and what you hear is up to how your subs interact with the geometry of the room; acoustics. If you want better, you'll need to add the likes of a product known as "DSPeaker, Anti-Mode, 8033" to deal with modes and room treatments to deal with nulls.

Hope the above is found helpful.
Reply
Reply
post #16 of 42
12/6/12 at 9:34am
Quote:
Originally Posted by JHAz View Post

be aware that a null caused by room interaction cannot be EQed out. when you increase the level ot the relevant frequency the corresponding standing wave that is subtratcing from the direct sound increases exactly as much. It's a lose lose situation that's best addressed, AFAIK, with sub placement etc. You can overheat your amps but you cannot elimiate the dip. Audyssey knows that and has a limit on how far it will EQ something that looks like a dip. Six dB if memory serves . . .

I understand that. The miniDSP is to smooth out the uneveness introduced by changing the phase and level. In particular to smooth out a bump around 40hz.

Cheers,
OldMovieNut
Reply
Reply
post #17 of 42
12/6/12 at 9:36am
Thread Starter 
Quote:
Originally Posted by BeeMan458 View Post

.......................................eek.gif......................................eek.gif.................................
Start around 67dB as opposed to 95dB.
The following is my drill. Do with it what you will.
Setting dual subs, I use the sub's gain to separately (all other subs turned off) set each subwoofer's test tone output, at a measured 39" or 1m, to 67dB. If I were running four subs, I'd dial it down to 65dB @ 1m. Doing this offsets the combined subs room gain. Now that the subs have been balanced at 1m, balance their output a second time at the main listening position so the output of each sub is individually the same at the main listening position; 65dB. The combined total of all the subs running, at main listening position, to keep Audyssey happy, should be in the 70-73dB. In my opinion, the subs are now balanced and ready for Audyssey to be run.
The above is an imperfect science/art as each room has it's own room acoustic characteristics (gain) so the final result will be dictated by how the room responds to the combined subs output; the need for user intervention.
When Audyssey has been run, go into the AVR's main menu, to the Audio section, bring up the subwoofer slider and adjust the subwoofer channel, all subs running, at main listening position, to 73dB. Yes, that's an odd number which is purposefully intended to split the difference between 70dB and 75dB; the two reining standards. Now the subs are balanced and what you hear is up to how your subs interact with the geometry of the room; acoustics. If you want better, you'll need to add the likes of a product known as "DSPeaker, Anti-Mode, 8033" to deal with modes and room treatments to deal with nulls.
Hope the above is found helpful.

So forget the behringer EQ altogether?
Reply
Reply
post #18 of 42
12/6/12 at 10:29am
Quote:
Originally Posted by BeeMan458 View Post

.......................................eek.gif......................................eek.gif.................................
Start around 67dB as opposed to 95dB.
The following is my drill. Do with it what you will.
Setting dual subs, I use the sub's gain to separately (all other subs turned off) set each subwoofer's test tone output, at a measured 39" or 1m, to 67dB. If I were running four subs, I'd dial it down to 65dB @ 1m. Doing this offsets the combined subs room gain. Now that the subs have been balanced at 1m, balance their output a second time at the main listening position so the output of each sub is individually the same at the main listening position; 65dB. .
What is the purpose of gain matching them, then re-adjusting at the listening position? I know there was a debate somewhere on which is better, but doesn't the adjustment at the LP negate the prior 1M balancing?
Reply
Reply
post #19 of 42
12/6/12 at 11:51am
Quote:
Originally Posted by michaelscott73 View Post

What is the purpose of gain matching them, then re-adjusting at the listening position? I know there was a debate somewhere on which is better, but doesn't the adjustment at the LP negate the prior 1M balancing?
More pertinent is that each sub will have a different response at the LP, so while gain matching them at the LP with pink noise may seem to give a useful result a single response peak throws the reading out of whack anyway. You want each sub to have the same output, arrange them for the flattest result at the LP, and finally EQ to flat. That means balancing them not even at 1m but nearfield, to take the room out of the equation. That also means using identical subs; mixing different subs just muddles things up even more.
Reply
Reply
post #20 of 42
12/6/12 at 12:05pm
Quote:
Originally Posted by Bill Fitzmaurice View Post

More pertinent is that each sub will have a different response at the LP, so while gain matching them at the LP with pink noise may seem to give a useful result a single response peak throws the reading out of whack anyway. You want each sub to have the same output, arrange them for the flattest result at the LP, and finally EQ to flat. That means balancing them not even at 1m but nearfield, to take the room out of the equation. That also means using identical subs; mixing different subs just muddles things up even more.
I was gonna post 4 paragraphs explaining this, but you did it much more succinctly. smile.gif

Craig
Reply
Reply
post #21 of 42
12/6/12 at 12:19pm
Thread Starter 
Fantastic information. I didn't even want the 15s but the 18s fall off so bad at 40hz that I thought they would help. Should be an interesting weekend. Would the 15s possibly do better in a sealed box?
Reply
Reply
post #22 of 42
12/6/12 at 12:45pm
Quote:
Originally Posted by Locoelectrician View Post

Would the 15s possibly do better in a sealed box?
That depends on their T/S specs.
Reply
Reply
post #23 of 42
12/6/12 at 12:50pm
Quote:
Originally Posted by Locoelectrician View Post

So forget the behringer EQ altogether?
Which Behringer is it?
Reply
Reply
post #24 of 42
12/6/12 at 2:14pm
Locoelectrician,

have you looked at the multisub approach?

http://mehlau.net/audio/multisub_geddes/

http://seriousaudioblog.blogspot.com/2012/05/two-great-articles-on-multiple.html

It is the only thing that worked in my room. you would have to move some of your subs from the front stage though.
Edited by zheka - 12/6/12 at 2:20pm
Reply
Reply
post #25 of 42
12/6/12 at 3:51pm
Quote:
Originally Posted by Locoelectrician View Post

So forget the behringer EQ altogether?

Only if reading my post. tongue.gif

I have an Anti-Mode, 8033S II. So that's the direction of my conversation. As to balancing the subs, that's done before running any EQ algorithm, including Audyssey or any of the other built into the AVR room analyzer program.

Quote:
Originally Posted by michaelscott73 View Post

What is the purpose of gain matching them, then re-adjusting at the listening position? I know there was a debate somewhere on which is better, but doesn't the adjustment at the LP negate the prior 1M balancing?

If adjustments are needed at the main listening position, yes, to a certain extent, it does negate the prior 1m balancing. The first thing one wants to do is make sure each sub is putting out the same SPL information at 1m. Now that one has a reference or starting position, equalize them at the main listening position. It's how I balance the subs. In an ideal world, they should be close to the same when separately measured at the main listening position. Now any adjustments are minor as opposed to major. Getting up and back down, several times, is a pain so it's helpful to dial in a constant while close to the gain knob. wink.gif

Now run Anti-Mode, or your outboard EQ box of choice and then run Audyssey. Now flavor to taste with the AVR main Audio menu and your handheld sound meter. Once finished, you can run that spiffy room analyzing kit you put together using your favorite RTA and frequency sweeps for a before and after comparison.

(currently waiting for a hopeful last (stereo XLR > 3.5mm) patch cord to come in)

As I posted to Vardo, next time, I opt for the root canal.

.........................rolleyes.gif
Reply
Reply
post #26 of 42
12/6/12 at 4:11pm
Quote:
Originally Posted by BeeMan458 View Post

When Audyssey has been run, go into the AVR's main menu, to the Audio section, bring up the subwoofer slider and adjust the subwoofer channel, all subs running, at main listening position, to 73dB. Yes, that's an odd number which is purposefully intended to split the difference between 70dB and 75dB; the two reining standards. Now the subs are balanced and what you hear is up to how your subs interact with the geometry of the room; acoustics.
You are missing a couple of significant details. First, when you use the receiver's internal test tones to set the levels, you are *bypassing* the Audyssey EQ. That's right... the Audyssey filters are out of the circuit when the test tones are played. So, none of the room correction provided by Audyssey is accounted for in the level setting process doing it the way you propose above. Also, the Audyssey level setting calculations take into account all the measurement locations you've measured at, and it looks at a broad range of frequencies. Surely that should be a more "representative" calculation that a single point measurement taken by an SPL meter using a band-limited test tone. Why do you feel it is necessary to "re-level" the level settings set by Audyssey anyway?

If you want to verify Audyssey level settings, it would be much more useful to use an external test tone, such as the tones on Avia, DVE, or The 5.1 Audio Toolkit. At least you can set those up to go through the Audyssey filters.

Craig
Reply
Reply
post #27 of 42
12/6/12 at 4:12pm
Thread Starter 
Quote:
Originally Posted by craig john View Post

Which Behringer is it?

Ultragraph pro FBQ-3102. 18s are Behringer Eurolive VP1800s and 15s are older pyramid WX-15x. Receiver is AVR-3310ci and amps are crown xls202 & 402. To refresh, I used a signal generator and played frequencies that matched the bands on my EQ and then balanced them as best I could. Sounds fantastic in movies with the ability to feel action scenes nearly as well as hearing them. Not so impressive with music though as I think I'm missing so much in the deeper frequencies. We had 8 identical 15s which were astoundingly loud but they just took up way too much space. This all started when my son was playing a bass song (bassheads by bassnectar) on the 8 15s and then again on the 18s & 15s and it was quite disappointing. I could care less about that song sounding good but figured it was a great chance to learn something and set everything up as good as possible.
Edited by Locoelectrician - 12/6/12 at 4:20pm
Reply
Reply
post #28 of 42
12/6/12 at 4:27pm
The Ultragraph Pro FBQ-3102 is a 1/3 octave, 31-band graphic EQ. It's basically a glorified set of tone controls. It uses fixed filter center frequencies. If the center frequencies of the filters don't match up to the center frequencies of the modes in your room, they're worthless to you. Also, the "widths" (Q's) of the filters are not adjustable. If you have a narrow peak, you can only reduce a broad range of frequencies. It's a single point EQ and it can't do anything in the time domain to reduce ringing in your room. Bottom line... YES I would forget the Behringer EQ.

Audyssey has TENS OF THOUSANDS of filter taps that it can set in the frequency and time domain. You can measure multiple positions and it will do a "spatial average" of those positions to calculate the EQ filter taps. It isn't perfect by any stretch, but it will do a better job of EQ'ing your system than the Behringer. Additionally, It likely won't be helped by the Behringer. Given the tedious and labor intensive nature of measuring with an SPL meter and adjusting the Behringer, it just isn't worth it.

Craig
Reply
Reply
post #29 of 42
12/6/12 at 4:28pm
Thread Starter 
Quote:
Originally Posted by zheka View Post

Locoelectrician,
have you looked at the multisub approach?
http://mehlau.net/audio/multisub_geddes/
http://seriousaudioblog.blogspot.com/2012/05/two-great-articles-on-multiple.html
It is the only thing that worked in my room. you would have to move some of your subs from the front stage though.

Great reads, thanks.
Reply
Reply
post #30 of 42
12/6/12 at 4:30pm
Thread Starter 
Quote:
Originally Posted by craig john View Post

The Ultragraph Pro FBQ-3102 is a 1/3 octave, 31-band graphic EQ. It's basically a glorified set of tone controls. It uses fixed filter center frequencies. If the center frequencies of the filters don't match up to the center frequencies of the modes in your room, they're worthless to you. Also, the "widths" (Q's) of the filters are not adjustable. If you have a narrow peak, you can only reduce a broad range of frequencies. It's a single point EQ and it can't do anything in the time domain to reduce ringing in your room. Bottom line... YES I would forget the Behringer EQ.
Audyssey has TENS OF THOUSANDS of filter taps that it can set in the frequency and time domain. You can measure multiple positions and it will do a "spatial average" of those positions to calculate the EQ filter taps. It isn't perfect by any stretch, but it will do a better job of EQ'ing your system than the Behringer. Additionally, It likely won't be helped by the Behringer. Given the tedious and labor intensive nature of measuring with an SPL meter and adjusting the Behringer, it just isn't worth it.
Craig

Interesting, looks like the behringer will be reduced to "pretty blinking light" duty.
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home

AVS Top Picks

AVS › AVS Forum › Audio › Subwoofers, Bass, and Transducers › Audyssey with multiple, different size subs.