subwoofer discussion

[Jdunk54nl]

Phase and crossover matter. Here is a good whitepaper on how you'd use them manually if you don't have an integrated amp or AVR that handles that upstream.

@carewser if you haven't tried the techniques in this whitepaper before try them out and see what you think and if any of your subs do better or worse than others. With 10 you are in a pretty interesting position to cross compare different features to see what the actual difference it.
https://www.soundoctor.com/whitepapers/subs.htm

Maybe pick up a umik-1 microphone and install REW on your laptop and start measuring the differences between them all in your particular room and share with the group. That would be a pretty cool set of findings.

That article is good.
Here is another good article: https://www.merlijnvanveen.nl/en/st...gnment-the-foolproof-relative-absolute-method

There are now cheap options to measure impulse / phase / etc. Open Sound Meter is free software to use and then you would just need a xlr microphone to computer interface (Motu m2 or Focusrite 2i2 are reasonably priced) and and XLR microphone (Dayton-EMM6 is great for the price). After doing it this way, there is no more guessing if it is the best or not. You have measured phase and once they are the best you can get, they are the best you can get.

Another rudimentary way is to use band limited pink noise through the crossover region between subs and mains. Flip the polarity on one and delay the mains (or add distance to the sub if using an AVR) and listen to when it gets the quietest. This will be the best delay you can get without using microphones.

If you have a USB microphone, then you can take sweeps with an acoustic response to get some delay info, but I have never found this to be the best information. You can look at the RTA screen, when playing subs with mains, using periodic pink noise with averages for like 1 sec. Make changes and see how the overall response sums. Proper Delay will be when it sums the best.

On all of those notes, I strongly suggest your first microphone should be a xlr microphone with a interface. If you are serious about this audio stuff, a usb microphone just doesn't cut it and you will eventually reach it's limits. It is easier to get going with a usb microphone, but not by much.

 

[Aerith Gainsborough]

On all of those notes, I strongly suggest your first microphone should be a xlr microphone with a interface. If you are serious about this audio stuff, a usb microphone just doesn't cut it and you will eventually reach it's limits. It is easier to get going with a usb microphone, but not by much.

Why would an XLR-mic + random audio interface perform better than a USB mic (assuming both mics have calibration files)?
Espeically in a non-professional setting?

 

[Jdunk54nl]

Why would an XLR-mic + random audio interface perform better than a USB mic (assuming both mics have calibration files)?
Espeically in a non-professional setting?

As that article said, just being able to measure frequency isn't doing much. You need to be able to measure in the time domain and therefore need a reference loopback. You can not do this due to clockdrift issues in electronics with a usb microphone.

 

[pjug]

Good article. As he points out, phase correction on the subwoofer isn't really possible in a meaningful way, since the subwoofer is typically already delayed.

On this point: many subwoofers just have a 0 or 180 degree switch. I assume this is just polarity inversion not delay. If this is the only tool available, and 180 degrees gives a better crossover result than 0 degrees, it seems the 180 setting would be the better option.

 

[NYfan2]

On my reference system (room is 21.5 feet X 17.5 feet with a reinforced wood floor) I am using the subwoofer towers from the Infinity IRS Beta System, which are two stereo subwoofer towers about man height with 4 X 12 inch carbon fiber woofers each in sealed enclosures. One woofer in each tower has an accelerometer mounted to it and a special signal cable to provide feedback from the accelerometer to the single system subwoofer controller which has many system integration settings and also compares the woofer movement to the input signal in order to provide more accurate deep bass. The subwoofer controller gets its main input signal from a single ended connection at the system line stage preamp. The output of the subwoofer controller drives two Class D (IcePower) amplifiers to drive the subwoofers. The system line stage balanced output connection is used to drive the main system Class AB amplifier to drive the main speakers, which are very efficient dynamic Open Baffle speakers of my design. These main speakers both outperformed and have proven much more reliable than the former OEM main speakers of the IRS Beta System. In addition to the main speakers and main subwoofers, that system also has a small additional subwoofer (an active M&K push-pull sealed cabinet with a pair of 8 inch woofers) that is positioned to the side of and slightly behind the listening position. This combination of subwoofers provides deep accurate bass similar to a live performance, that at times you feel in your chest and stomach, and is immersive and contributes to the deep soundstage of that system. It is the best subwoofer system that I have ever heard (or not heard as the sound always seems to come from the main speakers). I also have a half dozen other audio systems with one or two cube type subwoofers each, and there really is no comparison to the reference system's bass when it comes to re-creating a live performance.

I love to see that setup, can you post some photo's ?

To keep this thread on topic, here is were members post their system:
https://www.audiosciencereview.com/forum/index.php?threads/the-kit-we-bought-our-current-setups.564/

 

[Jdunk54nl]

On this point: many subwoofers just have a 0 or 180 degree switch. I assume this is just polarity inversion not delay. If this is the only tool available, and 180 degrees gives a better crossover result than 0 degrees, it seems the 180 setting would be the better option.

Yes, sometimes the polarity swap (180 vs 0) gives better integration. You want your subs and mains to be both in phase AND in time. You can have either one by itself but you want both!

This graph shows both in phase and in time between subs and mains. In time means the phase graph has the same slopes. In Phase means they have the same degree of phase at that frequency.

This graph shows in phase at ~120hz (they all are about 190 degrees) but not in time between the orange, green, and blue lines (different slope for orange line).

I pulled this from this video:

This shows in time but out of phase. Both lines are parallel to each other so in time, but phase does not line up. They are about 180 degrees out of phase, so this would be perfect for a polarity swap!

 

[pjug]

Yes, sometimes the polarity swap (180 vs 0) gives better integration. You want your subs and mains to be both in phase AND in time. You can have either one by itself but you want both!

Also, it would seem that the phase issue will often be caused by the sub being closer to the listener than the main speakers. In this case adding delay to the sub (if this is an option) would be a proper correction, giving better frequency response and also better timing.

 

[Jdunk54nl]

Also, it would seem that the phase issue will often be caused by the sub being closer to the listener than the main speakers. In this case adding delay to the sub (if this is an option) would be a proper correction, giving better frequency response and also better timing.

You could potentially move your sub right next to you to get it's natural delay timed properly with the mains. But this is unlikely for most people. Usually subs are tucked into corners or against walls to solve other issues. In my experiences, I have had to always delay the mains to get them in time/phase with the subs.

 

[abdo123]

it's really pointless to talk about delays below 100Hz, when even 15ms delays are barely perceptually different at these frequencies.

Just focus on the steady state / magnitude response and you should be good to go.

 

[Jdunk54nl]

it's really pointless to talk about delays below 100Hz, when even 15ms delays are barely perceptually different at these frequencies.

Just focus on the steady state / magnitude response and you should be good to go.

I disagree based on my actual measurements using SMAART or open sound meter. Others that also do this for a living would disagree based on what they have to do to get things in phase and time. A couple articles already posted disagree with your statement. That is quite a bit of data that goes against that.

 

[abdo123]

I disagree based on my actual measurements using SMAART or open sound meter. Others that also do this for a living would disagree based on what they have to do to get things in phase and time. A couple articles already posted disagree with your statement. That is quite a bit of data that goes against that.

what's the point of fixing the phase at one point in space if you just move the microphone 1 cm to the left and it's completely ruined again?

If you don't believe me then listen to someone who did their PhD in reproduction of low frequencies in small rooms.

 

[ernestcarl]

If you have a USB microphone, then you can take sweeps with an acoustic response to get some delay info, but I have never found this to be the best information.

Not the best, sure, but it’s actually quite good enough for many applications if you know how to compensate. I have an XLR mic for measurements and calibration, but the USB gets more use out of me out of convenience. Certainly it works just as well for much of the stuff I find when it comes to measurements at home. Smaart and Open Sound Meter are great tools, but it’s hard enough to learn just one software application, so I’d rather have someone learn and start with what they already have on hand and are most familiar with.

 

[ernestcarl]

what's the point of fixing the phase at one point in space if you just move the microphone 1 cm to the left and it's completely ruined again?

If you don't believe me then listen to someone who did their PhD in reproduction of low frequencies in small rooms.

I believe Geddes was talking about multiple sub application there where one is overlapping multiple bass sources in a small room.

With one or limited subs, optimization at some designated listening area(s) can be done much better by viewing the individual phase information and lining them up properly using adjustable time delays.

 

[Jdunk54nl]

Not the best, sure, but it’s actually quite good enough for many applications if you know how to compensate. I have an XLR mic for measurements and calibration, but the USB gets more use out of me out of convenience. Certainly it works just as well for much of the stuff I find when it comes to measurements at home. Smaart and Open Sound Meter are great tools, but it’s hard enough to learn just one software application, so I’d rather have someone learn and start with what they already have on hand and are most familiar with.

Absolutely, if you already have a usb microphone, use it to learn REW. REW has a learning curve in and of itself. It is still one of my go to programs for moving mic measurements and EQ work. But for any time / phase stuff I open SMAART...and it has a whole new learning curve. I am glad it remembers settings so I don't have to learn every-time I use it how to set it up....

But for those looking to buy their first microphone. Get an XLR microphone as you will most likely reach the USB microphone limits if you are actually going to measure things. The cost difference is not much (like $100) compared to what we spend on audio things.

 

[abdo123]

I believe Geddes was talking about multiple sub application there where one is overlapping multiple bass sources in a small room.

With one or limited subs, optimization at some designated listening area(s) can be done much better by viewing the individual phase information and lining them up properly using adjustable time delays.

the thing is with my experience with crossovers is that it’s almost a certainty that once you measure outside the original measurement position you always have a dip in the crossover.

I’m trying to do it a la Geddes lately but i’m having difficulty deciding where the low pass filters for the subwoofers should be. It’s like learning to walk all over again.

 

[Jdunk54nl]

what's the point of fixing the phase at one point in space if you just move the microphone 1 cm to the left and it's completely ruined again?

If you don't believe me then listen to someone who did their PhD in reproduction of low frequencies in small rooms.

Subwoofer frequencies are large enough that 1cm shouldn't mess with them as much. There is a point where you start losing the good phase / timing relationship, but usually it covers a wider space at sub frequencies.

 

[ernestcarl]

the thing is with my experience with crossovers is that it’s almost a certainty that once you measure outside the original measurement position you always have a dip in the crossover.

I’m trying to do it a la Geddes lately but i’m having difficulty deciding where the low pass filters for the subwoofers should be. It’s like learning to walk all over again.

Small rooms can be very difficult as the room can often skew the phase of both the sub(s) and main monitors. The phase of my speakers and sub at the listening position do not necessarily resemble what I get with my “anechoic” measurements. A fair bit of phase manipulation/correction was required to straighten them just enough to align optimally with minimal to no cancellation.

What one needs are multiple measurements spatially that preserves time domain information to possibly get a better/optimal average summation.

If the phases can be made coherent across as wide a range possible, one may not even need to use any crossovers. In my case, I am able to run all my monitors full-range with no cancellation via FIR filtering, but the subwoofer still needs to be low passed at some point, as the higher it plays, the more skewed its phase becomes up there which causes cancellation at multiple points. Sealed monitors and subs can make things easier here… I believe Geddes uses all sealed speakers as well.

 

[ernestcarl]

I found this recent SMWTMS video where even Geddes says his speakers are relatively phase linear down to 100Hz:

The speaker systems many of us use here would likely not pass muster with him.

 

[carewser]

I have a Dayton RS 18" tuned to 18hz, and 9 cheap JBL 12s. All diy and EQ'ed flat. At reasonable to a little loud volumes it sounds great. If I try to 'show off', like make your eyes blurry, I haven't figured out how to tame the bass yet. I don't know if it's box resonance, room resonance or what. Since I rarely show off it hasn't been a big deal, but it does kind of suck when I'm like "check out what it can do!".. sound like butt, that's what. The clipping lights on the amp aren't going off and the subs aren't moving a ton.. anyways, those 5 minutes a year are just disappointing.

So let me get this straight, you have 10 and half feet of subwoofers and you can't make them all move?

Maybe upgrade to an amp that puts out gigawatts

 

[Beershaun]

Thank you for your answer. I have Anthem Room Correction (ARC) included in my Paradigm PW-AMP. I'm only interested on a 2.1 system (80% music and 20% movies)

I think you meant a 2.2 system. You want both subwoofers managed independently. Check how your ARC version handles bass. Since the Paradigm PW-AMP only has one sub output it can't manage your subwoofers independently. I mean you could plug in a Y-splitter and just "run what you brung" but your system isn't really able to tune it properly. So you'd get some combination of overly exaggerated frequencies where the two are augmenting each other and dropouts where the two are cancelling each other with no way for the system to understand that it's two different speakers.

I mean by all means plug it in and see what it sounds like. No one is going to get hurt.