Placing Two Subwoofers

[anjunadeep]

Hello! I was posting in the REW thread on AVS and a member messaged me and suggested I ask it here as it may be more in this forums wheelhouse.

I have a pair of subwoofers on order that should arrive in another couple months, but I thought I'd take the time to use my little JL Audio E110 to try some new placement options and see how they summed up. This little lass is sure easy to move!

So I tried a bunch of different locations and picked what seemed to be the three best spots individually, and then tried summing them in various combinations. Here is the results (individual measurement colors correspond with the color box/subwoofer in the image and the trace arithmetic sum is labelled by hand):

Here is just the sums overlaid for easier comparison:

"A" certainly has the most headroom below 40Hz, by far, until we meet at 20Hz. Perhaps this is because with both subwoofers aligned on that front wall they are coupling like an array? "A" is the worst in the 43-52Hz region though, and there is a tight null at 56Hz or so (I've heard those aren't really audible though). "C" has the best midbass headroom, but it seems really down around the crossover and in the deeper bass.

The interesting thing is this does seem to be pretty accurate, because I have the 1/3rd and 2/3rd along the wall uncorrected measurements from my previous ported duals and it looked like this:

My biggest gripe with this positioning was that 45-55Hz range. However, if I just replace that problem with 30-35/38Hz, where my house curve is greater, not sure what is better...

With my old setup, this was the pre and post EQ measurements. You can see I did have some success boosting that null, but it was headroom limited. I'm hoping my new subwoofers have considerably more headroom there, so maybe I could power through that - but I was hoping to find better placement for the second sub. There was nothing I could do about the null at 57Hz, so I didn't even attempt to EQ that... but the bigger dip at 45Hz-52Hz or so I did have some success boosting.

Questions:
Is my investigation worthwhile? Can I use a small sealed sub moving around the room to get an idea where to place two giant sealed ones?
Why when I sum responses, does the output actually go DOWN in some circumstances? Is this from cancellations? I.e for option "C" the individual outputs look very smooth between 30-40Hz, but when summed the summed total is less!? Is that something I could fix with delay? How would I figure that out?

Thanks for any help!

 

[Wayne A. Pflughaupt]

What is the white sections of the top and bottom walls? Windows or openings?

Regards,
Wayne

 

[anjunadeep]

What is the white sections of the top and bottom walls? Windows or openings?

Regards,
Wayne

Hi Wayne,

The top sections are windows, the bottom is an entryway to my office. It's one of those difficult living room setups that I'm trying to make the best of.
I snapped some photos for you just now: http://noesis.csoft.net/wayne/

Notice the furniture is a bit different than in my drawing, but, that drawing is more accurate to how the room will look in the near future (i have some new furniture on order). "Position 9" you see referenced in those graphs and drawings is essentially right next to that gray couch that runs along the window currently. I'd obviously rather everything be up front for cable management reasons, but, if it makes a huge difference I think I could get creative under the rug to get some wires over there. The room isn't winning any aesthetic awards anyways lol (and excuse the extra wire mess while I wait for the new subs..).

 

[Wayne A. Pflughaupt]

Thanks for the pictures, definitely a help!

For starters, I’d measure from the MLP. I'm probably in the minority on that these days, but I don’t see any reason to measure elsewhere unless you’re trying to average. Personally I don’t mess with that because movie bass is all about “boom.” It’s a big world and I’m sure there are exceptions, but generally less-than-perfect measurements at secondary locations is not going to make or break “boom.” If it does, then you can worry about it. My experience with large open rooms like yours (I used to have one) is that adjusting for the MLP will get perfectly acceptable bass in other seats, except for those near a boundary. And I did it with only a single sub location, which is also increasingly counter to current thought.

Any sub located near a wall opening will get an abysmal reading. In addition, the one and only time I experimented with mid-wall placement the reading was a lost cause, and your orange graph for “B-2/3 & Position 9” confirms that once again.

I’ve always had the best luck putting subs in the corner with the longest uninterrupted wall length in both directions, which in your picture is the top left. And what a surprise, the purple graph looks better than any of the other individual readings. Moving the purple sub closer to the corner and measuring from the MLP hopefully will get you an even better measurement. From there you might try moving the mid-wall sub closer to the corner and look at that summed graph.

Regards,
Wayne

 

[anjunadeep]

For starters, I’d measure from the MLP. I'm probably in the minority on that these days, but I don’t see any reason to measure elsewhere unless you’re trying to average. Personally I don’t mess with that because movie bass is all about “boom.” It’s a big world and I’m sure there are exceptions, but generally less-than-perfect measurements at secondary locations is not going to make or break “boom.” If it does, then you can worry about it. My experience with large open rooms like yours (I used to have one) is that adjusting for the MLP will get perfectly acceptable bass in other seats, except for those near a boundary. And I did it with only a single sub location, which is also increasingly counter to current thought.

Thank you, Wayne. I am measuring from the MLP. Historically I'd made a bit of a cloud and averaged it before EQing, but only within 1ft or so of my head. Basically my thinking is I don't want to EQ so tight that if I move my head 1inch it's outside the window lol.

Any sub located near a wall opening will get an abysmal reading. In addition, the one and only time I experimented with mid-wall placement the reading was a lost cause, and your orange graph for “B-2/3 & Position 9” confirms that once again.

Which one do you think looks the best when summed?
One thing I should note about these measurements is I never played with delay other than a quickie test. Should I try the alignment tool and see with delays added what happens?

If I just put the subs in the front, wall, it seems like I miss out on that 40-50Hz range pretty badly. If I place one along that window, then that particular sub seems to have a terrible suckout in the deep bass, but helps in the range the front wall does worst. Am I thinking about that correctly?

I’ve always had the best luck putting subs in the corner with the longest uninterrupted wall length in both directions, which in your picture is the top left. And what a surprise, the purple graph looks better than any of the other individual readings. Moving the purple sub closer to the corner and measuring from the MLP hopefully will get you an even better measurement. From there you might try moving the mid-wall sub closer to the corner and look at that summed graph.

I will try moving the sub on the TV wall closer to that bottom corner (left corner). I think I get more deep bass gain there. Maybe that combo would work well to offset the suckout I suffer in deep bass from the opposite sub in order to get the 40+Hz boom.

 

[JStewart]

Should I try the alignment tool and see with delays added what happens?

 

[Sonnie]

Agree... alignment is important... I'd definitely check that.

Multiple subs is a smart move to start with, but usually much better for a room with openings to help fight the suckouts.

Wayne's not alone with measuring the main listening position. That's all I ever measure, and I rarely move my mic... never had any issues moving my head a few inches during movie watching. It might effect the imaging a little on critical two-channel listening, but moving the mic a few inches to a foot doesn't help it in my experiences of numerous measurement trials over the years... it only makes things worse for two-channel in my room. It doesn't matter near as much with movies, as you have stuff going on all over the place and dialogue is typically pretty broad. I can move my head several inches left/right and center is still center because it's coming from the center speaker.

 

[Wayne A. Pflughaupt]

Basically my thinking is I don't want to EQ so tight that if I move my head 1inch it's outside the window lol.

You’ll get a different measurement anytime you move the mic. But it’s nothing that you’ll be able to hear with program material. At least that’s my experience, but as I said it’s a big world and I’ve only experienced a few rooms.

Which one do you think looks the best when summed?

To me, a good graph is one that can reasonably be equalized. “Reasonably equalized” means it can be accomplished without huge boosted or negative-gain filters. By that definition, frankly, none of the summed graphs look good to me.

You might try placing the second sub in or near the left (bottom) corner. Even with the inevitable poor reading, maybe you’ll get lucky and get enough of the 40-50 range to get a good sum.

Regards,
Wayne

 

[anjunadeep]

Work has been awful with end of year, I finally got a chance to play with the alignment tool a bit, and option A and D come out the best. Neither are great though. . D maybe looks a bit better in the graph, but I'm afraid of having localization issues.

One thing I pondered for a hot minute was doing position A and then buying something like a Rythmik dual 8" to place in "position 9" with it EQd to just make boom between 40-53Hz or so. Not sure if something like that would work well with the JTR RS1s I have on order though. I guess I could buy a third RS1 next year...

Agree... alignment is important... I'd definitely check that.

Multiple subs is a smart move to start with, but usually much better for a room with openings to help fight the suckouts.

With one subwoofer almost every spot is bad, two just makes it manageable in my room. A part of me wants to have three, but actually playing with the alignment tool I was struggling to get three to work well. I could solve the 40-50Hz problem with a third sub, but then would cause issues above 70Hz. Some of that could just be a result of how I'm doing this though.

To me, a good graph is one that can reasonably be equalized. “Reasonably equalized” means it can be accomplished without huge boosted or negative-gain filters. By that definition, frankly, none of the summed graphs look good to me.

They all look terrible. Maybe option D would take less EQ assuming a house curve? I'm worried about localization with both subwoofers on the right hand side like that though...

How the heck do people get these near perfect graphs?!

I'm in Miami, now I just gotta get you to fix my room lol. Alignment tool was a good call, it's for sure progress.

 

[anjunadeep]

Some more graphs:

Here is the alignment tool for Option A (1/3rd and 2/3rd along the front wall):

and here it is for Option D (right corner + position 9):

The front wall, since they are the same distance, didn't need any delay. The other setups did.

Educational and depressing at the same time lol.

 

[JStewart]

Feel free to post your REW mdat file if you like.

Just curios... Is the room sealed other than the 1 office entrance opening? Do you have the room dimensions? Is the sub ported or sealed? (Guessing ported).

 

[anjunadeep]

Feel free to post your REW mdat file if you like.

Just curios... Is the room sealed other than the 1 office entrance opening? Do you have the room dimensions? Is the sub ported or sealed? (Guessing ported).

That's very kind of you, I've exported it and attached it. I have the summations I tried, and also the original measurements. P1 is my code for main listening position.

Unfortunately my room is my living room and it's all open, you can see the layout in various photos here: http://noesis.csoft.net/wayne/

The subwoofer I'm using for these measurements is a JL Audio E110. It's actually sealed, but it has a ported liked roll off - it's a tiny tiny subwoofer for its output so I think they just limit that deep extension to keep the thing from blowing up. The subwoofers I have on order will be sealed as well, but much larger and with deeper extension, they are JTR RS1s. I've been using my E110 as a temp subwoofer until my bigger ones arrive, I used to have ported subwoofers but sold them. I figured in the meantime I could use my little JL to easily move about and test some different locations and learn about what might work best with the JTRs arrive.

 

[anjunadeep]

...another thing I'm thinking of doing is moving my seat back. I was looking at older measurements where I took a "cloud" and averaged them, and that dip at 40-50Hz is a bit better if I move my seat... Maybe something to try this weekend. The downside of that is I'm already 10ft away from a 85" TV, so moving away isn't ideal. Forward it gets worse.

 

[JStewart]

Unfortunately my room is my living room and it's all open, you can see the layout in various photos here

Very nice looking room sir

The subwoofer I'm using for these measurements is a JL Audio E110. It's actually sealed, but it has a ported liked roll off - it's a tiny tiny subwoofer for its output so I think they just limit that deep extension to keep the thing from blowing up.

The LF output is pretty surprising at those SPLs. I guess that's why JL has such a good reputation.

The subwoofers I have on order will be sealed as well, but much larger and with deeper extension, they are JTR RS1s. I've been using my E110 as a temp subwoofer until my bigger ones arrive, I used to have ported subwoofers but sold them. I figured in the meantime I could use my little JL to easily move about and test some different locations and learn about what might work best with the JTRs arrive.

At 117lbs each for those JTRs that's a good plan! I have a couple of 210RTs at 130lbs each IIRC. I'll bet they're easier to move around than those subs though.
Jeff makes some really great speakers. You're gonna love 'em. Neighbors, not so much.

another thing I'm thinking of doing is moving my seat back. I was looking at older measurements where I took a "cloud" and averaged them, and that dip at 40-50Hz is a bit better if I move my seat... Maybe something to try this weekend. The downside of that is I'm already 10ft away from a 85" TV, so moving away isn't ideal. Forward it gets worse.

I only got a brief look last night and at work today, holiday coming, yada yada. A couple of the combined locations looked like they'd be OK with a little EQ. Left corner and pos 9 was one after a little delay was applied. Their weaker and stronger outputs complemented each other. You obviously have a good handle on the process too.

Moving the seating has a downside as you say and I'd guess those JTRs will have headroom to spare. I also agree with you're focus on the mid-bass. We know what an instrument is supposed to sound like where an explosion is pretty much going to sound like an explosion, IMHO.

They all look terrible. Maybe option D would take less EQ assuming a house curve? I'm worried about localization with both subwoofers on the right hand side like that though...

A valid concern depending on XO, XO slope, output relative to the farther sub and, of course, your hearing acuity.

How the heck do people get these near perfect graphs?!

Not universally true of course, but I've personally had lots of perfect spl charts that didn't sound good. Please also refer to @Wayne A. Pflughaupt 's signature.

 

[anjunadeep]

Very nice looking room sir

Thanks!

The LF output is pretty surprising at those SPLs. I guess that's why JL has such a good reputation.

Oh yeah, and it's not even stressing in those sweeps. I started the sweeps at a lower volume than normal concerned, and quickly realized my usual -20MV sweep would be just fine for it.

At 117lbs each for those JTRs that's a good plan! I have a couple of 210RTs at 130lbs each IIRC. I'll bet they're easier to move around than those subs though.
Jeff makes some really great speakers. You're gonna love 'em. Neighbors, not so much.

I'm excited! One good thing I will say about my house is between the concrete construction and hurricane glass, basically nothing gets outside so I'm still on the neighbors cookie list.

I only got a brief look last night and at work today, holiday coming, yada yada. A couple of the combined locations looked like they'd be OK with a little EQ. Left corner and pos 9 was one after a little delay was applied. Their weaker and stronger outputs complemented each other. You obviously have a good handle on the process too.

Moving the seating has a downside as you say and I'd guess those JTRs will have headroom to spare. I also agree with you're focus on the mid-bass. We know what an instrument is supposed to sound like where an explosion is pretty much going to sound like an explosion, IMHO.

My last subs ran out of gas in that 40-50Hz range, but they only had a 960watt rms/2100w peak amp. The JTR seems to have that SP1-2400 Speakerpower amp in it, so hoping that extra headroom works well for me.

A valid concern depending on XO, XO slope, output relative to the farther sub and, of course, your hearing acuity.

I tried the JL E110 next to me for a while last week and it wasn't great. I could for sure localize things a bit. Maybe if I had a second in that left corner it would have balanced it out, not sure.

Not universally true of course, but I've personally had lots of perfect spl charts that didn't sound good. Please also refer to @Wayne A. Pflughaupt 's signature.

Yes. I've had bad luck with Audyssey personally for that reason: it will over-EQ my room, eat up gobs of headroom (it allows +9dB boost and that's before anyone turns DynamicEQ on, which it defaults in the on position). My last setup that 40-50Hz range sounded kinda lean though, it was noticeable. So, I'd either stick with them both up front and try to boost more via EQ, or potentially try having one sub closer to me (i.e. position 9).

Maybe if I level match them rather than gain match them that would also help, but of course then the sub up front would run out of headroom first.

 

[JStewart]

This is the 1/3 and 2/3 with pos9 added:

Might be worth some experimentation with the JL in pos9 after the JTRs arrive.

 

[anjunadeep]

This is the 1/3 and 2/3 with pos9 added:

View attachment 38052

Might be worth some experimentation with the JL in pos9 after the JTRs arrive.

Thank you for taking a look! Yeah, I was looking at that option too (although I think you selected a better delay, so I adjusted mine). One thing I would run into is the JL E110 wouldn't be able to keep up with the JTRs. So this is what that looks like with -6 and -12 reduction of the gain on that rear subwoofer.

1/3 + 2/3 = two subs in the front
Aligned sum 0 =. JL added at position 9 with no gain reduction
Aligned sum -6 =. JL added at position 9 with -6 gain reduction
Aligned sum -12 =. JL added at position 9 with -12 gain reduction

I believe -6dB is 25% of the power requirement and -12dB is 6.3% of the power requirement, but I think that would just be for the SAME subwoofer (same displacement and cone area) so I'm not really sure how that math works out comparing unlike subwoofers? Potentially I could even use the -12dB and then EQ the front larger subwoofers to kinda pickup some of that slack? Or maybe I even "gate" (?) the bandwidth of the JL, where it rolls off at 24dB/octave or something below 38Hz and above 60Hz? That might keep the JL from struggling trying to recreate deep bass I don't need more of and above 60Hz maybe would prevent localization issues?

I found some testing of the RS1 on Audioholics and the JL E110 from Databass:

Thanks again for the help.