Added 2nd Sub - REW Huge Success

[Magialisk]

So I used to run "only" one SVS CS-Ultra but because of the size of my room (see attached Room Layout) and my craving for bass you can *feel* I'd always wanted to try adding a second one. A couple of weeks ago I found a guy locally selling his for a great price and snapped it up. This morning I finally had the opportunity to take a few hours and calibrate everything just right and I couldn't wait to share the results.

First, I should note that I set the tuning point on both subs to 12 Hz by plugging 2 ports. I used to run my single sub in the 16 Hz mode but now I have more output than I know what to do with and since I don't have a subsonic filter this seemed like the best thing to do. I also set the target level to 69dB when measuring an individual sub and 75dB when measuring the 2 together. I was able to realize most of the expected 6dB gain between the two measurements.

I've attached 4 images of my results with REW:
The first is a comparison of the response of both subs measured independently. The corner loaded sub is in green and *really* excites the 43 Hz room mode. It's 20dB above my target which should be ~4 times as loud!! It also seems to totally break down between 65 and 80 Hz..? The other sub in red had a very nice response IMO, just a few peaks to tame.

The second image is the measured response of the two subs together along with REW's prediction of the "corrected" response after I manually entered some filters. I subscribe to the fewer filters with larger bandwidths "minimal EQ" school of thought, so I used only 3 filters to take care of all the peaks. I also broke *the law* and added a 4dB boost to 20Hz to help pad that shortcoming. I may go back and take that out because psychologically it bothers me having a boost in there, but so far no bottoming or clipping during testing. The exact filters that I set can be seen in the 3rd image.

Finally the last image shows the measured response when it was all finished. Don't ask what happened around 13 Hz... All afternoon this would happen where a null would come and go in between measurements even though nothing else was changing. I'm inclined to blame it on the RS meter, as even with the .cal file loaded I suspect it's not very accurate below 15 Hz.

In the end, I'm thrilled with this result as it's flatter than I've been able to get before with just the one sub, with the added bonus of lower frequency extension. The amount of air being moved is incredible compared to just the one sub before, and so far no signs of bottoming. Everything I can think to test says that adding the 2nd sub was a hugely positive move.

Hopefully I didn't completely bore everyone, I just really wanted to share my results and maybe even get some feedback. Thanks for listening.

[Wayne A. Pflughaupt]

Wow, the extension of those things is impressive! I need to get me some.

It looks like you'll be applying a single set of filters to both subs. As you can see, the "green" sub is dominating, which means that what I'll call "global equalzing" will drive the "red" sub's level further downward between 20-60 Hz while the opposite will be true between ~60-80 Hz.

There has been some discussion here recently about whether "global" EQ the best approach for seperated subs, or equalizing them seperately. A number of years back I had my subs separated like yours, and it drove me nuts. I was able to localize them from the air pressure they generated to my ears, and that did not go away until I had tweaked them within a dB or so of each other across the full operating range, with 1/6-octave sine waves. It was a grueling, time-consuming process, as I only had 1/3-octave EQs at my disposal.

Others here like Sonnie have been happy just equalizing multiple subs globally, as a system. So if you're happy with the results of global EQ, enjoy! That's much more fun than constant tweaking and fretting!

Regards,
Wayne

[Magialisk]

Hi Wayne, I'm so excited that you found my post! You and Brucek have been the biggest influences in turning me from a REW noob to a (fairly?) knowledgeable user. If you can believe it I actually once ran through both channels of the BFD for a single sub just so that I could access more than 12 filters. I mean heck the line kept getting flatter so it must be better right?

Regarding what you call "global EQ" it's funny you bring that up because I've been wondering those exact same thoughts. On the one hand, what's great about DB's is that adding a second "quieter" sound source will barely affect the existing louder source, so the loudest sub at any frequency dominates. This is very easily seen by looking at the purple combined graph, as it follows the upper envelope of either green or red almost exactly. On the other hand, I have concerns as to whether that means one sub is working much harder than the other over the majority of the spectrum, making the second sub just a "gap filler" for where the first is weak. Something tells me it should be better if they were each flat individually around 69-70dB and then combined to still be flat at 75dB. One of the main reasons I wanted the second sub was to reduce the load on my existing one and I wonder if this global EQ approach has really managed that at all?

My next project will be plotting the corrected response of both subs by themselves with my current filters to see what it looks like, and then with a new bank of filters seeing if I can start over and EQ them separately and still get this good of a response. When I'm done (maybe next week? this is a busy weekend) I'll post more charts and tell the story. Oh yes, and I'll follow the rules for posting charts next time too, my bad on that one.

Thanks for your interest and suggestions. I'm really glad you brought up the global EQ specifically as it just proves there's always more to learn and more to tinker with.

[Wayne A. Pflughaupt]

Thanks very much for the kind words, Marc.

If you can believe it I actually once ran through both channels of the BFD for a single sub just so that I could access more than 12 filters. I mean heck the line kept getting flatter so it must be better right?

Did you hear about the guy who bought a second BFD and daisy-chained them both so that he could have 48 filters to work with?

Just kidding. Gotcha!

On the other hand, I have concerns as to whether that means one sub is working much harder than the other over the majority of the spectrum, making the second sub just a "gap filler" for where the first is weak.

No, one is not working harder, at least not now, because they're both getting the same electrical signal. Keep in mind that the greater overall output from the "green" sub is coming from boundary reinforcement. IOW, it's getting the extra gain for "free."

Something tells me it should be better if they were each flat individually around 69-70dB and then combined to still be flat at 75dB.

Now, if you do that, then the "red" sub will be working harder, because you'll have boosted its electrical signal (via EQ) to make up for what the other is getting for "free" from the room.

One of the main reasons I wanted the second sub was to reduce the load on my existing one and I wonder if this global EQ approach has really managed that at all?

Well, you did mention that you realized 6 dB in overall gain, so it looks like you've indeed bought yourself some extra headroom. You're doing better in that regard than I did when mine were seperated, pretty much like yours with one by each front speaker (L/R), with the right one being in the corner. When I finally stacked them both in that corner, I was surprised to see that I got a 6 dB gain. Well, that's typically what you get when you have a single sub and add a second. IOW, when I had them seperated, the one that had been out of the corner wasn't contributing anything, SPL-wise. It was just providing "air pressure."

Thanks for your interest and suggestions. I'm really glad you brought up the global EQ specifically as it just proves there's always more to learn and more to tinker with.

LOL, you're going to burn so-o-o much time you'll probably end up throwing darts at my picture instead of thanking me! Well, at leasts its fun and easy now with REW - wasn't the case way back when I was running all those sine waves and plotting my charts on graph paper!

Oh yes, and I'll follow the rules for posting charts next time too, my bad on that one.

Well, your scaling is fine, but you're going through a lot of trouble doing them as screen shots. See that little "save" icon in the bottom left-hand corner of the graph? Click on that to save the chart as a jpg.

Regards,
Wayne

[Magialisk]

Wow, so much information in one post. Most of it makes sense but I have to admit that one thing has me a tad confused.

Wayne A. Pflughaupt wrote:

No, one is not working harder, at least not now, because they're both getting the same electrical signal.

That part makes sense to me. It's a Y-cable from the receiver, identical EQ settings, identical amp gains, identical from end to end so they're both working just as hard as each other. That puts my mind at rest over that issue. Now if only there was a way not to work my 110'' window (the one right next to the corner loaded sub) so hard.

Wayne A. Pflughaupt wrote:

Now, if you do that, then the "red" sub will be working harder, because you'll have boosted its electrical signal (via EQ) to make up for what the other is getting for "free" from the room.

Hmm... again this makes sense, but at the same time it's slightly confusing. If I were to make both subs flat it would simply be by reducing the effect of my filters on the red sub. So where green might need a 12dB cut at a particular frequency, red might only need 6dB. I would not be raising the overall level (amp gain) of the red sub to match the green, but would leave it the 2-3 dB down that it's not getting from being in a corner. On the one hand I can see that by reducing the cut of the filter the electrical signal was in effect "boosted" thus it's working a bit harder, but only on certain frequencies, and certainly still nowhere near as hard as it would be working without any EQ.

I guess my brain is having trouble wrapping around this question. Is it better to:
1.) Have an identical signal to both subs even if one is clearly dominant (4-8dB higher across most of the spectrum) and the other is only shoring up where the first is weak. *or*
2.) Have two individually flat subs that are within 2-3 dB of each other for the majority of the spectrum.
The red sub is still going to be lower than the green except where the green is weak, but the separation in strength would be much less, if that makes sense. But with the combined SPL mostly following the upper envelope, would it even matter?

Wayne A. Pflughaupt wrote:

Well, you did mention that you realized 6 dB in overall gain, so it looks like you've indeed bought yourself some extra headroom.

I have to say my main intent here was to relax the load on my original sub while preserving my desired output level. Any gain in output or low end extension was just a bonus. From what I can tell it looks like I realized about 4dB across the board when using the two subs together. I'm judging this from a few different spots. In the 15-18Hz area for example you'll see the individual subs are both pretty flat at about 71dB, but combined it's more like 76dB or a 5B gain. Also at 60Hz where the individual subs intersect around 78 dB, the combined graph shows about 82dB. Surprisingly even the highest of the green peaks (89dB @ 43 Hz) rises 3-4dB when combined with the red sub which was a whole 9dB lower at that frequency. Long story short, I'd say I did pretty well gaining the headroom I wanted.

Wayne A. Pflughaupt wrote:

See that little "save" icon in the bottom left-hand corner of the graph? Click on that to save the chart as a jpg.

That just epitomizes how much more I could potentially learn about REW. Thank you though for the pointer, and for all of the advice above.

- Marc

[Wayne A. Pflughaupt]

On the one hand I can see that by reducing the cut of the filter the electrical signal was in effect "boosted" thus it's working a bit harder, but only on certain frequencies...

Yup, reducing the cutting filter on the red one will have the same effect as boosting those frequencies. The net effect will be that the red sub sub will be working harder than the green one, because the green's level across that range remains depressed by the equalizer.

...and certainly still nowhere near as hard as it would be working without any EQ.

More oft than not, any equalizing places additional demands on a sub, both its amp and driver. Before equalizing, your operating sub level is determined by any peak in response that you might have (~43 Hz in your case). When you cut that peak, you now find that your sub is too quiet, so you have to turn it up to compensate. Naturally, that places more demands on the sub than before.

I guess my brain is having trouble wrapping around this question. Is it better to:
1.) Have an identical signal to both subs even if one is clearly dominant (4-8dB higher across most of the spectrum) and the other is only shoring up where the first is weak. *or*
2.) Have two individually flat subs that are within 2-3 dB of each other for the majority of the spectrum.
The red sub is still going to be lower than the green except where the green is weak, but the separation in strength would be much less, if that makes sense. But with the combined SPL mostly following the upper envelope, would it even matter?

See, already you're obsessing. You should have taken my advice to just stick a fork in it and sit back and enjoy.

But - as long as we're obsessing, it's best to simply try it both ways and see which you like better. Either way you do it, demand placed on one sub vs. the other will no longer be the same. You only get that with no equalization to either one (assuming you keep both their gain controls set the same).

But the main thing you're after by adding the second sub is additional headroom. You're still getting that no matter what.

Regards,
Wayne