Dual Subs - looking for some guidance, please

[Bigdaddy999]

I'm several rounds of measurement and tweaking so far, and would like some feedback on how things look on the numbers end.

I have dual subs in a 4200cf room with openings to a kitchen area. The subs are located just outside the mains. My hope is to leave them where they are, but there are other placement options if need be if I cannot get this to work well enough.

I have a BFD 1124 and am manually tweaking the REW results after doing the Optimize, etc. in an attempt to smooth out things a bit.

First, here's the measurement with the orignal settings, including some phase shifting of the subs in the processor and a xover at 80hz. (BTW fronts are B&W N804's. Subs are Velo HGS-12's.

orig 10-21 2 subs no phase correct.jpg

After realizing the crossover might be important to shift down due to some nasty stuff near 50 Hz, I ended up applying these filters using a 60hz xover:

Nov 13 17-29 filters man tweak.png

This resulted in the following curve (no smoothing)

nov 13 17-29 man tweak with corrections.jpg

Any suggestions on what to do from here? At this point, I think I'm over-correcting with some of these overlapping settings? No house curve has been loaded, and doesn't yet have the sound I want, but not sure what's missing. I plan to play with this during the coming week, so I'm reaching out for ideas. THANKS!

[brucek]

Yeah, I guess you could get away with less filters if you wanted to get creative. The response looks pretty good though.

I think it's more important to get the sound you like over any worry about following a script, so if you're not satisfied, play around with it some more.

I don't know if I follow the logic in lowering the crossover. The lower the crossover, the less control you have over the signal, since you can only eq the subs. If you drop the cross from 80Hz to 60Hz, then you've lost some eq control over that area.

Any filters above the crossover won't have the impact of those below the crossover, since the mains are now in the mix. Best to eq from the cross and lower, and then add the mains and see were you are..

brucek

[Wayne A. Pflughaupt]

Hi Bigdaddy,

Thanks for showing us your filter panel along with your response chart, as well as leaving on the filter effect – it tells us a lot.

Where to begin? You’re way over filtered. A good number of those are unnecessary.

First, I would eliminate flters #2, #4, #6, #7, and #10 – i.e., everything with a bandwidth less than the 5/60 setting. Filters that narrow accomplish nothing good (you might want to review the “How to recognize useless filters” section of my Minimal EQ article).

Next, since your crossover point is at 80 Hz, you can eliminate #11 at 93 Hz.

Your baseline response below 30 Hz looks pretty good, so I don’t think anything down there is useful.

Filter #8 at 63 Hz, I don’t see any reason for a boost there.

Keep in mind that REW’s auto equalizing is tailored towards addressing modal ringing, not overall response smoothing, which often requires some boosted filters (another topic I discussed in the above article). I don’t see how to avoid doing some boosting here, short of experiment with different placement options for your subs (occasionally separating them results in some response dips).

Plus, even with a majority of cut-only filters, you’ve ended up with a substantial boost below 25 Hz, which is going to tax your amplifier headroom. Whether or not this is a issue depends on how much reserve your sub amps currently have.

Overall, you have a rather problematic baseline curve. You’re facing either wholesale gain reduction below 35 Hz, using an assortment of filters, which is not good. Or, substantial boosting in the 40 Hz and 52 Hz areas. Also not good. I suggest trying some different placement options. One I would definitely look at is co-locating the subs in a corner, if that option’s available.

Regards,
Wayne

[Bigdaddy999]

Hi Guys, and thanks.

Wayne - the curve is the result of both the filters and the shift to 60hz xover. I played with that AND the phase, so I probably shifted too many things at one time. BUT, it appeared that the xover change eliminated that dip/peak around 50hz

So I can go back and try again with the simpler filter set. as I recall, REW suggested most of those, and I added 2 more, then tweaked a few of them further within their ranges to try to flatten out the dotted line resulting curve. was that last step not appropriate?

I'll try some of your suggestions. I'm pretty happy with the lack of ringing in the room, tho. I guess that's part of what I'm hearing with the corrections thus far. Less boomy sound?

Anyway, i'll get a new baseline at 80hz with the current phasing and start from there.

Thanks

[dave7]

I would also question how you are matching the two subs together and with the mains. This can be tricky sometimes as the subs can work against each other and hurt the mains around the crossover point.

Dual subs can really help even out the room response even before any EQ.

Although - I am admittedly far from an expert.

[Bigdaddy999]

Dave - I'm not quite sure what you're asking. The subs are both coming of LFE outputs, and the charts are ONLY the subs so far. One thing I can adjust with the processor is the phase angle of the sub outputs, and so far, I haven't really tried to "match" them with the mains. if the response curve for the subs is improved with the EQ, then I assume that changing the phase would not affect the sub's room response by themselves? But when I start to add back the mains, then I may see interactions that are phase-related?

In any case, maybe I'm not understanding what you're suggesting here, if different from this. Thanks!

[Sonnie]

Hey Bigdaddy... check out this Sub locations thread where I did some measurements for two subs. It might be of some help or possibly give you some ideas.

[dave7]

The subs can interact with each other relative to the listening/measurement position. If you adjust the phase of them both equally through the processor than you are correct, you won’t likely see any change when you measure them without the mains being active. But, adjusting the subs’ phase individually can alter your response.

I feel this is important because your sub(s) don't share the same main baffle as you mains so they need to be timed correctly in order for their output to match the other drivers in you room properly. I think of phase as time aligning the subs to each other and to the mains. (BTW – I know there is an inherent problem with this logic since often the subs are farther away than the mains…but I solder on anyway)

To begin, I adjust the phase while playing one sub with one main to get the smoothest response. You will see using REW or RplusD how phase alters the response mostly at the crossover point where the two speakers are outputting the same frequencies. FWIW, I was never successful doing this by ear or even using a rat-shack meter. The graphical display in REW/RplusD makes it painfully easy to do though. Then I adjust the second sub's phase to get a good response with the first sub.

As an alternative, you might get the two subs working well together without the mains by using their on-board phase adjustment, and then alter the phase through your processor to match the mains.

Here's another surprising illustration - I did all the above and got things adjusted really nicely between both mains and both subs in all kinds of combinations. Then, I measured the FR of my center along with the subs. I had a huge null around the crossover frequencies. I thought I simply had the XO set too high, but that wasn't the case. I thought to myself "****!" because everything was so good with the mains, what the heck was going on. It was Terry Montlick of TM Labs (whom I had hired for acoustical consulting) who said it could be a simple placement issue with my center.

He ended up being correct. I simply moved my center rearward a little bit and that null filled right in perfectly. Incidentally it ended up being equidistant to my listening position as the mains. The lesson here is that just because I had it all in phase and level adjusted for the mains didn't mean squat until I tested each speaker individually (not easy or even possible for most people). The resulting improvements were stunning. It is not hard to do all this adjusting, but it does invite all sorts of obsessive-compulsive jabs/comments/jokes from our better halves.

I am far from an expert on these matters. I learned this when I worked with Terry. This is just my experience. I get the idea that most folks don’t set up their single sub correctly and two is even more of a challenge. But the in-room results are usually better with two...at least in my room they are, and I am not referring to SPL…I get a much smoother response.

Sorry about the length of this post...

[Bigdaddy999]

Thanks, Sonnie. An interesting read but of course YMMV, as they say as I doubt I'll get a curve that looks like the one you achieved! But I will start fooling around more with location options when I get a chance w/o the wife in the house. She does NOT appreciate the subtleties on this stuff at all so symmetry is cool, but putting a sub, say, in the middle of the room, to hit that perfect balance would be "painful". Too bad you can't correct WAF with REW and a BFD! Now THAT'd be a product combo.

Dave - the subs I have offer a 0/180 switch only, and I've tried to place them in symmetric location relative to the fronts, and on a plane with the fronts and the center. I havent started to try to measure with the mains on as yet, but in effect I'm doing the "trivial" version of your "alternative" with both subs getting the same signal and in the same phase (assuming they're wired identically, of course). I like your suggestion of matching with each the mains, but the reality will be that I can control phase only for both subs in tandem from the processor. but I can control crossover for each of the 5.1 channels independently.

Thank you both for the food for thought. Will post some results when I get the chance to do more.

[Sonnie]

Bigdaddy999 wrote:

Too bad you can't correct WAF with REW and a BFD! Now THAT'd be a product combo.

Oh yeah... we'd be millionaires...