First time sub integration - dual Monolith 10" THX Select with KEF Reference 3

[morethinkingcapacityneede]

I recently got a pair of Monoprice Monolith 10" THX subwoofers, and had my first experience of subwoofer integration, with the KEF Reference 3. I used a MiniDSP nanoDIGI and two SMSL DO100 DACs, driven by SPDIF co-ax. Using a 80 Hz XO, with time alignment and sub eq (no eq applied to the KEF mains, just the 80Hz high-pass XO).

Subs are placed are placed diagonally, back left and front right.

The plot shows L+R main and 2 subs, with red trace being a frequency sweep, and blue taken using the MMM.

Any thoughts? I have about four room modes that I still need to fix, but I think it's definitely an improvement over the sub-less set-up.

 

[thcdru2k]

I would do summation alignment rather than just time alignment. There is a huge null right around your crossover. I’d adjust delays while measuring until the 60–100 Hz region has minimal cancellation, then reevaluate.

 

[Keith_W]

In addition to what @thcdru2k suggested (it's a very important suggestion!!!), you could consider moving your subs.

(From Welti and Olive's 2002 multi-sub paper).

You could also consider no high pass for your main speakers. That way you have four bass sources in the room, and more opportunity for smoother bass.

 

[morethinkingcapacityneede]

I would do summation alignment rather than just time alignment. There is a huge null right around your crossover. I’d adjust delays while measuring until the 60–100 Hz region has minimal cancellation, then reevaluate.

Thanks for the reply, and the suggestion. For sure, there's some phase effects.

I looked at the subs individual response, and I think the main dips in the L+R response actually are issue with individual subs response (see plot).

The plot has green L sub only, blue R sub only and red L+R as per first post offset by -4.9dB.

That being said, maybe there are phase issues at 52 Hz. Probably I am missing something

 

[morethinkingcapacityneede]

In addition to what @thcdru2k suggested (it's a very important suggestion!!!), you could consider moving your subs.

View attachment 538764

(From Welti and Olive's 2002 multi-sub paper).

You could also consider no high pass for your main speakers. That way you have four bass sources in the room, and more opportunity for smoother bass.

Thank you for the suggestion. Unfortunately, this is a living room, so sub placement options are rather limited, especially with these Monoliths... At least for right now, until a sofa goes

Unfortunately, the mains suffer suffer similar room modes as the subs above 50 Hz

 

[Keith_W]

The subs don't have to be front-to-back. They can be side-to-side as well. Anyway, one of the weaknesses of Welti/Olive is that only rectangular rooms were considered. Real rooms aren't always rectangular, and we don't always have complete freedom to place speakers and subs where we want. Front/rear diagonal is not so bad, it was discussed as an option in that paper. I can't remember if it was a 2nd or 3rd tier option.

 

[mglobe]

Question for the more knowledgeable than I. What about the nulls in the 150-180 range? That’s well above his 80hz xover. Suggests to me that playing with the locations of the mains and/or MLP might be worth the effort?

 

[Keith_W]

Question for the more knowledgeable than I. What about the nulls in the 150-180 range? That’s well above his 80hz xover. Suggests to me that playing with the locations of the mains and/or MLP might be worth the effort?

Could be room modes, could be SBIR, who knows. Not enough information in the thread so far. If he really wants to find out, he'll have to move his microphone and speakers.

 

[morethinkingcapacityneede]

I've attached a diagram of the room set-up, which hopefully might provide some clues. I also tried to predict the response using a simple calculation which shows the room modes in 2D (code from this thread: https://www.audiosciencereview.com/...ds/arbitrary-shape-room-mode-simulator.20805/), but haven't tried to follow the suggested locations.

I agree with the replies, more work is needed on speaker placement and MLP to smooth out the <200Hz region.

 

[janbth]

The diagram of your room set-up and 2D room sim may provide some hints about your measured response in #1. These response plots show some peaks at around 40 Hz and 85 Hz. Looking at the room sim (such simulations are always approximate) these probably match the figures at around 39 Hz and 82 Hz. They show that your subs are located in opposite polarity regions - red and green in the figure. However, the *peaks* in the response indicates that your subs play *out-of-phase*. Is that correct? Try to play them in mono (exactly the same in-phase signal). That should knock down the peaks. A basic multi-sub mode cancelling technique.