Bass in rooms below 150Hz? how many peaks might I expect to have?

RE: Bass in rooms below 150Hz? how many peaks might I expect to have?, posted on December 31, 2016 at 05:38:16
b.l.zeebub Audiophile Posts: 9361 Location: 52deg 28'N,1deg56'W Joined: April 17, 2006	CABS (Controlled Acoustic Bass System) might be an interesting read. Think of it as an active bass trap and it is mainly dealing with rectangular rooms but is also applicable to other shapes. The results (ie reduction of peaks) is quite impresssive. http://vbn.aau.dk/files/12831869/AC-phd.pdf http://vbn.aau.dk/files/62729248/LF_sound_field_control.pdf

Long reply, posted on December 29, 2016 at 14:02:47

Dave_K

Tim,

I used a Rane PEQ 55 for equalizing a pair of stereo subs (link below). It is a DSP-based PEQ but with analog controls and analog inputs & outputs. It has 10 filter bands and two channels that can be used as a 5 band stereo EQ or 10 band mono EQ. In their low frequency setting, the filter bands are adjustable from 12.5-200 Hz and 1/12 to 2 octaves and the gain is adjustable -24 dB to +12 dB, which makes them perfectly suited for bass equalization. Also, four of the 10 can be configured as high or low shelf filters which I found to be very useful. The analog controls make it very easy to set up. The PEQ 55 is discontinued but they get traded around on eBay for ~$500 which is in your price range.

I read in your profile that your room is 32x14x8 feet and L shaped. But the 32' might be a diagonal measurement, so the actual length of the room might be somewhat less, like 25-27 feet? If so you have a difficult room because the dimensions are close to being multiples of each other. In the rooms I've measured so far, the strongest modes have been the lowest order axial modes associated with the length and width of the room, 1-0-0 and 0-1-0. Given the length of your room, 1-0-0 is going to be down in the vicinity of 20 Hz and I don't think it will be much of a problem. 0-1-0 should be around 40 Hz. A concern I would have with your stated room dimensions is that 2-0-0 is also going to be in the vicinity of 40 Hz, and depending on the shape of the L you might have 1-1-0 in that same frequency range.

If your room measures like the theory predicts, there are two ways you can deal with those strong low order modes: with equalization and/or with careful placement. When setting up my first identical pair of subs, I restricted my placement options to the front 1/3 of the room and ended up with stereo subs placed on the front wall behind and just inside of the speakers. That required using A LOT of equalization: deep, narrow notch filters at 35 and 42 Hz, a shelf around 20 or 25 Hz to reduce excessive room/boundary gain, and a couple of milder correction bands up around 60-80 Hz used to get a decent blend with the main speakers.

And that was after building in as much bass trapping as I could reasonably accommodate given the room size & shape. I have a mix of broadband and tuned bass traps and they are very effective at cleaning up the waterfall/RT from the mid-bass on up but they don't have much effect on the 35 Hz and 42 Hz modes and they don't prevent dips and nulls due to 1/4 wavelength cancellations which are entirely determined by placement of the speakers, subs, and listening position.

After a few years my original pair of subs started to flake out and I had to get new ones, and during the transition I experimented with 4 subs (albeit not identical). Not long after I heard about Duke's Swarm. These things helped me let go of the audiophile mantra that subwoofers should be placed near the main speakers for best integration. I ended up repositioning my subs based on two simple principles. First, if you don't want to excite a room mode, place the sub in one of its anti-nodes. For example, if you place a sub at the midpoint of one room dimension, it won't excite the odd-order modes associated with that dimension. Or if you place it at 1/3 length of one room dimension, it won't excite the even-order modes associated with that dimension. Second, if you place two subs symmetrically on opposite walls, they will cancel all odd-order modes associated with the room dimension they span. Given that the strongest two modes in my room are 1-0-0 at 35 Hz and 0-1-0 at 42 Hz, I placed the two subs at the midpoints of the two long walls.

I had great results with this subwoofer placement when combined with small stand-mounted speakers. It gave me flat response from both subs from the measurement limit (10-15 Hz) up to 60 Hz where the crossover kicks in, aside from a narrow null at 35 Hz, which is perfect because my main speakers have enough output 35 Hz to excite that mode moderately but not strongly enough to cause noticeable boom. I no longer needed any equalization, so I sold the Rane. And it only took two subs, not four.

So, to make a long story short, my advice is to start with two subs and work on placement first, taking advantage of mode structure to avoid exciting your room's worst modes. Second, use bass trapping to reduce the effect of higher order modes and to generally reduce reverberation and tighten up the bass. If that's not satisfactory, then add additional subs and equalization.

Dave

http://www.rane.com/peq55.html

Maaate, thanks for the ideas., posted on December 29, 2016 at 23:38:09
Timbo in Oz Audiophile Posts: 23221 Location: Canberra - in the ACT - SE Australia Joined: January 30, 2002	At last an analogue in, analogue out, DSP sub/bass eq. Stereo to boot. Warmest Tim Bailey Skeptical Measurer & Audio Scrounger

Agree about room, but, posted on December 29, 2016 at 04:12:37
E-Stat Audiophile Posts: 37666 Joined: May 12, 2000 Contributor Since: April 5, 2002	Have you considered room treatments like bass traps? I get remarkably flat measured in room response up to 200 hz (third octave plot in gallery) without the need for EQ.

RE: Agree about room, but, posted on December 31, 2016 at 04:51:45
b.l.zeebub Audiophile Posts: 9361 Location: 52deg 28'N,1deg56'W Joined: April 17, 2006	3rd octave measurements are almost completely useless. Try 24th or 48th octave smoothing for a much more useful result.

Believe what you will, but..., posted on December 31, 2016 at 11:48:53

E-Stat
Audiophile

Posts: 37666
Joined: May 12, 2000
Contributor
Since: April 5, 2002

I've found third octave in room measurements quite useful for bass trap and speaker placement for my large dipoles. Achieving very linear results at ten to twenty hz increments results in a very neutral overall sound.

I have test tones at 1 hz increments, but find such resolution unnecessary for that task.

Sorry, but I am not moved by tauntings from The Devil. :)

I would love to, but it's also the Lounge & Dining room., posted on December 29, 2016 at 12:42:47

Timbo in Oz
Audiophile

Posts: 23221
Location: Canberra - in the ACT - SE Australia
Joined: January 30, 2002

Hi there!

What with two large bookshelves and the two TL subs (once built) in the front corners, there isn't any room for even just two corner bass traps.

That's with two sealed push-pull subs under each QUAD 63.

But 4 spaced subs should give me some smoothness. As I hope will the two large tall bookcases 2/3rds filled, with random spaces.

Here's a q'n for you? I have two stereo SS power amps, known to be unconditionally stable into 'stats. One is a Fostex 300 - 150watts pc/4 ohms continuous and the other is a Perreaux 6000C - 380 wpc/8. This is the one my techie wants me to use into the QUADs. He knows stats well.

But, IF I used the Fostex 300 on the QUADs (1st order passive-line-level HP at 150Hz) I could then use the Perreaux for the two sealed P-P subs, and a rebuilt HK 870 for the two TLs. This would give me a good deal more headroom into the sealed subs, which are going to be Eq'd.

I also have two modified LEAK ST20s to use - strapped in mono Pentode mode - giving 40WPC off a virtual 2-ohm tap - for Winter use only into the QUADs.

Your thoughts?!

TIA

Warmest

Tim Bailey

Skeptical Measurer & Audio Scrounger

Depends on the room dimensions., posted on December 27, 2016 at 15:45:50
Kal Rubinson Reviewer Posts: 12436 Location: New York Joined: June 5, 2002	You should run them through a mode calculator such as these: http://www.bobgolds.com/Mode/RoomModes.htm http://realtraps.com/modecalc.htm

It is beginning to dawn on me that investing in analogue paremetrics to Eq a 'storm' array , posted on December 27, 2016 at 16:14:32
Timbo in Oz Audiophile Posts: 23221 Location: Canberra - in the ACT - SE Australia Joined: January 30, 2002	Is not a high ROI path. I'll just have to save up for some MiniDSP magic and a good ADC - in the medium term. Much else to do - once I am well - xxxxx fingers. Warmest Tim Bailey Skeptical Measurer & Audio Scrounger

If you're looking at miniDSP for subs, this might be of interest., posted on December 27, 2016 at 20:11:14

andy_c
Audiophile

Posts: 1470
Joined: June 2, 2007

The software is free. The subs need to be in mono for this technique to work though. The idea is to get flat response at multiple listening positions simultaneously. You make a bunch of measurements in Room EQ Wizard, export them as text files and import them into the program. After some setup and running an optimization, it calculates the filter coefficients to load into the miniDSP to optimize performance.

Multi-Sub Optimizer

Thanks Kal, but it is L-shaped. I was aware of both those sites., posted on December 27, 2016 at 15:50:34
Timbo in Oz Audiophile Posts: 23221 Location: Canberra - in the ACT - SE Australia Joined: January 30, 2002	I don't know of any spreadsheet / Calculator for L-shaped rooms. Does anybody? Or is there a way to overlay L-shaped room dimensions onto / into those calculators? Warmest Tim Bailey Skeptical Measurer & Audio Scrounger

RE: Thanks Kal, but it is L-shaped. I was aware of both those sites., posted on December 27, 2016 at 16:55:34
Kal Rubinson Reviewer Posts: 12436 Location: New York Joined: June 5, 2002	1. I agree that using electromechanical as opposed to software filtering is like going back to using hand-tools. 2. There are modelling programs for non-rectangular rooms. I no longer used them because, today, one can actually measure the modes more easily than one can analyze and predict them. So, I have used CARA 2.0 in the past but, now, I just measure with REW or DiracLive Calibration.

Speaker Asylum

Bass in rooms below 150Hz? how many peaks might I expect to have?

124.171.163.231