Zing's New (Mismatched) Center Channel

[Zing]

If it were me, I would use a shelf EQ for treble and set the F3 to 16Khz and cut about 3 - 5 dB. I would test -5dB for about a week then start raising it 1dB per week and decide when it seems obvious the treble a little bright.

Truth be told, when we saw that rise at the end, I adjusted my 16K EQ to -6dB to compensate. Later in the night while Paul was listening to his reference material, he commented that a particular part of a particular song sounded dull and flat to him and he thought the -6dB adjustment was the culprit. So I turned off the EQ altogether and when he listened to that song again, the "problem" was fixed.

 

[Flint]

It is a balancing act with dome tweeters and their resonance. In my experience the best results with using an EQ is to split the difference on the peaks. If the peak appears to be 5dB, then 2.0 - 2.5dB cut typically works out best.

But resonance are tricky. Many present a null at the resonant frequency until the dome is mechanically energized, then it appears as a huge peak. The spectral decay plots show this and it is why the "split the difference" approach sometimes works best.

 

[AndySTL]

Bumpity :text-bump: :text-bump:


While waiting for the police to arrive yesterday, Paul and I got our geek on and Paul was successfully able to get my TrueRTA working again (it's still unknown what the specific problem was but it was obviously user error related).

Exsqueeze me?

 

[Zing]

As Chris & Jo left Monday morning, they noticed Paul's vehicle had been lightly sideswiped in hit-and-run fashion. At first, we all assumed someone came up our one-way street street a little too close to Paul's vehicle and kept going. But after some impressive investigation, a super sleuth that goes by the name of Zing deduced that it was a garbage truck that backed down the street that did the damage. The garbage company sent a rep who agreed with my assessment and set things in motion to have Paul taken care of ASAP.

But isn't great to start a post with "While waiting for the police to arrive..."?

 

[Flint]

Dang... I clearly have a one track mind.

I assumed you were eager to watch a concert video of the Police (Sting, Stewart, and Andy) while the testing was going on.

 

[Zing]

Interesting.


Green Trace: Left Sub only
Yellow Trace: Right Sub only
Blue Trace: Left & Right Subs

 

[yromj]

Interesting.


Green Trace: Left Sub only
Yellow Trace: Right Sub only
Blue Trace: Left & Right Subs

Apparently there is something asymmetrical in your room causing those two nulls at different frequencies. It doesn't take much of a difference either. What I find most interesting is that the null centered at 50Hz corresponds to ~11'. Are your ceilings that high?

John

 

[yromj]

Above 45Hz or so (where the left and right sub graphs cross), the right sub is detrimental to the output.

John

 

[Flint]

What I see are nulls at approximately 58Hz, 116Hz, 172Hz, and 250Hz... and peaks at 29Hz and 82Hz... see a pattern there? That aligns to a distance of 19.5 feet, or 8.75 feet, or any multiple thereof.

 

[Zing]

Asymmetrical? Yes. The rear portion of the right side wall is open to the kitchen.

Ceiling height is 9' to flat surface of the ceiling itself but I think it's worth noting that it's a beamed ceiling so maybe a fifth of the ceiling's surface area is 8.5'.

Room dimensions are 11.25' wide by 18.5' deep by 8.5-9.0' high. In addition to missing a third of the right side wall, about half of the rear wall is open to the dining room (which then opens to the living room...which is open to the hall...which leads to the kitchen...which leads to the HT).

I'm more disturbed about the width of the nulls than the nulls themselves.

 

[Zing]

and peaks at 29Hz

I was experimenting when these measurements were taken and the 25Hz EQ band was set to +1. I don't have similar sweeps without that 1dB boost to compare its effect.

 

[Razz]

What I see are nulls at approximately 58Hz, 116Hz, 172Hz, and 250Hz... and peaks at 29Hz and 82Hz... see a pattern there? That aligns to a distance of 19.5 feet, or 8.75 feet, or any multiple thereof.

uhhhhhhhh what?
:confusion-questionmarks:

 

[Zing]

uhhhhhhhh what?
:confusion-questionmarks:

Layman Translation? Unless you listen to music outdoors and 200' off the ground, every room has its problems.

 

[Flint]

What I see are nulls at approximately 58Hz, 116Hz, 172Hz, and 250Hz... and peaks at 29Hz and 82Hz... see a pattern there? That aligns to a distance of 19.5 feet, or 8.75 feet, or any multiple thereof.

uhhhhhhhh what?
:confusion-questionmarks:

Acoustics are insane because you have dozens of reflection surfaces working with and against each other. So, when you see mathematically calculated dimensions aligned with frequencies of modes and nodes, they may not be perfect multiples, but close to multiples of the root.

 

[Deleted member 133]

What I see are nulls at approximately 58Hz, 116Hz, 172Hz, and 250Hz... and peaks at 29Hz and 82Hz... see a pattern there? That aligns to a distance of 19.5 feet, or 8.75 feet, or any multiple thereof.

uhhhhhhhh what?
:confusion-questionmarks:

The frequencies are all ratios of one another - as you might well expect. For example 58Hz times two gives 116Hz, and divided by two gives 29HZ. It's a little less precise as you go to ever higher frequencies but you can see where the rest of the numbers come from.

Frequency and wavelength are related in that if you multiply them together you get the speed of sound. Since we know the speed of sound at the temperature for an average room, if we pick but one of the problematic measured frequencies (58Hz seems to be a prime candidate) then that gives a wavelength of 19.4 feet. You could also start with that distance and take multiples (or fractions) of it to calculate expected frequencies.

I know that Zing's room could present a challenge, because it is asymmetrical. However that might also prove to be it's strength - sonically. The problem is that esthetically, the best solution acoustically might not be the most pleasing to the eye, or functional as far as mobility within the room goes. The asymmetry gives an advantage in that it should help to minimize standing waves (in general) although it can lead to very specific nulls - as we see at 58Hz.

If I had carte blanche and was willing to simply play around with sub placement however I wanted, my first decision would be to place the two subs where they had no real relation to each other spatially (as much as possible) using my "prime numbers" method. Assuming that their height is fixed (although in practice that need not be the case) I'd have two dimensions to work with: distance from front and back walls, and distance from side walls (again knowing that Zing's missing some of those, in places, from his room.)

So for purely academic reasons, let's take Zing's dimensions of 11.25' x 18.5' (w/d). Note that prime numbers that encompass those two numbers are 1, 2, 3, 5, 7, 11, 13, 17, and 19.

For the left sub I'd start by placing it 3 feet from the front (making it 8.25 feet from the rear) and 5 feet from the left wall (making it 13.5 feet from the right). Take measurements and adjust the spacing in 1 inch increments / decrements to see if that improves. The key is that those dimensions, as much as possible, are not multiples of each other, and are as close to being prime numbers (in feet) as possible. (Having a room that is in itself a couple of prime numbers in dimension makes it even easier. (ie. if it were exactly 11 x 19 all the better.)

For the right sub I'd go with 5 feet from the front (6.25 feet from the missing rear wall) and 3 feet from the right wall (15.5' from the left wall). Go through the same measurement for the one sub - and make small adjustments as need be.

Then run them both together.

So what I've tried to do, in a nutshell, is to break the room distances between walls and subs down into prime numbers, and to make sure that they are not multiples of each other. The numbers I chose were not perfect (for example the right sub's 3 feet and 15.5 feet are very close multiples) however I like how they match up with the other dimensions / distances.

Again, this is for shits and giggles only. I suspect it will give good results BUT I'm not suggesting that it's a solution that Zing would want, or would like, for any of a number of other reasons.

However it does demonstrate what I mean by my prime numbers approach and there's still plenty of other numbers to play with - some of which just might both give good results and be practical.

Jeff

ps. If symmetry is wanted (say for esthetics), an approach would be to place each sub 3 feet from the front wall and 5 feet from each side wall. This gives a nice 5 - ~9 - 5 (feet from walls and subs) from side to side and 3 - 8 (not a prime, I know, but not a multiple) from front to back. The symmetry of the sub placement will be tempered by the room's asymmetry.

 

[Razz]

thanks guys... you guys know your shit...
I never dove into the science of room acoustics. Aesthetics are too important for me. But this is good stuff!

 

[Flint]

It is really quite simple...

 

[Deleted member 133]

And to make it even simpler there are a ton of online calculators available into which you plug your known variable and out pops the calculated solution - in multiple units so you don't have to mess around with doing the conversions yourself. (To think I had to crunch these numbers by hand back in university!) Here's but one example that I just found: http://www.sengpielaudio.com/calculator-wavelength.htm

Jeff

 

[Deleted member 133]

One other thing that I should have mentioned in my earlier post, is that I really like, as much as possible, to get the subs away from corners, and walls, where practical - especially when you've got a pair of beasts like Zing's that have absolutely no need for room reinforcement in order to put out a high enough SPL. I find it tends to smooth things out and reduces the chance of introducing mechanical room disturbances (buzzes, rattles, groans, what have you) into the mix. A lot of times we look at a room as an idealized beast, where acoustic waves simply bounce off with no interaction, when in fact they cause those walls to vibrate / deflect and generate their own "sound." Very few of us have purpose-built rooms where the walls are designed and built to be mechanically inert. I know I've got two places where I can excite some very noisy buzzes in my main HT at extremely low frequencies at very high output levels. Loud enough to be noticeable. I put up with it for now, but I'm sometimes tempted to rip out the walls in those two places and rebuild. And I don't have any real options left in terms of moving my multiple subs / speaker stacks around.

Jeff

 

[Razz]

great stuff.. but for another thread....

Zing....
you are rockin some serious speakers now and Im stoked for you!!!!!