Mains have wild response compared to subs, is this normal?

[VERTIGGO]

I've successfully used the Reckhorn B1 and BFD to get a smooth curve out of my sub, but the mains look hopeless. When I measure them together it's even worse. Is this just the way it works? Are these major drop-outs usually just accepted in an untreated room?

My sub gave me this (B1 with ~17Hz subsonic and moderate 20Hz boost):


A BFD gave me this:


And all together now...:

[robbo266317]

You will probably find they are room effects.
Try moving the mic up about 12 inches and measure again.

[VERTIGGO]

I have noticed it vary with the mic position, but really what I'm wondering is do most people just eq the sub carefully and ignore the higher freqs (i.e. mains and surrounds) or is it important to find some sort of 5.1 equalizer?

Another question, is how can I determine how much my sub can handle in the subsonic range? According to WinISD I can only run signals down to about 17 Hz (tune) before my excursion spikes. Is there any way to find out how close I am to that so I don't have to just sacrifice that entire (0-17) range?

[robbo266317]

The Higher frequencies are helped by room treatment.
eg see http://www.hometheatershack.com/foru...ap-review.html

In WinISD you can set the power level on the appropriate tab and then look at cone excursion or look at it's maximum power tab (under the normal transfer function tab)
The Reckhorn filter is used to prevent excess sub-bass excursion.

[brucek]

Is this just the way it works?

When you post a graph please use the standard axis of vertical = 45dB-105dB and horizontal = 15Hz-200Hz using the Graph Limits button in the top right corner of REW. Set the size to 800 wide using the save graph icon in the plots lower left corner.
Use a target level of 75dB. You're measuring too high.

When you add the mains you will get interaction with the sub at the crossover transition that can be helped with the subs phase control. You can also try some positioning of the mains or sub to get a smooth response. Once the frequency gets above ~100Hz, you will start to see some comb filtering that can be relieved by using 1/3rd octave smoothing. This will reveal the signal that your ears will hear, as sharp, narrow dips aren't generally perceivable.

brucek

[DrWho]

VERTIGGO wrote:

I've successfully used the Reckhorn B1 and BFD to get a smooth curve out of my sub, but the mains look hopeless. When I measure them together it's even worse. Is this just the way it works? Are these major drop-outs usually just accepted in an untreated room?

Even a well treated room will measure similarly....

What you're seeing in the frequency domain is the effects of multiple reflections summing together at the microphone....so if a reflection is in phase at a certain frequency then they sum, but if it's out of phase then they cancel.

However, if the reflections arrive late enough, then we hear them separately from the direct sound and perceive it as a separate reflection. If you have a lot of densely spaced reflections, then you hear it as reverb. If the reflections arrive quickly, then we can't distinguish it from the direct sound.....and that will sound blurred and unnatural. But in all these situations, an FFT with a long enough window can measure all sorts of craziness like you're seeing.

Some people would claim that you can just run a 1/3 octave smoothing on the plot and get an idea for what your ear is going to hear, but I would suggest that doesn't tell the whole story. A dip in the frequency response (like one made by a super sharp EQ) may not be audible, but if that dip is the result of a reflection, then it will likely be very audible.

One of the goals for the ETC chart is to provide a way to identify the arrival times and relative amplitudes of all the reflections so that the room can be tailored to have a natural sounding reverberant field and not blur the direct sound. It's easy to talk about, but implementing solutions can be quite difficult depending on the room.