MTRX Studio SPQ with Bass Management and Room Curve using REW

[ric982]

ETC / SPL - reflections / comb filtering.

[BScout]

Originally Posted by ric982

4" OC703.

Is near useless at 80Hz. Use your mic to measure different positions. Consider moving your sitting position front to back. Comb-filtering from the rear wall is rarely an issue (because, you, the listener, are sitting between the two.) Comb filtering from the floor, ceiling, or desk is the issue.
Room modes are an issue (because the wavelengths are soo big -- which is why you need to look at moving your listening position.)

An easy test: use stereo only (L and R). Pan L to R (using stereo only -- at phantom center, it should be equal out of both speakers) an 80Hz wave. If it dips, you have a problem. Realize that's going to happen with any source material and panning. Which is why you need to fix that phase issue. Once it's fixed, you can align the sub fairly easily.

It looks like the space behind the right speaker is vastly different than the left. That's going to cause a big issue at 80Hz (door? can't quite make out what's happening there.) Also, if it was me, I would not put your sub2 there. It's positioned very far from the actual hard boundary which means cancellation wave from rear. Also, not a strong recommendation to put a rear sub off center; against the left or right walls is better (better yet left or right corner.) Good luck.

[ric982]

Originally Posted by BScout

Is near useless at 80Hz. Use your mic to measure different positions. Consider moving your sitting position front to back. Comb-filtering from the rear wall is rarely an issue (because, you, the listener, are sitting between the two.) Comb filtering from the floor, ceiling, or desk is the issue.
Room modes are an issue (because the wavelengths are soo big -- which is why you need to look at moving your listening position.)

An easy test: use stereo only (L and R). Pan L to R (using stereo only -- at phantom center, it should be equal out of both speakers) an 80Hz wave. If it dips, you have a problem. Realize that's going to happen with any source material and panning. Which is why you need to fix that phase issue. Once it's fixed, you can align the sub fairly easily.

It looks like the space behind the right speaker is vastly different than the left. That's going to cause a big issue at 80Hz (door? can't quite make out what's happening there.) Also, if it was me, I would not put your sub2 there. It's positioned very far from the actual hard boundary which means cancellation wave from rear. Also, not a strong recommendation to put a rear sub off center; against the left or right walls is better (better yet left or right corner.) Good luck.

Yes, that's a door on the right. And there's also a 1' wide wall on one side of the rear closet - irregularities at both ends. Yes, 4" bottoms out about about 125 Hz. The cloud not up yet - it's 4" OC703 too, but it' has a 4" air gap so that will go lower. There are some big bass traps at the rear that aren't in the room yet that I'm hoping will deal with room modes. I think I'll want to get them in before I get too serious about what's happening at the moment. I'll probably be back here when I get there.

But thanks for you advice, though. And I think you've got me through understanding my calibration questions which is where this started.

[ric982]

Originally Posted by BScout

An easy test: use stereo only (L and R). Pan L to R (using stereo only -- at phantom center, it should be equal out of both speakers) an 80Hz wave. If it dips, you have a problem. Realize that's going to happen with any source material and panning. Which is why you need to fix that phase issue. Once it's fixed, you can align the sub fairly easily.

So I ran this with REW Full speaker noise as the source, BM on but sub's muted, so crossover is active but no subs. Mic pointed at ceiling with 90 deg cal at LP. I've got a mono aux track sending it's output to a stereo aux track and I'm using the mono tracks panner. REW SPL meter.

100% pan L or R reads 70 dB SPL. Panned to center, L+R reads 71.5 dB SPL, L 66.6 dB, R 66.6 dB. WLM on the stereo aux is showing 23.1 LUFS panned L, R, or Center.

If they are off phase, I would expect less as you say. In the measurements I took yesterday, at 80 Hz, no EQ L phase was at 162 deg, R at 101. I'm looking at the Overlay panel with Phases selected and the speakers of interest showing (like in the picture I attached before). Wait - need to switch to 80 Hz after calibration....

[ric982]

Ok so at 80 Hz.

100% Left 65.5 dB, 100% Right 62.6 dB. Panned center, 66.0 dB.

The attached SPL graph (no EQ) might also say much...
The 3 dB difference at 80 Hz is in the graphs as well the resonances.

The no-EQ (measured) and EQ (predicted) curves look the same in this region.

[audiogamble]

I have a question for anyone here who has done this calibration.

When designing the EQ in REW, there are a few options available for setting up the generic "Configurable PEQ". I am doing my best to match the specs of the SPQ card in here, but I am stuck on the setting for "Q/BW Format".

There are 4 options:

RBJ Q (half gain)
Classic Q
Symmetric Q
BW Octaves

Any idea what would be the most accurate selection for the spq card?

[BScout]

Originally Posted by audiogamble

I have a question for anyone here who has done this calibration.

When designing the EQ in REW, there are a few options available for setting up the generic "Configurable PEQ". I am doing my best to match the specs of the SPQ card in here, but I am stuck on the setting for "Q/BW Format".

There are 4 options:

RBJ Q (half gain)
Classic Q
Symmetric Q
BW Octaves

Any idea what would be the most accurate selection for the spq card?

It's Classic Q you want.
The only annoying thing with REW's "configurable" eq is you can't do shelves as part of the configuration (whereas some of the other presets have them.) So you can get 16 PEQ but shelves are kind of important and are left out. I always knock the number of filters available down to 14 in the configurable eq to leave a high and low shelf possible. So measure, apply a high and low shelf to get things better (using the generic eq), output a predicted measurement (that builds a measurement now using those two shelves.) Go to the predicted measurement, use the configurable eq with 14 PEQ filters.

When transferring the PEQ filters to SPQ, output eq filters as a text file and then copy from the notes section of the predicted measurement the two shelves (the shelves get put there when that measurement is "created")