Not sure if this is disturbing, or encouraging...

[glaufman]

So I finally got around to taking some scans at my cousin's place... I was time limited, so I didn't play around with trying to fix what I saw, figured I'd go back there at some point... I'm not sure if these scans disturb me greatly, or if this is easily fixed with the eq's built into his Marantz 5002... please help walk me through the logic...

First, he's in an apartment, and didn't much want to hear about proper placement, he already knew where everything was going to go, so that's less than ideal, although for the purposes of these measurements, it's not all that terrible either. Room treatment (much to my chagrin) is simply not an option with him. It's a do-the-best-you-can-with-all-limbs-tied etc.

I'm showing two scans, one showing the crossover point between the sub and fronts, and the other full range. In each case, these are averages of the actual scans taken... for each average, I averaged 2 scans each in 4 positions, so each scan here is an average of 8 scans, which nicely smooths out the comb filtering in full range which is why I didn't bother smoothing the curves. I can 2 distinct peaks in the sub response that could be addressed, but what I'm focusing on here is the shelf around 160-200 Hz... The BW of the sub was left wide open, with the AVR's Xover set to 100Hz, so I would not have expected an interaction at 160-240, but is that what I'm looking at? I had EQs on the AVR turned off, in a no-processing mode other than bass management... should this be fixable by just giving gain to the right eq bands in the AVR? Or is something more sinister going on here? (the LCD FP sort of hangs in the middle of the room, could be partially blocking one of the 2 fronts...) I carefully checked the phasing of the fronts, and the sub, with test tones from the THX demo disc... since he is a distance away, I'd like to have a good gameplan before I go back ot fix this to make the most of my time... anyway, here ar ethe scans..

[glaufman]

Forgot to mention I also carefully set levels using RS meter, dont' remember if we used tones from THX demo disc or the internal ones on the AVR...
Should be noted that after these scans were taken, we watched some demo material and decided the rears were overpowering the fronts... and increased the gain on the fronts to compensate... so I wonder if they were mis-set to begin with... but then the question would be why?

I guess this raises another question, which is, what tones to use? Are the ones in THX ok? Better to use internal tones on the AVR (depending on the AVR?) I assume pink noise is the route to go, but full or BW limited? What limits?

[Wayne A. Pflughaupt]

should this be fixable by just giving gain to the right eq bands in the AVR? Or is something more sinister going on here?

Doesn't look too bad actually, but the RS meter isn't terribly useful above ~5 kHz. You'll need to get a ECM8000 and Xenyx 802 set-up for accurate full range measurements.

But using the full-range graph just as a "for example," the peak centered at about 5 kHz is about the only thing that would need adjusting with the main speakers. The receiver's EQ could probably take care of it, with a few dB cut at 5 kHz.

Should be noted that after these scans were taken, we watched some demo material and decided the rears were overpowering the fronts... and increased the gain on the fronts to compensate... so I wonder if they were mis-set to begin with... but then the question would be why?

I hope your full range graphs weren't taken with all channels going at once. Full range measurements should only be taken one speaker at a time (perhaps w/sub).

The built-in tones in a receiver are sufficient for setting main-channel levels.

Regards,
Wayne

[glaufman]

Hi Wayne... Thanks for the response...

but the RS meter isn't terribly useful above ~5 kHz. You'll need to get a ECM8000 and Xenyx 802 set-up for accurate full range measurements.

No problem there... used RS standalone to set levels, switched to ECM for sweeps...

But using the full-range graph just as a "for example," the peak centered at about 5 kHz is about the only thing that would need adjusting with the main speakers. The receiver's EQ could probably take care of it, with a few dB cut at 5 kHz.

At the moment, I'm less concerned with that than I am the steep dropoff between 160 and 240 Hz...
I think that's far enough away from the xover point on the rcvr (100Hz) to worry me that it's not simply a level issue between fronts and sub... also, the steepness particularly disturbed me, as it seemed steep compared to what I imagine a built-into-the-avr-eq to provide?

I hope your full range graphs weren't taken with all channels going at once. Full range measurements should only be taken one speaker at a time (perhaps w/sub).

Well, yes and no, sort of...
Full range scans were taken with sub and front LR... in this case, shouldn't be an issue... as this dropoff was seen in all 8 scans taken in four different listening positions, I don't think it's a simple interaction between the 2 front speakers such as a phase mistake...

The built-in tones in a receiver are sufficient for setting main-channel levels.

That's rather a bold statement to make as a generality across all receivers, is it not?
Seriously, though, better to use the receiver's built in stuff, as opposed to a tone form a dvd or even REW's own generator?

[Wayne A. Pflughaupt]

At the moment, I'm less concerned with that than I am the steep dropoff between 160 and 240 Hz...

Well for one thing it looks like the sub level is way to high.

I think that's far enough away from the xover point on the rcvr (100Hz) to worry me that it's not simply a level issue between fronts and sub...

Not really. Most phase problems typically show up an octave or so above or below the crossover point. We really won't be able to tell what the problem is without separate mains-only / sub-only graphs.

also, the steepness particularly disturbed me, as it seemed steep compared to what I imagine a built-into-the-avr-eq to provide?

Yep. A receiver's EQ isn't powerful enough to equalize subwoofers.

That's rather a bold statement to make as a generality across all receivers, is it not?
Seriously, though, better to use the receiver's built in stuff, as opposed to a tone form a dvd or even REW's own generator?

Well, a receiver circulates the test tone around all the channels. Hard to beat that for convenience. I suppose a DVD that could do the same thing would be fine. Using REW - at the back of the receiver you'd have to manually move the cable for each speaker to the channel that's generating the test tone. Sounds like a hassle to me.

Regards,
Wayne

[glaufman]

Well for one thing it looks like the sub level is way to high.

You mean too high as compared to the mains, yes? Which could just as easily be said as the mains are too low, yes? I thought this, but discounted it because if it was that simple, it wouldn't occured at the crossover? Because the sub looks like it's sitting at an average of about 75db, which is considered just about right, no?

Most phase problems typically show up an octave or so above or below the crossover point.

So with Xover at 100 Hz, this problem starting at 160Hz, if you had to guess, you'd guess it would be a level issue, not a phase issue, even though you can't say for sure at the moment?

Yep. A receiver's EQ isn't powerful enough to equalize subwoofers.

I wouldn't have been looking to eq the sub, but the mains... increase the gain on the mains above 160 Hz, leave rest alone... (again, as a starter)...

Well, a receiver circulates the test tone around all the channels. Hard to beat that for convenience. I suppose a DVD that could do the same thing would be fine. Using REW - at the back of the receiver you'd have to manually move the cable for each speaker to the channel that's generating the test tone. Sounds like a hassle to me.

I was going to elaborate on my concerns, but I need to check something before I shove my foot further down my throat... suffice to say the disc cycled the signal around, I'm just unsure of the intention of those signals, and therefore whether they had the same spectral distribution, and therefore their appropriateness for using for this purpose...

[Wayne A. Pflughaupt]

You mean too high as compared to the mains, yes? Which could just as easily be said as the mains are too low, yes? I thought this, but discounted it because if it was that simple, it wouldn't occured at the crossover? Because the sub looks like it's sitting at an average of about 75db, which is considered just about right, no?

With the subs, what matters is the peaks. The loudest frequencies are what you'll base its level setting. Your mains, at least in your graph, are averaging ~65-66 dB,while your sub is peaking at 81 dB. Typically subs are run about 10 dB or so hotter than the mains (depending on the room and personal taste).

So with Xover at 100 Hz, this problem starting at 160Hz, if you had to guess, you'd guess it would be a level issue, not a phase issue, even though you can't say for sure at the moment?

One octave above 100 Hz is 200 Hz. So it could be a phase issue.

increase the gain on the mains above 160 Hz, leave rest alone... (again, as a starter)...

Not sure how you would adjust the level of the mains only above 160 Hz...

was going to elaborate on my concerns, but I need to check something before I shove my foot further down my throat... suffice to say the disc cycled the signal around, I'm just unsure of the intention of those signals, and therefore whether they had the same spectral distribution, and therefore their appropriateness for using for this purpose...

The intention of the signal is to use the SPL meter, and using the circulating tone, adjust the level of each speaker to the same level (e.g. 75 dB). It's spectral distribution (content?) isn't relevant, as long as it's in the frequency range of the speakers.

Regards,
Wayne

[glaufman]

I thought I replied to this already, but no matter, because I think some things just got through my thick head after staring at REW for a while...
Mainly, that a phasing issue doesn't necessarily imply a wiring mistake, but can simply be due to the relative positions of the sub/mains... yes?
Then, the phase interaction at any given frequency will be different, as it depends on pathlength as % of wavelength, yes? In other words the same phase issue presents itself as constructive interference at some points, destructive at others, and vanishes at others?
Therefore, it's possible that a falloff just below the xover, flattening at the xover and peaking just after the xover could all be phasing, and then the drop to the mains level is as the mains level out and the sub finishes its rolloff?
Of course, that could explain the iregularities, but the question would still remain what happened? I used the RS and cycling through the speakers set everything to 75db (+/-1), but running the sweep proves a vastly different level?

[glaufman]

Follow up... it's inconvenient to putt the avr out to get access to the multi inputs, is there (generally) another easy way to run the sub without the mains that I might be overlooking? I figure the mains could be rnu without the sub my simply turning the sub off temporarily...

[Wayne A. Pflughaupt]

is there (generally) another easy way to run the sub without the mains that I might be overlooking?

Unplug the speaker cables at the speakers.

Regards,
Wayne