Originally Posted by Mark Seaton
When I saw the noEQ measurement on the screen of the 18.2 I asked Craig if they had the HP filter off on the amp and he said they had checked the amp. I'm pretty sure this was more a function of an acoustic anomaly in the room around 14Hz than differences in the high pass.
There's an acoustic anomaly in virtually every room between 10-15 Hz. I gathered some posted FR graphs over the years and noted the dips in that area, which were almost universal.
Had the measurement hardware been capable and the traces extended lower, they would no doubt have shown the equally universal lack of dips/peaks below 10 Hz until ultimate roll off, that is, in the case of sealed, no HPF subs, as all other alignments will fall into the noise floor below 10 Hz regardless.
With that out of the way, in the case of the CHT subs, it's simply a matter of the roll off being too early for room gain to compensate. In taking the posted graphs and placing the known native response vs the GTG in-room response, it's plain to see that the SubMersive ruled the day because it has the best native response to work with room gain than any of the other entrants.
There's exactly the same room gain profile seen by both subs. It's just that the SM has the better starting point before room gain. 15-20dB difference at 10 Hz before room gain is huge.
One other consideration is that these were not frequency sweeps with any sort of windowing/noise immunity, they were RTA-type measurements. At low frequencies noise is a real issue, and given the roll off before any EQ, the response below ~15Hz was likely getting into the noise floor for both the vented and sealed CHT subs, as you see other vented subs and the F20s all hit a floor below which the response doesn't drop.
This is true and mostly overlooked by those who measure FR at the LP. The level should be raised to 100dB to help avoid this problem. Assuming the levels were properly calibrated, 75dB is not high enough to avoid landing in the noise floor below 20 Hz.
In the end the peak/dip swamps the difference in the roll of of the two subs as it occurs in the middle of the roll off.
I have to disagree with this one. The SM trace shows the dip to be, in my experience, although not ideal, not a significant problem, and placement would probably have a positive effect in a real world application vs the one of convenience for the GTG.