Critical subwoofer tip

I won't criticize anyone who sets theirs subs up by ear. Everyone has their own way to do things. That's all part of the fun.
However, I think it's kind of like watching a car drive past and guessing how fast it's going w/out using a radar gun.
I spent a couple more hours today tweaking my subwoofer(s) setting to try and improve the frequency response. There's no way I could have gotten to where I am without the aid of REW.
I'm pretty happy with the first image, as it's only out by a couple of dB (in one place) from what's acceptable using EQ/DSP (5dB).
Second image shows my previous setup compared to today's changes. It also includes how the 'Mains Only - without Subs' measure.
https://imgur.com/a/aXlCoXh

I found a better result in my system by playing a repetitive kick drum and trying to match the beat sharper than softer.

It depends on how the kick drum was eq'd. For some music, there is no low end at all and other's next to no point. Additionally, we often flipped the foot phase on pop recordings.

My favorite piece by ear is a pipe band recorded out of doors. When the phase is correct, focus dials in just like a camera.

Everyone rightfully cites REL subs as having high-level inputs---the sub taking it’s input from the output of a power amp. Often not mentioned is that the Rythmik subs having one of the company’s PEQ plate amps provide both low (line) -level and high-level inputs---low-level on RCA jacks, high-level on speaker cable binding posts. Rythmik’s XLR model plate amps forfeit the high-level inputs.

Rhthmik plate amps also provide a continuously-variable phase control, which is labeled "Delay". The knob can be placed anywhere between 0 (no delay) to 16ms (roughly 180 degrees of phase rotation). Delay can also be achieved by physically moving the sub, but the delay control allows one to find the optimal placement in terms of room modes, the control then aligning the loudspeaker/sub electronically.

If the woofer of a loudspeaker and the driver of a sub are both reproducing, say, 80HZ, and the 80Hz wave from the two are 180 degrees apart in phase (opposite polarity), the result will be cancellation centered at 80Hz, resulting in a deep hole in frequency response. It is no different than the tweeter and midrange driver of the loudspeaker being out of phase with each other, except that with tweeters and midrange drivers in a loudspeaker, the designer is free to use x/o parts to boost the output of the drivers to achieve a flat response.

In the bass region, the in-room frequency response is by far what matters the most to the ears.

You see, the ear’s time-domain resolution is very poor at long wavelengths: We cannot even detect the PRESENCE of bass energy from less than one wavelength, must hear multiple cycles to begin to detect pitch. So by the time we BEGIN to hear the bass, the room’s effects are all over it (think of how long the bass wavelengths are in relation to the in-room reflection path lengths), thereby swamping any initial phase coherence or lack thereof. The transients and overtones which give us the timing information of bass instruments are much further up the spectrum.

On the other hand, the ear is very good at hearing changes in sound pressure level at low frequencies. This is predicted by the way equal-loudness curves bunch up south of 100 Hz: A 3 dB change at 40 Hz can be perceptually comparable to a 6 dB change at 1 kHz.

To the extent that the phase of the subwoofer(s) affects the in-room frequency response, it matters. For those who have a phase control on their subwoofer’s amplifier, it is a useful tool for getting a good blend between mains and sub(s). It is not as powerful as the some of the other controls (such as gain and crossover frequency), but it is still useful. Adjust the phase control for smoothest frequency response (after adjusting the gain and frequency controls) and you will have made the biggest improvement it has to offer in the area that matters the most.

Duke
subwoofer manufacturer

I wouldn’t advocate against the use of REW or other methods of DSP available if you have them. 
If you are attempting to dial in your sub to a system, phasing does have some importance, maybe especially for odd room acoustics or multi sub environments. 

I found the utilization of a sub/main sound wave null measurement at the listening position for a particular set of frequencies near crossover point,  then reversing polarity of the main speakers didn’t achieve the anticipated result. I used two identical dB meters, and varied the locations taking notes. I don’t have a perfect listening room, that was quite obvious. 
I found at the lowest apparent dB a general agreement, then main speaker phase reversal  would indeed show an approximate ‘highest output’  but the ‘walk around’ test still felt somewhat disjointed. 
After retrying this with an ear also to the timing of impact, I found an interesting thing that slightly shorter sub delay (I’d have to guess of several msec) built a more cohesive setup- both at the listening position and from various parts of the room and in the adjacent areas. 
This was using several types of recordings with dynamic sharp impact at an appropriate low Hz level output as called for in the situation. 
One would think it not to be so critical maybe especially using several subs. But this was the way I approached setup after finding the other methods to not quite nail it. 
It’s obviously only a suggestion to those trying to dial in using the sub/speaker bass null method that they may also find a slightly shorter (or maybe longer, depending) delay to attain the desired result. Cheers,
RW