Ok, bear with me, here. Since, it's a coaxial, I've been taking my measurements right up on the cone, with the mic flush with the baffle, about.
On DIY audio, there's a thread I was reading about pro audio coaxials, why would you want a raggedy pro coaxial? They were saying some of those dips, like the one at 5k, were often inherent in the sound, due to an internal reflection. I also noticed the Celestion coaxial has something very similar, a couple deep, narrow notches. So, I wanted to back off the mic placement to see if that stuff in the prior measurement was a reflection to the mic. I have to confess, I also made a couple tweaks to the hi cut filter to take down a bit more top...it was just a bit too much. But, it's mostly comparable.
So, this is three feet out, 1/6th smoothing with the scale back to how I usually have it. Generally, I consider it a victory if I can contain the waviness to a 5 or 6 db range, I've got some peaks at this setting, but generally pretty good...especially considering how flat that phase is. There's still that something at 5khz, so I guess it's part of the driver...just at the bottom of my 5 or 6 db window. Now it could be deeper and just covered up by the smoothing, so here's another look.
Now, that is 1/48th smoothing, and we can see that the 5khz, although it has another 5 or 6 db to plunge, it's super narrow, comparable to some of the waviness in the bass due to letting the room into the measurement. It's certainly something I don't hear, but does seem to be part of the driver, rather than a reflection off the mic.
Really happy with the impulse at 3 feet, too.
While I was at it, I tried to get something close to the listening position. This one is pretty smoothed to show the general idea...
Ok, the blue one is both speakers on, and the yellow is just one. Now, I am convinced my Dayton USB test mic is not a true omnidirectional. It has struck me as having some strong cardiod tendencies at the highs, so with both speakers on, mic pointed at the center, it gives a droop in the 4k-8k which I don't think is real. Hence, I added the single speaker with the mic pointed at it, to give the general sloping downward, target curve like response I've ended up with at the listening position.
But, I like the blue curve to try and show where the bass is. I've circled the general 45 hz area of where -3db was modeled to occur. If one looks at 68 db as a peak, and perhaps "flat" is the 84 db line around the 600-1.5k area, then that's about right. I guess I'd have to measure outside to know for sure. Who knows? I tend to always have that dip in the 100-200hz range in this room.
I may REALLY be ready to order my resistors, now....probably not....
