A question about low freq responses...

[karl k]

Over the last several months, I've been trying to put together some main L/R speakers. They, for the most part, have been progressing nicely but I've run into a snag on the low end. Up til last weekend, I've been limited to testing with Speaker Workshop in my living room and while useful only down to 1KHz or so due to reflections corrupting the recorded signal, it did alow the mid high x-over point to be worked on with a fair amount of success. I took nearfield measurements for the low end and the port, cut and spliced everything together to make a nice graph. Well, to make a long story short, I went to my workplace to do some testing in a large enough room to measure the low end without reflections being a problem. I also wanted to measure from a distance that was more compairable to the actual listening distance in my livingroom. So I get there, unload and setup everything, put the speaker on one forklift and raise it 10' in the air. I put the mic(Behringer ECM 8000 test mic w/o calib.) on a stand, on another forklift 9' away and raised it to be level with the midrange on the speaker. I get all the volumes set and begin testing.

Now for the speaker...


2ea Peerless CSX 10" woofer(wired in series), in a vented enclosure 130L tuned to 26Hz

1ea 250wt plate amp, variable active x-over 40-160Hz, 2nd order Q=.275, x-over bypass switch.

1ea Peerless CSX 6 1/2" midwoofer in sealed enclosure 6.55L with stuffing. No high pass, 2nd order L/R low pass at 3KHz.

1ea HiVi RT2E-A planar tweeter angled vertically down Approx 10deg. 2nd order Butterworth at 3KHz

The drivers are offset from each other to facilitate time alignment. Woofer to mid = 1.5" and mid to tweeter = 2.25".

Here's a pic of the speaker in question...(see IM000052)


And now for the question...(see farfield)


If you look at the response from 300Hz and down, you'll notice a series of peaks and dips. Normally, I would attribute these to reflections in the room. But in studying the graph today, I noticed a pattern that cant be mear coincidence. I know the dip at 45Hz is most likely an impedence spike as it shows up on another graph. You'll notice the rest of the dips are spaced about every 25Hz apart. There's one at 22, 45, 70, 95, and 150. Can anyone explain what this might be? I thought about box resonance, have ruled out standing waves in the room(too large and too short of play/rec time for waves to build), interference from the midbass(peaks still exist in the same place and magnitude while only woofers are playing) and have just about ruled out reflections as the reflections would have to be very large and opposite in phase to create dips of this magnitude(I think). I just haven't learned enough about figuring resonances in box's yet to know if that's the cause or not.

I'll take any help I can get on this matter.