Today things started coming together - I finished wiring up the extra I/O connectors for the crossover test boards and also finished the setup and configuration for the new test cart with the Earthworks microphone and using Fuzzmeasure, my old favorite, for test software.
Here's a shot of the cart with the microphone stand in the foreground...
It's relatively compact, easy to move around, and follows my traditional signal path, using the Motu interface and its excellent microphone preamp with 48V phantom power running the Earthworks M30.
Here's a few of the test cabinets with the crossovers connected behind...
When the crossovers are "finalized", I'll create a small cabinet for them and an internal wiring setup with a single input connector. Note that I added double thickness riser plates with binding posts on both input and output for these crossover boards, in additional to the Euroterminal connector for I/O.
Here's a view of the full setup under test:
Well, how do things look? Or rather that is, test?
I think some important things were accomplished with this build.
Today I did tests at the "traditional" 1 meter, which for a speaker this size can present issue of adequate integration of the drivers- that was something I wanted to examine, and did so using a measurement height range of 35" (a low seating position, and obviously quite nearfield!) and an upper position of 48". At a reasonable listening distance, this correlates with much greater extremes of height. So, for a speaker of this size, it's a bit of an acid test, given the driver spacing.
in the range of +/- 15 degrees in the horizontal axis, there was almost no difference, so I went with on axis for these first checks.
Half Hamming window was used, with a restricted duration to reduce room interaction on measurements, so the bass range is truncated a bit in measurements.
47" from floor:
35" from floor
Something I noticed right away is that the value of L9 is probably actually off in spite of the LCR meter checks, because wiht the 3uF parallel cap, the notch should be tuned to 31kHz, but it's more like 28-29 kHz here. So L9 needs to be tweaked.
But considering the measurement distance and the change in vertical angle, things look pretty good, though with the T34B-4 tweeter I am seeing the slight rise in the top end for the higher "elevation" measurement which I expected based on differences between the T34A and T34B data sheets. However, the driver integration looks quite good as regards the crossover point, and a quick check produced deep nulls with reverse phase. Note, the ripples in the midrange appear to be due to cabinet diffraction, and are quite sensitive in location and level with different heights. The 37" off the floor measurement was the worst.
Two comments about "off axis".
With a 1m measurement distance and a measurement height of 37", one is actually getting fairly off axis on the T34B, sitting up there at about 60" above the floor. A 1" tweeter would certainly do better, but not have the dynamic range and low distortion of the T34 series.
Now, what about off axis? +/- 30 degrees looks pretty good, but as you approach 40 degrees, the limitations of the midrange driver come into play... as well as the 34mm tweeter.
Considering the drivers, this is in line with what I'd expect, and is pretty decent all in all.
After I examine L9 and see about a tweak, I'll be doing more measurements. But right at this point it looks like the driver and crossover integration is meeting the targets I'd hoped for, in a reasonably wide axis in front of the speaker.
Of course, the important thing is how does it sound?
I have only had the opportunity to play some files from the MacBook Pro test computer, via the Motu interface and the Behringer test amplifier. so far, I planned just glorious living mono listening to "The Best of Fourplay" and "Play Bach: The 50th Anniversary Recording" by Jacque Loussier, but shortly after putting Fourplay on, I noticed ET, who has been supervising this effort, nodding his head and.... foot tapping. That's almost unnerving to see... then he insisted on some Stanley Clarke. Apparently he has a thing for "Bass Folk Song #10". After hearing with the crossover updates, I can see why. I guess that was a good sign...
Here's a shot of the cart with the microphone stand in the foreground...
It's relatively compact, easy to move around, and follows my traditional signal path, using the Motu interface and its excellent microphone preamp with 48V phantom power running the Earthworks M30.
Here's a few of the test cabinets with the crossovers connected behind...
When the crossovers are "finalized", I'll create a small cabinet for them and an internal wiring setup with a single input connector. Note that I added double thickness riser plates with binding posts on both input and output for these crossover boards, in additional to the Euroterminal connector for I/O.
Here's a view of the full setup under test:
Well, how do things look? Or rather that is, test?
I think some important things were accomplished with this build.
Today I did tests at the "traditional" 1 meter, which for a speaker this size can present issue of adequate integration of the drivers- that was something I wanted to examine, and did so using a measurement height range of 35" (a low seating position, and obviously quite nearfield!) and an upper position of 48". At a reasonable listening distance, this correlates with much greater extremes of height. So, for a speaker of this size, it's a bit of an acid test, given the driver spacing.
in the range of +/- 15 degrees in the horizontal axis, there was almost no difference, so I went with on axis for these first checks.
Half Hamming window was used, with a restricted duration to reduce room interaction on measurements, so the bass range is truncated a bit in measurements.
47" from floor:
35" from floor
Something I noticed right away is that the value of L9 is probably actually off in spite of the LCR meter checks, because wiht the 3uF parallel cap, the notch should be tuned to 31kHz, but it's more like 28-29 kHz here. So L9 needs to be tweaked.
But considering the measurement distance and the change in vertical angle, things look pretty good, though with the T34B-4 tweeter I am seeing the slight rise in the top end for the higher "elevation" measurement which I expected based on differences between the T34A and T34B data sheets. However, the driver integration looks quite good as regards the crossover point, and a quick check produced deep nulls with reverse phase. Note, the ripples in the midrange appear to be due to cabinet diffraction, and are quite sensitive in location and level with different heights. The 37" off the floor measurement was the worst.
Two comments about "off axis".
With a 1m measurement distance and a measurement height of 37", one is actually getting fairly off axis on the T34B, sitting up there at about 60" above the floor. A 1" tweeter would certainly do better, but not have the dynamic range and low distortion of the T34 series.
Now, what about off axis? +/- 30 degrees looks pretty good, but as you approach 40 degrees, the limitations of the midrange driver come into play... as well as the 34mm tweeter.
Considering the drivers, this is in line with what I'd expect, and is pretty decent all in all.
After I examine L9 and see about a tweak, I'll be doing more measurements. But right at this point it looks like the driver and crossover integration is meeting the targets I'd hoped for, in a reasonably wide axis in front of the speaker.
Of course, the important thing is how does it sound?
I have only had the opportunity to play some files from the MacBook Pro test computer, via the Motu interface and the Behringer test amplifier. so far, I planned just glorious living mono listening to "The Best of Fourplay" and "Play Bach: The 50th Anniversary Recording" by Jacque Loussier, but shortly after putting Fourplay on, I noticed ET, who has been supervising this effort, nodding his head and.... foot tapping. That's almost unnerving to see... then he insisted on some Stanley Clarke. Apparently he has a thing for "Bass Folk Song #10". After hearing with the crossover updates, I can see why. I guess that was a good sign...
Comment