How are you controlling reflection/re-radiation of sound through the midbass driver (other than damping material)? I'm currently looking at using curved sides, coincidentally similar to what John Krutke is detailing on his blog.
If you use the correct internal absorption materials, it is a non-issue. You can directly refer to the A-mounted co-efficient vs. frequency absorption charts of various materials, and pick a type and thickness for each parallel wall group based upon this data. Proper material and application will mean that nearly 100 percent of sound within the intended range will not be reflected or reverberated back to the midrange diaphragm. I use, typically, 8lb/ft^3 density mineral wool board, as is also commonly used in room treatment construction. After all, high density fiberglass and mineral wool of this sort are commonly used in anechoic chamber construction to prevent almost all sound in the intended range from being able to reverberate or reflect. The high density fiberglass/mineral wool is so effective, in fact, that if you use too much, it will even damp the
desired resonance at LF in a sealed enclosure, causing LF to reduce in SPL.
1) Linkwitz appears to favour sealed systems, with EQ to increase bass. What are your thoughts on this, and what steps did you take to keep group delay and port noise low with your design?
Group delay in a typical ported alignment, so far as I can tell, is a non-issue. I am not aware of specific credible perceptual study that establishes the levels of delay at subject here in LF to be substantial for musical playback. I have conducted double-blinded ABX tests on prepared samples of music, simulating sealed group delay curve, ported group delay curve, and comparing them to the original. The differences are so small when using a highly linear monitor headphone(with zero room induced effects), that I can not consider it even a factor in a real room. You can PM me for the samples in order to conduct your own ABX if you wish. Note: you can easily achieve the 'sealed' sound with a ported design by applying negative E.Q.(rolling off LF gradually), but have a higher dynamic range at LF with a proper ported system.
Port noise -- this is one 8" woofer per channel. A 3" port PVC port was used. No port noise has ever been induced, that I found to be audible, on musical playback. You can play sustained LF sine waves at higher SPL near the port tuning frequency and hear some minor chaffing. For larger woofers, a larger port must be used, or a large radius must be used on the inside and outside of the port terminations to reduce turbulence along the edges of the vent. Port compression should also be considered, if using large or very high excursion woofers. Even a 6" diameter port may not be sufficient for some [extreme] woofers available today. This must all be balanced in the initial design stages. Obviously, if you are limited to the port size because of other issues, and this limitation you have to make will cause substantial port compression at higher SPL, then you may as well use a sealed system with EQ.
2) I take it the port is rear firing? Was this an attempt to aid omni directional response of bass frequencies, or just a practicality issue with the box design?
Practical box design.
Finally, for your future designs, are you planning on using the same tweeter, or putting a pair of tweeters front/back ala Linkwitz's Orion++?
I will not use the same tweeter in any future design. In the next design, it is intended to use extremely wide dispersion(no more than 3db deviation at +/- 75 degrees at 15khz) tweeters, one on the back, one on the front, and with extremely large front edge radius(6") to have minimum diffraction effects on and off axis, preserving a linear response. I intended to this long before Linkwitz published the Orion ++ information. Also, he is not using tweeters with sufficiently wide dispersion for my purposes. My net result will be almost the same room effect acoustically, as an omnipolar tweeter.
-Chris
