I preferred the 2way BR with electrical XO. BTW thanks @drewmc for the improvement suggestions on the BP6 version.
Both designs work and are very listenable, and surprising for $30, but they both have their flaws and some explanation is required. In the 100Hz-500KHz range the TC6028 sounds better than one PC83 and in the >3Khz range the TD20F sounds better than the PC83. In the <80Hz range the TC6028 BP6 sounds better than the TC6028 BR.
I tried to use the pure acoustic roll offs that are inherent in both designs to reduce component costs. The BP6+Fullrange has 3 pure acoustic slopes, while the 2way has 1 pure acoustic slope + 2 electrical slopes. The pure acoustic slopes, like the sealed Fullrange LF rolloff at 180Hz, are due to decreased efficiency. The same power is going to the driver at LF and it still moves alot, generates decreasing output, can exceed Xmax and the SQ suffers. You would not notice that SQ issue from the swept sine FR test. It also occurs in the BP6 below its VLF rolloff (needs a HP to prevent low power Xmax) and above its LF rolloff (MF residue, needs a LP). Music with a high VLF content would unload the TC6028 (both BP6 and BR), xmax may be exceeded, and SQ would suffer across its entire range. I was not expecting the "in band" SQ to suffer as the drivers see their "out of band" frequencies, and I was incorrectly expecting just an acoustic efficiency drop.
The 2way electrical XO actually decreases the power to the drivers in the range where they misbehave (out of band). So less misbehaviour and the SQ is better. In my next project (modular 3way) I plan on making the electrical XO dominant to avoid using the pure acoustic slopes. So every driver will have a bandpass and will only be fully powered for its intended "in band" well behaved range.
Last edited by DonVK; 25th September 2018 at 02:08 PM.