Some of the mid drivers have failed open circuit so I have chopped one apart to find out what the issue is. Despite quite a lot of gigs the coils show no sign of overheating, the internal construction of the 3fe25's is also quite interesting with a copper shorting cap on the pole and some kind of kapton like polymer former. A high level of technology for the price. The failure is where the tinsle lead is touched by the paper mache on the outside of the terminal block, this area is coroded and no longer conducting. To remedy this I will construct new midranges with glue over this area to protect it. I wish Faital would do a sealed back version 
.
I also took apart a compression driver to check the dome and voice coil. Seems some liquid got down the horn at some point but I doubt this has impacted performance.
Once I have completed the repairs to the tops I'm going to amend the crossover to reduce complexity and reduce the amount of EQ that is done passively. I'm also going to complete polar measurements in order to define mixed FIR/IIR filters that are optimised for listening window power response rather than just on axis. Im going to try and avoid boosting any acoustic notch as at the moment I feel I should be able to get more output from the compression drivers before limiting.
. I also took apart a compression driver to check the dome and voice coil. Seems some liquid got down the horn at some point but I doubt this has impacted performance.
Once I have completed the repairs to the tops I'm going to amend the crossover to reduce complexity and reduce the amount of EQ that is done passively. I'm also going to complete polar measurements in order to define mixed FIR/IIR filters that are optimised for listening window power response rather than just on axis. Im going to try and avoid boosting any acoustic notch as at the moment I feel I should be able to get more output from the compression drivers before limiting.
