Thanks Mal,
I have to say I tested the Auricaps thoroughly about a year ago and found the following:
1. With 20V or more of DC across them, they perform superbly, giving excellent bass and lovely, luscious midrange and a clear top end. For this reason, they are my choice for output coupler on the GK-1, which has 48V of DC across it.
2. With less than 20V, and particularly sub one volt, their performance became muddy.
At low DC bias, my favorite is the RTX from RelCap, which delivers excellent performance across the musical spectrum with only millivolts across it. However, it was only slightly better than the Evox-Rifa which I normally fit to all kits, and which therefore gets the nod for the stocker.
The primary technical specs for good coupling caps appear to be dielectric absorption, which should be lower than 0.05 and is best with polystyrene, closely followed by polypropylene; and slew rate, which relates to transient attack, and should be around 100V/uS minimum. In this connection film caps (particularly the film and foil caps) are superior to electros, , but every year now electros, particularly the dry types like BGs, are improving. and Cyril Brenneman of EW believes that the non-polar caps are VERY good, almost as good as quality film caps. It is my belief that most capacitor differences reflect variations in DA and SR, but I suspect also that Q enter into it, being a measure of characteristic impedance. This last quality clearly makes caps topology-sensitive, that is, diff'rent strokes for diff'rent folks....
Having said this, YMMV, and Malcolm takes a slightly different, but entirely credible view. This happens in audio a lot, and accounts for the 'house' sound of companies like Cary. The only definitive course is to build it progressively, noting the 'sound' as you go, but it's awkward and potentially harmful to the pcbs. I'm afraid it's your call, Fred, and watch those pants closely!
Cheers,
Hugh