It has been suggest to me several times that the best sound can be had by completely removing and jumpering the output coupling caps. And I believe that this is probably true. But in spite of having no problem with ripping out the complete GLC filter I just don't want to go without any DC blocking caps on the outputs. Probably silly of me, because when I measured for DC just before those caps I got 2.5mV. But still, I did include some caps here.
I've been experimenting with different configurations in my DCD-1290's and 2700's and at this point I use Nichicon ES Muse NPE, Panasonic ECW polypropylene and Wima MKS2 caps here. A single 220µF/50v ES with a 2.2µF ECW in parallel measured lower in ESR than the series/parallel setup that is OEM. The non-populated position is jumpered. With the other players I also added some 0.47µF, 0.22µF, 0.1µF and 0.01µF soldered to the legs of the 220µF & 2.2µF.
Having the 2 extra parallel pads is a luxury that I made good use of. There is a single 220µF ES in one of the 470µF positions (C345 & C346), a 2.2µF + 0.22µF ECW in one of the former 4.7µF positions (C347 & 348), and in the 2 extra positions I added a 0.47µF + 0.1µF ECW (C391 &392) and some 0.047µF, 0.022µF & 0.01µF in the final spots (C389 & 390).
Lastly, there is a 10 ohm resistor in the signal path after these caps, R335 & 336, followed by some 150 ohm (R051 & 052) just before the output plugs. The 10 ohm are not present in the otherwise nearly identical analog section of the DCD-2700, although the 150 ohm are. In the DCD-2700 those 150 ohm resistors are shown as fusible 1/4w and that is what I used. However in the DCD3000/S10 they are shown as 1w but look exactly the same as the rest of the resistors in terms of physical size.
In my case I removed and jumpered the 10 ohm, and replaced the 150 ohm with SFR-25 400mW metal film.
Lastly lastly, the RY-12W-K relays are available at Mouser and I replaced the relay with a new one.
Cheers,
James