Dette er sakset fra Vertex AQ's sider om modifikasjonene som er gjort i MK2-versjonen av dac'en deres:
Med tanke på at de brukte Duelund før så er det jo for det første oppsiktsvekkende at de har skiftet til elektrolytter, og for det andre at de har brukt en så stor kapasitans. MEN så til spørsmålet: Hva er egentlig "invers parallellkobling" av kondensatorer?Summer 2015 – The MkII Modifications
The dac-1 and dac-1.5 designs were upgraded to MkII status in August 2015. The main upgrade was a revision to the passive output stage, with a change of configuration, size and type of output capacitors used. For quite some time we’d wondered about systematic faults in capacitors, and also if some of the old-school rules about capacitors were perhaps incorrect. In the case of DC blocking, generally the conventional calculations say a DC blocking capacitor value of 2-3uf is all that’s required to have no significant phase shift above about 20Hz, and any residual phase shift below this frequency will not be audible. Well even very small phase shifts above about 10Hz seem now, to our ears, quite noticeable in a revealing system. And on consulting with our maths guru, Dr Gareth Humphreys-Jones, he confirmed that whenever any bass content was present, even small amounts below 20Hz, combined with a circuit that can add any phase shift at these frequencies, will shift all higher frequencies up and down with it. And this would produce intermodulation products right through the audio band. Bingo!
Next in the story is that this year Audio Note released their new Kaisei capacitors, built along the lines of the old Black Gates. So we tried them and they’re simply brilliant, especially when used in inverse parallel. So now, the output stage in both the dac-1.0 and dac-1.5 has, per channel, two 100uf Kaisei capacitors in inverse parallel.