Tried myself few opamps lately in I/V and LPF stages, like: OPA1602, LME49720, MUSES8820, OPA2604, OPA1642, OPA2132PA and MUSES02; FETs in I/V and bipolars in LPF.
I bought OPA1642 because is a step up in every aspect from OPA2134 and TI recommends it as an upgrade in audio chain: https://e2e.ti.com/support/amplifiers/precision_amplifiers/w/design_notes/upgrading-op-amps-in-audio-equipment. If anyone else experimented with this opamp, please let me/us know your conclusions.
Haven't seen people trying the below combination, so...I'll just post my thoughts here:
- In LPF, MUSES02 and MUSES8820 are very similar; also OPA1602 is about the same too (perhaps not so detailed).
- The only FET I tried in LPF was OPA2604 and this is quite neutral, not bright at all and not very detailed...a bit dark I'd say and might be good to pair with bright headphones.
- Best soundstage is with OPA2604 in I/V, but oscillates a lot and can't be a keeper. Not very detailed, but soundstage was great.
- OPA2132PA + MUSES02 is detailed and a bit bright with good soundstage.
- OPA1642 and MUSES8820 has a very similar sound with the OPA2132PA+MUSES02 from above, maybe a little bit more mid-range (saxophone)...but very difficult to conclude 100%. Sound is clean and definitely more detailed then NE5532, a little bit more bright I'd say, but a bit more spacious and relaxed.
As I couldn't keep MUSES02 for myself (borrowed from a friend) now I'm using 4 x OPA1642 in I/V, 2 x MUSES8820 in LPF, 2 x MUSES8820 in headamp input stage (yes, I replaced LME49720 with MUSES8820) and original LM4562 in buffer stage RCA/XLR.
Also, I realized that LM4562 is included into schematic only because of volume pot from the middle. :) So, anyone can bypass the 3 x LM4562, based on "best opamp is no opamp" and get the outputs directly from LPF opamps; output voltage will remain unchanged, just don't forget to add the output capacitors for any possible DC voltage (by default there is no DC on LPF outputs, but in case of PSU or opamp failure it will be a lot of DC on output pins!).
Also, just measured the other day with my Picoscope the PSU ripple + noise (RMS) and it's only about few hundreds of uVolts on each rail (250-500uV), so this looks pretty OK for me. BTW, if someone replaced the main PSU capacitors with bigger ones, please let us know the values for ripple + noise (RMS). Thanks!
Anyway, best upgrade for my E1 was the GOSS band around the trafo, which completely removed the barely audible hum. BTW, the hum I was hearing initially to my E1 I was able to hear it to another E1 of a friend of mine too, having the original PCB (non-MUSES edition, without gain jumpers), so K550 are really sensitive for this job. :)
Headphones used for testing: DT880/600, K701 and K550.
Regards and happy modding,
Edited by raoultrifan - 7/13/15 at 12:20pm