Hello Everyone,
for my Preamp Project (https://www.diyaudio.com/forums/analog-line-level/359465-time-preamp.html) I've decided to start with the design of the phono stage. I've designed this one based loosely on an old circuit I found at my dad's electronics workshop and some guessed knowledge 🙂
Before I buy components and try to build it, I'd like to hear your opinion. At the moment, it's just a MM stage, I'm planning on adding an MC preamp too.
I'll attach a picture and the Spice File of the schematic, I'm happy to use your comments for improvement 🙂
Regards
Lukas
EDIT: The OP U4 is just outputting a reverse RIAA curve for testing purposes.
for my Preamp Project (https://www.diyaudio.com/forums/analog-line-level/359465-time-preamp.html) I've decided to start with the design of the phono stage. I've designed this one based loosely on an old circuit I found at my dad's electronics workshop and some guessed knowledge 🙂
Before I buy components and try to build it, I'd like to hear your opinion. At the moment, it's just a MM stage, I'm planning on adding an MC preamp too.
I'll attach a picture and the Spice File of the schematic, I'm happy to use your comments for improvement 🙂
Regards
Lukas
EDIT: The OP U4 is just outputting a reverse RIAA curve for testing purposes.
Attachments
Member
Joined 2009
Paid Member
It will probably work or be fixable.
Complex but some apparent non-optimums. You run big BJTs at ~~1mA for <1k noise level and then have R5 at 3k. And this is for 47k needles? IIRC the best noise performance for BJTs at that impedance comes with 20uA-500uA device current.
(Which means the '5532 is already pretty good, without stuff in front.)
R8R9 add a flat gain of 6 at the end. Yes, U1 gives midband gain about 17 which is a bit low, but it also stops rolling-off near 32kHz (instead of infinity) which makes a few dB error at the top of the audio band.
I don't understand why a DC-servo at the output, after passing through three R-C networks before that.
Complex but some apparent non-optimums. You run big BJTs at ~~1mA for <1k noise level and then have R5 at 3k. And this is for 47k needles? IIRC the best noise performance for BJTs at that impedance comes with 20uA-500uA device current.
(Which means the '5532 is already pretty good, without stuff in front.)
R8R9 add a flat gain of 6 at the end. Yes, U1 gives midband gain about 17 which is a bit low, but it also stops rolling-off near 32kHz (instead of infinity) which makes a few dB error at the top of the audio band.
I don't understand why a DC-servo at the output, after passing through three R-C networks before that.