I/V mod-nelp needed

[dreamth]

I started a few weeks ago to mod an old and cheap Phillips cd player based on TDA1543 dac and lm833 as i/v stage .I finished the analog mod after the i/v and was completely happy with the sound but one idea came into my mind just because i have read a lot on it but never saw an embodiment of it. So i started with simulating first a simple circuit just to see how it reacts and it simulate well.I don't know how to make a signal look like a sampled current output in a simple way so i just assumed linear operation for the start.
Then i built it(using 4 2sc1048 trasistors) and modified capacitor values, decoupled the biasing resistor and although i can clearly hear the music , there's a mid frequency like a few kilohertz carrier (4..5khz)over the sound.It's not very powerfull , but it does not vary with the capacitive feedback or decoupling and most important, it's there even when NOT PLAYING like it's an oscillation of some sort.



Can you help me with some ideas how to get rid of that sound?
Again...i have no access to measuring instruments for that thing.
A simple idea that i didn't try yet would be to place like 100pf feedback from the op-amp output over the transistor to the emitter input and also decouple it with another 100pf to the ground at that emitter input, but it's just shooting without aiming.

 

[Ken Newton]

It's not clear to me what you were hoping to gain from your hybrid I/V circuit. It sort of seems that you're hoping to combine an grounded-base I/V with an feedback virtual-ground tranimpedance op-amp I/V to some advantage which is not immediately apparent. I'm not surprised that your circuit is experiencing oscillation.

Without access to test equipment, I don't think you are very likely to solve the problem. I suggest you simply restore the original op-amp I/V circuit, and enjoy the improved sound you said your getting from your other modifications.

 

[dreamth]

I've seen claims that this techniques is used in v-out video dac converters inside to separate the output capacitance of the dac to the transimpedance stage so that the integrating op-amp can use less feedback capacitance and subsequently have better transient response.
In the end I'm doing this just because i can and because I'm curious of the outcome.

 

[dreamth]

First of all, i wanted to parallel both sections of a double op-amp liike lm833 and cascoding looked like the only possibility to deal the differences in offset between op-amps.

 

[dreamth]

Problem solved with no measuring tools as i do it for 17 years here and there, from time to time... The rest of the analog toy is in the next photos.


I didn't pass 6 R&D technician jobs in the last 5 months or so...and i know why: that is because the engineers are running the interviews...They always ask me a tricky question and just because I'm not Einstein, and they can't work without an Einstein by their side, while paying him with peanuts...