Passive IV PCM1794A gain & bal to se circuit help needed

[benproiii]

I build my DAC based on two PCM1794A in mono configuration with current outputs in parallel, that feeds into a 25 ohm resistor to do IV conversion then it goes into a long tail pair to convert the differential signal to single ended and add some gain to the signal. right now Im using a single MAT12 transistor 4 CCS's. I want to improve the THD of the circuit but i don't want to add negative feedback. Im thinking about adding a CFP transistor as I was told that it reduces distortion, would any benefit come from increasing bias current right now its at 4-5mA per transistor and the LTP top resistor is 1.74K. when I increase the bias current does that resistor also have to be changed? should half the current go through the 1K74 resistor and the other half go through the MAT12?? I attached a picture of the circuit that I want to use, will this circuit produce minimal distortion when used in single-ended mode? what improvements could be made?

Q5 and Q2 are MAT12 in real life and right now Rin7 and Rin8 are 249ohm and R9 and R10 are not connected in the real circuit
also V3 and V4 are green LED's in the real circuit

is adding a CFP worth it for this circuit?

 

[Ken Newton]

1. What does the THD presently measure? Specify the output signal amplitude of that measurement.

2. You say that you don't want to utilize negative feedback, however, CFP configurations inherently utilize negative feedback. In addition, Q1-Q5, and Q2-Q6 in your schematic already form two Compound-Feedback-Pair circuits. Please clarify?

 

[benproiii]

I wanted the THD to be low enough that i can get 24-bits of resolution out of it, do you think it would be worth it too use a current to voltage converter with a BJT instead of the simple 25 ohm resistor, has anyone heard the difference between the two? is my differential converter good enough for the PCM1794? or should some improvement be made to it?, i just want the circuit to be the best it could be