• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Improving S/N ratio of Photodiodes in CD Players


Senior Member
Nov 26, 2018
Dear Friends,
in many first gen. CD Players - but also a few later models - the RF-Amp for the photodiodes receiving the laser signal
was mounted directly on the laser head. This resulted in a very clean Eyepattern as the weak current of the photodiodes
did only travel a short path before being amplified. But in 80% of all players, the RF-Amp sits on the main PCB.
The current travels at least 3 inches, the Eyepattern is always noisier.
No problem, as long as a disc is within specs.
But it becomes a problem if the disc is a (Re)-Writable or heavily scratched,
making the Eyepattern even more noisier...
And especially in an old player, this noisy signal is the base for the analog(!) servo-circuitry.

I have a third gen. player with Sony´s CX2019 RF-Amp, a standard IC in many players
from the early to late eighties. The RF-Amp is on the main PCB.
I have maintained it well but the Eyepattern is noisy and I can crosscheck between several new laser heads.

I have the mechanical skills putting the RF-Amp on the laser assembly but it is too complicated,
as the RF-Amp-section includes trimmers. So Instead I would like to amplify the photodiode
signals before they reach the RF-Amp.

My question:
The very first thing an RF-Amp does is an I/V conversion - see attachment, a CX2019 RF-Amp from Sony.
The Inputs 5 and 6 receive the current of the four main diodes A-D (they are summed in pairs) and
Inputs 8 and 11 for the E-F Photo-Diodes tracking signals, typical for 3-beam lasers.

I assume theoretically I could amplify the diode signals that leave the laser using Opamps,
(let´s say by a factor of ten) then send them over the cables (with a high S/N ratio) to the main board,
then dampen the signals by the same ratio before they enter the RF-Amp.
But can this be done with current, not voltage?
And will I get a better S/N?

If so, there could be another side effect: With some light amplification, the laserpower could be reduced - the laserdiode lasts longer.

Ideas and help appreciated!
I attached the circuit of the CX20109 from two sources, Nakamichi (where the values of the internal feedback resistors are given)
and Sony, with a typical implementation.
All the best, Herbert
CX20109-RF-Amp-Internes Schaltbild-Nakamichi.png
CX20109-RF-Amp-Internes Schaltbild-Sony.png
Last edited:


Senior Member
Nov 26, 2018
Hm - no one? In fact I do want to do the same as in the attached schematic,
taken from a first gen. Nakamichi OMS-5 / 7:
Q301 amplifies the diode current before ot is sent to U 101 and U 102 on the main pcb.
But in my case, instead of U 101 / 102 it will be a CX20109, doing an I/V conversion first.
Top Bottom