Tuner meter and stereo signal light on Sony STR-6045

Hi everybody,

I have a strange problem that does not make any sense to me and I thought I could use some help from the members of this wonderful forum.

I did quite some work and learning on this little receiver (Sony STR-6045) and it is sounds amazing now. But strangely after a day or two after putting back the cover, the radio signal meter just stops working and after a week the stereo light will just light up in between stations. One evening, things were back to normal and the next day, it was behaving like this again.

Every time I remove the cover and take some measurements (good stations will give about 130 mV on the stereo light), it will start working again after 2 minutes! And it will keep working for about another 2 days with its cover on.

What I did so far:

  • Changed the stereo incandescent light to a LED (see this post)
  • Verified that the needle of the meter was not stuck
  • Changed the electrolytic capacitor on the tuner meter circuit thinking it could have a problem charging.
  • Moved the receiver to different locations in the house in case all of this could be caused by weak signal.

I usually start by posting the circuit but since it is so weird, I just want to know if any one has ever seen such erratic behaviors.
Joined 2010
Paid Member
Perhaps you have disturbed or broken a power supply or signal connection to the indicator cicuit, such that you now have intermittent connections in the meter circuit.
You could measure the voltages in operation (usually with respect to ground potential) after the zener and at the regulated power output connector and any subsquent connections, to verify that the zener and indicator LEDs are correct type and still functional in the modes where the LED supply should be low/high. Since LEDs draw less current than incandescent globes, I'd expect circuit voltages to be a fair bit higher and this will likely interfere with the indications you expect.
Last edited:
Thank you Ian,

This conversion to LED is suspicious to me too as it is the only thing that changed in this area. And as I said, once I open the case to measure voltages, everything comes back to normal. So measuring does not help.

Not sure either the signal meter and the stereo light issues are related. They could be 2 different problems.

Could the heat inside cause this malfunction? It would not explain why when I power the unit the next day, the problem is still present though.
That sounds like it but how come it would work for about 2 days with the case closed? And I can't see anything loose or stressed.

And things don't return to normal immediately. It takes about 5 minutes after that I try to measure the voltages at the terminals of the meter.

Very weird!
Osvaldo, let's explore your suggestions a bit. I used a resistor of 1000 ohms and then a zener diode to shut off completely the LED as there was always a bit of current bleeding. As below:

Should I have put the zener first then the resistor? Does it matter?

For the full circuit, you can see it here. Is the resistor you are talking about R319?
Joined 2007
Paid Member
Just quickly skimming through this... and it looks to me like the bulb could be a vital part of the circuit.

I think an LED would light between stations because there is a conductive path via R318 (I think that's the number... its blurry)

From the original configuration of a bulb try replacing R319 (the resistor across the bulb) with a 150 ohm. That gives approx the same current as the bulb which seems to be about 33ma based on the voltages shown.

When Q303 is on and which is the state to light the bulb you will have about 4.5 volts across the new 150 ohm. An LED across that in series with a 470 ohm or 1k or whatever suits to get the brightness you want should work.

When Q303 is off there is only around 300mv across the 150 ohm which will not light the LED.
Thank you all for your suggestions and analysis. I have left the receiver with its case open for a few days now on my bench and the stereo light misbehaves again but not the signal meter. Which leads me to think that the stereo light is a circuit issue and the meter a thermal issue.

Mooly, I will not change R319 but I will make a different LED conversion for the stereo light in such a way it sees the same load as with an incandescent bulb (4.5V, 40mA) and see what happens. This is what a user on the AK forum proposed (see this link).
Joined 2007
Paid Member
I'm not a member there so can't see the attachment :) but it looks like others must have had this issue.

I looked at the voltages on the circuit which show 14.8v as a rail voltage applied to R316 and 4.7 volts shown on the other side of the resistor. Current has to (14.8-4.7)/300 which was 33ma. It may not be all that critical and I wonder if R319 across the bulb is to maintain a current in the event the bulb fails.

So a modified bulb might look a bit like this. LED current approx 1ma here:

Screenshot 2024-05-15 172525.png


  • Sony Tuner.asc
    1.4 KB · Views: 12
Thank you Mooly for this. You were a good teacher, I had already designed my own with LTSPICE. For understanding, I have modified yours so our designs look alike.

First, I was confused as to why R318 was short circuited but I assume it is because of conducting transistors. Right?

Also, I think that we need to remove the V2 source as it is 0V with the stereo signal on (value in parenthesis in the schematic). The design I had copied from the other forum was like this. Can you please confirm if it makes sense? This gives about 1.5mA to the led but can be adjusted by varying R_LOAD.

Capture d’écran du 2024-05-15 21-43-16.png

Not sure if both designs (yours and mine) work but the important thing is that it shuts off the LED when there is a bit of current bleeding.


  • Sony Tuner JRB.asc
    1.3 KB · Views: 6
Jrbastien, in your circuit, the current through the LED is very heavily dependent on the exact value of the shunt resistor - changing that resistor by a few ohms greatly changes the LED current. Mooly's circuit is much less sensitive to exact resistor values. I'd recommend using Mooly's configuration.
Actually, I thought it could be of some advantage to be able to change the LED current. If 1mA is too dim for instance. Anyhow, I will try both designs and will let you know.

I'm waiting on an order of warm white leds for this as I don't like the color of the ones I found at my local store.