Telefunken RC760

[HRmeteohub]

I have already started a topic under restoration section, but my RC760 has got technical problems, so here we are. I am not counting bad belts and dirty switches/pots under this. There are some pictures on that thread. Let me start: There was a certain amount of glue holding C760 and surrounding elements together, probably to increase stability and prevent tearing of the capacitor off the board during transit and use. Long story short - the glue solidified in dark brown brittle mass. After that, a lot of corrosion was under it, and R1000 got very hot due to poor ventilation(see below).
This was the situation after I have removed the "glue" using small pliers.



There was corrosion above and under board. Clear sign that something went wrong - capacitors have leaked but there is still work to be done.



When turned on, the box produced distinct humming sound (50Hz in EU/Russia/Japan, 60Hz in USA/Japan). The humming has gone away after I have replaced 220uF and 2200uF filter capacitors. Also, after several minutes, R1002(R1000?) overheated and unit went silent until it was left to cool off. Unfortunately, I have a service manual that does not list this 270 ohm 1/4W resistor... as well as T904, T905 and other elements around it. Go figure.



I have replaced both the capacitors and resistor. There is the board pictured from above, after the electrolytic caps are replaced, and 270 ohm is waiting underside to be cleansed.


Now I have to find another schematic diagram or draw this part myself. Most of them on the 'net are exactly this one. It seems to be a copy from Portugal.



There are two transistors, T904 (2SA966) and T905 (2SC1815), coupled together via this (270ohm, R1000) resistor. Base of T905 is connected to ground via 5k6 resistor, and connected to S004 via 10k ohm resistor. Therefore the T905 is driving T904 and it is basically turning power to IC701 (power amp) when the S004 is in appropriate position. To put it in perspective, the 270 ohm resistor gets rather hot in a mater of minute or so, and that heat is probably the reason why the glue got so brown and brittle, why the capacitors leaked, and why the box will go silent in minutes if I do not find what is wrong (the resistor will burn through, and T904 will loose drive and turn off the power to IC701).

 

[baddboybill]

I cant see scheme good but is there possibly a diode near by as that might cause issue if diode bad

 

[HRmeteohub]

Hi BaddBoyBill! Well, the diode on a schema is LED (power indicator) and it works.

So I pulled out trusted instrument and it says that the current through the base of T904 (2SA966) is about 45 mA and the voltage drop is about 11V. That gives about 0.5W power dissipation on that resistor - R1000 (270 ohm) that is connected to base of T904. Way too much for 1/4W resistor. It turns out my previous rant about glue being the culprit turns on me.

No wonder it was charred, capacitors leaked, and the board is baked. You see, the flux on the board is bad, as in not usable - it won't stick. It was baked and oxidized. Even the vias on the PCB and pins of the elements are oxidized and evene the new flux will not stick to them. I had to remove the flux with wick, then scrape the pins and PCB to get a good contact with new flux. If the glue that was not there to dissipate that heat, it would burn the board and turn it to coal!

Since the T904 and T905 are basically inverting switch, I look at the 2SA966 (3W, 60V, 1.5A PNP transistor) as probable culprit. It is leaking - probably. I'll try to replace it with BD140, because I do not have 2SA966 at hand, and see what happens. I feel that it is hard to carry over the outlook of this circuit, maybe if I draw it?

Another thing is why these elements are not present in the schema that we can find on the net. Probably the S004-1 switch was to weak for 400-500 mA the power amp is drawing, and it was cheaper to put this circuit rather than find/order new switch, one capable of switching more current. I have found some <a href="http://2.bp.blogspot.com/-Zgz2HvebaDQ/VTx0_mvyQ1I/AAAAAAAAYUQ/0DHgcOVS0JE/s1600/IMGP4806.JPG">image</a>s of TFK RC760 on a <a href="http://magnitolia.blogspot.hr/2015/04/grundig-rr-446-telefunken-rc-760.html">russian site</a> and the board from 1987 looks the same. So I guess they changed it rather early in production (since my RC760 is from summer 1986).

On the other hand, I do like a challange...

 

[HRmeteohub]

Well, so far no good news. I have tested T904 (2SA966) and T905 (2SC1815) and they are both OK. I have found the schema I was looking for, and the (modified) power section is in grey background:


So, it looks like the engineer who designed the circuit wanted really to saturate T904, and perhaps in haste forgot to calculate power dissipation on that resistor. Never mind, the T904 is saturated with any resistor smaller than 750 ohm on it's base, so I have placed on 560 ohm, tested to see if it is OK and closed the subject.
Back to finding the new belts and testing the tape unit.

 

[HRmeteohub]

There it is. R1000 is now 470 ohm, as I have found one 1/2 W resistor from 1980-s. The T904 is still opened "hard", but there is no 1/2 W dissipation on R1000. I did not had capacitors from 80's so I placed low-ESR, modern versions, but as they are in power filtration segment, it will only gain better results. I also used 3300uF/16V instead 2200uF/16V as they are of similar size, and larger capacitor will give smoother power rail. I did not measure the 50Hz hum on the audio output, as the result was pretty obvious "by ear". I also increased C905 from 220uF to 470uF, since it stands after T904, right on the power amp chip.

I have placed the glue over C706, C905 and T904, as it was, but kept the R1000 free to cool. I have raised R1000 it a bit so it does not cook surrounding elements. The board under R1000 is half-baked, and it would not last long if resistor was heating it any more. The temp of the resistor is now about 15 degrees above ambient, so I guess that would not be a problem after all, but still.

The belts are on the way, now I have more cleaning to do.

Just to remind you, how it looked after glue was removed:

 

[HRmeteohub]

So, my technical aspect of customization of TFK RC760 looks like this:

I have decided to use "Wide stereo" switch as a selector of Bluetooth module audio signal. That leaves me with free Line-in RCA connectors for connecting other audio inputs. Perhaps this is not such a good idea, but the mod can be easily reversed, so there!


Now, the audio amp I have choosen uses PAM8610 module:

Unfortunately the signal levels between original radio/tape preamp, and BT audio module, are rather different from the expected signal level of amp module. BT module pushes the amp right to the top. Hmmm.

So, I'll try to use some resistors (as in image) in order to attenuate the signal from BT module, but I need to adjust amplification level of amp module, because radio can not cause the distortion module. Frankly, Pioneer 20W 4 ohm speakers I have choosen are "barely trying" when I turn the volume all the way when using radio.

Now, that could mean that I need to amplify the radio/tape signal, and decrease the audio from BT module. Also, I have to try how will it work with other external sources. All should be ok, as They should push about 0.7 - 1.0V but not more. BT module is probably pushing more since it is meant to drive headphones, and not amps.

I see in data sheet that PAM6810 gain can adjusted using DC voltage on pin 5. After scouring the web and looking at the datasheet I see that there is no way to change that on board because the pins 3, 4, and 5 are tied together - to Analog Vdd (5 V), and then coupled with small resistor capacitor to ground. Basically that means the voltage on the pin is at the maximum, and no further increase in gain should be expected. If I understand the datasheets correctly.



Now, the next step is to get the 12V PSU in the box, and make 5V for the BT module.