Sony TA-5650 signal path, need help!!

Tim5

Active Member
Hi y'all,
I'm going to start a restoration of a Ta-5650 V-fet (my 3rd already), but this time I want to get the best out of it, and do some improvements of the components which are in the signal path, better resistors (metal film, metal oxide), Nichicon KZ Muse or Elna Silmic II and polypropylene's up to about 3,3uF.
I have made a start already up to the power amp section, there I got stuck.
Please also check to progress made up-till now, I've used green pen to trace.
Thanks....
 

Attachments

  • Ta-5650 001.jpg
    Ta-5650 001.jpg
    88.2 KB · Views: 36
  • Ta-5650 002.jpg
    Ta-5650 002.jpg
    81.8 KB · Views: 25
  • Ta-5650 003.jpg
    Ta-5650 003.jpg
    70.2 KB · Views: 22
  • Ta-5650 004.jpg
    Ta-5650 004.jpg
    97.6 KB · Views: 23
  • Ta-5650 005.jpg
    Ta-5650 005.jpg
    103.8 KB · Views: 31
  • Ta-5650 006.jpg
    Ta-5650 006.jpg
    109.2 KB · Views: 28
You seem to have it right (pictures kinda fuzzy). At the amp, you traced the negative half of the signal, you just need to use a similar path on the upper side and join together after the outputs. Then through the output inductor and protection relay, then the speaker switch and speaker terminals... FWIW I can understand why, but don't like series speaker switching - speakers A + B are a series hookup....


attachment.php
 

Attachments

  • Ta-5650edit.jpg
    Ta-5650edit.jpg
    196 KB · Views: 106
The problem here is the notion of 'signal path'. You would be wise to forget about such a concept altogether as it does not really exist. Many if not all parts of an amp pass some version of the signal, be it voltage or current or both and they all need attention based on the actual circuit function and what each part in it does. If you don't understgand the circuits. then I am afraid your efforts will at best be partially successful and based on guesswork. Eg. the main filter caps are DIRECTLY in the outpout signal current path... and also in the rectifier current path.
 
If you could post a clearer (bigger?) schematic, I could circle the caps that I would replace with film or low leakage electrolytics, and those I would replace with "audio grade".

Lee.
 
If you don't understgand the circuits. then I am afraid your efforts will at best be partially successful and based on guesswork.

I do not understand all there is to now about a circuit, but I have to start somewhere I guess, it is a learning process, especially since I have no background in electronics.
Off course not only the parts in the signal path will get (guesswork) attention.
 
Last edited:
I do not understand all there is to now about a circuit, but I have to start somewhere I guess, it is a learning process, especially since I have no background in electronics.
Off course not only the parts in the signal path will get (guesswork) attention.

As i said, forget the idea of 'signal path'. This is a good start. Rather, look at signal loops (i.e. current loops). Where does the signal current go and where it returns (usually thorugh ground).
In this manner, you can identify components in each loop of the amplifier, namely input and output loops of every stage. Depending what you want the current to do, i.e. drive the other stage or produce a voltage across a resistor or some other ocmponent to drive another stage, you can select components that are more or less important regarding potential change with better quality parts. Often there will be components (like capacitors) that are not in series with the signal (as input or output caps would be, these are obvious candidates for better parts) but rather in parallel with a component developing a voltage drop. Sometimes there is cap there intended to keep the AC component of the voltage drop (i.e. audio component!) constant - usually a capacitor - this is a candidate for component change, while the DC setting component (say a resistor, diode or zener) is not. As I said, it comes down to knowing the basics of circuits, and this is actually VERY difficult to deduce from a complex collection of basic circuits such as an amplifier, especially an unusual one like a VFET power amplifier.Even experienced designers may miss some aspects of VFETs, like the fact that every pin on these devices can be considered an input.
 
Thanks ilimzn for your input.
I do understand your point of view on this matter.
The first thing I'm trying to achieve is to keep my V-fet amps alive and while I'm at it do some improvements the best possible way I can, step by step.
Full recap, diodes, trim-pots, updates as per service bulletin (resistors) wire-wraps.....
 
If you could post a clearer (bigger?) schematic, I could circle the caps that I would replace with film or low leakage electrolytics, and those I would replace with "audio grade".

Lee.

I was thinking to replace C101-151, C107-157, C205-255, C213-263, C218-268, C230-280 and C301-351 with audio grade, the ones up to 3,3 uF with polypropylene, all others with Panasonic FC.
Uprate the main caps CC420,421 with Blue Chemi-Con 15000uF 100V (E36D101HPN153MC92M), because I had some good results with a TA-4650 uprating them to 10000uF, double factory value.
bayfix_com_1353x2xebbc4ea8a40.jpg

bayfix_com_1353x2x20538b8e9f0.jpg

bayfix_com_1353x2x20538b8e9f1.jpg

bayfix_com_1353x2x20538b8e9f2.jpg


Thanks Lee :thmbsp:
 
I've circled some caps on the schematics you sent me. If I were restoring this for a customer, anything circled in green I would use either Nichicon KLs, or if possible squeeze a film cap in there. The only problem is film caps get real big real quick once you get over certain values. WIMA caps are square and upright, but they have the disadvantage of short leads that can't be bent that far. Stacked film such as Panasonic ECQ or Cornell Dubilier DME have longer leads that cap be bent shaped to fit easier.

Anything else circled in red, I would use Nichicon KWs. I didn't circle anything on the power supply schematic. Is there any evidence of the caps running hot here? If so, I would using Nichicon PWs (105C vs 85C), if not the KWs again. The big 10000uf caps, either KWs again, or Panasonic snap-in like the TU-P series, depending what fits in easiest.

Hope this helps,

Lee.
 

Attachments

  • SonyTA-5650-01.jpg
    SonyTA-5650-01.jpg
    118.6 KB · Views: 25
  • SonyTA-5650-02.jpg
    SonyTA-5650-02.jpg
    127.8 KB · Views: 22
  • SonyTA-5650-03.jpg
    SonyTA-5650-03.jpg
    176.1 KB · Views: 22
I've circled some caps on the schematics you sent me. If I were restoring this for a customer, anything circled in green I would use either Nichicon KLs, or if possible squeeze a film cap in there. The only problem is film caps get real big real quick once you get over certain values. WIMA caps are square and upright, but they have the disadvantage of short leads that can't be bent that far. Stacked film such as Panasonic ECQ or Cornell Dubilier DME have longer leads that cap be bent shaped to fit easier.

Anything else circled in red, I would use Nichicon KWs. I didn't circle anything on the power supply schematic. Is there any evidence of the caps running hot here? If so, I would using Nichicon PWs (105C vs 85C), if not the KWs again. The big 10000uf caps, either KWs again, or Panasonic snap-in like the TU-P series, depending what fits in easiest.

Hope this helps,

Lee.

Thank you very much for taking your time to sort this out :thmbsp:
I will order the parts and get started.
 
Back
Top Bottom