• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Sony TA-4650 Restoration

and bias setting is probably counter clockwise to plus mv on the dmm and clockwise to zero. Not noted in any sony manuals of same I've read.. blew some stuff myself till I tested the ************ vr pots.
 
Do you think I've damaged or killed the other capacitors? Do I have to recap everything, or just the two capacitors that fried?
 
ahess said:
DR, the pots were measured off the PCB.

EW, you mean KSC not KSA, right?
Click to expand...
Yeah...fixed.

ahess said:
Wow! I just discovered something really weird. Looking over my pictures of the amp before I replaced the caps, I saw that caps c420 and C 421 were 50v/47uF. C421 also has a red marker slash across the top . On the mounting diagram they are listed as 100v/100 uF!

I compare the mounting diagram to the schematic, and saw that while the schematic is listed as former type the mounting diagram is not. The parts list at the end of the service manual also differentiates between the values of some caps for the former model. I suspect I installed the wrong values in some of these locations. I need to look over the board tonight.

Could this have been the reason I blew the speaker protection transistors and these caps? Could this have been responsible for inverting the voltages?
Click to expand...
All I see are 50V caps. The voltages for these caps comes from the 4700uf 50V supply caps and is padded by 1K resistors, so while it's a good idea to replace caps with what you find installed (if you're sure that is what came from the factory), I don't see how the 50V vs 100V would be a problem here.
 
It's not just V, but the uF which are way off.
Here is what was in the original (which is also what is listed in the schematic) vs what I put in (going by the erroneous mounting diagram):

Cap #s -- original (good) values -- what I put in:

C303/C353 -- 47uF/10v -- 100uF/10v

C306/C307/C356/C357- - 100uF/6.3v -- 10uF/16v

C409- - 47uF/10v -- 100uF/10v

C420/C421- - 47uF/50v -- 100uF/50v

So:
Could this have been responsible for frying the transistors and C417, C420, C421 caps?

Since those caps got fried, do you think I've damaged or killed the other capacitors? Do I have to recap everything, or just those caps that fried?

Thanks, everyone, for your help with this. I feel sorta stoopid that I did this, but at least I found something that might explain the problems I've been having.
 
ahess said:
So:
Could this have been responsible for frying the transistors and C417, C420, C421 caps?
Click to expand...
No.
Since those caps got fried, do you think I've damaged or killed the other capacitors?
Click to expand...
Doubtful. Electrolytic caps are a tough beast. And nothing I have read here leads to a plausible explanation as to why they smoked, unless they were incorrectly installed, or a terrible error was made in something else you did.
 
Last edited:
ahess said:
It's not just V, but the uF which are way off.
Here is what was in the original (which is also what is listed in the schematic) vs what I put in (going by the erroneous mounting diagram):

Cap #s -- original (good) values -- what I put in:

C303/C353 -- 47uF/10v -- 100uF/10v

C306/C307/C356/C357- - 100uF/6.3v -- 10uF/16v

C409- - 47uF/10v -- 100uF/10v

C420/C421- - 47uF/50v -- 100uF/50v

...
Click to expand...

The Service Manual for the 4650 shows Former Values and New Values for each cap you listed. The values you pulled are the former values and you replaced them with the new listed values. There is also a note about PCB's being interchangeable between former and new models so maybe you got a transitional model where a new PCB was being used with former cap values.

As a general practice, shouldn't items be replaced with the same values unless a service bulletin or such specifically identifies a part that needs an updated value? When I read the service manual I interpret it to mean if you have the newer model with newer values then replace the parts with the newer specified values but if you have a former model then replace with former values. Is this correct?

Sorry for being so verbose but I have both a 4650 and 5650 waiting for a refurb and I want to be clear on this point.
 
Last edited:
That's my understanding. I think I have a straight-up former model. My mistake was referring to the values in the "mounting diagram" which was printed with the New Values.
 
When I did my 4650, what caps came out is what went in value wise. Mine had the barrier strip/screw terminal speaker connections. It had the resistor change already done by the factory (or by a board swap by Sony service). C209 has the silkscreen backwards. SN 800765.
 
Just noticed something--D406-- on the PS looks like a transistor that's been neutered.The middle lead's been snipped! Looks original.
 
Eeet eeesss diode. (A zener, to be specific.)

Check the schematic. That's why it's designated D406.
 
I HAVE FOUND THE PROBLEM!!

After scrutinizing the schematic over and over, rechecking the values of all the parts I’ve put in, checking my solder joints with a magnifying glass, trying to figure out what would cause this strange reversal of current from the PS to the amp – I found the reason for the fried parts.



<waiting for dramatic effect>



https://photos.google.com/photo/AF1QipPpYiW7JBhYbxvDFG978BBzuTMmnNKwxVZ-yF85



On the far left, you see the cap with the bulging top (one of my victims) but below that is plug #15. Notice how the female part is offset—there is an open pin at the bottom. From looking at the diagram for the board, it seems as if that would account for things bulging and smoking, right? Is there anything I should be checking, now that I know what happened?



So, I’m relieved that I found something to account for the inverted voltage and fried parts, but feeling pretty stooped.
 
Just happens to be a diode in a TO-92 body. Unused leads are nothing new to the electronics industry.
 
Did correcting connector 15 fix your issue?
FYI, I recently bought a 4650. I didn't even power it up before I did a diode and cap operation.
It sounds really good on headphones, but the bias voltage jumping around over 50mV has me worried,
Solution: buy another 4650 and see what an unmolested one does. Haven't unpacked it yet, that's for the weekend...
 
Last edited:
I'll find out on Monday when I connect the amp.

BTW, the transistors are in and the PS is running. I did measure all the voltages for the PS and was really bummed when the voltages were WAAY too low.
Then I realized I hadn't turned it on....
I've got to slow down.

Voltages from the PS are 48 (pins 1 & 6), 97( pins 2 & 5), 20 to D board, 97 to A board. Ok?
 
Last edited:
I fired up my new to me 4650, it appears to be original and unmolested. It is a later serial number than my first one (recapped, re-dioded, repaired, with jumpy bias). Voltages measured at the PS are around spec. Bias was adjusted to ~50 mV, and wanders around 25mV over 5 hours, as measured by my Iwatsu VOAC 7602 bench meter using the very nice statistic function. Room temperature is pretty constant.
I do plan to replace the double diodes, and maybe after A/B comparison listening, the caps.
 
Last edited:
That blog makes for great reading, especially with Google Translation: "Seems to grandfather their relics that went to heaven is increasing rapidly"! The PS section mentions voltages that are in my ballpark, so I'm not going to sweat it too much. I plan to check the amp voltages, especially around the vfets, before installing the vfets. Then adjust...and listen.
 
ahess said:
That blog makes for great reading, especially with Google Translation: "Seems to grandfather their relics that went to heaven is increasing rapidly"! The PS section mentions voltages that are in my ballpark, so I'm not going to sweat it too much. I plan to check the amp voltages, especially around the vfets, before installing the vfets. Then adjust...and listen.
Click to expand...

I got stuck on that same sentence and had to read it three times to figure it out.
Seems the relics of grandfathers who went to heaven are increasing rapidly.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom