• 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 VFET Amp Restoration

@PE9ZZ interesting, I could not find it in the bulletins either, but I have seen it mentioned on DIYaudio a few times. For now I'm going to leave the originals in because I'm eager to get it running and off my desk.

Fired up the power supply today with the proper diode mod, everything seems to work fine. B4+ adjusts nicely to 20.00v and B+ is around 91v, which (afaik) is the expected value without the VFETs installed. Gonna connect the amp board today, check all the voltages on the transistor sockets, and then power it up with outputs if all looks good. Since this TA-4650 was working when I got it, don't expect any issues unless I screwed more things up.
 
In case anyone doing a restoration/recap stumbles on this thread in the future, the PEH200MA4470MU2 caps from Mouser are a perfect filter cap replacement and can be screwed into the power supply board (the lead spacing isn't identical but it's close enough to work). New metric screws that fit with the caps will of course be necessary, and there needs to be some padding added between the cap and the bracket because the new ones are smaller in diameter.
 
AlexHempel said:
In case anyone doing a restoration/recap stumbles on this thread in the future, the PEH200MA4470MU2 caps from Mouser are a perfect filter cap replacement and can be screwed into the power supply board (the lead spacing isn't identical but it's close enough to work). New metric screws that fit with the caps will of course be necessary, and there needs to be some padding added between the cap and the bracket because the new ones are smaller in diameter.
Click to expand...

Thanks for posting this. I found a 4650 and am in the process of replacing the caps, diodes, etc. I am going to order these replacements - I was just too lazy to look for this specific replacement.

The resistor replacement is in a number of places across various sites and the only "service bulletin" I could find that mentions it is in a foreign language. I tried to attach it but it is too large but its the last service bulletin on hifi engine for the 4650. Also found a picture of it on someone else's restoration they documented.

img_8754.jpg
 
That must be another service bulletin. There are several, lemme go through them. If I find it I will translate the Dutch. It is prolly explained.

Edit: well, it isn't. The text reads: "Below a list of changes that have been introduced after first production to improve reliability. It is recommended to include these changes in the service manuals." and then the resistor values and bias setting ("Previous value", "New value"). Also how to test a V-FET (G-S diode junction test) and mods for other amps.
 

Attachments

Last edited:
Huh, interesting. Well, I went ahead and adjusted each side for >20 gate-source voltage differential and powered on. It works! Bias adjusted to about 75mv, it drifts a few mv up and down. More concerningly, DC offset is pretty high. Left channel is about 20mv, right gets up to 40mv. Since there's no offset adjustment pot, how would I go about fixing this? Replacing small transistors on the amp board? I have verified that the problem is in the amp section and not the pre.

I've been listening for a few hours and it sounds quite good. Runs nice and cool, no audible distortion, hums, etc. I will post some pictures later today.

High-pitch vocals seem to have great presence and clarity, cutting through the mix better than with my other amplifiers. I wonder if this could be a result of high-frequency distortion caused by the dc offset? The low end doesn't seem as tight or strong as my NC400s, though. These are very rough impressions, as I do not really have a proper A/B set up right now and half my amps are in an NYC storage locker due to corona. Curious if anyone else has thoughts on the 4650 sound.
 
I did an A/B shootout between a MOSFET and VFET 4650 and couldn't hear any difference. But that was just the output power transistors. The owner did the same and also could not hear any difference. That said, doesn't say much about how the amp sounds itself. But then I think the "sounding" of an amp is overrated. If you can hear it it is broken. Like when I did a shootout with my KT88 Dynaclone vs. my regular Philips DFA880 ("glasbak vs. zandbak", glass vs. sand in Dutch, bak being container). The Philips won but that was only because I am not happy with the design of the Dynaclone. Once I'm finished with it things may be different.
 
Well as it turns out, the resistors had already been updated in mine. Just waiting on my filter caps now. Thanks again.
 
One other thing: the new caps are a bit taller, so you'll need to replace the grounding post at the front of the power supply with a jumper or something.
 
Nice job..

I have a TAN-5550. Very nice mid and high. Very sharp image and clean. But not as warm as some of my other amps.

Attached is the Sony V-FET amp service bulletin with English description...
 

Attachments

  • bulletin-Sony-TAN5550.jpg
    bulletin-Sony-TAN5550.jpg
    108.2 KB · Views: 57
Ok, a not great update: I turned the TA-4650 on yesterday and got no sound output. I verified that my DAC was working correctly, but still no dice, so I put it on the DBT. Unfortunately, it's clearly shorted somewhere -- 100w bulb fully lit. Interestingly enough, no magic smoke, no breaker tripping, and all fuses still intact. Visual inspection doesn't reveal any bulged caps or signs of overheating, but I was careful not to leave it running for long once I realized something was up. Also, the speaker protection relay still pulls in. I removed the VFETs, expecting them to be toast, but they all test good using the standard diode and resistance tests. Perhaps the VFETs are fried in a way that I'd need a curve tracer to verify? Or there's a short somewhere else in the amp that's incidentally preventing sound output. Anyways, I don't expect this to be an easy fix, so the TA-4650 is probably gonna be shelved for at least a few weeks, and my trusty bulletproof Hypex amps are going back into the system. If worst comes to worst, I'll get a chance to try the MOSFET mod!
 
Do not use DBT on VFET. It will prevent the VFET from getting the proper bias voltage to stop large bias current.

Instead, mark the VFET for their position, remove them, bring up the amp using DBT, check that there is proper VFET bias voltage...

There's a thread somewhere here on how to bring uyp a VFET amp... try searching "VFET"....

Good luck...
 
Ah, you are right! I knew that when I did the initial restoration, but I totally forgot in the months since. Luckily it was only a 100w bulb, or I might have blown my outputs! I will check all the voltages with my DMM this weekend and do everything properly.
 
So, after my DBT mishap I tested all the transistors (good) and never got around to reassembling the amp. The original issue wasn't even the amp, it was my DAC, which glitched in a way I had never experienced before and refused to output sound. So if I can't get this fixed I'm going to feel especially stupid.

Today I finally got back around to reassembling the TA-4650. The outputs test good with the diode tester, and all the power supply voltages are good, as are the voltages on the transistor sockets. So, I figured I would be in the clear, but nope. It powers up fine, relay pulls in, DC offset is about -20mv on both channels (same as before) and bias seems normal (climbs quite a bit, but I remember this amp being slow to warm up). But, after being left on for a minute or two, the protection relay will sometimes release, and I cannot figure out why. I attached my scope to watch both speaker channels while also monitoring my DMM to check bias, everything seems normal. It seems like it might be an intermittent fault, because it's only happened three times and does not seem to occur within a consistent timeframe. If anyone has ideas, I'd love to hear em. Perhaps I messed up something in the protection circuitry? On the bright side, it's a good chance to improve my diagnostic skills.
 
I have one of these in my collection but it is not in working condition so I was looking through your posts while I decide if I want to try restoring it.
 
I need to get around to fixing the TA-4650, which I left back in California (currently in NYC). I remember liking the sound and it would make a good solid state reference amp for when I don't wanna run my 300 watt Class A Pioneer M-22. Since I made this thread, I've gotten into tubes and learned a ton about electronics. At this point, I'm usually capable of component-level troubleshooting, so I'm fairly confident I could get it working. Was in a bit over my head previously.
 
You must log in or register to reply here.
Back
Top Bottom