In this video, we do a deep inspection and tear down on a Sony TA-4650 V-FET amplifier and see if it is a viable candidate for a full restoration from the ground up... Will it work?
I have a ta-5650 that I still have to take a look at. I also have a 4650 that supposedly has been restored but I have never opened it up to check. The amp seems pretty easy to work on. I am always amazed at the lenghts you go to in your restorations. It will be interesting to see what is causing the distortion.
As a customer looking to do a full restore on this unit or similar, would need to be really keen. I dare say the cost of the full restore would out way the cost of the unit. I came close to buying the TA-8650 but they wanted the earth. I then looked at the TA-N7B , cost wise is even worse. I ended up buying the ta-3200f power and ta-2000f pre at a fraction of the cost, admit ably not V fets.
You could get away with a few hours work to just do the service bulletins and doing the basics but being VFET, the units are unique and their prices are going up rapidly as unrestored units die and disappear. What the units are worth really comes down to VFET desirability and collectability. Part cost and time is not object for these restores to me :)
Hello! Could you please share a link to the service bulletin you are referring to in the video? The one that I found doesn't have the info about the necessary resistors' parameters for replacement. I believe they are different for each grade of V-FETs? That's why I would really appreciate you sharing the link! I'm starting restoration of my 4650 and will follow your videos thoroughly. Keep it up and good luck!
There are 2 service bulletins, this is the one for the resistors and bias change. Just take it slow and leave out the vfets until you confirm all the voltages and do the service bulletins. The other bulletin covers testing the VFETs and setting the bias voltage to maximum before reinstalling. Link www.robuust.nl/~maarten/forum/Sony%20VFET%20bulletin%20[MAE].pdf
@@whatcouldgowrong7914 Thank you very much, I appreciate it! By any chance, do you know if any components should be replaced in the TAN-8550? Or is there a same kind of a note for it? According to my research, it all goes down to replacing the VD1221's and recapping + using the correct resistors to match the grade of V-FETs according to the regular manual. Seems kinda strange to me that 4650 and others have some changes and have to have their resistors replaced for the bigger value and the 8550 seemingly does not.
@@musicscratchpony I don’t have any experience with the 8550 (yet) but im sure some people have experience on audiokarma. General practice for me is to replace any adjustment pots and those double diodes at a bare minimum. But if you want to be really really sure the whole unit should be gone through testing transistors, capacitors etc. It used to just be recapping vintage gear but we’re approaching 50yrs old now and transistors are becoming a big problem also with age. I believe the resistor changes were in relation to the bias change and a design modification. It was just a value change but I generally use higher power resistors because technology has shrunk them drastically
That pot setting is bizarre and a little nuts If " It's clicking so that's fine " what is the point of it ? Potentiometers as I'm sure you know are for finding maximum and minimums or 'peaking' or for setting a precise voltage/current. Just wind the pot all the way to the end to give the maximum voltage is wrong because a piece of wire would be better, unless this works in a very odd way that's nonsense. Trimmers should never be maxed out. Can you explain please ?.......cheers
VFETs are depletion mode devices unlike most transistors which means you want maximum drive voltage on the bias / gate for first power up instead of none. Without high bias voltage, the VFETs will fully conduct and blow themselves up. From this point you then set the bias to 45mV :) the pots are 2.2k which replaced the factory ones. That is why 1-2ohms is normal from source to drain with no power 🙂
@@andymouse Yes in the video it is powered up with the highest gate voltage to stop the VFETs conducting, then I back off the trimpots and eventually the bias starts to climb which you can see in realtime :)
