Sony TA-4650 repair question

[jvdk]

Hi all

I have a TA-4650 on my bench that according to the owner blew a fuse. When I received the amp I removed the v-fets and tested them, they all seem to be ok (see table below). With the v-fets removed I did some checks and noticed that the +20V and the +75V were both missing. After a lot of troubleshooting it turned out to be a defective 470K resistor in the power supply. Replaced the resistor as well as a few prone to failure BJTs and all electrolytics on the PS board except the large filter caps. Also did the essential modification for added stability and the power supply tests fine now. Recapped the power amp board and checked the voltages on the v-fet terminals and adjusted the bias trimmers to max gate-source voltage (about 21V). So far so good. I installed the V-Fets and powered on via my 75W dim-bulb tester. The lamp lit a bit too bright for my taste and I turned power off. Removed the v-fets and tested them again, same values as before, so still seem to be fine. I don't know what the idle power consumption is of this amp, and maybe I'm getting a bit paranoid. Can a dim-bulb tester be used on this type of amp? I read somewhere that a variac cannot be used and can cause destruction of the precious transistors... I don't want to blow this one ;-)

Note that the amp comes out of protection when powered on with the vfets removed and the preamp tests good.

Your opinion on this is highly appreciated!

	Gate->Source	Gate->Drain	Source->Gate	Drain->Gate	Drain->Source	Source->Drain
2SJ18	264K	260K	infinite	infinite	1.74	1.92
2SJ18	292K	297K	infinite	infinite	1.92	2.12
2SK60	437K	431K	infinite	infinite	1.67	1.45
2SK60	427K	423K	infinite	infinite	1.84	1.67

 

[dr*audio]

I read in the service manual not to use a variac. A dim bulb tester will also reduce the mains voltage. I would connect a meter across the Source resistors to measure current, and power it up normally, since the voltages in the unit seem ok without the outputs. Do not connect a load. Get ready to shut it down if the current is too high; more than 50mA to be conservative.

 

[jvdk]

Thank you, that was very helpful. I installed the Vfets of 1 channel and observed voltage over source resistors, after power on it peaked for a second to around 100mV then dropped to zero and the relay klicked on. Checked bias, was zero as expected, then adjusted to 45mV and let the amp on for about 15 minutes and readjusted. Repeated for the other channel and everything went well. Amp plays beautifully! I have put a lot of work in this one but it was worth the effort (albeit a nervy one).

Thanks!