specific problem? -50.6 at the -19v output. There's plenty of AWR-101 PSU write ups on here that cover it. sigh... one more time....
R18 (24k) & R17 (15k) form a voltage divider that takes pin 14's -51.5v and gives out -19.8 volts.
C23 is added on for several reasons. Stability, ramping up gently...
R19 (150 ohms ) is a bit superfluous, 0.9v drop (-51.5 - -50.6) says 6 mA load. It's a 1/4 watt resistor, so maybe it's supposed to pop on a fault.The transistor will pop sooner. Sometimes it's supposed to share the heat of reducing the voltage, instead of the transistor handling all the heat.
The workhorse is Q6 2sa720. the emitter will follow the base voltage MINUS the (about) 0.6v Vbe, so the C23 voltage of -19.8 applied to the
base of Q6 will give -19.1v at the emitter (pin 15)
-50.6v says Q6 is shorted.
replacement is usually a ksa1220, but you can TEMPORARILY get away with using ksa992 as it's dissipating 0.19w and the a992 can handle 0.5w max.
ksa1220 is being discontinued, so TTA-004B (757-TTA004BQ @ $0.48) is fine.
Re: power supply. my rant:
all new electrolytics, they have been cooking and working hard for years.
ditto for semiconductors, BOTH transistors AND diodes!! They have worked hard and heated up for years.
And all are made with older processes and materials - we have learned a lot in the intervening decades.
power supply #1
protection #2
power amps #3
why? when those three things are done the issues that remain are NOT destructive.. may SOUND horrible, but no blowups...
The first time you dither around, and in the meantime a bad trimmer or a 2sc1451 cause the amp to erupt in smoke and flames, You'll change your priorities.
AND there CAN be 2sc1451 IN THE SX-950!! Amp transistors Q7 & Q8 are listed at 2sc1628 but the BOM admits that 2sc1451's could be used.