I wanted to take a moment and highlight a Sansui G8700DB we had in recently for repair.
The 8700DB is very similar to the 9700. With two phono inputs, stepped attenuator, 160W/Ch output at .025%dB THD the G-8700DB was a real performer. Typical of the G series with the big dual front knobs and driver topographies. This unit came in with a totally fried driver section, defective right driver and in protection. I’ve seen very similar failures in almost all the G-series.
On this unit the resistors array for the bias and 3rd driver had completely burnt up, carbonizing some sections of the PCB and marring the heatsink. Resistors R26-35+ had become either totally open or fused and caught fire. One of the major issues with fusible resistors. 2nd order BJT’s in the left along with several series R’s were open. I noted at least 25% of the intact fusible resistors were 10-15% off their rated tolerance wheres as carbon films noted along 3-5% tolerances.
Judging from mis-matched out-put devices, BJT’s and capacitors a repair was attempted at least some point in this units life. Along with the noted defects above the right channel output devices TO-3 were shorted between emitter and bases along with open emitter resistors.
On both F-3074 drivers all existing fusible resistors were replaced with 1/2W and 2W MOX(1W<) and carbon film KOA resistors. The left bank 3115/14 assemblies had their .33 5W emitter resistors replaced and all TO-3 (2SD665/2SB645) replaced with 5% matched MJ15022/23 devices.
All the electrolytic were replaced with audio grade KA and low impedance PW Nichicons’ C17/18 Ceramics .022 replaced with high grade WIMA polypropylenes, along with the C11 .047 mylar with a high grade WIMA PP film. Original bias/offset potentiometers replaced with a precision bourns potentiometer.
New first order differential pair transistors, bias and 2/3order transistors replaced. BJT counter parts with 2SA939 /2071 MJE350/340 heavy duty devices and driver transistors to MJ15033/32, TR13 1904 with a tabbed KSC3503. New heatsink applied to the missing devices on the left channel.
I’ve noted several issues with the 939 transistors, along with damage above the F3070/3072 power supply/protect assemblies. TR02 (939) was replaced again with a MJE350 OnSemi BJT device and its associated D02 (4148)D602 (Ultra-Fast 4005) and R06 due to excessive current draw (Over 65+ initial Offset)
Final output results are well within expected ranges. 170.20 W/Ch @ 8OHM w/ 0.023%dB
100Hz L/R 28.53Vrms @.0076%dB THD+N
1000Hz L/R 28.53Vrms @ .0075%dB THD+N
10000Hz L/R 28.53Vrms @ .011%dB THD+N
Total Output (continous @ 8OHM) 160WCh @ 8OHM w/ 0.025%dB THD
1000Hz L/R 36.90.30Vrms 170.20 W/Ch @ 8OHM w/ 0.023%dB THD+N Before Clipping plateau.
1000Hz L/R 38.30Vrms 183.361W/Ch @ 8OHM w/ 0.030%dB THD+N Before Clipping plateau.