Sansui AU-D9 – Power transistors not amplifying after restoration

[GESMAN]

Hi everyone,


I'm working on a Sansui AU-D9 that had several issues, most of which I've resolved:

• Replaced leaky capacitors
• Removed the infamous "black flags"
• Replaced the burnt output transistors
• Replaced the drivers
• Also replaced the distortion correction stage transistors

Here's what I’ve already checked:

• Verified all flame resistance
• Checked voltages across preamp, differential stage, and power stage
• Found inconsistent voltages on two transistors (I can provide details)
• The protection relay engages normally
• All solder joints and PCB traces appear fine

The current issue is that the output transistors do not amplify.


Using an oscilloscope, I can confirm that the signal reaches the base of the output transistors, but it does not get amplified. At the output, I can only see the signal from the distortion correction stage, as if the power transistors are not conducting at all.


Bias is set according to the service manual, and the power supply voltages are correct.


Has anyone experienced a similar issue with this model? Or any suggestions on what to check next?
I'm happy to share measurements, circuit photos, or relevant parts of the schematic if that helps.


Thanks in advance for any help!

 

[davidsrsb]

Are your transistors genuine?

 

[GESMAN]

The final transistors are not the original 1169/2273 but I replaced them with a pair of original 2SA1295/2SC3264 Sanken,
the drivers are original Toshiba, the distortion correctors are from ISC Semiconductors, all the transistors are purchased from a reliable shop; I have not replaced any other transistors so far

 

[GESMAN]

I have a copy of original transistors (1169/2773) and I tried with those too but it didn't work

 

[BSST]

Hi GESMAN,

Welcome to the forum!

Is my attachment your amp?

Why does your first post feature the jpg? Is there a problem in that area? Do the voltages in red correspond to your unit? I hope your amp presents with the voltages in black.

Your first post implies the bias currents are nominal for both the main and correction outputs? True for both channels?

Can you post scope photos for both correction and main amp outputs, preferably in a dual-trace photo?

Thanks!

 

[rayma]

Please post the entire schematic, with all DC voltages and scope photos of the output.
Have you tested the output transistor emitter resistors for value?

 

[GESMAN]

Hi BSST,
thank you for your interest in my post, I ask you and the others to have a lot of patience with me, I am not yet equipped to perform double-track measurements, my laboratory corner is very limited so I will do the tests you ask me one at a time and for this advice I thank you;

I think that the final transistors do not amplify, why do I make this statement?
I connected my 8Ohm dummy load to the output terminals and I gave volume until it started to clip, but this did not happen because I went all the way down with the volume knob;
I measured the output voltage which is around 12Vac, so I deduce that only the distortion correctors work (which are 25W transistors);

the jpg I attached refers to the measurements taken on the entire power board; the measurements in red are those that do not correspond, especially those inside the red circle;
the kDZ2 is a 33V zener diode, to have -49.6V that zener should be 11V, or am I doing something wrong?

 

[GESMAN]

Please post the entire schematic, with all DC voltages and scope photos of the output.
Have you tested the output transistor emitter resistors for value?

yes I measured the emitter resistors and they are good

I forgot to say, I don't have the 0.4V voltage at the emitter, and at the base of the finals I have 580mV rather than 600mV, all this by setting bias to 10mV and DC 0V

 

[rayma]

Most output stages do not have voltage gain because they are emitter followers.

 

[BSST]

About 11V for the Zener seems about right. I'm trying to pick my way through the parts list, but can't find sensical results. Has that part been changed in earlier repair? Same situation in both channels?

Would you report all base and emitter voltages for the output transistors, as rayma suggested? (i.e. Q13 through Q20). It's hard to imagine how bias currents could be in accordance with SM, yet not pass sine wave signals.

 

[GESMAN]

Most output stages do not have voltage gain because they are emitter followers.

So not having that output voltage shouldn't be a problem?

 

[GESMAN]

About 11V for the Zener seems about right. I'm trying to pick my way through the parts list, but can't find sensical results. Has that part been changed in earlier repair? Same situation in both channels?

I didn't change it

if you look at page 7 of the DM there are all the transistors and diodes used in the amplifier; as zener they are indicated 13-15-27V but not 33V as in reality; if you look at page 3 there is the list of materials used on the F3258 board and there the 33V zener is reported.

 

[GESMAN]

Measurements taken with the amp on for 15/20 minutes
RIGHT Channel (11mV bias)

Q13_Vb = 1,689V - Ve_= 1,103V
Q14_Vb= -1,675V - Ve_= -1,091V
Q15_Vb= 1,103V - Ve_= 604mV
Q16_Vb= -1,094V - Ve_= -574mV
Q17_Vb= 604mV - Ve_= 11mV
Q18_Vb= -580mV - Ve_= -1mV
Q19_Vb= 625mV - Ve_= 147mV
Q20_Vb= -651V - Ve_= -128mV

LEFT Channel (11mV bias)

Q13_Vb = 1,690V - Ve_= 1,104V
Q14_Vb= -1,652V - Ve_= -1,077V
Q15_Vb= 1,102V - Ve_= 584mV
Q16_Vb= -1,077V - Ve_= -583mV
Q17_Vb= 585mV - Ve_= -1mV
Q18_Vb= -580mV - Ve_= 12mV
Q19_Vb= 640mV - Ve_= 162mV
Q20_Vb= -626V - Ve_= -108mV

 

[GESMAN]

@rayma said

 

[BSST]

I sketched out the right channel output section and noted your noted measured voltages. (Left channel looked very similar.) All DC voltages looked reasonable, showing roughly 0.6V drop across each b-e junction.

I can't fathom how DC voltages can drop in reasonable fashion through the cascade of transistors (eg. Q13,Q15,Q17), and fail to simultaneously transfer the sine wave drive through each b-e junction.

I suggest using your scope to step through each stage, with scope set for DC coupling. Drive amp input with sine wave to deliver about 0.2V pk-pk at base of Q13. Your scope should show dropping DC voltage at each junction (base, then emitter) as you step in order from Q13, Q15, Q17 followed by emitter, then base of Q18, Q16, Q14. The sine wave should follow supper-imposed on the DC bias, with only minor drops in AC amplitude.

If you find anything at odds with this description, please describe. Please attach photos if helpful.

Good luck!

BTW, the SM that @rayma has contributed seems to show a 33V Zener, so perhaps that mystery is resolved.

 

[wg_ski]

If you’ve got the right vbe’s and bias in the output stage, but not getting a signal at the speaker out it may just be a speaker relay that’s failing to close.

 

[rayma]

Yes, scope the output before the speaker switches and selector.

 

[GESMAN]

If you’ve got the right vbe’s and bias in the output stage, but not getting a signal at the speaker out it may just be a speaker relay that’s failing to close.

the signal arrives at the output terminals

 

[BSST]

From your original post:

Using an oscilloscope, I can confirm that the signal reaches the base of the output transistors, but it does not get amplified. At the output, I can only see the signal from the distortion correction stage, as if the power transistors are not conducting at all.

It seems you're probing both the correction stage and the main power section. If you're seeing signal at the correction stage, it has to originate from Q13 and Q14 emitters. But bias seems drive Q17 and Q18, but no audio showing there? What are we overlooking?

Do you find DC (and AC) across R43 and R44? Try to probe directly on the resistor leads so that you exercise the path to each emitter connection. And confirm continuity from each emitter to junction of R43, R44.

 

[BSST]

Riffing on this theme, measuring from the emitter of Q17 at the package body to board edge terminal 13 should show a few tenths ohm, i.e. R43 + winding resistance of transformer L1; similar for emitter of Q18.