Please help! Pioneer QX 949 Offset set to zero but unable to set bias appropriately

[petehall347]

yes .. as a rule there should be 600mv between base emitter .
with 1.1v at base there should be 500mv at emitters .. not enough to turn outputs on ..
this is why i query the 150 ohm emitter resistors for the drivers .

 

[backmd]

Output transistor voltages

righto according to the drivers base voltages +- 1.1v bias should be low.as emitters should be at 500mv +-
check r42 and r40 . 150 ohm 1/2 w Checked the driver emitter resistors in series and they measure 295 ohms respectively. Spec is 300 ohms
and voltages of drivers q10 q12 ... collectors and emitters ..
We have a real clear schematic now.
Spec Measured
Base of 2SB528 Q12 -1.2 -1.1
Base of 2SD358 Q10 1.2 1.1
R38 -28.0 -27.5
Base of Q6 -41.4 -42.1
Base of Q8 1.8 1.56
Junction between
R34 and R36 19 19.6
R18 .07 0.0
R14 -.12 .56
Q12 emitter -0.67
Q10 emitter 0.64
Q12 collector -34.3
Q10 collector 34.3
So the drivers have around 0.4 volts forward bias?

Pin 27 base of Q16 0.66
Pin 28 emitter of Q16 -.037
Pin 30 base of Q14 0.64
Pin 31 emitter of Q14 .051
So the power outputs have a 0.60 to 0.69 volt forward bias?

 

[petehall347]

660−37=623 ...623mv
640−51=589 .....589mv ..

for some reason the drivers aren't dropping enough voltage base to emitter ..
unless i am missing something i suggest change the drivers ..
only after confirming those 150 ohm resistors .

 

[backmd]

So is the output starting to turn on too much

660−37=623 ...623mv
640−51=589 .....589mv ..

for some reason the drivers aren't dropping enough voltage base to emitter ..
unless i am missing something i suggest change the drivers ..
only after confirming those 150 ohm resistors .

Let me see if I understand. With 623mv of bias is the PNP transistor turning on too much and causing too much current flow ie 78mv versus 20 mv across the emitter resistors? The 2SD357 may have read good on one range and questionable on another by pin 26. So I will replace both drivers in the questionable channel.

 

[petehall347]

to be sure measure bases of outputs on another channel that is set to the correct 20mv .
it however doesn't excuse the low voltage drops base emitter on the drivers ..
i am pretty sure changing the drivers will sort it ...
if not a regroup is called for ...

 

[backmd]

Ingenious!

Thanks to you and ecluser, I think I understand that the eb voltage on the drivers must be 600mv. And if it is 400mv or so, that could be due to aging on the transistors, thereby developing leaky junctions. I have purchased the NTE equivalents for the drivers and will attempt to solder them in without completely removing the circuit board. As you well know removing the board is quite tedious.

 

[petehall347]

i wouldn't be using nte ...
there must be something better than those .. maybe there is a current parts list floating around ..

 

[backmd]

150 ohm resistors

660−37=623 ...623mv
640−51=589 .....589mv ..

for some reason the drivers aren't dropping enough voltage base to emitter ..
unless i am missing something i suggest change the drivers ..
only after confirming those 150 ohm resistors .

One measures 147 and the other measures 148 ohms

 

[petehall347]

close enough ..

 

[backmd]

Courage and variac

close enough ..

Am watching the current as I slowly increase voltage on the variac, am also monitoring the bias and offset simultaneously. Went to about 60 volts and then had to read the schematic again to verify there is basically no load across the outputs. The loads are in the k ohm range. It is taking all my courage to first adjust the bias and then the offset. I am feeling timid, since the current is in the 0.4 amp range on the variac at 60 volts. Bummer; 38mv offset and 110mv bias
Let's repeat the Sencore TF26 Cricket transistor tester analysis since obviously I am missing something
And now the outer board:
36-38 PNP good
33-35 NPN good
30-32 NPN good
27-29 PNP NO GOOD
2SB727 PNP good by pin 38
2SD357 NPN good on one range, questionable on another by pin 36
2SD358 NPN good by pin 29
2SB528 PNP good by pin 27
2SC1451 NPN good by pin 30
2SC1451 NPN good by pin 35
Apparently the PNP power output just slowly died! I wonder why?

 

[backmd]

Time to compare the "good" bias channel to the bad bias channel

Am watching the current as I slowly increase voltage on the variac, am also monitoring the bias and offset simultaneously. Went to about 60 volts and then had to read the schematic again to verify there is basically no load across the outputs. The loads are in the k ohm range. It is taking all my courage to first adjust the bias and then the offset. I am feeling timid, since the current is in the 0.4 amp range on the variac at 60 volts. Bummer; 38mv offset and 110mv bias
Let's repeat the Sencore TF26 Cricket transistor tester analysis since obviously I am missing something
And now the outer board:
36-38 PNP good
33-35 NPN good
30-32 NPN good
27-29 PNP NO GOOD
2SB727 PNP good by pin 38
2SD357 NPN good on one range, questionable on another by pin 36
2SD358 NPN good by pin 29
2SB528 PNP good by pin 27
2SC1451 NPN good by pin 30
2SC1451 NPN good by pin 35
Apparently the PNP power output just slowly died! I wonder why?

Replaced the PNP with NTE285, but still have excessive bias, albeit the offset came down, then retested the PNP output which tested good on gain? See above
Spec Measured Good channel
Base of 2SB528 Q12 -1.2 was -1.1 now -1.0
Base of 2SD358 Q10 1.2 was 1.1 now 1.2
R38 -28.0 was -27.5 now -27.5
Base of Q6 -41.4 was -42.1 now -42.2
Base of Q8 1.8 was1.56 now 1.46
Junction between
R34 and R32 19 was19.6 now 19.7
R18 .07 was 0.0 now .058
R14 -.12 was.56 now -.014
Q12 emitter was -0.67 now –0.65
Q10 emitter was 0.64 now 0.70
Q12 collector was -34.3 now –35.0
Q10 collector was 34.3 now 35.0
Pin 27 base of Q16 was -0.66 now- 0.64
Pin 28 emitter of Q16 was -.037 now-.05
Pin 30 base of Q14 was 0.64 now 0.69
Pin 31 emitter of Q14 was .051 now .09

Is C18 good? Yes 118uf!

 

[petehall347]

did you compare voltages of the drivers in a behaving channel yet ?

 

[backmd]

Driver voltage comparison between good biased and too high biased channels

Spec Measured Good channel
Base of 2SB528 Q12-1.2 was -1.1 now -1.0 -1.16
Base of 2SD358 Q101.2 was 1.1 now 1.2 1.20
R38 -28.0 was -27.5 now -27.5
Base of Q6 -41.4 was -42.1 now -42.2
Base of Q8 1.8 was1.56 now 1.46
Junction between
R34 and R32 19 was19.6 now 19.7
R18 .07 was 0.0 now .058
R14 -.12 was.56 now -.014
Q12 emitter was -0.67 now –0.65 -0.55
Q10 emitter was 0.64 now 0.70 0.58
Q12 collector was -34.3 now –35.0
Q10 collector was 34.3 now 35.0
Pin 27 base of Q16 was -0.66 now- 0.64
Pin 28 emitter of Q16 was -.037 now-.05
Pin 30 base of Q14 was 0.64 now 0.69
Pin 31 emitter of Q14 was .051 now .09
8mv offset and 132mv bias
Is C18 good? Yes 118uf!
No big differences. Is the NPN power output transistor leaky?

 

[backmd]

So the power outputs have 0.59 to 0.60 forward bias

Point tested Specification Measured voltage before and after PNP output replacement Good channel voltages
Base of 2SB528 Q12-1.2 was -1.1 now -1.0 -1.16
Base of 2SD358 Q10 1.2 was 1.1 now 1.2 1.20
R38 -28.0 was -27.5 now -27.5
Base of Q6 -41.4 was -42.1 now -42.2
Base of Q8 1.8 was1.56 now 1.46
Junction between
R34 and R32 19 was19.6 now 19.7
R18 .07 was 0.0 now .058
R14 -.12 was.56 now -.014
Q12 emitter was -0.67 now –0.65 -0.55
Q10 emitter was 0.64 now 0.70 0.58
Q12 collector was -34.3 now –35.0
Q10 collector was 34.3 now 35.0
Pin 27 base of Q16 was -0.66 now- 0.64
Pin 28 emitter of Q16 was -.037 now-.05
Pin 30 base of Q14 was 0.64 now 0.69
Pin 31 emitter of Q14 was .051 now .09
8mv offset and 132mv bias
Is C18 good? Yes 118uf!
No big differences. Is the NPN power output transistor leaky?[/QUOTE]

So if the power outputs have 0.59 and 0.60 volts forward bias for Q16 and Q14 respectively, they likely are properly biased? So if the emitter current exceeds 20ma then the power transistor is leaky?

 

[backmd]

The working channel is compared to the high bias channel below

did you compare voltages of the drivers in a behaving channel yet ?

Yes I did compare the voltages in the above 31, 33 and 34 passages. However it is difficult to arrange the data in columns that are clear.

Protection board is pretty messed up but not fubar.

 

[backmd]

On the home stretch!

After replacing a transistor on the protection board and inspecting all those sky blue capacitors, and tightening the 7.5VAC 3amp fuse, we have 4 channel output and the relay clicks appropriately. After this diligent work we have the classical WFMT station at 4.5 on the signal meter and in stereo! Many many thanks to PeteHall347 for all the insight and wonderful help to achieve basically a working receiver.

Just thought of something Base of Q8 is getting progressively lower, now is 19 percent lower than spec, possible causes C14 becoming capsistor, just measured and it is 3.6uf and spec is 3.3uf, within specification. Also check Q6, 2sc869. Addendum, it is fine on the Sencore TF26 based upon gain readings only. I have not got a clue.

 

[backmd]

Why would a power transistor with 0.6 volts not bias properly?

Why would a power transistor with 0.6 volt emitter base voltage not bias properly and have too much emitter current?

 

[backmd]

All fixed! VEry Very Happy!

Why would a power transistor with 0.6 volt emitter base voltage not bias properly and have too much emitter current?

Yes, C14 measures 0.0uf and should be 3.3uf. 100uf cap at 50v soldered in and connections verified! Better channel balance but still not equal! C12 is supposed to be 330uf at 10volts and is zero!
Replaced cap with 470uf at 16 volts and volume is now equal! Burning 40.8 watts!
Bias set at 20mv and offset is at zero mv. Many Many Thanks to PeteHall347

 

[petehall347]

Yes, C14 measures 0.0uf and should be 3.3uf. 100uf cap at 50v soldered in and connections verified! Better channel balance but still not equal! C12 is supposed to be 330uf at 10volts and is zero!
Replaced cap with 470uf at 16 volts and volume is now equal! Burning 40.8 watts!
Bias set at 20mv and offset is at zero mv. Many Many Thanks to PeteHall347

yes those caps would screw it up as you found out ...
congrats to you . you did all the work .

 

[backmd]

Yes, but I still cannot figure out the theory.

The only thing I can think of is that when C12 has zero uf, it throws off the base of Q8 which we measured as out of spec. When the base of Q8 is off, then this affects the bias of the drivers, which affect the output transistors.
This base of Q8 is also key for AC or signal amplication and the L front channel was about 20db decreased compared to the other 3 channels. Only after using AC or signal analysis could we fix the problem.

Still perplexed and confused.