Okay, I'll explain. The voltages you measured show 12 mA through xtr 8, but requiring a base current of 2 mA of xtr 8. This is typical of xtr operating in reverse mode. Xtr is not destroyed because xtr 4 limits C_E voltage of xtr 8 below E_B (E_C) reverse breakdown voltages.
Xtr 8 is not capable of pulling up voltage enough because reverse hfe is low.
Maybe other channel xtr 8 is just good enough to make it work. Maybe other channel xtr 8 is of different type.
Can you confirm that the actual transistor used for xtr 8 has C and E pinout matching that of pcb.
Xtr 8 is not capable of pulling up voltage enough because reverse hfe is low.
Maybe other channel xtr 8 is just good enough to make it work. Maybe other channel xtr 8 is of different type.
Can you confirm that the actual transistor used for xtr 8 has C and E pinout matching that of pcb.
Xtr8 is passing 12.7mA. It is in saturation. Xtr3 is also passing the 12.7mA, but needing 12.9v from E to C because of limited base current or low hfe.
Xtr16 is sinking 14.0mA.
So where is the additional 1.3mA coming from?
Maybe disconnecting Xtr4, Xtr5, Xtr13 and Xtr14 will reveal if they are the source of the extra current.
An alternate thought - change the 470 ohm resistor attached to base of Xtr8 to 1K. Thus providing Xtr3 with additional base current.
Or increase supply voltage beyond 10 volts.
Xtr16 is sinking 14.0mA.
So where is the additional 1.3mA coming from?
Maybe disconnecting Xtr4, Xtr5, Xtr13 and Xtr14 will reveal if they are the source of the extra current.
An alternate thought - change the 470 ohm resistor attached to base of Xtr8 to 1K. Thus providing Xtr3 with additional base current.
Or increase supply voltage beyond 10 volts.
Last edited:
2N4250 Transistor | Alltronics
Alltronics appears to have a very few 2N4250 transistors left in stock.
A search seems to show that a 2N4248 is a substitute for the 2N4250 and that is more commonly available.
---Gary
My mistake. Xtr8 (2N4250?) is not defective or weak. In fact, it is operating in heavy saturation, where Ib is almost same current as Ic. No need to replace this device. It is doing what it is told.
The cascode device Xtr3 is the one I would question. But if there is an additional current sneaking into the path, then I would check Xtr4, Xtr5, Xtr13 and Xtr14.
Maybe measuring voltages of working channel will reveal where differences are at.
The cascode device Xtr3 is the one I would question. But if there is an additional current sneaking into the path, then I would check Xtr4, Xtr5, Xtr13 and Xtr14.
Maybe measuring voltages of working channel will reveal where differences are at.
Here are readings at 30.7+/- volts. I think I may need to hook the outputs back up to the broken side before it is of real help.
Volts30 with working
I put an updated copy of the voltage chart in post one.
http://www.diyaudio.com/forums/pass-labs/271817-threshold-4000-repair-need-some-help.html#post4267129
http://www.diyaudio.com/forums/pass-labs/271817-threshold-4000-repair-need-some-help.html#post4267129
Voltage on base of Q22 (and Q16) should be 2 diode drops higher than voltage at collector of Q21, i.e.
2 x 0.7v above -28.9v = -27.5v
Your measurements show -24.3 or 4.6v across the two diodes.
Conclusion: One (or both) of the two 1N914 diodes connected between base of Q22 and collector of Q21 is open.
Suppose to have ~2 ma thru Q22 and ~6 mA thru Q16. Measurements show ~15 mA thru Q22 and ~42 mA thru Q16.
2 x 0.7v above -28.9v = -27.5v
Your measurements show -24.3 or 4.6v across the two diodes.
Conclusion: One (or both) of the two 1N914 diodes connected between base of Q22 and collector of Q21 is open.
Suppose to have ~2 ma thru Q22 and ~6 mA thru Q16. Measurements show ~15 mA thru Q22 and ~42 mA thru Q16.