I did replace with the same tranasitor on one channel only with 2SC3419 - Y . The other one on the other channel was fine so i left it there. Same result though.. No bias reading.
I was expecting to hear and see crossover distortion but i don't so i'm out of ideas. the bias trimmers are set to maximum (100 ohms) at this time so that i can have minimum bias.
The only thing i can do is measure b-e voltages on all stages up to the the predriver and driver transistors to see if there is any difference in b-e voltage when i adjust the trimmer.
As a test i measured b-e voltage at 2SC3419 and with bias trimmer at 100ohm (max) it was at 600mv. When i started adjusting the trimmer that went up.
So i plan to measure b-e voltages on all the other pairs and see how the circuit behaves while i make (small) adjustments to the bias trimmer. I don't know what else to do to be honest. And since it has fried the outputs multiple times i need to make sure the bias circuitry works.
All components test fine right now and i don't see any clipping, assymetry, oscillation or zero crossing distortion and in theory it works fine without producing a ton of heat.
So yeah.. i don't know what to do...
Measuring Vbe tells you little, as far as output stage bias goes. It’s the bias voltage spread that matters. If you are making measurements with retarded supply because of using a dim bulb for the AC you will not be measuring correct voltages and potentially never be able to create enough output stage bias current.
The easiest way for us to help you is for you to mark up the schematic with the voltages that you measure comparing against what they show, all of them!!
The easiest way for us to help you is for you to mark up the schematic with the voltages that you measure comparing against what they show, all of them!!
Last edited:
Have you taken this into account? The manual states that the bias should be set starting from cold condition to between 0.3 and 0.4mV.
It should slowly rise to +/- 18mV after about half an hour.
So, if you set the bias at +/- 18mV when cold it will rise to levels far too high for the amp to survive.
Edit: This would also explain why you read close to zero bias when cold and think the bias cannot be set. Keep monitoring the bias points
and look if the voltage slowly rises in time.
It should slowly rise to +/- 18mV after about half an hour.
So, if you set the bias at +/- 18mV when cold it will rise to levels far too high for the amp to survive.
Edit: This would also explain why you read close to zero bias when cold and think the bias cannot be set. Keep monitoring the bias points
and look if the voltage slowly rises in time.
I agree, not something you want to do and not pay attention.
Based on the method as described in the manual, probably not doable if using a limited AC supply, and probably what is being experienced.
Can’t say I remember such a bias setting method inferring this type of thermal bias stability.
Based on the method as described in the manual, probably not doable if using a limited AC supply, and probably what is being experienced.
Can’t say I remember such a bias setting method inferring this type of thermal bias stability.
I did just that here is the schematic with the voltages measured from ground, without a dimbulb, Bias set to a minimum and the amp was cold (attached pdf of the image below).
You will see that with the bias trimmer all the way down the negative side is very low.
On the previous repair, when i tried to adjust bias based on the readings from the emitters to ground on each stage to reach the voltages the schematics states it fried the outputs in 5-10 minutes of playback at a low volume. i did that because i could not measure bias. So now i have left the bias to a minimum until i figure out what else is wrong here.
I cannot set the bias at all it is always zero
After 30 minutes (while on the dimbulb) it should show something but it does not. No matter where i turn thr trimmer. The trimmer works and on the driver and pre driver stages the voltages do change when i turn the bias trimmer. Also another note is that before the channel was fried i could measure the 3-4mv on those resistors when absolutelty cold. Now i cannot. Reading always zero even after 30 mins.
I agree, that confused me too.
measurements look okay.
Looks like you have some DC offset, you could adjust R3264 to set it as close to 0V as you can get (with respect to ground), (what does it measure now?)
measure at T017 (with respect to ground) which is the o/p and feedback point. then the numbers ( across C2268 with respect to ground) will be more symetrical.
Just going to have to go for it, turn the bias pot up (R3300) a bit at a time while monitoring the OPS bias, DMM across TO13,TO14 or TO15,TO16 to see how much OPS bias you get, go slow. use lead clips, not probes that will slip off and short out stuff. Attach/detach lead clips when power is off only
Looks like you have some DC offset, you could adjust R3264 to set it as close to 0V as you can get (with respect to ground), (what does it measure now?)
measure at T017 (with respect to ground) which is the o/p and feedback point. then the numbers ( across C2268 with respect to ground) will be more symetrical.
Just going to have to go for it, turn the bias pot up (R3300) a bit at a time while monitoring the OPS bias, DMM across TO13,TO14 or TO15,TO16 to see how much OPS bias you get, go slow. use lead clips, not probes that will slip off and short out stuff. Attach/detach lead clips when power is off only
To follow up, what are the voltages at base and emitter of Q 7266 with respect to C 2265? If there's still ~ 0v b-e and voltage present across R3294, then there's open along the R3294 path. Hence, no bias current possible in output transistors!
One other failure scenario: Base-emitter junction is shorted, collector open. Still no available drive to bias spreader.
One other failure scenario: Base-emitter junction is shorted, collector open. Still no available drive to bias spreader.
Last edited:
The voltage gain of the amp without feedback is so massively high that once the loop is closed (feedback in place) the crossover distortion is reduced to very low levels. If you know what to listen for I think you would hear it at zero bias though. You need a pure very low level sine and crossover distortion will show as a slight roughness in the tone.
Where are you up to now with all this? If you have no bias and it is not adjusting then the first quick measurement to take is across C2267 and C2268 which are across the bias generator derived voltage.
On the faulty channel what voltage do you see with the preset on both minimum and maximum? So two readings.
There is! Maybe i did not mark it down properly on the schematic i attached. Emiter voltage is approximately 1.4v .
Base and collector voltages are approximately 54v so it is close enough to what the schematic says.
Right now dc offset is at about 20mv on one channel and 30mv on the other. The one im showing on the schematic is the 30mv one and i cannot get it to go any lower.
As far as measurements are concerned i find it very puzzling that the negative side is so much lower than the positive. Maybe because i haven't set the bias yet because when i do set it, i see changes in voltages there.
But no matter how much i turn the bias trimmer, even all the way, i get no reading on the output resistors on either channel.
The amp plays fine and the oscilloscope doesn't show any crossover distortion or oscillation or clipping.
With the readings im measuring i thought i would see something but I don't.
And i dont understand how this could be happening.
I took measurements on the one channel C2268 (previously burning up all the time). Across this capacitor the voltage is:
2.65volts || bias trimmer set to maximum value (100ohms lowest bias)
3.25volts || bias trimmer set to minimum value (0 ohms highest bias)
Since the measurements are similar and i have zero bias on both channels with no way to adjust them, i did not check the other channel on capacitor C2267 (the one i repaired today and it's the first time it blew. Only the output transistors where shorted on that one, and i replaced the 4x 10 ohm resistors at their bases and the emitter resistors 2x 0.18ohms as a precaution). But i can measure that one too if you want me to.
Last edited:
Alex said:
There is maybe i did not mark it down properly on the schematic i attached. Emiter viltage is approximately 1.4v . Base and collector voltages are approximately 54v so it is close enough. """
I don't think I conveyed my point: There must be voltage drop across R3294 if there is any available supply power to the bias spreader and the output transistors. The supply current must flow through R3294. This issue may be why you can't produce any bias current.
Alex said:
There is maybe i did not mark it down properly on the schematic i attached. Emiter viltage is approximately 1.4v . Base and collector voltages are approximately 54v so it is close enough. """
I don't think I conveyed my point: There must be voltage drop across R3294 if there is any available supply power to the bias spreader and the output transistors. The supply current must flow through R3294. This issue may be why you can't produce any bias current.
Maybe.
If Q7266 is functional, there should be about 0.6V base-emitter voltage.
If there isn't, the base junction might be shorted, and the collector open. No current to bias spreader.
Or, R3294 or its conducting traces might be open or too high resistance. Base-emitter voltage is then low because there's no current through R3294. No current to bias spreader.
Since you confirm 150 ohm resistance, open traces are a possibility. What is b-e voltage? If it's too low, why?
This issue has to be sorted IMHO.
If Q7266 is functional, there should be about 0.6V base-emitter voltage.
If there isn't, the base junction might be shorted, and the collector open. No current to bias spreader.
Or, R3294 or its conducting traces might be open or too high resistance. Base-emitter voltage is then low because there's no current through R3294. No current to bias spreader.
Since you confirm 150 ohm resistance, open traces are a possibility. What is b-e voltage? If it's too low, why?
This issue has to be sorted IMHO.
When the Q7266 issue is solved and the above is correct, the problem has to be found in the output transistor circuit or perhaps beyond.
What are the emitter readings on 7222, 7274, 7276 and 7278 with respect to ground and the bias pot all the way up and all the way down?
Again two reading for each transistor.
Hugo
I took a closer, more careful look at your post. If you observe 1.6V directly across the resistor leads and you confirm 150 ohm, that does indeed suggest a bit more than 10mA flowing. Question, where is it flowing? If through collector and into bias spreader, what voltage do you observe across R3302 in the spreader?
Just saw Hugo's post. Points well taken.